From 03efed8a2a1b8e00164eb4720a82a7dd5e368a8e Mon Sep 17 00:00:00 2001 From: Tiger Yang Date: Sat, 28 May 2011 00:34:19 +0800 Subject: [PATCH 0001/3089] ocfs2: Bugfix for hard readonly mount ocfs2 cannot currently mount a device that is readonly at the media ("hard readonly"). Fix the broken places. see detail: http://oss.oracle.com/bugzilla/show_bug.cgi?id=1322 [ Description edited -- Joel ] Signed-off-by: Tiger Yang Reviewed-by: Sunil Mushran Signed-off-by: Joel Becker --- fs/ocfs2/dlmglue.c | 21 +++++++++++++++------ fs/ocfs2/super.c | 3 ++- 2 files changed, 17 insertions(+), 7 deletions(-) diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c index 7642d7ca73e52..da103f555a621 100644 --- a/fs/ocfs2/dlmglue.c +++ b/fs/ocfs2/dlmglue.c @@ -1692,7 +1692,7 @@ int ocfs2_open_lock(struct inode *inode) mlog(0, "inode %llu take PRMODE open lock\n", (unsigned long long)OCFS2_I(inode)->ip_blkno); - if (ocfs2_mount_local(osb)) + if (ocfs2_is_hard_readonly(osb) || ocfs2_mount_local(osb)) goto out; lockres = &OCFS2_I(inode)->ip_open_lockres; @@ -1718,6 +1718,12 @@ int ocfs2_try_open_lock(struct inode *inode, int write) (unsigned long long)OCFS2_I(inode)->ip_blkno, write ? "EXMODE" : "PRMODE"); + if (ocfs2_is_hard_readonly(osb)) { + if (write) + status = -EROFS; + goto out; + } + if (ocfs2_mount_local(osb)) goto out; @@ -2298,7 +2304,7 @@ int ocfs2_inode_lock_full_nested(struct inode *inode, if (ocfs2_is_hard_readonly(osb)) { if (ex) status = -EROFS; - goto bail; + goto getbh; } if (ocfs2_mount_local(osb)) @@ -2356,7 +2362,7 @@ local: mlog_errno(status); goto bail; } - +getbh: if (ret_bh) { status = ocfs2_assign_bh(inode, ret_bh, local_bh); if (status < 0) { @@ -2628,8 +2634,11 @@ int ocfs2_dentry_lock(struct dentry *dentry, int ex) BUG_ON(!dl); - if (ocfs2_is_hard_readonly(osb)) - return -EROFS; + if (ocfs2_is_hard_readonly(osb)) { + if (ex) + return -EROFS; + return 0; + } if (ocfs2_mount_local(osb)) return 0; @@ -2647,7 +2656,7 @@ void ocfs2_dentry_unlock(struct dentry *dentry, int ex) struct ocfs2_dentry_lock *dl = dentry->d_fsdata; struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb); - if (!ocfs2_mount_local(osb)) + if (!ocfs2_is_hard_readonly(osb) && !ocfs2_mount_local(osb)) ocfs2_cluster_unlock(osb, &dl->dl_lockres, level); } diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index cdbaf5e97308f..029c4cd8a6915 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1974,7 +1974,8 @@ static void ocfs2_dismount_volume(struct super_block *sb, int mnt_err) * If we failed before we got a uuid_str yet, we can't stop * heartbeat. Otherwise, do it. */ - if (!mnt_err && !ocfs2_mount_local(osb) && osb->uuid_str) + if (!mnt_err && !ocfs2_mount_local(osb) && osb->uuid_str && + !ocfs2_is_hard_readonly(osb)) hangup_needed = 1; if (osb->cconn) -- GitLab From 3d75be7c4771c7e4d5b5fa586a599af8473de32c Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Sun, 29 May 2011 22:56:31 +0300 Subject: [PATCH 0002/3089] ocfs2: checking the wrong variable in ocfs2_move_extent() "new_phys_cpos" is always a valid pointer here. ocfs2_probe_alloc_group() allocates "*new_phys_cpos". Signed-off-by: Dan Carpenter Signed-off-by: Joel Becker --- fs/ocfs2/move_extents.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c index cd9427023d2e7..d3433d60dbb94 100644 --- a/fs/ocfs2/move_extents.c +++ b/fs/ocfs2/move_extents.c @@ -746,7 +746,7 @@ static int ocfs2_move_extent(struct ocfs2_move_extents_context *context, */ ocfs2_probe_alloc_group(inode, gd_bh, &goal_bit, len, move_max_hop, new_phys_cpos); - if (!new_phys_cpos) { + if (!*new_phys_cpos) { ret = -ENOSPC; goto out_commit; } -- GitLab From 87f0d5c8db7aad85b9120c26723fdc63cd84a460 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Sun, 29 May 2011 22:57:16 +0300 Subject: [PATCH 0003/3089] ocfs2: null deref on allocation error The original code had a null derefence in the error handling. Signed-off-by: Dan Carpenter Signed-off-by: Joel Becker --- fs/ocfs2/ioctl.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c index bc91072b72196..d9a6ce7792669 100644 --- a/fs/ocfs2/ioctl.c +++ b/fs/ocfs2/ioctl.c @@ -381,7 +381,7 @@ int ocfs2_info_handle_freeinode(struct inode *inode, if (!oifi) { status = -ENOMEM; mlog_errno(status); - goto bail; + goto out_err; } if (o2info_from_user(*oifi, req)) @@ -431,7 +431,7 @@ bail: o2info_set_request_error(&oifi->ifi_req, req); kfree(oifi); - +out_err: return status; } @@ -666,7 +666,7 @@ int ocfs2_info_handle_freefrag(struct inode *inode, if (!oiff) { status = -ENOMEM; mlog_errno(status); - goto bail; + goto out_err; } if (o2info_from_user(*oiff, req)) @@ -716,7 +716,7 @@ bail: o2info_set_request_error(&oiff->iff_req, req); kfree(oiff); - +out_err: return status; } -- GitLab From 730e663bd82c1a10a85ff00728d34152a5a67ec8 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Mon, 30 May 2011 21:58:05 +0900 Subject: [PATCH 0004/3089] ocfs2: use proper little-endian bitops Using __test_and_{set,clear}_bit_le() with ignoring its return value can be replaced with __{set,clear}_bit_le(). Signed-off-by: Akinobu Mita Cc: Joel Becker Cc: Mark Fasheh Cc: ocfs2-devel@oss.oracle.com Signed-off-by: Joel Becker --- fs/ocfs2/ocfs2.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 409285854f647..bcde467acca39 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -836,13 +836,13 @@ static inline unsigned int ocfs2_clusters_to_megabytes(struct super_block *sb, static inline void _ocfs2_set_bit(unsigned int bit, unsigned long *bitmap) { - __test_and_set_bit_le(bit, bitmap); + __set_bit_le(bit, bitmap); } #define ocfs2_set_bit(bit, addr) _ocfs2_set_bit((bit), (unsigned long *)(addr)) static inline void _ocfs2_clear_bit(unsigned int bit, unsigned long *bitmap) { - __test_and_clear_bit_le(bit, bitmap); + __clear_bit_le(bit, bitmap); } #define ocfs2_clear_bit(bit, addr) _ocfs2_clear_bit((bit), (unsigned long *)(addr)) -- GitLab From bf61549a2d8e0326f5d6e4d1718883a7212d725f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 09:58:05 +0200 Subject: [PATCH 0005/3089] x86: Fix memblock_x86_check_reserved_size() use in efi_reserve_boot_services() The return value of memblock_x86_check_reserved_size() doesn't indicate whether there's an overlapping reservatoin or not. It indicates whether the caller needs to iterate further to discover all reserved portions of the specified area. efi_reserve_boot_esrvices() wants to check whether the boot services area overlaps with other reservations but incorrectly used membloc_x86_check_reserved_size(). Use memblock_is_region_reserved() instead. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310457490-3356-2-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: H. Peter Anvin --- arch/x86/platform/efi/efi.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index 899e393d8e732..a4c322ca1a5d5 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -325,8 +325,7 @@ void __init efi_reserve_boot_services(void) if ((start+size >= virt_to_phys(_text) && start <= virt_to_phys(_end)) || !e820_all_mapped(start, start+size, E820_RAM) || - memblock_x86_check_reserved_size(&start, &size, - 1<num_pages = 0; memblock_dbg(PFX "Could not reserve boot range " -- GitLab From 53348f27168534561c0c814843bbf181314374f4 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 09:58:06 +0200 Subject: [PATCH 0006/3089] bootmem: Fix __free_pages_bootmem() to use @order properly a226f6c899 (FRV: Clean up bootmem allocator's page freeing algorithm) separated out __free_pages_bootmem() from free_all_bootmem_core(). __free_pages_bootmem() takes @order argument but it assumes @order is either 0 or ilog2(BITS_PER_LONG). Note that all the current users match that assumption and this doesn't cause actual problems. Fix it by using 1 << order instead of BITS_PER_LONG. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310457490-3356-3-git-send-email-tj@kernel.org Cc: David Howells Signed-off-by: H. Peter Anvin --- mm/page_alloc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 9119faae6e6a9..b6da6ed818a8f 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -705,10 +705,10 @@ void __meminit __free_pages_bootmem(struct page *page, unsigned int order) int loop; prefetchw(page); - for (loop = 0; loop < BITS_PER_LONG; loop++) { + for (loop = 0; loop < (1 << order); loop++) { struct page *p = &page[loop]; - if (loop + 1 < BITS_PER_LONG) + if (loop + 1 < (1 << order)) prefetchw(p + 1); __ClearPageReserved(p); set_page_count(p, 0); -- GitLab From 15fb09722df32b7685be1cbcac198bb556ddaffe Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 09:58:07 +0200 Subject: [PATCH 0007/3089] memblock: Use MEMBLOCK_ALLOC_ACCESSIBLE instead of ANYWHERE in memblock_alloc_try_nid() After node affine allocation fails, memblock_alloc_try_nid() calls memblock_alloc_base() with @max_addr set to MEMBLOCK_ALLOC_ANYWHERE. This is inconsistent with memblock_alloc() and what the function's sole user - sparc/mm/init_64 - expects, although it doesn't make any difference as sparc64 doesn't have highmem and ACCESSIBLE equals ANYWHERE. This patch makes memblock_alloc_try_nid() use ACCESSIBLE instead of ANYWHERE. This isn't complete as node affine allocation doesn't consider memblock.current_limit. It will be handled with future changes. This patch doesn't introduce any behavior difference. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310457490-3356-4-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- mm/memblock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/memblock.c b/mm/memblock.c index a0562d1a6ad42..87e512dc8018b 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -612,7 +612,7 @@ phys_addr_t __init memblock_alloc_try_nid(phys_addr_t size, phys_addr_t align, i if (res) return res; - return memblock_alloc_base(size, align, MEMBLOCK_ALLOC_ANYWHERE); + return memblock_alloc_base(size, align, MEMBLOCK_ALLOC_ACCESSIBLE); } -- GitLab From 348968eb151e2569ad0ebe19b2f9c3c25b5c816a Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 09:58:08 +0200 Subject: [PATCH 0008/3089] memblock: Use round_up/down() instead of memblock_align_up/down() Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310457490-3356-5-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- mm/memblock.c | 21 +++++---------------- 1 file changed, 5 insertions(+), 16 deletions(-) diff --git a/mm/memblock.c b/mm/memblock.c index 87e512dc8018b..9882a88d4a104 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -41,17 +41,6 @@ static inline const char *memblock_type_name(struct memblock_type *type) /* * Address comparison utilities */ - -static phys_addr_t __init_memblock memblock_align_down(phys_addr_t addr, phys_addr_t size) -{ - return addr & ~(size - 1); -} - -static phys_addr_t __init_memblock memblock_align_up(phys_addr_t addr, phys_addr_t size) -{ - return (addr + (size - 1)) & ~(size - 1); -} - static unsigned long __init_memblock memblock_addrs_overlap(phys_addr_t base1, phys_addr_t size1, phys_addr_t base2, phys_addr_t size2) { @@ -87,7 +76,7 @@ static phys_addr_t __init_memblock memblock_find_region(phys_addr_t start, phys_ if (end < size) return MEMBLOCK_ERROR; - base = memblock_align_down((end - size), align); + base = round_down(end - size, align); /* Prevent allocations returning 0 as it's also used to * indicate an allocation failure @@ -102,7 +91,7 @@ static phys_addr_t __init_memblock memblock_find_region(phys_addr_t start, phys_ res_base = memblock.reserved.regions[j].base; if (res_base < size) break; - base = memblock_align_down(res_base - size, align); + base = round_down(res_base - size, align); } return MEMBLOCK_ERROR; @@ -486,7 +475,7 @@ phys_addr_t __init __memblock_alloc_base(phys_addr_t size, phys_addr_t align, ph /* We align the size to limit fragmentation. Without this, a lot of * small allocs quickly eat up the whole reserve array on sparc */ - size = memblock_align_up(size, align); + size = round_up(size, align); found = memblock_find_base(size, align, 0, max_addr); if (found != MEMBLOCK_ERROR && @@ -562,7 +551,7 @@ static phys_addr_t __init memblock_alloc_nid_region(struct memblock_region *mp, start = mp->base; end = start + mp->size; - start = memblock_align_up(start, align); + start = round_up(start, align); while (start < end) { phys_addr_t this_end; int this_nid; @@ -590,7 +579,7 @@ phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int n /* We align the size to limit fragmentation. Without this, a lot of * small allocs quickly eat up the whole reserve array on sparc */ - size = memblock_align_up(size, align); + size = round_up(size, align); /* We do a bottom-up search for a region with the right * nid since that's easier considering how memblock_nid_range() -- GitLab From 1f5026a7e21e409c2b9dd54f6dfb9446511fb7c5 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 09:58:09 +0200 Subject: [PATCH 0009/3089] memblock: Kill MEMBLOCK_ERROR 25818f0f28 (memblock: Make MEMBLOCK_ERROR be 0) thankfully made MEMBLOCK_ERROR 0 and there already are codes which expect error return to be 0. There's no point in keeping MEMBLOCK_ERROR around. End its misery. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310457490-3356-6-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- arch/x86/kernel/aperture_64.c | 2 +- arch/x86/kernel/check.c | 2 +- arch/x86/kernel/e820.c | 2 +- arch/x86/kernel/setup.c | 4 ++-- arch/x86/kernel/trampoline.c | 2 +- arch/x86/mm/init.c | 2 +- arch/x86/mm/memblock.c | 6 +++--- arch/x86/mm/numa.c | 6 +++--- arch/x86/mm/numa_32.c | 4 ++-- arch/x86/mm/numa_emulation.c | 2 +- include/linux/memblock.h | 4 +--- kernel/printk.c | 2 +- mm/memblock.c | 21 ++++++++++----------- mm/nobootmem.c | 2 +- mm/page_alloc.c | 4 ++-- 15 files changed, 31 insertions(+), 34 deletions(-) diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index 3d2661ca6542f..56363082bbdf3 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -88,7 +88,7 @@ static u32 __init allocate_aperture(void) */ addr = memblock_find_in_range(GART_MIN_ADDR, GART_MAX_ADDR, aper_size, aper_size); - if (addr == MEMBLOCK_ERROR || addr + aper_size > GART_MAX_ADDR) { + if (!addr || addr + aper_size > GART_MAX_ADDR) { printk(KERN_ERR "Cannot allocate aperture memory hole (%lx,%uK)\n", addr, aper_size>>10); diff --git a/arch/x86/kernel/check.c b/arch/x86/kernel/check.c index 452932d347307..95680fc4df5c1 100644 --- a/arch/x86/kernel/check.c +++ b/arch/x86/kernel/check.c @@ -86,7 +86,7 @@ void __init setup_bios_corruption_check(void) u64 size; addr = memblock_x86_find_in_range_size(addr, &size, PAGE_SIZE); - if (addr == MEMBLOCK_ERROR) + if (!addr) break; if (addr >= corruption_check_size) diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 3e2ef84253164..0f9ff58d06d7e 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -745,7 +745,7 @@ u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align) for (start = startt; ; start += size) { start = memblock_x86_find_in_range_size(start, &size, align); - if (start == MEMBLOCK_ERROR) + if (!start) return 0; if (size >= sizet) break; diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index afaf38447ef5f..31ffe20d5d278 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -331,7 +331,7 @@ static void __init relocate_initrd(void) ramdisk_here = memblock_find_in_range(0, end_of_lowmem, area_size, PAGE_SIZE); - if (ramdisk_here == MEMBLOCK_ERROR) + if (!ramdisk_here) panic("Cannot find place for new RAMDISK of size %lld\n", ramdisk_size); @@ -554,7 +554,7 @@ static void __init reserve_crashkernel(void) crash_base = memblock_find_in_range(alignment, CRASH_KERNEL_ADDR_MAX, crash_size, alignment); - if (crash_base == MEMBLOCK_ERROR) { + if (!crash_base) { pr_info("crashkernel reservation failed - No suitable area found.\n"); return; } diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c index a91ae7709b49e..a1f13ddb06e05 100644 --- a/arch/x86/kernel/trampoline.c +++ b/arch/x86/kernel/trampoline.c @@ -14,7 +14,7 @@ void __init setup_trampolines(void) /* Has to be in very low memory so we can execute real-mode AP code. */ mem = memblock_find_in_range(0, 1<<20, size, PAGE_SIZE); - if (mem == MEMBLOCK_ERROR) + if (!mem) panic("Cannot allocate trampoline\n"); x86_trampoline_base = __va(mem); diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index 30326443ab81d..13cf05a616057 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -68,7 +68,7 @@ static void __init find_early_table_space(unsigned long end, int use_pse, #endif base = memblock_find_in_range(start, good_end, tables, PAGE_SIZE); - if (base == MEMBLOCK_ERROR) + if (!base) panic("Cannot find space for the kernel page tables"); pgt_buf_start = base >> PAGE_SHIFT; diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 992da5ec5a64d..e126117d1b03b 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -66,7 +66,7 @@ u64 __init memblock_x86_find_in_range_size(u64 start, u64 *sizep, u64 align) return addr; } - return MEMBLOCK_ERROR; + return 0; } static __init struct range *find_range_array(int count) @@ -78,7 +78,7 @@ static __init struct range *find_range_array(int count) end = memblock.current_limit; mem = memblock_find_in_range(0, end, size, sizeof(struct range)); - if (mem == MEMBLOCK_ERROR) + if (!mem) panic("can not find more space for range array"); /* @@ -274,7 +274,7 @@ u64 __init memblock_x86_find_in_range_node(int nid, u64 start, u64 end, u64 size { u64 addr; addr = find_memory_core_early(nid, size, align, start, end); - if (addr != MEMBLOCK_ERROR) + if (addr) return addr; /* Fallback, should already have start end within node range */ diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c index fbeaaf416610d..fa1015de5cc0e 100644 --- a/arch/x86/mm/numa.c +++ b/arch/x86/mm/numa.c @@ -226,10 +226,10 @@ static void __init setup_node_data(int nid, u64 start, u64 end) } else { nd_pa = memblock_x86_find_in_range_node(nid, nd_low, nd_high, nd_size, SMP_CACHE_BYTES); - if (nd_pa == MEMBLOCK_ERROR) + if (!nd_pa) nd_pa = memblock_find_in_range(nd_low, nd_high, nd_size, SMP_CACHE_BYTES); - if (nd_pa == MEMBLOCK_ERROR) { + if (!nd_pa) { pr_err("Cannot find %zu bytes in node %d\n", nd_size, nid); return; @@ -395,7 +395,7 @@ static int __init numa_alloc_distance(void) phys = memblock_find_in_range(0, PFN_PHYS(max_pfn_mapped), size, PAGE_SIZE); - if (phys == MEMBLOCK_ERROR) { + if (!phys) { pr_warning("NUMA: Warning: can't allocate distance table!\n"); /* don't retry until explicitly reset */ numa_distance = (void *)1LU; diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c index 3adebe7e536ac..58878b536ef2a 100644 --- a/arch/x86/mm/numa_32.c +++ b/arch/x86/mm/numa_32.c @@ -199,7 +199,7 @@ void __init init_alloc_remap(int nid, u64 start, u64 end) /* allocate node memory and the lowmem remap area */ node_pa = memblock_find_in_range(start, end, size, LARGE_PAGE_BYTES); - if (node_pa == MEMBLOCK_ERROR) { + if (!node_pa) { pr_warning("remap_alloc: failed to allocate %lu bytes for node %d\n", size, nid); return; @@ -209,7 +209,7 @@ void __init init_alloc_remap(int nid, u64 start, u64 end) remap_pa = memblock_find_in_range(min_low_pfn << PAGE_SHIFT, max_low_pfn << PAGE_SHIFT, size, LARGE_PAGE_BYTES); - if (remap_pa == MEMBLOCK_ERROR) { + if (!remap_pa) { pr_warning("remap_alloc: failed to allocate %lu bytes remap area for node %d\n", size, nid); memblock_x86_free_range(node_pa, node_pa + size); diff --git a/arch/x86/mm/numa_emulation.c b/arch/x86/mm/numa_emulation.c index d0ed086b6247d..e3d471c20cdca 100644 --- a/arch/x86/mm/numa_emulation.c +++ b/arch/x86/mm/numa_emulation.c @@ -351,7 +351,7 @@ void __init numa_emulation(struct numa_meminfo *numa_meminfo, int numa_dist_cnt) phys = memblock_find_in_range(0, PFN_PHYS(max_pfn_mapped), phys_size, PAGE_SIZE); - if (phys == MEMBLOCK_ERROR) { + if (!phys) { pr_warning("NUMA: Warning: can't allocate copy of distance table, disabling emulation\n"); goto no_emu; } diff --git a/include/linux/memblock.h b/include/linux/memblock.h index 7525e38c434d6..d235ec5fe6783 100644 --- a/include/linux/memblock.h +++ b/include/linux/memblock.h @@ -2,8 +2,6 @@ #define _LINUX_MEMBLOCK_H #ifdef __KERNEL__ -#define MEMBLOCK_ERROR 0 - #ifdef CONFIG_HAVE_MEMBLOCK /* * Logical memory blocks. @@ -164,7 +162,7 @@ static inline unsigned long memblock_region_reserved_end_pfn(const struct memblo #else static inline phys_addr_t memblock_alloc(phys_addr_t size, phys_addr_t align) { - return MEMBLOCK_ERROR; + return 0; } #endif /* CONFIG_HAVE_MEMBLOCK */ diff --git a/kernel/printk.c b/kernel/printk.c index 35185392173f5..b1d5a6174d652 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -199,7 +199,7 @@ void __init setup_log_buf(int early) unsigned long mem; mem = memblock_alloc(new_log_buf_len, PAGE_SIZE); - if (mem == MEMBLOCK_ERROR) + if (!mem) return; new_log_buf = __va(mem); } else { diff --git a/mm/memblock.c b/mm/memblock.c index 9882a88d4a104..1969936613465 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -74,7 +74,7 @@ static phys_addr_t __init_memblock memblock_find_region(phys_addr_t start, phys_ /* In case, huge size is requested */ if (end < size) - return MEMBLOCK_ERROR; + return 0; base = round_down(end - size, align); @@ -94,7 +94,7 @@ static phys_addr_t __init_memblock memblock_find_region(phys_addr_t start, phys_ base = round_down(res_base - size, align); } - return MEMBLOCK_ERROR; + return 0; } static phys_addr_t __init_memblock memblock_find_base(phys_addr_t size, @@ -126,10 +126,10 @@ static phys_addr_t __init_memblock memblock_find_base(phys_addr_t size, if (bottom >= top) continue; found = memblock_find_region(bottom, top, size, align); - if (found != MEMBLOCK_ERROR) + if (found) return found; } - return MEMBLOCK_ERROR; + return 0; } /* @@ -214,10 +214,10 @@ static int __init_memblock memblock_double_array(struct memblock_type *type) */ if (use_slab) { new_array = kmalloc(new_size, GFP_KERNEL); - addr = new_array == NULL ? MEMBLOCK_ERROR : __pa(new_array); + addr = new_array ? __pa(new_array) : 0; } else addr = memblock_find_base(new_size, sizeof(phys_addr_t), 0, MEMBLOCK_ALLOC_ACCESSIBLE); - if (addr == MEMBLOCK_ERROR) { + if (!addr) { pr_err("memblock: Failed to double %s array from %ld to %ld entries !\n", memblock_type_name(type), type->max, type->max * 2); return -1; @@ -478,8 +478,7 @@ phys_addr_t __init __memblock_alloc_base(phys_addr_t size, phys_addr_t align, ph size = round_up(size, align); found = memblock_find_base(size, align, 0, max_addr); - if (found != MEMBLOCK_ERROR && - !memblock_add_region(&memblock.reserved, found, size)) + if (found && !memblock_add_region(&memblock.reserved, found, size)) return found; return 0; @@ -559,14 +558,14 @@ static phys_addr_t __init memblock_alloc_nid_region(struct memblock_region *mp, this_end = memblock_nid_range(start, end, &this_nid); if (this_nid == nid) { phys_addr_t ret = memblock_find_region(start, this_end, size, align); - if (ret != MEMBLOCK_ERROR && + if (ret && !memblock_add_region(&memblock.reserved, ret, size)) return ret; } start = this_end; } - return MEMBLOCK_ERROR; + return 0; } phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int nid) @@ -588,7 +587,7 @@ phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int n for (i = 0; i < mem->cnt; i++) { phys_addr_t ret = memblock_alloc_nid_region(&mem->regions[i], size, align, nid); - if (ret != MEMBLOCK_ERROR) + if (ret) return ret; } diff --git a/mm/nobootmem.c b/mm/nobootmem.c index 6e93dc7f25863..5b0eb06ecb4e4 100644 --- a/mm/nobootmem.c +++ b/mm/nobootmem.c @@ -43,7 +43,7 @@ static void * __init __alloc_memory_core_early(int nid, u64 size, u64 align, addr = find_memory_core_early(nid, size, align, goal, limit); - if (addr == MEMBLOCK_ERROR) + if (!addr) return NULL; ptr = phys_to_virt(addr); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index b6da6ed818a8f..c7f0e5be4a31d 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3878,13 +3878,13 @@ u64 __init find_memory_core_early(int nid, u64 size, u64 align, addr = memblock_find_in_range(final_start, final_end, size, align); - if (addr == MEMBLOCK_ERROR) + if (!addr) continue; return addr; } - return MEMBLOCK_ERROR; + return 0; } #endif -- GitLab From fc769a8e70a3348d5de49e5f69f6aff810157360 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 09:58:10 +0200 Subject: [PATCH 0010/3089] memblock: Replace memblock_find_base() with memblock_find_in_range() memblock_find_base() is a static function with two callers in memblock.c and memblock_find_in_range() is a wrapper around it which just changes the types and order of parameters. Make memblock_find_in_range() take phys_addr_t instead of u64 for consistency and replace memblock_find_base() with it. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310457490-3356-7-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- include/linux/memblock.h | 3 ++- mm/memblock.c | 19 +++++++------------ 2 files changed, 9 insertions(+), 13 deletions(-) diff --git a/include/linux/memblock.h b/include/linux/memblock.h index d235ec5fe6783..349688899cb06 100644 --- a/include/linux/memblock.h +++ b/include/linux/memblock.h @@ -46,7 +46,8 @@ extern int memblock_can_resize; #define memblock_dbg(fmt, ...) \ if (memblock_debug) printk(KERN_INFO pr_fmt(fmt), ##__VA_ARGS__) -u64 memblock_find_in_range(u64 start, u64 end, u64 size, u64 align); +phys_addr_t memblock_find_in_range(phys_addr_t start, phys_addr_t end, + phys_addr_t size, phys_addr_t align); int memblock_free_reserved_regions(void); int memblock_reserve_reserved_regions(void); diff --git a/mm/memblock.c b/mm/memblock.c index 1969936613465..0f9626f01b5e6 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -97,8 +97,11 @@ static phys_addr_t __init_memblock memblock_find_region(phys_addr_t start, phys_ return 0; } -static phys_addr_t __init_memblock memblock_find_base(phys_addr_t size, - phys_addr_t align, phys_addr_t start, phys_addr_t end) +/* + * Find a free area with specified alignment in a specific range. + */ +phys_addr_t __init_memblock memblock_find_in_range(phys_addr_t start, phys_addr_t end, + phys_addr_t size, phys_addr_t align) { long i; @@ -132,14 +135,6 @@ static phys_addr_t __init_memblock memblock_find_base(phys_addr_t size, return 0; } -/* - * Find a free area with specified alignment in a specific range. - */ -u64 __init_memblock memblock_find_in_range(u64 start, u64 end, u64 size, u64 align) -{ - return memblock_find_base(size, align, start, end); -} - /* * Free memblock.reserved.regions */ @@ -216,7 +211,7 @@ static int __init_memblock memblock_double_array(struct memblock_type *type) new_array = kmalloc(new_size, GFP_KERNEL); addr = new_array ? __pa(new_array) : 0; } else - addr = memblock_find_base(new_size, sizeof(phys_addr_t), 0, MEMBLOCK_ALLOC_ACCESSIBLE); + addr = memblock_find_in_range(0, MEMBLOCK_ALLOC_ACCESSIBLE, new_size, sizeof(phys_addr_t)); if (!addr) { pr_err("memblock: Failed to double %s array from %ld to %ld entries !\n", memblock_type_name(type), type->max, type->max * 2); @@ -477,7 +472,7 @@ phys_addr_t __init __memblock_alloc_base(phys_addr_t size, phys_addr_t align, ph */ size = round_up(size, align); - found = memblock_find_base(size, align, 0, max_addr); + found = memblock_find_in_range(0, max_addr, size, align); if (found && !memblock_add_region(&memblock.reserved, found, size)) return found; -- GitLab From 5dfe8660a3d7f1ee1265c3536433ee53da3f98a3 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 14 Jul 2011 09:46:10 +0200 Subject: [PATCH 0011/3089] bootmem: Replace work_with_active_regions() with for_each_mem_pfn_range() Callback based iteration is cumbersome and much less useful than for_each_*() iterator. This patch implements for_each_mem_pfn_range() which replaces work_with_active_regions(). All the current users of work_with_active_regions() are converted. This simplifies walking over early_node_map and will allow converting internal logics in page_alloc to use iterator instead of walking early_node_map directly, which in turn will enable moving node information to memblock. powerpc change is only compile tested. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/20110714074610.GD3455@htj.dyndns.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- arch/powerpc/mm/numa.c | 50 ++++++++++++--------------------------- arch/x86/mm/memblock.c | 23 ++++-------------- drivers/pci/intel-iommu.c | 24 +++++++------------ include/linux/mm.h | 22 +++++++++++++++-- mm/page_alloc.c | 40 +++++++++++++++++++++---------- 5 files changed, 76 insertions(+), 83 deletions(-) diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c index 2164006fe170f..6f06ea53bca28 100644 --- a/arch/powerpc/mm/numa.c +++ b/arch/powerpc/mm/numa.c @@ -127,45 +127,25 @@ static int __cpuinit fake_numa_create_new_node(unsigned long end_pfn, } /* - * get_active_region_work_fn - A helper function for get_node_active_region - * Returns datax set to the start_pfn and end_pfn if they contain - * the initial value of datax->start_pfn between them - * @start_pfn: start page(inclusive) of region to check - * @end_pfn: end page(exclusive) of region to check - * @datax: comes in with ->start_pfn set to value to search for and - * goes out with active range if it contains it - * Returns 1 if search value is in range else 0 - */ -static int __init get_active_region_work_fn(unsigned long start_pfn, - unsigned long end_pfn, void *datax) -{ - struct node_active_region *data; - data = (struct node_active_region *)datax; - - if (start_pfn <= data->start_pfn && end_pfn > data->start_pfn) { - data->start_pfn = start_pfn; - data->end_pfn = end_pfn; - return 1; - } - return 0; - -} - -/* - * get_node_active_region - Return active region containing start_pfn + * get_node_active_region - Return active region containing pfn * Active range returned is empty if none found. - * @start_pfn: The page to return the region for. - * @node_ar: Returned set to the active region containing start_pfn + * @pfn: The page to return the region for + * @node_ar: Returned set to the active region containing @pfn */ -static void __init get_node_active_region(unsigned long start_pfn, - struct node_active_region *node_ar) +static void __init get_node_active_region(unsigned long pfn, + struct node_active_region *node_ar) { - int nid = early_pfn_to_nid(start_pfn); + unsigned long start_pfn, end_pfn; + int i, nid; - node_ar->nid = nid; - node_ar->start_pfn = start_pfn; - node_ar->end_pfn = start_pfn; - work_with_active_regions(nid, get_active_region_work_fn, node_ar); + for_each_mem_pfn_range(i, MAX_NUMNODES, &start_pfn, &end_pfn, &nid) { + if (pfn >= start_pfn && pfn < end_pfn) { + node_ar->nid = nid; + node_ar->start_pfn = start_pfn; + node_ar->end_pfn = end_pfn; + break; + } + } } static void map_cpu_to_node(int cpu, int node) diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index e126117d1b03b..da0d5c84586ef 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -115,28 +115,13 @@ static void __init memblock_x86_subtract_reserved(struct range *range, int az) memblock_reserve_reserved_regions(); } -struct count_data { - int nr; -}; - -static int __init count_work_fn(unsigned long start_pfn, - unsigned long end_pfn, void *datax) -{ - struct count_data *data = datax; - - data->nr++; - - return 0; -} - static int __init count_early_node_map(int nodeid) { - struct count_data data; - - data.nr = 0; - work_with_active_regions(nodeid, count_work_fn, &data); + int i, cnt = 0; - return data.nr; + for_each_mem_pfn_range(i, nodeid, NULL, NULL, NULL) + cnt++; + return cnt; } int __init __get_free_all_memory_range(struct range **rangep, int nodeid, diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c index f02c34d26d1b0..8ec352077e1ac 100644 --- a/drivers/pci/intel-iommu.c +++ b/drivers/pci/intel-iommu.c @@ -2178,18 +2178,6 @@ static inline void iommu_prepare_isa(void) static int md_domain_init(struct dmar_domain *domain, int guest_width); -static int __init si_domain_work_fn(unsigned long start_pfn, - unsigned long end_pfn, void *datax) -{ - int *ret = datax; - - *ret = iommu_domain_identity_map(si_domain, - (uint64_t)start_pfn << PAGE_SHIFT, - (uint64_t)end_pfn << PAGE_SHIFT); - return *ret; - -} - static int __init si_domain_init(int hw) { struct dmar_drhd_unit *drhd; @@ -2221,9 +2209,15 @@ static int __init si_domain_init(int hw) return 0; for_each_online_node(nid) { - work_with_active_regions(nid, si_domain_work_fn, &ret); - if (ret) - return ret; + unsigned long start_pfn, end_pfn; + int i; + + for_each_mem_pfn_range(i, nid, &start_pfn, &end_pfn, NULL) { + ret = iommu_domain_identity_map(si_domain, + PFN_PHYS(start_pfn), PFN_PHYS(end_pfn)); + if (ret) + return ret; + } } return 0; diff --git a/include/linux/mm.h b/include/linux/mm.h index c70a326b8f267..57e4c9ffdff81 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1327,9 +1327,27 @@ int add_from_early_node_map(struct range *range, int az, int nr_range, int nid); u64 __init find_memory_core_early(int nid, u64 size, u64 align, u64 goal, u64 limit); -typedef int (*work_fn_t)(unsigned long, unsigned long, void *); -extern void work_with_active_regions(int nid, work_fn_t work_fn, void *data); extern void sparse_memory_present_with_active_regions(int nid); + +extern void __next_mem_pfn_range(int *idx, int nid, + unsigned long *out_start_pfn, + unsigned long *out_end_pfn, int *out_nid); + +/** + * for_each_mem_pfn_range - early memory pfn range iterator + * @i: an integer used as loop variable + * @nid: node selector, %MAX_NUMNODES for all nodes + * @p_start: ptr to ulong for start pfn of the range, can be %NULL + * @p_end: ptr to ulong for end pfn of the range, can be %NULL + * @p_nid: ptr to int for nid of the range, can be %NULL + * + * Walks over configured memory ranges. Available after early_node_map is + * populated. + */ +#define for_each_mem_pfn_range(i, nid, p_start, p_end, p_nid) \ + for (i = -1, __next_mem_pfn_range(&i, nid, p_start, p_end, p_nid); \ + i >= 0; __next_mem_pfn_range(&i, nid, p_start, p_end, p_nid)) + #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */ #if !defined(CONFIG_ARCH_POPULATES_NODE_MAP) && \ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index c7f0e5be4a31d..69fffabf61b7f 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3903,18 +3903,6 @@ int __init add_from_early_node_map(struct range *range, int az, return nr_range; } -void __init work_with_active_regions(int nid, work_fn_t work_fn, void *data) -{ - int i; - int ret; - - for_each_active_range_index_in_nid(i, nid) { - ret = work_fn(early_node_map[i].start_pfn, - early_node_map[i].end_pfn, data); - if (ret) - break; - } -} /** * sparse_memory_present_with_active_regions - Call memory_present for each active range * @nid: The node to call memory_present for. If MAX_NUMNODES, all nodes will be used. @@ -4421,6 +4409,34 @@ static inline void setup_nr_node_ids(void) } #endif +/* + * Common iterator interface used to define for_each_mem_pfn_range(). + */ +void __meminit __next_mem_pfn_range(int *idx, int nid, + unsigned long *out_start_pfn, + unsigned long *out_end_pfn, int *out_nid) +{ + struct node_active_region *r = NULL; + + while (++*idx < nr_nodemap_entries) { + if (nid == MAX_NUMNODES || nid == early_node_map[*idx].nid) { + r = &early_node_map[*idx]; + break; + } + } + if (!r) { + *idx = -1; + return; + } + + if (out_start_pfn) + *out_start_pfn = r->start_pfn; + if (out_end_pfn) + *out_end_pfn = r->end_pfn; + if (out_nid) + *out_nid = r->nid; +} + /** * add_active_range - Register a range of PFNs backed by physical memory * @nid: The node ID the range resides on -- GitLab From 96e907d1360240d1958fe8ce3a3ac640733330d4 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 10:46:29 +0200 Subject: [PATCH 0012/3089] bootmem: Reimplement __absent_pages_in_range() using for_each_mem_pfn_range() __absent_pages_in_range() was needlessly complex. Reimplement it using for_each_mem_pfn_range(). Also, update zone_absent_pages_in_node() such that it doesn't call __absent_pages_in_range() with @zone_start_pfn which is larger than @zone_end_pfn. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310460395-30913-3-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- mm/page_alloc.c | 54 +++++++++++-------------------------------------- 1 file changed, 12 insertions(+), 42 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 69fffabf61b7f..3092a97268d16 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -4044,46 +4044,16 @@ unsigned long __meminit __absent_pages_in_range(int nid, unsigned long range_start_pfn, unsigned long range_end_pfn) { - int i = 0; - unsigned long prev_end_pfn = 0, hole_pages = 0; - unsigned long start_pfn; - - /* Find the end_pfn of the first active range of pfns in the node */ - i = first_active_region_index_in_nid(nid); - if (i == -1) - return 0; - - prev_end_pfn = min(early_node_map[i].start_pfn, range_end_pfn); - - /* Account for ranges before physical memory on this node */ - if (early_node_map[i].start_pfn > range_start_pfn) - hole_pages = prev_end_pfn - range_start_pfn; - - /* Find all holes for the zone within the node */ - for (; i != -1; i = next_active_region_index_in_nid(i, nid)) { - - /* No need to continue if prev_end_pfn is outside the zone */ - if (prev_end_pfn >= range_end_pfn) - break; - - /* Make sure the end of the zone is not within the hole */ - start_pfn = min(early_node_map[i].start_pfn, range_end_pfn); - prev_end_pfn = max(prev_end_pfn, range_start_pfn); + unsigned long nr_absent = range_end_pfn - range_start_pfn; + unsigned long start_pfn, end_pfn; + int i; - /* Update the hole size cound and move on */ - if (start_pfn > range_start_pfn) { - BUG_ON(prev_end_pfn > start_pfn); - hole_pages += start_pfn - prev_end_pfn; - } - prev_end_pfn = early_node_map[i].end_pfn; + for_each_mem_pfn_range(i, nid, &start_pfn, &end_pfn, NULL) { + start_pfn = clamp(start_pfn, range_start_pfn, range_end_pfn); + end_pfn = clamp(end_pfn, range_start_pfn, range_end_pfn); + nr_absent -= end_pfn - start_pfn; } - - /* Account for ranges past physical memory on this node */ - if (range_end_pfn > prev_end_pfn) - hole_pages += range_end_pfn - - max(range_start_pfn, prev_end_pfn); - - return hole_pages; + return nr_absent; } /** @@ -4104,14 +4074,14 @@ static unsigned long __meminit zone_absent_pages_in_node(int nid, unsigned long zone_type, unsigned long *ignored) { + unsigned long zone_low = arch_zone_lowest_possible_pfn[zone_type]; + unsigned long zone_high = arch_zone_highest_possible_pfn[zone_type]; unsigned long node_start_pfn, node_end_pfn; unsigned long zone_start_pfn, zone_end_pfn; get_pfn_range_for_nid(nid, &node_start_pfn, &node_end_pfn); - zone_start_pfn = max(arch_zone_lowest_possible_pfn[zone_type], - node_start_pfn); - zone_end_pfn = min(arch_zone_highest_possible_pfn[zone_type], - node_end_pfn); + zone_start_pfn = clamp(node_start_pfn, zone_low, zone_high); + zone_end_pfn = clamp(node_end_pfn, zone_low, zone_high); adjust_zone_range_for_zone_movable(nid, zone_type, node_start_pfn, node_end_pfn, -- GitLab From c13291a536b835b2ab278ab201f2cb1ce22f2785 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 10:46:30 +0200 Subject: [PATCH 0013/3089] bootmem: Use for_each_mem_pfn_range() in page_alloc.c The previous patch added for_each_mem_pfn_range() which is more versatile than for_each_active_range_index_in_nid(). This patch replaces for_each_active_range_index_in_nid() and open coded early_node_map[] walks with for_each_mem_pfn_range(). All conversions in this patch are straight-forward and shouldn't cause any functional difference. After the conversions, for_each_active_range_index_in_nid() doesn't have any user left and is removed. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310460395-30913-4-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- mm/page_alloc.c | 165 ++++++++++++++++-------------------------------- 1 file changed, 56 insertions(+), 109 deletions(-) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 3092a97268d16..902f03a4fd6ba 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3711,34 +3711,6 @@ __meminit int init_currently_empty_zone(struct zone *zone, } #ifdef CONFIG_ARCH_POPULATES_NODE_MAP -/* - * Basic iterator support. Return the first range of PFNs for a node - * Note: nid == MAX_NUMNODES returns first region regardless of node - */ -static int __meminit first_active_region_index_in_nid(int nid) -{ - int i; - - for (i = 0; i < nr_nodemap_entries; i++) - if (nid == MAX_NUMNODES || early_node_map[i].nid == nid) - return i; - - return -1; -} - -/* - * Basic iterator support. Return the next active range of PFNs for a node - * Note: nid == MAX_NUMNODES returns next region regardless of node - */ -static int __meminit next_active_region_index_in_nid(int index, int nid) -{ - for (index = index + 1; index < nr_nodemap_entries; index++) - if (nid == MAX_NUMNODES || early_node_map[index].nid == nid) - return index; - - return -1; -} - #ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID /* * Required by SPARSEMEM. Given a PFN, return what node the PFN is on. @@ -3748,15 +3720,12 @@ static int __meminit next_active_region_index_in_nid(int index, int nid) */ int __meminit __early_pfn_to_nid(unsigned long pfn) { - int i; - - for (i = 0; i < nr_nodemap_entries; i++) { - unsigned long start_pfn = early_node_map[i].start_pfn; - unsigned long end_pfn = early_node_map[i].end_pfn; + unsigned long start_pfn, end_pfn; + int i, nid; + for_each_mem_pfn_range(i, MAX_NUMNODES, &start_pfn, &end_pfn, &nid) if (start_pfn <= pfn && pfn < end_pfn) - return early_node_map[i].nid; - } + return nid; /* This is a memory hole */ return -1; } @@ -3785,11 +3754,6 @@ bool __meminit early_pfn_in_nid(unsigned long pfn, int node) } #endif -/* Basic iterator support to walk early_node_map[] */ -#define for_each_active_range_index_in_nid(i, nid) \ - for (i = first_active_region_index_in_nid(nid); i != -1; \ - i = next_active_region_index_in_nid(i, nid)) - /** * free_bootmem_with_active_regions - Call free_bootmem_node for each active range * @nid: The node to free memory on. If MAX_NUMNODES, all nodes are freed. @@ -3799,25 +3763,19 @@ bool __meminit early_pfn_in_nid(unsigned long pfn, int node) * add_active_ranges() contain no holes and may be freed, this * this function may be used instead of calling free_bootmem() manually. */ -void __init free_bootmem_with_active_regions(int nid, - unsigned long max_low_pfn) +void __init free_bootmem_with_active_regions(int nid, unsigned long max_low_pfn) { - int i; - - for_each_active_range_index_in_nid(i, nid) { - unsigned long size_pages = 0; - unsigned long end_pfn = early_node_map[i].end_pfn; - - if (early_node_map[i].start_pfn >= max_low_pfn) - continue; + unsigned long start_pfn, end_pfn; + int i, this_nid; - if (end_pfn > max_low_pfn) - end_pfn = max_low_pfn; + for_each_mem_pfn_range(i, nid, &start_pfn, &end_pfn, &this_nid) { + start_pfn = min(start_pfn, max_low_pfn); + end_pfn = min(end_pfn, max_low_pfn); - size_pages = end_pfn - early_node_map[i].start_pfn; - free_bootmem_node(NODE_DATA(early_node_map[i].nid), - PFN_PHYS(early_node_map[i].start_pfn), - size_pages << PAGE_SHIFT); + if (start_pfn < end_pfn) + free_bootmem_node(NODE_DATA(this_nid), + PFN_PHYS(start_pfn), + (end_pfn - start_pfn) << PAGE_SHIFT); } } @@ -3891,15 +3849,12 @@ u64 __init find_memory_core_early(int nid, u64 size, u64 align, int __init add_from_early_node_map(struct range *range, int az, int nr_range, int nid) { + unsigned long start_pfn, end_pfn; int i; - u64 start, end; /* need to go over early_node_map to find out good range for node */ - for_each_active_range_index_in_nid(i, nid) { - start = early_node_map[i].start_pfn; - end = early_node_map[i].end_pfn; - nr_range = add_range(range, az, nr_range, start, end); - } + for_each_mem_pfn_range(i, nid, &start_pfn, &end_pfn, NULL) + nr_range = add_range(range, az, nr_range, start_pfn, end_pfn); return nr_range; } @@ -3913,12 +3868,11 @@ int __init add_from_early_node_map(struct range *range, int az, */ void __init sparse_memory_present_with_active_regions(int nid) { - int i; + unsigned long start_pfn, end_pfn; + int i, this_nid; - for_each_active_range_index_in_nid(i, nid) - memory_present(early_node_map[i].nid, - early_node_map[i].start_pfn, - early_node_map[i].end_pfn); + for_each_mem_pfn_range(i, nid, &start_pfn, &end_pfn, &this_nid) + memory_present(this_nid, start_pfn, end_pfn); } /** @@ -3935,13 +3889,15 @@ void __init sparse_memory_present_with_active_regions(int nid) void __meminit get_pfn_range_for_nid(unsigned int nid, unsigned long *start_pfn, unsigned long *end_pfn) { + unsigned long this_start_pfn, this_end_pfn; int i; + *start_pfn = -1UL; *end_pfn = 0; - for_each_active_range_index_in_nid(i, nid) { - *start_pfn = min(*start_pfn, early_node_map[i].start_pfn); - *end_pfn = max(*end_pfn, early_node_map[i].end_pfn); + for_each_mem_pfn_range(i, nid, &this_start_pfn, &this_end_pfn, NULL) { + *start_pfn = min(*start_pfn, this_start_pfn); + *end_pfn = max(*end_pfn, this_end_pfn); } if (*start_pfn == -1UL) @@ -4484,6 +4440,7 @@ void __init add_active_range(unsigned int nid, unsigned long start_pfn, void __init remove_active_range(unsigned int nid, unsigned long start_pfn, unsigned long end_pfn) { + unsigned long this_start_pfn, this_end_pfn; int i, j; int removed = 0; @@ -4491,26 +4448,22 @@ void __init remove_active_range(unsigned int nid, unsigned long start_pfn, nid, start_pfn, end_pfn); /* Find the old active region end and shrink */ - for_each_active_range_index_in_nid(i, nid) { - if (early_node_map[i].start_pfn >= start_pfn && - early_node_map[i].end_pfn <= end_pfn) { + for_each_mem_pfn_range(i, nid, &this_start_pfn, &this_end_pfn, NULL) { + if (this_start_pfn >= start_pfn && this_end_pfn <= end_pfn) { /* clear it */ early_node_map[i].start_pfn = 0; early_node_map[i].end_pfn = 0; removed = 1; continue; } - if (early_node_map[i].start_pfn < start_pfn && - early_node_map[i].end_pfn > start_pfn) { - unsigned long temp_end_pfn = early_node_map[i].end_pfn; + if (this_start_pfn < start_pfn && this_end_pfn > start_pfn) { early_node_map[i].end_pfn = start_pfn; - if (temp_end_pfn > end_pfn) - add_active_range(nid, end_pfn, temp_end_pfn); + if (this_end_pfn > end_pfn) + add_active_range(nid, end_pfn, this_end_pfn); continue; } - if (early_node_map[i].start_pfn >= start_pfn && - early_node_map[i].end_pfn > end_pfn && - early_node_map[i].start_pfn < end_pfn) { + if (this_start_pfn >= start_pfn && this_end_pfn > end_pfn && + this_start_pfn < end_pfn) { early_node_map[i].start_pfn = end_pfn; continue; } @@ -4593,15 +4546,11 @@ void __init sort_node_map(void) unsigned long __init node_map_pfn_alignment(void) { unsigned long accl_mask = 0, last_end = 0; + unsigned long start, end, mask; int last_nid = -1; - int i; - - for_each_active_range_index_in_nid(i, MAX_NUMNODES) { - int nid = early_node_map[i].nid; - unsigned long start = early_node_map[i].start_pfn; - unsigned long end = early_node_map[i].end_pfn; - unsigned long mask; + int i, nid; + for_each_mem_pfn_range(i, MAX_NUMNODES, &start, &end, &nid) { if (!start || last_nid < 0 || last_nid == nid) { last_nid = nid; last_end = end; @@ -4628,12 +4577,12 @@ unsigned long __init node_map_pfn_alignment(void) /* Find the lowest pfn for a node */ static unsigned long __init find_min_pfn_for_node(int nid) { - int i; unsigned long min_pfn = ULONG_MAX; + unsigned long start_pfn; + int i; - /* Assuming a sorted map, the first range found has the starting pfn */ - for_each_active_range_index_in_nid(i, nid) - min_pfn = min(min_pfn, early_node_map[i].start_pfn); + for_each_mem_pfn_range(i, nid, &start_pfn, NULL, NULL) + min_pfn = min(min_pfn, start_pfn); if (min_pfn == ULONG_MAX) { printk(KERN_WARNING @@ -4662,15 +4611,16 @@ unsigned long __init find_min_pfn_with_active_regions(void) */ static unsigned long __init early_calculate_totalpages(void) { - int i; unsigned long totalpages = 0; + unsigned long start_pfn, end_pfn; + int i, nid; + + for_each_mem_pfn_range(i, MAX_NUMNODES, &start_pfn, &end_pfn, &nid) { + unsigned long pages = end_pfn - start_pfn; - for (i = 0; i < nr_nodemap_entries; i++) { - unsigned long pages = early_node_map[i].end_pfn - - early_node_map[i].start_pfn; totalpages += pages; if (pages) - node_set_state(early_node_map[i].nid, N_HIGH_MEMORY); + node_set_state(nid, N_HIGH_MEMORY); } return totalpages; } @@ -4725,6 +4675,8 @@ restart: /* Spread kernelcore memory as evenly as possible throughout nodes */ kernelcore_node = required_kernelcore / usable_nodes; for_each_node_state(nid, N_HIGH_MEMORY) { + unsigned long start_pfn, end_pfn; + /* * Recalculate kernelcore_node if the division per node * now exceeds what is necessary to satisfy the requested @@ -4741,13 +4693,10 @@ restart: kernelcore_remaining = kernelcore_node; /* Go through each range of PFNs within this node */ - for_each_active_range_index_in_nid(i, nid) { - unsigned long start_pfn, end_pfn; + for_each_mem_pfn_range(i, nid, &start_pfn, &end_pfn, NULL) { unsigned long size_pages; - start_pfn = max(early_node_map[i].start_pfn, - zone_movable_pfn[nid]); - end_pfn = early_node_map[i].end_pfn; + start_pfn = max(start_pfn, zone_movable_pfn[nid]); if (start_pfn >= end_pfn) continue; @@ -4849,8 +4798,8 @@ static void check_for_regular_memory(pg_data_t *pgdat) */ void __init free_area_init_nodes(unsigned long *max_zone_pfn) { - unsigned long nid; - int i; + unsigned long start_pfn, end_pfn; + int i, nid; /* Sort early_node_map as initialisation assumes it is sorted */ sort_node_map(); @@ -4900,11 +4849,9 @@ void __init free_area_init_nodes(unsigned long *max_zone_pfn) } /* Print out the early_node_map[] */ - printk("early_node_map[%d] active PFN ranges\n", nr_nodemap_entries); - for (i = 0; i < nr_nodemap_entries; i++) - printk(" %3d: %0#10lx -> %0#10lx\n", early_node_map[i].nid, - early_node_map[i].start_pfn, - early_node_map[i].end_pfn); + printk("Early memory PFN ranges\n"); + for_each_mem_pfn_range(i, MAX_NUMNODES, &start_pfn, &end_pfn, &nid) + printk(" %3d: %0#10lx -> %0#10lx\n", nid, start_pfn, end_pfn); /* Initialise every node */ mminit_verify_pageflags_layout(); -- GitLab From b2fea988f4f3b38ff4edfc1556a843c91932804c Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 10:46:31 +0200 Subject: [PATCH 0014/3089] memblock: Improve generic memblock_nid_range() using for_each_mem_pfn_range() Given an address range, memblock_nid_range() determines the node the start of the range belongs to and upto where the range stays in the same node. It's implemented by calling get_pfn_range_for_nid(), which determines min and max pfns for a given node, for each node and testing whether start address falls in there. This is not only inefficient but also incorrect when nodes interleave as min-max ranges for nodes overlap. This patch reimplements memblock_nid_range() using for_each_mem_pfn_range(). It's simpler, walks the mem ranges once and can find the exact range the start address falls in. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310460395-30913-5-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- mm/memblock.c | 20 +++----------------- 1 file changed, 3 insertions(+), 17 deletions(-) diff --git a/mm/memblock.c b/mm/memblock.c index 0f9626f01b5e6..97f3486ce4d64 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -511,28 +511,14 @@ phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align) phys_addr_t __weak __init memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid) { #ifdef CONFIG_ARCH_POPULATES_NODE_MAP - /* - * This code originates from sparc which really wants use to walk by addresses - * and returns the nid. This is not very convenient for early_pfn_map[] users - * as the map isn't sorted yet, and it really wants to be walked by nid. - * - * For now, I implement the inefficient method below which walks the early - * map multiple times. Eventually we may want to use an ARCH config option - * to implement a completely different method for both case. - */ unsigned long start_pfn, end_pfn; int i; - for (i = 0; i < MAX_NUMNODES; i++) { - get_pfn_range_for_nid(i, &start_pfn, &end_pfn); - if (start < PFN_PHYS(start_pfn) || start >= PFN_PHYS(end_pfn)) - continue; - *nid = i; - return min(end, PFN_PHYS(end_pfn)); - } + for_each_mem_pfn_range(i, MAX_NUMNODES, &start_pfn, &end_pfn, nid) + if (start >= PFN_PHYS(start_pfn) && start < PFN_PHYS(end_pfn)) + return min(end, PFN_PHYS(end_pfn)); #endif *nid = 0; - return end; } -- GitLab From f9b18db3b1cedc75e5d002a4d7097891c3399736 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 10:46:32 +0200 Subject: [PATCH 0015/3089] memblock: Don't allow archs to override memblock_nid_range() memblock_nid_range() is used to implement memblock_[try_]alloc_nid(). The generic version determines the range by walking early_node_map with for_each_mem_pfn_range(). The generic version is defined __weak to allow arch override. Currently, only sparc overrides it; however, with the previous update to the generic implementation, there isn't much to be gained with arch override. Sparc would behave exactly the same with the generic implementation. This patch disallows arch override for memblock_nid_range() and make both generic and sparc versions static. sparc is only compile tested. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310460395-30913-6-git-send-email-tj@kernel.org Cc: "David S. Miller" Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- arch/sparc/mm/init_64.c | 4 ++-- include/linux/memblock.h | 1 - mm/memblock.c | 2 +- 3 files changed, 3 insertions(+), 4 deletions(-) diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c index 3fd8e18bed80a..8415f614ce0c2 100644 --- a/arch/sparc/mm/init_64.c +++ b/arch/sparc/mm/init_64.c @@ -785,7 +785,7 @@ static int find_node(unsigned long addr) return -1; } -u64 memblock_nid_range(u64 start, u64 end, int *nid) +static u64 memblock_nid_range(u64 start, u64 end, int *nid) { *nid = find_node(start); start += PAGE_SIZE; @@ -803,7 +803,7 @@ u64 memblock_nid_range(u64 start, u64 end, int *nid) return start; } #else -u64 memblock_nid_range(u64 start, u64 end, int *nid) +static u64 memblock_nid_range(u64 start, u64 end, int *nid) { *nid = 0; return end; diff --git a/include/linux/memblock.h b/include/linux/memblock.h index 349688899cb06..329ffb26c1c98 100644 --- a/include/linux/memblock.h +++ b/include/linux/memblock.h @@ -89,7 +89,6 @@ extern int memblock_is_region_reserved(phys_addr_t base, phys_addr_t size); extern void memblock_dump_all(void); /* Provided by the architecture */ -extern phys_addr_t memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid); extern int memblock_memory_can_coalesce(phys_addr_t addr1, phys_addr_t size1, phys_addr_t addr2, phys_addr_t size2); diff --git a/mm/memblock.c b/mm/memblock.c index 97f3486ce4d64..22cd999b0d4e5 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -508,7 +508,7 @@ phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align) * have been done to populate it. */ -phys_addr_t __weak __init memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid) +static phys_addr_t __init memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid) { #ifdef CONFIG_ARCH_POPULATES_NODE_MAP unsigned long start_pfn, end_pfn; -- GitLab From 34e1845548418e5cecee0568ba721e1f089c092c Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 10:46:33 +0200 Subject: [PATCH 0016/3089] memblock: Make memblock_alloc_[try_]nid() top-down NUMA aware memblock alloc functions - memblock_alloc_[try_]nid() - weren't properly top-down because memblock_nid_range() scanned forward. This patch reverses memblock_nid_range(), renames it to memblock_nid_range_rev() and updates related functions to implement proper top-down allocation. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310460395-30913-7-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- mm/memblock.c | 29 +++++++++++------------------ 1 file changed, 11 insertions(+), 18 deletions(-) diff --git a/mm/memblock.c b/mm/memblock.c index 22cd999b0d4e5..447cf64304bac 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -499,27 +499,26 @@ phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align) /* - * Additional node-local allocators. Search for node memory is bottom up - * and walks memblock regions within that node bottom-up as well, but allocation - * within an memblock region is top-down. XXX I plan to fix that at some stage + * Additional node-local top-down allocators. * * WARNING: Only available after early_node_map[] has been populated, * on some architectures, that is after all the calls to add_active_range() * have been done to populate it. */ -static phys_addr_t __init memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid) +static phys_addr_t __init memblock_nid_range_rev(phys_addr_t start, + phys_addr_t end, int *nid) { #ifdef CONFIG_ARCH_POPULATES_NODE_MAP unsigned long start_pfn, end_pfn; int i; for_each_mem_pfn_range(i, MAX_NUMNODES, &start_pfn, &end_pfn, nid) - if (start >= PFN_PHYS(start_pfn) && start < PFN_PHYS(end_pfn)) - return min(end, PFN_PHYS(end_pfn)); + if (end > PFN_PHYS(start_pfn) && end <= PFN_PHYS(end_pfn)) + return max(start, PFN_PHYS(start_pfn)); #endif *nid = 0; - return end; + return start; } static phys_addr_t __init memblock_alloc_nid_region(struct memblock_region *mp, @@ -531,21 +530,19 @@ static phys_addr_t __init memblock_alloc_nid_region(struct memblock_region *mp, start = mp->base; end = start + mp->size; - start = round_up(start, align); while (start < end) { - phys_addr_t this_end; + phys_addr_t this_start; int this_nid; - this_end = memblock_nid_range(start, end, &this_nid); + this_start = memblock_nid_range_rev(start, end, &this_nid); if (this_nid == nid) { - phys_addr_t ret = memblock_find_region(start, this_end, size, align); + phys_addr_t ret = memblock_find_region(this_start, end, size, align); if (ret && !memblock_add_region(&memblock.reserved, ret, size)) return ret; } - start = this_end; + end = this_start; } - return 0; } @@ -561,11 +558,7 @@ phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int n */ size = round_up(size, align); - /* We do a bottom-up search for a region with the right - * nid since that's easier considering how memblock_nid_range() - * works - */ - for (i = 0; i < mem->cnt; i++) { + for (i = mem->cnt - 1; i >= 0; i--) { phys_addr_t ret = memblock_alloc_nid_region(&mem->regions[i], size, align, nid); if (ret) -- GitLab From e64980405cc6aa74ef178d8d9aa4018c867ceed1 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 10:46:34 +0200 Subject: [PATCH 0017/3089] memblock: Separate out memblock_find_in_range_node() Node affine memblock allocation logic is currently implemented across memblock_alloc_nid() and memblock_alloc_nid_region(). This reorganizes it such that it resembles that of non-NUMA allocation API. Area finding is collected and moved into new exported function memblock_find_in_range_node() which is symmetrical to non-NUMA counterpart - it handles @start/@end and understands ANYWHERE and ACCESSIBLE. memblock_alloc_nid() now simply calls memblock_find_in_range_node() and reserves the returned area. This makes memblock_alloc[_try]_nid() observe ACCESSIBLE limit on node affine allocations too (again, this doesn't make any difference for the current sole user - sparc64). Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310460395-30913-8-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- include/linux/memblock.h | 4 +++ mm/memblock.c | 57 ++++++++++++++++++++++------------------ 2 files changed, 36 insertions(+), 25 deletions(-) diff --git a/include/linux/memblock.h b/include/linux/memblock.h index 329ffb26c1c98..7400d029df487 100644 --- a/include/linux/memblock.h +++ b/include/linux/memblock.h @@ -61,6 +61,10 @@ extern long memblock_reserve(phys_addr_t base, phys_addr_t size); /* The numa aware allocator is only available if * CONFIG_ARCH_POPULATES_NODE_MAP is set */ +extern phys_addr_t memblock_find_in_range_node(phys_addr_t start, + phys_addr_t end, + phys_addr_t size, + phys_addr_t align, int nid); extern phys_addr_t memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int nid); extern phys_addr_t memblock_alloc_try_nid(phys_addr_t size, phys_addr_t align, diff --git a/mm/memblock.c b/mm/memblock.c index 447cf64304bac..a8edb422795bc 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -521,49 +521,56 @@ static phys_addr_t __init memblock_nid_range_rev(phys_addr_t start, return start; } -static phys_addr_t __init memblock_alloc_nid_region(struct memblock_region *mp, +phys_addr_t __init memblock_find_in_range_node(phys_addr_t start, + phys_addr_t end, phys_addr_t size, phys_addr_t align, int nid) { - phys_addr_t start, end; + struct memblock_type *mem = &memblock.memory; + int i; - start = mp->base; - end = start + mp->size; + BUG_ON(0 == size); - while (start < end) { - phys_addr_t this_start; - int this_nid; + /* Pump up max_addr */ + if (end == MEMBLOCK_ALLOC_ACCESSIBLE) + end = memblock.current_limit; - this_start = memblock_nid_range_rev(start, end, &this_nid); - if (this_nid == nid) { - phys_addr_t ret = memblock_find_region(this_start, end, size, align); - if (ret && - !memblock_add_region(&memblock.reserved, ret, size)) - return ret; + for (i = mem->cnt - 1; i >= 0; i--) { + struct memblock_region *r = &mem->regions[i]; + phys_addr_t base = max(start, r->base); + phys_addr_t top = min(end, r->base + r->size); + + while (base < top) { + phys_addr_t tbase, ret; + int tnid; + + tbase = memblock_nid_range_rev(base, top, &tnid); + if (nid == MAX_NUMNODES || tnid == nid) { + ret = memblock_find_region(tbase, top, size, align); + if (ret) + return ret; + } + top = tbase; } - end = this_start; } + return 0; } phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int nid) { - struct memblock_type *mem = &memblock.memory; - int i; - - BUG_ON(0 == size); + phys_addr_t found; - /* We align the size to limit fragmentation. Without this, a lot of + /* + * We align the size to limit fragmentation. Without this, a lot of * small allocs quickly eat up the whole reserve array on sparc */ size = round_up(size, align); - for (i = mem->cnt - 1; i >= 0; i--) { - phys_addr_t ret = memblock_alloc_nid_region(&mem->regions[i], - size, align, nid); - if (ret) - return ret; - } + found = memblock_find_in_range_node(0, MEMBLOCK_ALLOC_ACCESSIBLE, + size, align, nid); + if (found && !memblock_add_region(&memblock.reserved, found, size)) + return found; return 0; } -- GitLab From eb40c4c27f1722f058e4713ccfedebac577d5190 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 10:46:35 +0200 Subject: [PATCH 0018/3089] memblock, x86: Replace memblock_x86_find_in_range_node() with generic memblock calls With the previous changes, generic NUMA aware memblock API has feature parity with memblock_x86_find_in_range_node(). There currently are two users - x86 setup_node_data() and __alloc_memory_core_early() in nobootmem.c. This patch converts the former to use memblock_alloc_nid() and the latter memblock_find_range_in_node(), and kills memblock_x86_find_in_range_node() and related functions including find_memory_early_core_early() in page_alloc.c. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310460395-30913-9-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 1 - arch/x86/mm/memblock.c | 15 -------- arch/x86/mm/numa.c | 9 +---- include/linux/mm.h | 2 - mm/nobootmem.c | 3 +- mm/page_alloc.c | 67 --------------------------------- 6 files changed, 2 insertions(+), 95 deletions(-) diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index 0cd3800f33b9d..161792ec524f7 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -15,7 +15,6 @@ int get_free_all_memory_range(struct range **rangep, int nodeid); void memblock_x86_register_active_regions(int nid, unsigned long start_pfn, unsigned long last_pfn); u64 memblock_x86_hole_size(u64 start, u64 end); -u64 memblock_x86_find_in_range_node(int nid, u64 start, u64 end, u64 size, u64 align); u64 memblock_x86_free_memory_in_range(u64 addr, u64 limit); u64 memblock_x86_memory_in_range(u64 addr, u64 limit); bool memblock_x86_check_reserved_size(u64 *addrp, u64 *sizep, u64 align); diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index da0d5c84586ef..e4569f85b390c 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -251,21 +251,6 @@ void __init memblock_x86_free_range(u64 start, u64 end) memblock_free(start, end - start); } -/* - * Need to call this function after memblock_x86_register_active_regions, - * so early_node_map[] is filled already. - */ -u64 __init memblock_x86_find_in_range_node(int nid, u64 start, u64 end, u64 size, u64 align) -{ - u64 addr; - addr = find_memory_core_early(nid, size, align, start, end); - if (addr) - return addr; - - /* Fallback, should already have start end within node range */ - return memblock_find_in_range(start, end, size, align); -} - /* * Finds an active region in the address range from start_pfn to last_pfn and * returns its range in ei_startpfn and ei_endpfn for the memblock entry. diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c index fa1015de5cc0e..824efadc57415 100644 --- a/arch/x86/mm/numa.c +++ b/arch/x86/mm/numa.c @@ -192,8 +192,6 @@ int __init numa_add_memblk(int nid, u64 start, u64 end) /* Initialize NODE_DATA for a node on the local memory */ static void __init setup_node_data(int nid, u64 start, u64 end) { - const u64 nd_low = PFN_PHYS(MAX_DMA_PFN); - const u64 nd_high = PFN_PHYS(max_pfn_mapped); const size_t nd_size = roundup(sizeof(pg_data_t), PAGE_SIZE); bool remapped = false; u64 nd_pa; @@ -224,17 +222,12 @@ static void __init setup_node_data(int nid, u64 start, u64 end) nd_pa = __pa(nd); remapped = true; } else { - nd_pa = memblock_x86_find_in_range_node(nid, nd_low, nd_high, - nd_size, SMP_CACHE_BYTES); - if (!nd_pa) - nd_pa = memblock_find_in_range(nd_low, nd_high, - nd_size, SMP_CACHE_BYTES); + nd_pa = memblock_alloc_nid(nd_size, SMP_CACHE_BYTES, nid); if (!nd_pa) { pr_err("Cannot find %zu bytes in node %d\n", nd_size, nid); return; } - memblock_x86_reserve_range(nd_pa, nd_pa + nd_size, "NODE_DATA"); nd = __va(nd_pa); } diff --git a/include/linux/mm.h b/include/linux/mm.h index 57e4c9ffdff81..9ebc65ae68633 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1325,8 +1325,6 @@ extern void free_bootmem_with_active_regions(int nid, unsigned long max_low_pfn); int add_from_early_node_map(struct range *range, int az, int nr_range, int nid); -u64 __init find_memory_core_early(int nid, u64 size, u64 align, - u64 goal, u64 limit); extern void sparse_memory_present_with_active_regions(int nid); extern void __next_mem_pfn_range(int *idx, int nid, diff --git a/mm/nobootmem.c b/mm/nobootmem.c index 5b0eb06ecb4e4..c78162668bc45 100644 --- a/mm/nobootmem.c +++ b/mm/nobootmem.c @@ -41,8 +41,7 @@ static void * __init __alloc_memory_core_early(int nid, u64 size, u64 align, if (limit > memblock.current_limit) limit = memblock.current_limit; - addr = find_memory_core_early(nid, size, align, goal, limit); - + addr = memblock_find_in_range_node(goal, limit, size, align, nid); if (!addr) return NULL; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 902f03a4fd6ba..8ab5e5e7fdad9 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3779,73 +3779,6 @@ void __init free_bootmem_with_active_regions(int nid, unsigned long max_low_pfn) } } -#ifdef CONFIG_HAVE_MEMBLOCK -/* - * Basic iterator support. Return the last range of PFNs for a node - * Note: nid == MAX_NUMNODES returns last region regardless of node - */ -static int __meminit last_active_region_index_in_nid(int nid) -{ - int i; - - for (i = nr_nodemap_entries - 1; i >= 0; i--) - if (nid == MAX_NUMNODES || early_node_map[i].nid == nid) - return i; - - return -1; -} - -/* - * Basic iterator support. Return the previous active range of PFNs for a node - * Note: nid == MAX_NUMNODES returns next region regardless of node - */ -static int __meminit previous_active_region_index_in_nid(int index, int nid) -{ - for (index = index - 1; index >= 0; index--) - if (nid == MAX_NUMNODES || early_node_map[index].nid == nid) - return index; - - return -1; -} - -#define for_each_active_range_index_in_nid_reverse(i, nid) \ - for (i = last_active_region_index_in_nid(nid); i != -1; \ - i = previous_active_region_index_in_nid(i, nid)) - -u64 __init find_memory_core_early(int nid, u64 size, u64 align, - u64 goal, u64 limit) -{ - int i; - - /* Need to go over early_node_map to find out good range for node */ - for_each_active_range_index_in_nid_reverse(i, nid) { - u64 addr; - u64 ei_start, ei_last; - u64 final_start, final_end; - - ei_last = early_node_map[i].end_pfn; - ei_last <<= PAGE_SHIFT; - ei_start = early_node_map[i].start_pfn; - ei_start <<= PAGE_SHIFT; - - final_start = max(ei_start, goal); - final_end = min(ei_last, limit); - - if (final_start >= final_end) - continue; - - addr = memblock_find_in_range(final_start, final_end, size, align); - - if (!addr) - continue; - - return addr; - } - - return 0; -} -#endif - int __init add_from_early_node_map(struct range *range, int az, int nr_range, int nid) { -- GitLab From ed7b56a799cade11f458cd83e1150af54a66b7e8 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 11:15:54 +0200 Subject: [PATCH 0019/3089] memblock: Remove memblock_memory_can_coalesce() Arch could implement memblock_memor_can_coalesce() to veto merging of adjacent or overlapping memblock regions; however, no arch did and any vetoing would trigger WARN_ON(). Memblock regions are supposed to deal with proper memory anyway. Remove the unused hook. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310462166-31469-2-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- include/linux/memblock.h | 4 ---- mm/memblock.c | 29 ----------------------------- 2 files changed, 33 deletions(-) diff --git a/include/linux/memblock.h b/include/linux/memblock.h index 7400d029df487..aa5df9e11fff9 100644 --- a/include/linux/memblock.h +++ b/include/linux/memblock.h @@ -92,10 +92,6 @@ extern int memblock_is_region_reserved(phys_addr_t base, phys_addr_t size); extern void memblock_dump_all(void); -/* Provided by the architecture */ -extern int memblock_memory_can_coalesce(phys_addr_t addr1, phys_addr_t size1, - phys_addr_t addr2, phys_addr_t size2); - /** * memblock_set_current_limit - Set the current allocation limit to allow * limiting allocations to what is currently diff --git a/mm/memblock.c b/mm/memblock.c index a8edb422795bc..bd3a3a9591d44 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -251,12 +251,6 @@ static int __init_memblock memblock_double_array(struct memblock_type *type) return 0; } -extern int __init_memblock __weak memblock_memory_can_coalesce(phys_addr_t addr1, phys_addr_t size1, - phys_addr_t addr2, phys_addr_t size2) -{ - return 1; -} - static long __init_memblock memblock_add_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size) { @@ -282,17 +276,6 @@ static long __init_memblock memblock_add_region(struct memblock_type *type, * of a block. */ if (base < rgn->base && end >= rgn->base) { - /* If we can't coalesce, create a new block */ - if (!memblock_memory_can_coalesce(base, size, - rgn->base, - rgn->size)) { - /* Overlap & can't coalesce are mutually - * exclusive, if you do that, be prepared - * for trouble - */ - WARN_ON(end != rgn->base); - goto new_block; - } /* We extend the bottom of the block down to our * base */ @@ -316,17 +299,6 @@ static long __init_memblock memblock_add_region(struct memblock_type *type, * top of a block */ if (base <= rend && end >= rend) { - /* If we can't coalesce, create a new block */ - if (!memblock_memory_can_coalesce(rgn->base, - rgn->size, - base, size)) { - /* Overlap & can't coalesce are mutually - * exclusive, if you do that, be prepared - * for trouble - */ - WARN_ON(rend != base); - goto new_block; - } /* We adjust our base down to enclose the * original block and destroy it. It will be * part of our new allocation. Since we've @@ -349,7 +321,6 @@ static long __init_memblock memblock_add_region(struct memblock_type *type, return 0; } - new_block: /* If we are out of space, we fail. It's too late to resize the array * but then this shouldn't have happened in the first place. */ -- GitLab From 784656f9c680d334e7b4cdb6951c5c913e5a26bf Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 11:15:55 +0200 Subject: [PATCH 0020/3089] memblock: Reimplement memblock_add_region() memblock_add_region() carefully checked for merge and overlap conditions while adding a new region, which is complicated and makes it difficult to allow arbitrary overlaps or add more merge conditions (e.g. node ID). This re-implements memblock_add_region() such that insertion is done in two steps - all non-overlapping portions of new area are inserted as separate regions first and then memblock_merge_regions() scan and merge all neighbouring compatible regions. This makes addition logic simpler and more versatile and enables adding node information to memblock. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310462166-31469-3-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- mm/memblock.c | 195 ++++++++++++++++++++++++++++---------------------- 1 file changed, 110 insertions(+), 85 deletions(-) diff --git a/mm/memblock.c b/mm/memblock.c index bd3a3a9591d44..992aa1807473e 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -251,117 +251,142 @@ static int __init_memblock memblock_double_array(struct memblock_type *type) return 0; } -static long __init_memblock memblock_add_region(struct memblock_type *type, - phys_addr_t base, phys_addr_t size) +/** + * memblock_merge_regions - merge neighboring compatible regions + * @type: memblock type to scan + * + * Scan @type and merge neighboring compatible regions. + */ +static void __init_memblock memblock_merge_regions(struct memblock_type *type) { - phys_addr_t end = base + size; - int i, slot = -1; + int i = 0; - /* First try and coalesce this MEMBLOCK with others */ - for (i = 0; i < type->cnt; i++) { - struct memblock_region *rgn = &type->regions[i]; - phys_addr_t rend = rgn->base + rgn->size; + /* cnt never goes below 1 */ + while (i < type->cnt - 1) { + struct memblock_region *this = &type->regions[i]; + struct memblock_region *next = &type->regions[i + 1]; - /* Exit if there's no possible hits */ - if (rgn->base > end || rgn->size == 0) - break; - - /* Check if we are fully enclosed within an existing - * block - */ - if (rgn->base <= base && rend >= end) - return 0; + if (this->base + this->size != next->base) { + BUG_ON(this->base + this->size > next->base); + i++; + continue; + } - /* Check if we overlap or are adjacent with the bottom - * of a block. - */ - if (base < rgn->base && end >= rgn->base) { - /* We extend the bottom of the block down to our - * base - */ - rgn->base = base; - rgn->size = rend - base; + this->size += next->size; + memmove(next, next + 1, (type->cnt - (i + 1)) * sizeof(*next)); + type->cnt--; + } +} - /* Return if we have nothing else to allocate - * (fully coalesced) - */ - if (rend >= end) - return 0; +/** + * memblock_insert_region - insert new memblock region + * @type: memblock type to insert into + * @idx: index for the insertion point + * @base: base address of the new region + * @size: size of the new region + * + * Insert new memblock region [@base,@base+@size) into @type at @idx. + * @type must already have extra room to accomodate the new region. + */ +static void __init_memblock memblock_insert_region(struct memblock_type *type, + int idx, phys_addr_t base, + phys_addr_t size) +{ + struct memblock_region *rgn = &type->regions[idx]; - /* We continue processing from the end of the - * coalesced block. - */ - base = rend; - size = end - base; - } + BUG_ON(type->cnt >= type->max); + memmove(rgn + 1, rgn, (type->cnt - idx) * sizeof(*rgn)); + rgn->base = base; + rgn->size = size; + type->cnt++; +} - /* Now check if we overlap or are adjacent with the - * top of a block - */ - if (base <= rend && end >= rend) { - /* We adjust our base down to enclose the - * original block and destroy it. It will be - * part of our new allocation. Since we've - * freed an entry, we know we won't fail - * to allocate one later, so we won't risk - * losing the original block allocation. - */ - size += (base - rgn->base); - base = rgn->base; - memblock_remove_region(type, i--); - } - } +/** + * memblock_add_region - add new memblock region + * @type: memblock type to add new region into + * @base: base address of the new region + * @size: size of the new region + * + * Add new memblock region [@base,@base+@size) into @type. The new region + * is allowed to overlap with existing ones - overlaps don't affect already + * existing regions. @type is guaranteed to be minimal (all neighbouring + * compatible regions are merged) after the addition. + * + * RETURNS: + * 0 on success, -errno on failure. + */ +static long __init_memblock memblock_add_region(struct memblock_type *type, + phys_addr_t base, phys_addr_t size) +{ + bool insert = false; + phys_addr_t obase = base, end = base + size; + int i, nr_new; - /* If the array is empty, special case, replace the fake - * filler region and return - */ - if ((type->cnt == 1) && (type->regions[0].size == 0)) { + /* special case for empty array */ + if (type->regions[0].size == 0) { + WARN_ON(type->cnt != 1); type->regions[0].base = base; type->regions[0].size = size; return 0; } - - /* If we are out of space, we fail. It's too late to resize the array - * but then this shouldn't have happened in the first place. +repeat: + /* + * The following is executed twice. Once with %false @insert and + * then with %true. The first counts the number of regions needed + * to accomodate the new area. The second actually inserts them. */ - if (WARN_ON(type->cnt >= type->max)) - return -1; + base = obase; + nr_new = 0; + + for (i = 0; i < type->cnt; i++) { + struct memblock_region *rgn = &type->regions[i]; + phys_addr_t rbase = rgn->base; + phys_addr_t rend = rbase + rgn->size; - /* Couldn't coalesce the MEMBLOCK, so add it to the sorted table. */ - for (i = type->cnt - 1; i >= 0; i--) { - if (base < type->regions[i].base) { - type->regions[i+1].base = type->regions[i].base; - type->regions[i+1].size = type->regions[i].size; - } else { - type->regions[i+1].base = base; - type->regions[i+1].size = size; - slot = i + 1; + if (rbase >= end) break; + if (rend <= base) + continue; + /* + * @rgn overlaps. If it separates the lower part of new + * area, insert that portion. + */ + if (rbase > base) { + nr_new++; + if (insert) + memblock_insert_region(type, i++, base, + rbase - base); } + /* area below @rend is dealt with, forget about it */ + base = min(rend, end); } - if (base < type->regions[0].base) { - type->regions[0].base = base; - type->regions[0].size = size; - slot = 0; + + /* insert the remaining portion */ + if (base < end) { + nr_new++; + if (insert) + memblock_insert_region(type, i, base, end - base); } - type->cnt++; - /* The array is full ? Try to resize it. If that fails, we undo - * our allocation and return an error + /* + * If this was the first round, resize array and repeat for actual + * insertions; otherwise, merge and return. */ - if (type->cnt == type->max && memblock_double_array(type)) { - BUG_ON(slot < 0); - memblock_remove_region(type, slot); - return -1; + if (!insert) { + while (type->cnt + nr_new > type->max) + if (memblock_double_array(type) < 0) + return -ENOMEM; + insert = true; + goto repeat; + } else { + memblock_merge_regions(type); + return 0; } - - return 0; } long __init_memblock memblock_add(phys_addr_t base, phys_addr_t size) { return memblock_add_region(&memblock.memory, base, size); - } static long __init_memblock __memblock_remove(struct memblock_type *type, -- GitLab From 67e24bcb725cabd15ef577bf301275d03d6086d7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 14 Jul 2011 11:42:03 +0200 Subject: [PATCH 0021/3089] memblock: Use __meminit[data] instead of __init[data] From 19ab281ed67b87a6623d725237a7333ca79f1e75 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 14 Jul 2011 11:22:16 +0200 memblock will be extended to include early_node_map[], which is also used during memory hotplug. Make memblock use __meminit[data] instead of __init[data] so that memory hotplug code can safely reference it. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/20110714094203.GE3455@htj.dyndns.org Reported-by: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- include/linux/memblock.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/memblock.h b/include/linux/memblock.h index aa5df9e11fff9..434b958a4f5fb 100644 --- a/include/linux/memblock.h +++ b/include/linux/memblock.h @@ -152,8 +152,8 @@ static inline unsigned long memblock_region_reserved_end_pfn(const struct memblo #ifdef ARCH_DISCARD_MEMBLOCK -#define __init_memblock __init -#define __initdata_memblock __initdata +#define __init_memblock __meminit +#define __initdata_memblock __meminitdata #else #define __init_memblock #define __initdata_memblock -- GitLab From 7c0caeb866b0f648d91bb75b8bc6f86af95bb033 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 14 Jul 2011 11:43:42 +0200 Subject: [PATCH 0022/3089] memblock: Add optional region->nid From 83103b92f3234ec830852bbc5c45911bd6cbdb20 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 14 Jul 2011 11:22:16 +0200 Add optional region->nid which can be enabled by arch using CONFIG_HAVE_MEMBLOCK_NODE_MAP. When enabled, memblock also carries NUMA node information and replaces early_node_map[]. Newly added memblocks have MAX_NUMNODES as nid. Arch can then call memblock_set_node() to set node information. memblock takes care of merging and node affine allocations w.r.t. node information. When MEMBLOCK_NODE_MAP is enabled, early_node_map[], related data structures and functions to manipulate and iterate it are disabled. memblock version of __next_mem_pfn_range() is provided such that for_each_mem_pfn_range() behaves the same and its users don't have to be updated. -v2: Yinghai spotted section mismatch caused by missing __init_memblock in memblock_set_node(). Fixed. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/20110714094342.GF3455@htj.dyndns.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- include/linux/memblock.h | 26 +++++++ include/linux/mm.h | 2 + mm/Kconfig | 3 + mm/memblock.c | 142 ++++++++++++++++++++++++++++++++++----- mm/page_alloc.c | 47 +++++++------ 5 files changed, 183 insertions(+), 37 deletions(-) diff --git a/include/linux/memblock.h b/include/linux/memblock.h index 434b958a4f5fb..c36a55d3c1c2d 100644 --- a/include/linux/memblock.h +++ b/include/linux/memblock.h @@ -24,6 +24,9 @@ struct memblock_region { phys_addr_t base; phys_addr_t size; +#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP + int nid; +#endif }; struct memblock_type { @@ -58,6 +61,29 @@ extern long memblock_remove(phys_addr_t base, phys_addr_t size); extern long memblock_free(phys_addr_t base, phys_addr_t size); extern long memblock_reserve(phys_addr_t base, phys_addr_t size); +#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP +extern int memblock_set_node(phys_addr_t base, phys_addr_t size, int nid); + +static inline void memblock_set_region_node(struct memblock_region *r, int nid) +{ + r->nid = nid; +} + +static inline int memblock_get_region_node(const struct memblock_region *r) +{ + return r->nid; +} +#else +static inline void memblock_set_region_node(struct memblock_region *r, int nid) +{ +} + +static inline int memblock_get_region_node(const struct memblock_region *r) +{ + return 0; +} +#endif /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */ + /* The numa aware allocator is only available if * CONFIG_ARCH_POPULATES_NODE_MAP is set */ diff --git a/include/linux/mm.h b/include/linux/mm.h index 9ebc65ae68633..ceb1e4a1a7362 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1307,12 +1307,14 @@ extern void free_area_init_node(int nid, unsigned long * zones_size, * CONFIG_ARCH_POPULATES_NODE_MAP */ extern void free_area_init_nodes(unsigned long *max_zone_pfn); +#ifndef CONFIG_HAVE_MEMBLOCK_NODE_MAP extern void add_active_range(unsigned int nid, unsigned long start_pfn, unsigned long end_pfn); extern void remove_active_range(unsigned int nid, unsigned long start_pfn, unsigned long end_pfn); extern void remove_all_active_ranges(void); void sort_node_map(void); +#endif unsigned long node_map_pfn_alignment(void); unsigned long __absent_pages_in_range(int nid, unsigned long start_pfn, unsigned long end_pfn); diff --git a/mm/Kconfig b/mm/Kconfig index 8ca47a5ee9c85..30a5d4792b833 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -131,6 +131,9 @@ config SPARSEMEM_VMEMMAP config HAVE_MEMBLOCK boolean +config HAVE_MEMBLOCK_NODE_MAP + boolean + # eventually, we can have this option just 'select SPARSEMEM' config MEMORY_HOTPLUG bool "Allow for memory hot-add" diff --git a/mm/memblock.c b/mm/memblock.c index 992aa1807473e..e815f4b758099 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -161,12 +161,8 @@ int __init_memblock memblock_reserve_reserved_regions(void) static void __init_memblock memblock_remove_region(struct memblock_type *type, unsigned long r) { - unsigned long i; - - for (i = r; i < type->cnt - 1; i++) { - type->regions[i].base = type->regions[i + 1].base; - type->regions[i].size = type->regions[i + 1].size; - } + memmove(&type->regions[r], &type->regions[r + 1], + (type->cnt - (r + 1)) * sizeof(type->regions[r])); type->cnt--; /* Special case for empty arrays */ @@ -174,6 +170,7 @@ static void __init_memblock memblock_remove_region(struct memblock_type *type, u type->cnt = 1; type->regions[0].base = 0; type->regions[0].size = 0; + memblock_set_region_node(&type->regions[0], MAX_NUMNODES); } } @@ -266,7 +263,9 @@ static void __init_memblock memblock_merge_regions(struct memblock_type *type) struct memblock_region *this = &type->regions[i]; struct memblock_region *next = &type->regions[i + 1]; - if (this->base + this->size != next->base) { + if (this->base + this->size != next->base || + memblock_get_region_node(this) != + memblock_get_region_node(next)) { BUG_ON(this->base + this->size > next->base); i++; continue; @@ -290,7 +289,7 @@ static void __init_memblock memblock_merge_regions(struct memblock_type *type) */ static void __init_memblock memblock_insert_region(struct memblock_type *type, int idx, phys_addr_t base, - phys_addr_t size) + phys_addr_t size, int nid) { struct memblock_region *rgn = &type->regions[idx]; @@ -298,6 +297,7 @@ static void __init_memblock memblock_insert_region(struct memblock_type *type, memmove(rgn + 1, rgn, (type->cnt - idx) * sizeof(*rgn)); rgn->base = base; rgn->size = size; + memblock_set_region_node(rgn, nid); type->cnt++; } @@ -327,6 +327,7 @@ static long __init_memblock memblock_add_region(struct memblock_type *type, WARN_ON(type->cnt != 1); type->regions[0].base = base; type->regions[0].size = size; + memblock_set_region_node(&type->regions[0], MAX_NUMNODES); return 0; } repeat: @@ -355,7 +356,7 @@ repeat: nr_new++; if (insert) memblock_insert_region(type, i++, base, - rbase - base); + rbase - base, MAX_NUMNODES); } /* area below @rend is dealt with, forget about it */ base = min(rend, end); @@ -365,7 +366,8 @@ repeat: if (base < end) { nr_new++; if (insert) - memblock_insert_region(type, i, base, end - base); + memblock_insert_region(type, i, base, end - base, + MAX_NUMNODES); } /* @@ -459,6 +461,101 @@ long __init_memblock memblock_reserve(phys_addr_t base, phys_addr_t size) return memblock_add_region(_rgn, base, size); } +#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP +/* + * Common iterator interface used to define for_each_mem_range(). + */ +void __init_memblock __next_mem_pfn_range(int *idx, int nid, + unsigned long *out_start_pfn, + unsigned long *out_end_pfn, int *out_nid) +{ + struct memblock_type *type = &memblock.memory; + struct memblock_region *r; + + while (++*idx < type->cnt) { + r = &type->regions[*idx]; + + if (PFN_UP(r->base) >= PFN_DOWN(r->base + r->size)) + continue; + if (nid == MAX_NUMNODES || nid == r->nid) + break; + } + if (*idx >= type->cnt) { + *idx = -1; + return; + } + + if (out_start_pfn) + *out_start_pfn = PFN_UP(r->base); + if (out_end_pfn) + *out_end_pfn = PFN_DOWN(r->base + r->size); + if (out_nid) + *out_nid = r->nid; +} + +/** + * memblock_set_node - set node ID on memblock regions + * @base: base of area to set node ID for + * @size: size of area to set node ID for + * @nid: node ID to set + * + * Set the nid of memblock memory regions in [@base,@base+@size) to @nid. + * Regions which cross the area boundaries are split as necessary. + * + * RETURNS: + * 0 on success, -errno on failure. + */ +int __init_memblock memblock_set_node(phys_addr_t base, phys_addr_t size, + int nid) +{ + struct memblock_type *type = &memblock.memory; + phys_addr_t end = base + size; + int i; + + /* we'll create at most two more regions */ + while (type->cnt + 2 > type->max) + if (memblock_double_array(type) < 0) + return -ENOMEM; + + for (i = 0; i < type->cnt; i++) { + struct memblock_region *rgn = &type->regions[i]; + phys_addr_t rbase = rgn->base; + phys_addr_t rend = rbase + rgn->size; + + if (rbase >= end) + break; + if (rend <= base) + continue; + + if (rbase < base) { + /* + * @rgn intersects from below. Split and continue + * to process the next region - the new top half. + */ + rgn->base = base; + rgn->size = rend - rgn->base; + memblock_insert_region(type, i, rbase, base - rbase, + rgn->nid); + } else if (rend > end) { + /* + * @rgn intersects from above. Split and redo the + * current region - the new bottom half. + */ + rgn->base = end; + rgn->size = rend - rgn->base; + memblock_insert_region(type, i--, rbase, end - rbase, + rgn->nid); + } else { + /* @rgn is fully contained, set ->nid */ + rgn->nid = nid; + } + } + + memblock_merge_regions(type); + return 0; +} +#endif /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */ + phys_addr_t __init __memblock_alloc_base(phys_addr_t size, phys_addr_t align, phys_addr_t max_addr) { phys_addr_t found; @@ -689,19 +786,26 @@ void __init_memblock memblock_set_current_limit(phys_addr_t limit) memblock.current_limit = limit; } -static void __init_memblock memblock_dump(struct memblock_type *region, char *name) +static void __init_memblock memblock_dump(struct memblock_type *type, char *name) { unsigned long long base, size; int i; - pr_info(" %s.cnt = 0x%lx\n", name, region->cnt); - - for (i = 0; i < region->cnt; i++) { - base = region->regions[i].base; - size = region->regions[i].size; + pr_info(" %s.cnt = 0x%lx\n", name, type->cnt); - pr_info(" %s[%#x]\t[%#016llx-%#016llx], %#llx bytes\n", - name, i, base, base + size - 1, size); + for (i = 0; i < type->cnt; i++) { + struct memblock_region *rgn = &type->regions[i]; + char nid_buf[32] = ""; + + base = rgn->base; + size = rgn->size; +#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP + if (memblock_get_region_node(rgn) != MAX_NUMNODES) + snprintf(nid_buf, sizeof(nid_buf), " on node %d", + memblock_get_region_node(rgn)); +#endif + pr_info(" %s[%#x]\t[%#016llx-%#016llx], %#llx bytes%s\n", + name, i, base, base + size - 1, size, nid_buf); } } @@ -759,11 +863,13 @@ void __init memblock_init(void) */ memblock.memory.regions[0].base = 0; memblock.memory.regions[0].size = 0; + memblock_set_region_node(&memblock.memory.regions[0], MAX_NUMNODES); memblock.memory.cnt = 1; /* Ditto. */ memblock.reserved.regions[0].base = 0; memblock.reserved.regions[0].size = 0; + memblock_set_region_node(&memblock.reserved.regions[0], MAX_NUMNODES); memblock.reserved.cnt = 1; memblock.current_limit = MEMBLOCK_ALLOC_ANYWHERE; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 8ab5e5e7fdad9..3c7ea45ffba96 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -182,28 +182,31 @@ static unsigned long __meminitdata nr_all_pages; static unsigned long __meminitdata dma_reserve; #ifdef CONFIG_ARCH_POPULATES_NODE_MAP - /* - * MAX_ACTIVE_REGIONS determines the maximum number of distinct - * ranges of memory (RAM) that may be registered with add_active_range(). - * Ranges passed to add_active_range() will be merged if possible - * so the number of times add_active_range() can be called is - * related to the number of nodes and the number of holes - */ - #ifdef CONFIG_MAX_ACTIVE_REGIONS - /* Allow an architecture to set MAX_ACTIVE_REGIONS to save memory */ - #define MAX_ACTIVE_REGIONS CONFIG_MAX_ACTIVE_REGIONS - #else - #if MAX_NUMNODES >= 32 - /* If there can be many nodes, allow up to 50 holes per node */ - #define MAX_ACTIVE_REGIONS (MAX_NUMNODES*50) + #ifndef CONFIG_HAVE_MEMBLOCK_NODE_MAP + /* + * MAX_ACTIVE_REGIONS determines the maximum number of distinct ranges + * of memory (RAM) that may be registered with add_active_range(). + * Ranges passed to add_active_range() will be merged if possible so + * the number of times add_active_range() can be called is related to + * the number of nodes and the number of holes + */ + #ifdef CONFIG_MAX_ACTIVE_REGIONS + /* Allow an architecture to set MAX_ACTIVE_REGIONS to save memory */ + #define MAX_ACTIVE_REGIONS CONFIG_MAX_ACTIVE_REGIONS #else - /* By default, allow up to 256 distinct regions */ - #define MAX_ACTIVE_REGIONS 256 + #if MAX_NUMNODES >= 32 + /* If there can be many nodes, allow up to 50 holes per node */ + #define MAX_ACTIVE_REGIONS (MAX_NUMNODES*50) + #else + /* By default, allow up to 256 distinct regions */ + #define MAX_ACTIVE_REGIONS 256 + #endif #endif - #endif - static struct node_active_region __meminitdata early_node_map[MAX_ACTIVE_REGIONS]; - static int __meminitdata nr_nodemap_entries; + static struct node_active_region __meminitdata early_node_map[MAX_ACTIVE_REGIONS]; + static int __meminitdata nr_nodemap_entries; +#endif /* !CONFIG_HAVE_MEMBLOCK_NODE_MAP */ + static unsigned long __meminitdata arch_zone_lowest_possible_pfn[MAX_NR_ZONES]; static unsigned long __meminitdata arch_zone_highest_possible_pfn[MAX_NR_ZONES]; static unsigned long __initdata required_kernelcore; @@ -4268,6 +4271,7 @@ static inline void setup_nr_node_ids(void) } #endif +#ifndef CONFIG_HAVE_MEMBLOCK_NODE_MAP /* * Common iterator interface used to define for_each_mem_pfn_range(). */ @@ -4456,6 +4460,11 @@ void __init sort_node_map(void) sizeof(struct node_active_region), cmp_node_active_region, NULL); } +#else /* !CONFIG_HAVE_MEMBLOCK_NODE_MAP */ +static inline void sort_node_map(void) +{ +} +#endif /** * node_map_pfn_alignment - determine the maximum internode alignment -- GitLab From 0608f70c78a384c2f225f2de226ca057a196f108 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 14 Jul 2011 11:44:23 +0200 Subject: [PATCH 0023/3089] x86: Use HAVE_MEMBLOCK_NODE_MAP From 5732e1247898d67cbf837585150fe9f68974671d Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 14 Jul 2011 11:22:16 +0200 Convert x86 to HAVE_MEMBLOCK_NODE_MAP. The only difference in memory handling is that allocations can't no longer cross node boundaries whether they're node affine or not, which shouldn't matter at all. This conversion will enable further simplification of boot memory handling. -v2: Fix build failure on !NUMA configurations discovered by hpa. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/20110714094423.GG3455@htj.dyndns.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: H. Peter Anvin --- arch/x86/Kconfig | 1 + arch/x86/include/asm/memblock.h | 2 -- arch/x86/mm/init_32.c | 8 ++++---- arch/x86/mm/init_64.c | 2 +- arch/x86/mm/memblock.c | 14 -------------- arch/x86/mm/numa.c | 15 +++++++-------- 6 files changed, 13 insertions(+), 29 deletions(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index da349723d4115..97f08941dd792 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -25,6 +25,7 @@ config X86 select HAVE_IOREMAP_PROT select HAVE_KPROBES select HAVE_MEMBLOCK + select HAVE_MEMBLOCK_NODE_MAP select ARCH_WANT_OPTIONAL_GPIOLIB select ARCH_WANT_FRAME_POINTERS select HAVE_DMA_ATTRS diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index 161792ec524f7..1460db219a8b8 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -12,8 +12,6 @@ int __get_free_all_memory_range(struct range **range, int nodeid, unsigned long start_pfn, unsigned long end_pfn); int get_free_all_memory_range(struct range **rangep, int nodeid); -void memblock_x86_register_active_regions(int nid, unsigned long start_pfn, - unsigned long last_pfn); u64 memblock_x86_hole_size(u64 start, u64 end); u64 memblock_x86_free_memory_in_range(u64 addr, u64 limit); u64 memblock_x86_memory_in_range(u64 addr, u64 limit); diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 29f7c6d981790..5d173db93c4ed 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -650,18 +650,18 @@ void __init initmem_init(void) highstart_pfn = highend_pfn = max_pfn; if (max_pfn > max_low_pfn) highstart_pfn = max_low_pfn; - memblock_x86_register_active_regions(0, 0, highend_pfn); - sparse_memory_present_with_active_regions(0); printk(KERN_NOTICE "%ldMB HIGHMEM available.\n", pages_to_mb(highend_pfn - highstart_pfn)); num_physpages = highend_pfn; high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1; #else - memblock_x86_register_active_regions(0, 0, max_low_pfn); - sparse_memory_present_with_active_regions(0); num_physpages = max_low_pfn; high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1; #endif + + memblock_set_node(0, (phys_addr_t)ULLONG_MAX, 0); + sparse_memory_present_with_active_regions(0); + #ifdef CONFIG_FLATMEM max_mapnr = num_physpages; #endif diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index d865c4aeec55f..7fb064cbdcec1 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -607,7 +607,7 @@ kernel_physical_mapping_init(unsigned long start, #ifndef CONFIG_NUMA void __init initmem_init(void) { - memblock_x86_register_active_regions(0, 0, max_pfn); + memblock_set_node(0, (phys_addr_t)ULLONG_MAX, 0); } #endif diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index e4569f85b390c..97fbc39739341 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -283,20 +283,6 @@ static int __init memblock_x86_find_active_region(const struct memblock_region * return 1; } -/* Walk the memblock.memory map and register active regions within a node */ -void __init memblock_x86_register_active_regions(int nid, unsigned long start_pfn, - unsigned long last_pfn) -{ - unsigned long ei_startpfn; - unsigned long ei_endpfn; - struct memblock_region *r; - - for_each_memblock(memory, r) - if (memblock_x86_find_active_region(r, start_pfn, last_pfn, - &ei_startpfn, &ei_endpfn)) - add_active_range(nid, ei_startpfn, ei_endpfn); -} - /* * Find the hole size (in bytes) in the memory range. * @start: starting address of the memory range to scan diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c index 824efadc57415..f4a40bdb2e4e2 100644 --- a/arch/x86/mm/numa.c +++ b/arch/x86/mm/numa.c @@ -498,13 +498,10 @@ static int __init numa_register_memblks(struct numa_meminfo *mi) if (WARN_ON(nodes_empty(node_possible_map))) return -EINVAL; - for (i = 0; i < mi->nr_blks; i++) - memblock_x86_register_active_regions(mi->blk[i].nid, - mi->blk[i].start >> PAGE_SHIFT, - mi->blk[i].end >> PAGE_SHIFT); - - /* for out of order entries */ - sort_node_map(); + for (i = 0; i < mi->nr_blks; i++) { + struct numa_memblk *mb = &mi->blk[i]; + memblock_set_node(mb->start, mb->end - mb->start, mb->nid); + } /* * If sections array is gonna be used for pfn -> nid mapping, check @@ -538,6 +535,8 @@ static int __init numa_register_memblks(struct numa_meminfo *mi) setup_node_data(nid, start, end); } + /* Dump memblock with node info and return. */ + memblock_dump_all(); return 0; } @@ -575,7 +574,7 @@ static int __init numa_init(int (*init_func)(void)) nodes_clear(node_possible_map); nodes_clear(node_online_map); memset(&numa_meminfo, 0, sizeof(numa_meminfo)); - remove_all_active_ranges(); + WARN_ON(memblock_set_node(0, ULLONG_MAX, MAX_NUMNODES)); numa_reset_distance(); ret = init_func(); -- GitLab From ab5d140b9eafae402aa3e673a63c5ef6164a9dd2 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 11:15:58 +0200 Subject: [PATCH 0024/3089] x86: Use __memblock_alloc_base() in early_reserve_e820() early_reserve_e820() implements its own ad-hoc early allocator using memblock_x86_find_in_range_size(). Use __memblock_alloc_base() instead and remove the unnecessary @startt parameter (it's top-down allocation anyway). Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310462166-31469-6-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/e820.h | 2 +- arch/x86/kernel/e820.c | 30 ++++++------------------------ arch/x86/kernel/mpparse.c | 6 ++---- 3 files changed, 9 insertions(+), 29 deletions(-) diff --git a/arch/x86/include/asm/e820.h b/arch/x86/include/asm/e820.h index 908b96957d88a..37782566af241 100644 --- a/arch/x86/include/asm/e820.h +++ b/arch/x86/include/asm/e820.h @@ -117,7 +117,7 @@ static inline void early_memtest(unsigned long start, unsigned long end) extern unsigned long e820_end_of_ram_pfn(void); extern unsigned long e820_end_of_low_ram_pfn(void); -extern u64 early_reserve_e820(u64 startt, u64 sizet, u64 align); +extern u64 early_reserve_e820(u64 sizet, u64 align); void memblock_x86_fill(void); void memblock_find_dma_reserve(void); diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 0f9ff58d06d7e..b99d9402ae8bc 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -737,35 +737,17 @@ core_initcall(e820_mark_nvs_memory); /* * pre allocated 4k and reserved it in memblock and e820_saved */ -u64 __init early_reserve_e820(u64 startt, u64 sizet, u64 align) +u64 __init early_reserve_e820(u64 size, u64 align) { - u64 size = 0; u64 addr; - u64 start; - for (start = startt; ; start += size) { - start = memblock_x86_find_in_range_size(start, &size, align); - if (!start) - return 0; - if (size >= sizet) - break; + addr = __memblock_alloc_base(size, align, MEMBLOCK_ALLOC_ACCESSIBLE); + if (addr) { + e820_update_range_saved(addr, size, E820_RAM, E820_RESERVED); + printk(KERN_INFO "update e820_saved for early_reserve_e820\n"); + update_e820_saved(); } -#ifdef CONFIG_X86_32 - if (start >= MAXMEM) - return 0; - if (start + size > MAXMEM) - size = MAXMEM - start; -#endif - - addr = round_down(start + size - sizet, align); - if (addr < start) - return 0; - memblock_x86_reserve_range(addr, addr + sizet, "new next"); - e820_update_range_saved(addr, sizet, E820_RAM, E820_RESERVED); - printk(KERN_INFO "update e820_saved for early_reserve_e820\n"); - update_e820_saved(); - return addr; } diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index 9103b89c145a5..8faeaa0ed2cc0 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -836,10 +836,8 @@ early_param("alloc_mptable", parse_alloc_mptable_opt); void __init early_reserve_e820_mpc_new(void) { - if (enable_update_mptable && alloc_mptable) { - u64 startt = 0; - mpc_new_phys = early_reserve_e820(startt, mpc_new_length, 4); - } + if (enable_update_mptable && alloc_mptable) + mpc_new_phys = early_reserve_e820(mpc_new_length, 4); } static int __init update_mp_table(void) -- GitLab From 35fd0808d7d8d001cd72f112e3bca84664b596a3 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 11:15:59 +0200 Subject: [PATCH 0025/3089] memblock: Implement for_each_free_mem_range() Implement for_each_free_mem_range() which iterates over free memory areas according to memblock (memory && !reserved). This will be used to simplify memblock users. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310462166-31469-7-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Signed-off-by: H. Peter Anvin --- include/linux/memblock.h | 20 +++++++++++ mm/memblock.c | 76 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 96 insertions(+) diff --git a/include/linux/memblock.h b/include/linux/memblock.h index c36a55d3c1c2d..31def584ccebd 100644 --- a/include/linux/memblock.h +++ b/include/linux/memblock.h @@ -61,6 +61,26 @@ extern long memblock_remove(phys_addr_t base, phys_addr_t size); extern long memblock_free(phys_addr_t base, phys_addr_t size); extern long memblock_reserve(phys_addr_t base, phys_addr_t size); +extern void __next_free_mem_range(u64 *idx, int nid, phys_addr_t *out_start, + phys_addr_t *out_end, int *out_nid); + +/** + * for_each_free_mem_range - iterate through free memblock areas + * @i: u64 used as loop variable + * @nid: node selector, %MAX_NUMNODES for all nodes + * @p_start: ptr to phys_addr_t for start address of the range, can be %NULL + * @p_end: ptr to phys_addr_t for end address of the range, can be %NULL + * @p_nid: ptr to int for nid of the range, can be %NULL + * + * Walks over free (memory && !reserved) areas of memblock. Available as + * soon as memblock is initialized. + */ +#define for_each_free_mem_range(i, nid, p_start, p_end, p_nid) \ + for (i = 0, \ + __next_free_mem_range(&i, nid, p_start, p_end, p_nid); \ + i != (u64)ULLONG_MAX; \ + __next_free_mem_range(&i, nid, p_start, p_end, p_nid)) + #ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP extern int memblock_set_node(phys_addr_t base, phys_addr_t size, int nid); diff --git a/mm/memblock.c b/mm/memblock.c index e815f4b758099..c4a8750406fc9 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -461,6 +461,82 @@ long __init_memblock memblock_reserve(phys_addr_t base, phys_addr_t size) return memblock_add_region(_rgn, base, size); } +/** + * __next_free_mem_range - next function for for_each_free_mem_range() + * @idx: pointer to u64 loop variable + * @nid: nid: node selector, %MAX_NUMNODES for all nodes + * @p_start: ptr to phys_addr_t for start address of the range, can be %NULL + * @p_end: ptr to phys_addr_t for end address of the range, can be %NULL + * @p_nid: ptr to int for nid of the range, can be %NULL + * + * Find the first free area from *@idx which matches @nid, fill the out + * parameters, and update *@idx for the next iteration. The lower 32bit of + * *@idx contains index into memory region and the upper 32bit indexes the + * areas before each reserved region. For example, if reserved regions + * look like the following, + * + * 0:[0-16), 1:[32-48), 2:[128-130) + * + * The upper 32bit indexes the following regions. + * + * 0:[0-0), 1:[16-32), 2:[48-128), 3:[130-MAX) + * + * As both region arrays are sorted, the function advances the two indices + * in lockstep and returns each intersection. + */ +void __init_memblock __next_free_mem_range(u64 *idx, int nid, + phys_addr_t *out_start, + phys_addr_t *out_end, int *out_nid) +{ + struct memblock_type *mem = &memblock.memory; + struct memblock_type *rsv = &memblock.reserved; + int mi = *idx & 0xffffffff; + int ri = *idx >> 32; + + for ( ; mi < mem->cnt; mi++) { + struct memblock_region *m = &mem->regions[mi]; + phys_addr_t m_start = m->base; + phys_addr_t m_end = m->base + m->size; + + /* only memory regions are associated with nodes, check it */ + if (nid != MAX_NUMNODES && nid != memblock_get_region_node(m)) + continue; + + /* scan areas before each reservation for intersection */ + for ( ; ri < rsv->cnt + 1; ri++) { + struct memblock_region *r = &rsv->regions[ri]; + phys_addr_t r_start = ri ? r[-1].base + r[-1].size : 0; + phys_addr_t r_end = ri < rsv->cnt ? r->base : ULLONG_MAX; + + /* if ri advanced past mi, break out to advance mi */ + if (r_start >= m_end) + break; + /* if the two regions intersect, we're done */ + if (m_start < r_end) { + if (out_start) + *out_start = max(m_start, r_start); + if (out_end) + *out_end = min(m_end, r_end); + if (out_nid) + *out_nid = memblock_get_region_node(m); + /* + * The region which ends first is advanced + * for the next iteration. + */ + if (m_end <= r_end) + mi++; + else + ri++; + *idx = (u32)mi | (u64)ri << 32; + return; + } + } + } + + /* signal end of iteration */ + *idx = ULLONG_MAX; +} + #ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP /* * Common iterator interface used to define for_each_mem_range(). -- GitLab From 8d89ac808417e92a33fb5fa3c86352016643775a Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 11:16:00 +0200 Subject: [PATCH 0026/3089] x86: Replace memblock_x86_find_in_range_size() with for_each_free_mem_range() setup_bios_corruption_check() and memtest do_one_pass() open code memblock free area iteration using memblock_x86_find_in_range_size(). Convert them to use for_each_free_mem_range() instead. This leaves memblock_x86_find_in_range_size() and memblock_x86_check_reserved_size() unused. Kill them. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310462166-31469-8-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 3 -- arch/x86/kernel/check.c | 34 ++++++++---------- arch/x86/mm/memblock.c | 62 --------------------------------- arch/x86/mm/memtest.c | 31 +++++++---------- 4 files changed, 28 insertions(+), 102 deletions(-) diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index 1460db219a8b8..d2a5a59bd358c 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -3,8 +3,6 @@ #define ARCH_DISCARD_MEMBLOCK -u64 memblock_x86_find_in_range_size(u64 start, u64 *sizep, u64 align); - void memblock_x86_reserve_range(u64 start, u64 end, char *name); void memblock_x86_free_range(u64 start, u64 end); struct range; @@ -15,6 +13,5 @@ int get_free_all_memory_range(struct range **rangep, int nodeid); u64 memblock_x86_hole_size(u64 start, u64 end); u64 memblock_x86_free_memory_in_range(u64 addr, u64 limit); u64 memblock_x86_memory_in_range(u64 addr, u64 limit); -bool memblock_x86_check_reserved_size(u64 *addrp, u64 *sizep, u64 align); #endif diff --git a/arch/x86/kernel/check.c b/arch/x86/kernel/check.c index 95680fc4df5c1..621cd23bb4e75 100644 --- a/arch/x86/kernel/check.c +++ b/arch/x86/kernel/check.c @@ -62,7 +62,8 @@ early_param("memory_corruption_check_size", set_corruption_check_size); void __init setup_bios_corruption_check(void) { - u64 addr = PAGE_SIZE; /* assume first page is reserved anyway */ + phys_addr_t start, end; + u64 i; if (memory_corruption_check == -1) { memory_corruption_check = @@ -82,28 +83,23 @@ void __init setup_bios_corruption_check(void) corruption_check_size = round_up(corruption_check_size, PAGE_SIZE); - while (addr < corruption_check_size && num_scan_areas < MAX_SCAN_AREAS) { - u64 size; - addr = memblock_x86_find_in_range_size(addr, &size, PAGE_SIZE); + for_each_free_mem_range(i, MAX_NUMNODES, &start, &end, NULL) { + start = clamp_t(phys_addr_t, round_up(start, PAGE_SIZE), + PAGE_SIZE, corruption_check_size); + end = clamp_t(phys_addr_t, round_down(end, PAGE_SIZE), + PAGE_SIZE, corruption_check_size); + if (start >= end) + continue; - if (!addr) - break; - - if (addr >= corruption_check_size) - break; - - if ((addr + size) > corruption_check_size) - size = corruption_check_size - addr; - - memblock_x86_reserve_range(addr, addr + size, "SCAN RAM"); - scan_areas[num_scan_areas].addr = addr; - scan_areas[num_scan_areas].size = size; - num_scan_areas++; + memblock_x86_reserve_range(start, end, "SCAN RAM"); + scan_areas[num_scan_areas].addr = start; + scan_areas[num_scan_areas].size = end - start; /* Assume we've already mapped this early memory */ - memset(__va(addr), 0, size); + memset(__va(start), 0, end - start); - addr += size; + if (++num_scan_areas >= MAX_SCAN_AREAS) + break; } if (num_scan_areas) diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 97fbc39739341..648d47d52a862 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -7,68 +7,6 @@ #include #include -/* Check for already reserved areas */ -bool __init memblock_x86_check_reserved_size(u64 *addrp, u64 *sizep, u64 align) -{ - struct memblock_region *r; - u64 addr = *addrp, last; - u64 size = *sizep; - bool changed = false; - -again: - last = addr + size; - for_each_memblock(reserved, r) { - if (last > r->base && addr < r->base) { - size = r->base - addr; - changed = true; - goto again; - } - if (last > (r->base + r->size) && addr < (r->base + r->size)) { - addr = round_up(r->base + r->size, align); - size = last - addr; - changed = true; - goto again; - } - if (last <= (r->base + r->size) && addr >= r->base) { - *sizep = 0; - return false; - } - } - if (changed) { - *addrp = addr; - *sizep = size; - } - return changed; -} - -/* - * Find next free range after start, and size is returned in *sizep - */ -u64 __init memblock_x86_find_in_range_size(u64 start, u64 *sizep, u64 align) -{ - struct memblock_region *r; - - for_each_memblock(memory, r) { - u64 ei_start = r->base; - u64 ei_last = ei_start + r->size; - u64 addr; - - addr = round_up(ei_start, align); - if (addr < start) - addr = round_up(start, align); - if (addr >= ei_last) - continue; - *sizep = ei_last - addr; - while (memblock_x86_check_reserved_size(&addr, sizep, align)) - ; - - if (*sizep) - return addr; - } - - return 0; -} - static __init struct range *find_range_array(int count) { u64 end, size, mem; diff --git a/arch/x86/mm/memtest.c b/arch/x86/mm/memtest.c index 92faf3a1c53e2..46a5ff25eda40 100644 --- a/arch/x86/mm/memtest.c +++ b/arch/x86/mm/memtest.c @@ -70,24 +70,19 @@ static void __init memtest(u64 pattern, u64 start_phys, u64 size) static void __init do_one_pass(u64 pattern, u64 start, u64 end) { - u64 size = 0; - - while (start < end) { - start = memblock_x86_find_in_range_size(start, &size, 1); - - /* done ? */ - if (start >= end) - break; - if (start + size > end) - size = end - start; - - printk(KERN_INFO " %010llx - %010llx pattern %016llx\n", - (unsigned long long) start, - (unsigned long long) start + size, - (unsigned long long) cpu_to_be64(pattern)); - memtest(pattern, start, size); - - start += size; + u64 i; + phys_addr_t this_start, this_end; + + for_each_free_mem_range(i, MAX_NUMNODES, &this_start, &this_end, NULL) { + this_start = clamp_t(phys_addr_t, this_start, start, end); + this_end = clamp_t(phys_addr_t, this_end, start, end); + if (this_start < this_end) { + printk(KERN_INFO " %010llx - %010llx pattern %016llx\n", + (unsigned long long)this_start, + (unsigned long long)this_end, + (unsigned long long)cpu_to_be64(pattern)); + memtest(pattern, this_start, this_end - this_start); + } } } -- GitLab From 64a02daacbc880bac1d6b3aeefbcd226a9341fa7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 11:16:01 +0200 Subject: [PATCH 0027/3089] memblock, x86: Make free_all_memory_core_early() explicitly free lowmem only nomemblock is currently used only by x86 and on x86_32 free_all_memory_core_early() silently freed only the low mem because get_free_all_memory_range() in arch/x86/mm/memblock.c implicitly limited range to max_low_pfn. Rename free_all_memory_core_early() to free_low_memory_core_early() and make it call __get_free_all_memory_range() and limit the range to max_low_pfn explicitly. This makes things clearer and also is consistent with the bootmem behavior. This leaves get_free_all_memory_range() without any user. Kill it. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310462166-31469-9-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 1 - arch/x86/mm/memblock.c | 10 ---------- arch/x86/mm/numa_64.c | 2 +- include/linux/bootmem.h | 2 +- mm/nobootmem.c | 8 ++++---- 5 files changed, 6 insertions(+), 17 deletions(-) diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index d2a5a59bd358c..6c72ecaee577b 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -8,7 +8,6 @@ void memblock_x86_free_range(u64 start, u64 end); struct range; int __get_free_all_memory_range(struct range **range, int nodeid, unsigned long start_pfn, unsigned long end_pfn); -int get_free_all_memory_range(struct range **rangep, int nodeid); u64 memblock_x86_hole_size(u64 start, u64 end); u64 memblock_x86_free_memory_in_range(u64 addr, u64 limit); diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 648d47d52a862..0e8442a9baff3 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -89,16 +89,6 @@ int __init __get_free_all_memory_range(struct range **rangep, int nodeid, return nr_range; } -int __init get_free_all_memory_range(struct range **rangep, int nodeid) -{ - unsigned long end_pfn = -1UL; - -#ifdef CONFIG_X86_32 - end_pfn = max_low_pfn; -#endif - return __get_free_all_memory_range(rangep, nodeid, 0, end_pfn); -} - static u64 __init __memblock_x86_memory_in_range(u64 addr, u64 limit, bool get_free) { int i, count; diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index dd27f401f0a03..92e27119ee1a0 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -19,7 +19,7 @@ unsigned long __init numa_free_all_bootmem(void) for_each_online_node(i) pages += free_all_bootmem_node(NODE_DATA(i)); - pages += free_all_memory_core_early(MAX_NUMNODES); + pages += free_low_memory_core_early(MAX_NUMNODES); return pages; } diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h index ab344a5211057..66d3e954eb6c0 100644 --- a/include/linux/bootmem.h +++ b/include/linux/bootmem.h @@ -44,7 +44,7 @@ extern unsigned long init_bootmem_node(pg_data_t *pgdat, unsigned long endpfn); extern unsigned long init_bootmem(unsigned long addr, unsigned long memend); -unsigned long free_all_memory_core_early(int nodeid); +extern unsigned long free_low_memory_core_early(int nodeid); extern unsigned long free_all_bootmem_node(pg_data_t *pgdat); extern unsigned long free_all_bootmem(void); diff --git a/mm/nobootmem.c b/mm/nobootmem.c index c78162668bc45..2037a8a047614 100644 --- a/mm/nobootmem.c +++ b/mm/nobootmem.c @@ -106,7 +106,7 @@ static void __init __free_pages_memory(unsigned long start, unsigned long end) __free_pages_bootmem(pfn_to_page(i), 0); } -unsigned long __init free_all_memory_core_early(int nodeid) +unsigned long __init free_low_memory_core_early(int nodeid) { int i; u64 start, end; @@ -114,7 +114,7 @@ unsigned long __init free_all_memory_core_early(int nodeid) struct range *range = NULL; int nr_range; - nr_range = get_free_all_memory_range(&range, nodeid); + nr_range = __get_free_all_memory_range(&range, nodeid, 0, max_low_pfn); for (i = 0; i < nr_range; i++) { start = range[i].start; @@ -136,7 +136,7 @@ unsigned long __init free_all_bootmem_node(pg_data_t *pgdat) { register_page_bootmem_info_node(pgdat); - /* free_all_memory_core_early(MAX_NUMNODES) will be called later */ + /* free_low_memory_core_early(MAX_NUMNODES) will be called later */ return 0; } @@ -154,7 +154,7 @@ unsigned long __init free_all_bootmem(void) * Use MAX_NUMNODES will make sure all ranges in early_node_map[] * will be used instead of only Node0 related */ - return free_all_memory_core_early(MAX_NUMNODES); + return free_low_memory_core_early(MAX_NUMNODES); } /** -- GitLab From 8a9ca34c11e1695dab7aff3cfa7780fbfe76b2f8 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 11:16:02 +0200 Subject: [PATCH 0028/3089] memblock, x86: Replace __get_free_all_memory_range() with for_each_free_mem_range() __get_free_all_memory_range() walks memblock, calculates free memory areas and fills in the specified range. It can be easily replaced with for_each_free_mem_range(). Convert free_low_memory_core_early() and add_highpages_with_active_regions() to for_each_free_mem_range(). This leaves __get_free_all_memory_range() without any user. Kill it and related functions. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310462166-31469-10-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 3 -- arch/x86/mm/init_32.c | 28 ++++++---------- arch/x86/mm/memblock.c | 59 --------------------------------- mm/nobootmem.c | 28 +++++++++------- 4 files changed, 27 insertions(+), 91 deletions(-) diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index 6c72ecaee577b..bc9e44b0812d4 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -5,9 +5,6 @@ void memblock_x86_reserve_range(u64 start, u64 end, char *name); void memblock_x86_free_range(u64 start, u64 end); -struct range; -int __get_free_all_memory_range(struct range **range, int nodeid, - unsigned long start_pfn, unsigned long end_pfn); u64 memblock_x86_hole_size(u64 start, u64 end); u64 memblock_x86_free_memory_in_range(u64 addr, u64 limit); diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 5d173db93c4ed..0c1da394a6347 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -427,23 +427,17 @@ static void __init add_one_highpage_init(struct page *page) void __init add_highpages_with_active_regions(int nid, unsigned long start_pfn, unsigned long end_pfn) { - struct range *range; - int nr_range; - int i; - - nr_range = __get_free_all_memory_range(&range, nid, start_pfn, end_pfn); - - for (i = 0; i < nr_range; i++) { - struct page *page; - int node_pfn; - - for (node_pfn = range[i].start; node_pfn < range[i].end; - node_pfn++) { - if (!pfn_valid(node_pfn)) - continue; - page = pfn_to_page(node_pfn); - add_one_highpage_init(page); - } + phys_addr_t start, end; + u64 i; + + for_each_free_mem_range(i, nid, &start, &end, NULL) { + unsigned long pfn = clamp_t(unsigned long, PFN_UP(start), + start_pfn, end_pfn); + unsigned long e_pfn = clamp_t(unsigned long, PFN_DOWN(end), + start_pfn, end_pfn); + for ( ; pfn < e_pfn; pfn++) + if (pfn_valid(pfn)) + add_one_highpage_init(pfn_to_page(pfn)); } } #else diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 0e8442a9baff3..4107c1a32b789 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -30,65 +30,6 @@ static __init struct range *find_range_array(int count) return range; } -static void __init memblock_x86_subtract_reserved(struct range *range, int az) -{ - u64 final_start, final_end; - struct memblock_region *r; - - /* Take out region array itself at first*/ - memblock_free_reserved_regions(); - - memblock_dbg("Subtract (%ld early reservations)\n", memblock.reserved.cnt); - - for_each_memblock(reserved, r) { - memblock_dbg(" [%010llx-%010llx]\n", (u64)r->base, (u64)r->base + r->size - 1); - final_start = PFN_DOWN(r->base); - final_end = PFN_UP(r->base + r->size); - if (final_start >= final_end) - continue; - subtract_range(range, az, final_start, final_end); - } - - /* Put region array back ? */ - memblock_reserve_reserved_regions(); -} - -static int __init count_early_node_map(int nodeid) -{ - int i, cnt = 0; - - for_each_mem_pfn_range(i, nodeid, NULL, NULL, NULL) - cnt++; - return cnt; -} - -int __init __get_free_all_memory_range(struct range **rangep, int nodeid, - unsigned long start_pfn, unsigned long end_pfn) -{ - int count; - struct range *range; - int nr_range; - - count = (memblock.reserved.cnt + count_early_node_map(nodeid)) * 2; - - range = find_range_array(count); - nr_range = 0; - - /* - * Use early_node_map[] and memblock.reserved.region to get range array - * at first - */ - nr_range = add_from_early_node_map(range, count, nr_range, nodeid); - subtract_range(range, count, 0, start_pfn); - subtract_range(range, count, end_pfn, -1ULL); - - memblock_x86_subtract_reserved(range, count); - nr_range = clean_sort_range(range, count); - - *rangep = range; - return nr_range; -} - static u64 __init __memblock_x86_memory_in_range(u64 addr, u64 limit, bool get_free) { int i, count; diff --git a/mm/nobootmem.c b/mm/nobootmem.c index 2037a8a047614..7075bc00fa84c 100644 --- a/mm/nobootmem.c +++ b/mm/nobootmem.c @@ -108,21 +108,25 @@ static void __init __free_pages_memory(unsigned long start, unsigned long end) unsigned long __init free_low_memory_core_early(int nodeid) { - int i; - u64 start, end; unsigned long count = 0; - struct range *range = NULL; - int nr_range; - - nr_range = __get_free_all_memory_range(&range, nodeid, 0, max_low_pfn); - - for (i = 0; i < nr_range; i++) { - start = range[i].start; - end = range[i].end; - count += end - start; - __free_pages_memory(start, end); + phys_addr_t start, end; + u64 i; + + /* free reserved array temporarily so that it's treated as free area */ + memblock_free_reserved_regions(); + + for_each_free_mem_range(i, MAX_NUMNODES, &start, &end, NULL) { + unsigned long start_pfn = PFN_UP(start); + unsigned long end_pfn = min_t(unsigned long, + PFN_DOWN(end), max_low_pfn); + if (start_pfn < end_pfn) { + __free_pages_memory(start_pfn, end_pfn); + count += end_pfn - start_pfn; + } } + /* put region array back? */ + memblock_reserve_reserved_regions(); return count; } -- GitLab From 6b5d41a1b97f5529284f16170211b87fd60264c0 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 11:16:03 +0200 Subject: [PATCH 0029/3089] memblock, x86: Reimplement memblock_find_dma_reserve() using iterators memblock_find_dma_reserve() wants to find out how much memory is reserved under MAX_DMA_PFN. memblock_x86_memory_[free_]in_range() are used to find out the amounts of all available and free memory in the area, which are then subtracted to find out the amount of reservation. memblock_x86_memblock_[free_]in_range() are implemented using __memblock_x86_memory_in_range() which builds ranges from memblock and then count them, which is rather unnecessarily complex. This patch open codes the counting logic directly in memblock_find_dma_reserve() using memblock iterators and removes now unused __memblock_x86_memory_in_range() and find_range_array(). Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310462166-31469-11-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 2 - arch/x86/kernel/e820.c | 25 ++++++++-- arch/x86/mm/memblock.c | 87 --------------------------------- 3 files changed, 20 insertions(+), 94 deletions(-) diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index bc9e44b0812d4..a0cc7d66ac55b 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -7,7 +7,5 @@ void memblock_x86_reserve_range(u64 start, u64 end, char *name); void memblock_x86_free_range(u64 start, u64 end); u64 memblock_x86_hole_size(u64 start, u64 end); -u64 memblock_x86_free_memory_in_range(u64 addr, u64 limit); -u64 memblock_x86_memory_in_range(u64 addr, u64 limit); #endif diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index b99d9402ae8bc..84475f1e22012 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -1093,15 +1093,30 @@ void __init memblock_x86_fill(void) void __init memblock_find_dma_reserve(void) { #ifdef CONFIG_X86_64 - u64 free_size_pfn; - u64 mem_size_pfn; + u64 nr_pages = 0, nr_free_pages = 0; + unsigned long start_pfn, end_pfn; + phys_addr_t start, end; + int i; + u64 u; + /* * need to find out used area below MAX_DMA_PFN * need to use memblock to get free size in [0, MAX_DMA_PFN] * at first, and assume boot_mem will not take below MAX_DMA_PFN */ - mem_size_pfn = memblock_x86_memory_in_range(0, MAX_DMA_PFN << PAGE_SHIFT) >> PAGE_SHIFT; - free_size_pfn = memblock_x86_free_memory_in_range(0, MAX_DMA_PFN << PAGE_SHIFT) >> PAGE_SHIFT; - set_dma_reserve(mem_size_pfn - free_size_pfn); + for_each_mem_pfn_range(i, MAX_NUMNODES, &start_pfn, &end_pfn, NULL) { + start_pfn = min_t(unsigned long, start_pfn, MAX_DMA_PFN); + end_pfn = min_t(unsigned long, end_pfn, MAX_DMA_PFN); + nr_pages += end_pfn - start_pfn; + } + + for_each_free_mem_range(u, MAX_NUMNODES, &start, &end, NULL) { + start_pfn = min_t(unsigned long, PFN_UP(start), MAX_DMA_PFN); + end_pfn = min_t(unsigned long, PFN_DOWN(end), MAX_DMA_PFN); + if (start_pfn < end_pfn) + nr_free_pages += end_pfn - start_pfn; + } + + set_dma_reserve(nr_pages - nr_free_pages); #endif } diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index 4107c1a32b789..a9d0972df10af 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -7,93 +7,6 @@ #include #include -static __init struct range *find_range_array(int count) -{ - u64 end, size, mem; - struct range *range; - - size = sizeof(struct range) * count; - end = memblock.current_limit; - - mem = memblock_find_in_range(0, end, size, sizeof(struct range)); - if (!mem) - panic("can not find more space for range array"); - - /* - * This range is tempoaray, so don't reserve it, it will not be - * overlapped because We will not alloccate new buffer before - * We discard this one - */ - range = __va(mem); - memset(range, 0, size); - - return range; -} - -static u64 __init __memblock_x86_memory_in_range(u64 addr, u64 limit, bool get_free) -{ - int i, count; - struct range *range; - int nr_range; - u64 final_start, final_end; - u64 free_size; - struct memblock_region *r; - - count = (memblock.reserved.cnt + memblock.memory.cnt) * 2; - - range = find_range_array(count); - nr_range = 0; - - addr = PFN_UP(addr); - limit = PFN_DOWN(limit); - - for_each_memblock(memory, r) { - final_start = PFN_UP(r->base); - final_end = PFN_DOWN(r->base + r->size); - if (final_start >= final_end) - continue; - if (final_start >= limit || final_end <= addr) - continue; - - nr_range = add_range(range, count, nr_range, final_start, final_end); - } - subtract_range(range, count, 0, addr); - subtract_range(range, count, limit, -1ULL); - - /* Subtract memblock.reserved.region in range ? */ - if (!get_free) - goto sort_and_count_them; - for_each_memblock(reserved, r) { - final_start = PFN_DOWN(r->base); - final_end = PFN_UP(r->base + r->size); - if (final_start >= final_end) - continue; - if (final_start >= limit || final_end <= addr) - continue; - - subtract_range(range, count, final_start, final_end); - } - -sort_and_count_them: - nr_range = clean_sort_range(range, count); - - free_size = 0; - for (i = 0; i < nr_range; i++) - free_size += range[i].end - range[i].start; - - return free_size << PAGE_SHIFT; -} - -u64 __init memblock_x86_free_memory_in_range(u64 addr, u64 limit) -{ - return __memblock_x86_memory_in_range(addr, limit, true); -} - -u64 __init memblock_x86_memory_in_range(u64 addr, u64 limit) -{ - return __memblock_x86_memory_in_range(addr, limit, false); -} - void __init memblock_x86_reserve_range(u64 start, u64 end, char *name) { if (start == end) -- GitLab From 474b881bf4ee86aba55d46a4fdf293de32cba91b Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 11:16:04 +0200 Subject: [PATCH 0030/3089] x86: Use absent_pages_in_range() instead of memblock_x86_hole_size() memblock_x86_hole_size() calculates the total size of holes in a given range according to memblock and is used by numa emulation code and numa_meminfo_cover_memory(). Since conversion to MEMBLOCK_NODE_MAP, absent_pages_in_range() also uses memblock and gives the same result. This patch replaces memblock_x86_hole_size() uses with absent_pages_in_range(). After the conversion the x86 function doesn't have any user left and is killed. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310462166-31469-12-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 2 -- arch/x86/mm/memblock.c | 52 --------------------------------- arch/x86/mm/numa.c | 4 +-- arch/x86/mm/numa_emulation.c | 30 +++++++++++-------- 4 files changed, 20 insertions(+), 68 deletions(-) diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index a0cc7d66ac55b..17a882e90ada0 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -6,6 +6,4 @@ void memblock_x86_reserve_range(u64 start, u64 end, char *name); void memblock_x86_free_range(u64 start, u64 end); -u64 memblock_x86_hole_size(u64 start, u64 end); - #endif diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c index a9d0972df10af..7325c5d8ace56 100644 --- a/arch/x86/mm/memblock.c +++ b/arch/x86/mm/memblock.c @@ -32,55 +32,3 @@ void __init memblock_x86_free_range(u64 start, u64 end) memblock_free(start, end - start); } - -/* - * Finds an active region in the address range from start_pfn to last_pfn and - * returns its range in ei_startpfn and ei_endpfn for the memblock entry. - */ -static int __init memblock_x86_find_active_region(const struct memblock_region *ei, - unsigned long start_pfn, - unsigned long last_pfn, - unsigned long *ei_startpfn, - unsigned long *ei_endpfn) -{ - u64 align = PAGE_SIZE; - - *ei_startpfn = round_up(ei->base, align) >> PAGE_SHIFT; - *ei_endpfn = round_down(ei->base + ei->size, align) >> PAGE_SHIFT; - - /* Skip map entries smaller than a page */ - if (*ei_startpfn >= *ei_endpfn) - return 0; - - /* Skip if map is outside the node */ - if (*ei_endpfn <= start_pfn || *ei_startpfn >= last_pfn) - return 0; - - /* Check for overlaps */ - if (*ei_startpfn < start_pfn) - *ei_startpfn = start_pfn; - if (*ei_endpfn > last_pfn) - *ei_endpfn = last_pfn; - - return 1; -} - -/* - * Find the hole size (in bytes) in the memory range. - * @start: starting address of the memory range to scan - * @end: ending address of the memory range to scan - */ -u64 __init memblock_x86_hole_size(u64 start, u64 end) -{ - unsigned long start_pfn = start >> PAGE_SHIFT; - unsigned long last_pfn = end >> PAGE_SHIFT; - unsigned long ei_startpfn, ei_endpfn, ram = 0; - struct memblock_region *r; - - for_each_memblock(memory, r) - if (memblock_x86_find_active_region(r, start_pfn, last_pfn, - &ei_startpfn, &ei_endpfn)) - ram += ei_endpfn - ei_startpfn; - - return end - start - ((u64)ram << PAGE_SHIFT); -} diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c index f4a40bdb2e4e2..88e562729967f 100644 --- a/arch/x86/mm/numa.c +++ b/arch/x86/mm/numa.c @@ -475,8 +475,8 @@ static bool __init numa_meminfo_cover_memory(const struct numa_meminfo *mi) numaram = 0; } - e820ram = max_pfn - (memblock_x86_hole_size(0, - PFN_PHYS(max_pfn)) >> PAGE_SHIFT); + e820ram = max_pfn - absent_pages_in_range(0, max_pfn); + /* We seem to lose 3 pages somewhere. Allow 1M of slack. */ if ((s64)(e820ram - numaram) >= (1 << (20 - PAGE_SHIFT))) { printk(KERN_ERR "NUMA: nodes only cover %LuMB of your %LuMB e820 RAM. Not used.\n", diff --git a/arch/x86/mm/numa_emulation.c b/arch/x86/mm/numa_emulation.c index e3d471c20cdca..971fe70549b36 100644 --- a/arch/x86/mm/numa_emulation.c +++ b/arch/x86/mm/numa_emulation.c @@ -28,6 +28,16 @@ static int __init emu_find_memblk_by_nid(int nid, const struct numa_meminfo *mi) return -ENOENT; } +static u64 mem_hole_size(u64 start, u64 end) +{ + unsigned long start_pfn = PFN_UP(start); + unsigned long end_pfn = PFN_DOWN(end); + + if (start_pfn < end_pfn) + return PFN_PHYS(absent_pages_in_range(start_pfn, end_pfn)); + return 0; +} + /* * Sets up nid to range from @start to @end. The return value is -errno if * something went wrong, 0 otherwise. @@ -89,7 +99,7 @@ static int __init split_nodes_interleave(struct numa_meminfo *ei, * Calculate target node size. x86_32 freaks on __udivdi3() so do * the division in ulong number of pages and convert back. */ - size = max_addr - addr - memblock_x86_hole_size(addr, max_addr); + size = max_addr - addr - mem_hole_size(addr, max_addr); size = PFN_PHYS((unsigned long)(size >> PAGE_SHIFT) / nr_nodes); /* @@ -135,8 +145,7 @@ static int __init split_nodes_interleave(struct numa_meminfo *ei, * Continue to add memory to this fake node if its * non-reserved memory is less than the per-node size. */ - while (end - start - - memblock_x86_hole_size(start, end) < size) { + while (end - start - mem_hole_size(start, end) < size) { end += FAKE_NODE_MIN_SIZE; if (end > limit) { end = limit; @@ -150,7 +159,7 @@ static int __init split_nodes_interleave(struct numa_meminfo *ei, * this one must extend to the boundary. */ if (end < dma32_end && dma32_end - end - - memblock_x86_hole_size(end, dma32_end) < FAKE_NODE_MIN_SIZE) + mem_hole_size(end, dma32_end) < FAKE_NODE_MIN_SIZE) end = dma32_end; /* @@ -158,8 +167,7 @@ static int __init split_nodes_interleave(struct numa_meminfo *ei, * next node, this one must extend to the end of the * physical node. */ - if (limit - end - - memblock_x86_hole_size(end, limit) < size) + if (limit - end - mem_hole_size(end, limit) < size) end = limit; ret = emu_setup_memblk(ei, pi, nid++ % nr_nodes, @@ -180,7 +188,7 @@ static u64 __init find_end_of_node(u64 start, u64 max_addr, u64 size) { u64 end = start + size; - while (end - start - memblock_x86_hole_size(start, end) < size) { + while (end - start - mem_hole_size(start, end) < size) { end += FAKE_NODE_MIN_SIZE; if (end > max_addr) { end = max_addr; @@ -211,8 +219,7 @@ static int __init split_nodes_size_interleave(struct numa_meminfo *ei, * creates a uniform distribution of node sizes across the entire * machine (but not necessarily over physical nodes). */ - min_size = (max_addr - addr - memblock_x86_hole_size(addr, max_addr)) / - MAX_NUMNODES; + min_size = (max_addr - addr - mem_hole_size(addr, max_addr)) / MAX_NUMNODES; min_size = max(min_size, FAKE_NODE_MIN_SIZE); if ((min_size & FAKE_NODE_MIN_HASH_MASK) < min_size) min_size = (min_size + FAKE_NODE_MIN_SIZE) & @@ -252,7 +259,7 @@ static int __init split_nodes_size_interleave(struct numa_meminfo *ei, * this one must extend to the boundary. */ if (end < dma32_end && dma32_end - end - - memblock_x86_hole_size(end, dma32_end) < FAKE_NODE_MIN_SIZE) + mem_hole_size(end, dma32_end) < FAKE_NODE_MIN_SIZE) end = dma32_end; /* @@ -260,8 +267,7 @@ static int __init split_nodes_size_interleave(struct numa_meminfo *ei, * next node, this one must extend to the end of the * physical node. */ - if (limit - end - - memblock_x86_hole_size(end, limit) < size) + if (limit - end - mem_hole_size(end, limit) < size) end = limit; ret = emu_setup_memblk(ei, pi, nid++ % MAX_NUMNODES, -- GitLab From c378ddd53f9b8832a46fd4fec050a97fc2269858 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 14 Jul 2011 11:46:03 +0200 Subject: [PATCH 0031/3089] memblock, x86: Make ARCH_DISCARD_MEMBLOCK a config option From 6839454ae63f1eb21e515c10229ca95c22955fec Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 14 Jul 2011 11:22:17 +0200 Make ARCH_DISCARD_MEMBLOCK a config option so that it can be handled together with other MEMBLOCK options. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/20110714094603.GH3455@htj.dyndns.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: H. Peter Anvin --- arch/x86/Kconfig | 1 + arch/x86/include/asm/memblock.h | 2 -- include/linux/memblock.h | 2 +- mm/Kconfig | 3 +++ mm/memblock.c | 2 +- 5 files changed, 6 insertions(+), 4 deletions(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 97f08941dd792..28116d4f7b64c 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -26,6 +26,7 @@ config X86 select HAVE_KPROBES select HAVE_MEMBLOCK select HAVE_MEMBLOCK_NODE_MAP + select ARCH_DISCARD_MEMBLOCK select ARCH_WANT_OPTIONAL_GPIOLIB select ARCH_WANT_FRAME_POINTERS select HAVE_DMA_ATTRS diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h index 17a882e90ada0..bc5667081aea7 100644 --- a/arch/x86/include/asm/memblock.h +++ b/arch/x86/include/asm/memblock.h @@ -1,8 +1,6 @@ #ifndef _X86_MEMBLOCK_H #define _X86_MEMBLOCK_H -#define ARCH_DISCARD_MEMBLOCK - void memblock_x86_reserve_range(u64 start, u64 end, char *name); void memblock_x86_free_range(u64 start, u64 end); diff --git a/include/linux/memblock.h b/include/linux/memblock.h index 31def584ccebd..2491355bb6e48 100644 --- a/include/linux/memblock.h +++ b/include/linux/memblock.h @@ -197,7 +197,7 @@ static inline unsigned long memblock_region_reserved_end_pfn(const struct memblo region++) -#ifdef ARCH_DISCARD_MEMBLOCK +#ifdef CONFIG_ARCH_DISCARD_MEMBLOCK #define __init_memblock __meminit #define __initdata_memblock __meminitdata #else diff --git a/mm/Kconfig b/mm/Kconfig index 30a5d4792b833..7c5697116fcf8 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -134,6 +134,9 @@ config HAVE_MEMBLOCK config HAVE_MEMBLOCK_NODE_MAP boolean +config ARCH_DISCARD_MEMBLOCK + boolean + # eventually, we can have this option just 'select SPARSEMEM' config MEMORY_HOTPLUG bool "Allow for memory hot-add" diff --git a/mm/memblock.c b/mm/memblock.c index c4a8750406fc9..ebc6119f12804 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -959,7 +959,7 @@ static int __init early_memblock(char *p) } early_param("memblock", early_memblock); -#if defined(CONFIG_DEBUG_FS) && !defined(ARCH_DISCARD_MEMBLOCK) +#if defined(CONFIG_DEBUG_FS) && !defined(CONFIG_ARCH_DISCARD_MEMBLOCK) static int memblock_debug_show(struct seq_file *m, void *private) { -- GitLab From 24aa07882b672fff2da2f5c955759f0bd13d32d5 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 12 Jul 2011 11:16:06 +0200 Subject: [PATCH 0032/3089] memblock, x86: Replace memblock_x86_reserve/free_range() with generic ones Other than sanity check and debug message, the x86 specific version of memblock reserve/free functions are simple wrappers around the generic versions - memblock_reserve/free(). This patch adds debug messages with caller identification to the generic versions and replaces x86 specific ones and kills them. arch/x86/include/asm/memblock.h and arch/x86/mm/memblock.c are empty after this change and removed. Signed-off-by: Tejun Heo Link: http://lkml.kernel.org/r/1310462166-31469-14-git-send-email-tj@kernel.org Cc: Yinghai Lu Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Ingo Molnar Cc: "H. Peter Anvin" Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/memblock.h | 7 ------- arch/x86/kernel/aperture_64.c | 2 +- arch/x86/kernel/check.c | 2 +- arch/x86/kernel/head.c | 2 +- arch/x86/kernel/head32.c | 5 +++-- arch/x86/kernel/head64.c | 5 +++-- arch/x86/kernel/mpparse.c | 6 ++---- arch/x86/kernel/setup.c | 17 ++++++++--------- arch/x86/kernel/trampoline.c | 2 +- arch/x86/mm/Makefile | 2 -- arch/x86/mm/init.c | 6 +++--- arch/x86/mm/memblock.c | 34 --------------------------------- arch/x86/mm/memtest.c | 2 +- arch/x86/mm/numa.c | 5 ++--- arch/x86/mm/numa_32.c | 6 +++--- arch/x86/mm/numa_emulation.c | 4 ++-- arch/x86/platform/efi/efi.c | 6 ++---- arch/x86/xen/mmu.c | 12 ++++-------- arch/x86/xen/setup.c | 7 +++---- include/linux/memblock.h | 2 -- mm/memblock.c | 5 +++++ mm/nobootmem.c | 6 +++--- 22 files changed, 48 insertions(+), 97 deletions(-) delete mode 100644 arch/x86/include/asm/memblock.h delete mode 100644 arch/x86/mm/memblock.c diff --git a/arch/x86/include/asm/memblock.h b/arch/x86/include/asm/memblock.h deleted file mode 100644 index bc5667081aea7..0000000000000 --- a/arch/x86/include/asm/memblock.h +++ /dev/null @@ -1,7 +0,0 @@ -#ifndef _X86_MEMBLOCK_H -#define _X86_MEMBLOCK_H - -void memblock_x86_reserve_range(u64 start, u64 end, char *name); -void memblock_x86_free_range(u64 start, u64 end); - -#endif diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index 56363082bbdf3..6e76c191a8357 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -94,7 +94,7 @@ static u32 __init allocate_aperture(void) addr, aper_size>>10); return 0; } - memblock_x86_reserve_range(addr, addr + aper_size, "aperture64"); + memblock_reserve(addr, aper_size); /* * Kmemleak should not scan this block as it may not be mapped via the * kernel direct mapping. diff --git a/arch/x86/kernel/check.c b/arch/x86/kernel/check.c index 621cd23bb4e75..5da1269e8ddcb 100644 --- a/arch/x86/kernel/check.c +++ b/arch/x86/kernel/check.c @@ -91,7 +91,7 @@ void __init setup_bios_corruption_check(void) if (start >= end) continue; - memblock_x86_reserve_range(start, end, "SCAN RAM"); + memblock_reserve(start, end - start); scan_areas[num_scan_areas].addr = start; scan_areas[num_scan_areas].size = end - start; diff --git a/arch/x86/kernel/head.c b/arch/x86/kernel/head.c index af0699ba48cfb..48d9d4ea1020f 100644 --- a/arch/x86/kernel/head.c +++ b/arch/x86/kernel/head.c @@ -52,5 +52,5 @@ void __init reserve_ebda_region(void) lowmem = 0x9f000; /* reserve all memory between lowmem and the 1MB mark */ - memblock_x86_reserve_range(lowmem, 0x100000, "* BIOS reserved"); + memblock_reserve(lowmem, 0x100000 - lowmem); } diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index 3bb08509a7a15..be9282bcda72a 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -33,7 +33,8 @@ void __init i386_start_kernel(void) { memblock_init(); - memblock_x86_reserve_range(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); + memblock_reserve(__pa_symbol(&_text), + __pa_symbol(&__bss_stop) - __pa_symbol(&_text)); #ifdef CONFIG_BLK_DEV_INITRD /* Reserve INITRD */ @@ -42,7 +43,7 @@ void __init i386_start_kernel(void) u64 ramdisk_image = boot_params.hdr.ramdisk_image; u64 ramdisk_size = boot_params.hdr.ramdisk_size; u64 ramdisk_end = PAGE_ALIGN(ramdisk_image + ramdisk_size); - memblock_x86_reserve_range(ramdisk_image, ramdisk_end, "RAMDISK"); + memblock_reserve(ramdisk_image, ramdisk_end - ramdisk_image); } #endif diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 5655c2272adb8..fd25b11549b89 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -100,7 +100,8 @@ void __init x86_64_start_reservations(char *real_mode_data) memblock_init(); - memblock_x86_reserve_range(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); + memblock_reserve(__pa_symbol(&_text), + __pa_symbol(&__bss_stop) - __pa_symbol(&_text)); #ifdef CONFIG_BLK_DEV_INITRD /* Reserve INITRD */ @@ -109,7 +110,7 @@ void __init x86_64_start_reservations(char *real_mode_data) unsigned long ramdisk_image = boot_params.hdr.ramdisk_image; unsigned long ramdisk_size = boot_params.hdr.ramdisk_size; unsigned long ramdisk_end = PAGE_ALIGN(ramdisk_image + ramdisk_size); - memblock_x86_reserve_range(ramdisk_image, ramdisk_end, "RAMDISK"); + memblock_reserve(ramdisk_image, ramdisk_end - ramdisk_image); } #endif diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index 8faeaa0ed2cc0..a6b79c16ec78c 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -564,9 +564,7 @@ void __init default_get_smp_config(unsigned int early) static void __init smp_reserve_memory(struct mpf_intel *mpf) { - unsigned long size = get_mpc_size(mpf->physptr); - - memblock_x86_reserve_range(mpf->physptr, mpf->physptr+size, "* MP-table mpc"); + memblock_reserve(mpf->physptr, get_mpc_size(mpf->physptr)); } static int __init smp_scan_config(unsigned long base, unsigned long length) @@ -595,7 +593,7 @@ static int __init smp_scan_config(unsigned long base, unsigned long length) mpf, (u64)virt_to_phys(mpf)); mem = virt_to_phys(mpf); - memblock_x86_reserve_range(mem, mem + sizeof(*mpf), "* MP-table mpf"); + memblock_reserve(mem, sizeof(*mpf)); if (mpf->physptr) smp_reserve_memory(mpf); diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 31ffe20d5d278..97d227ec995d1 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -306,7 +306,8 @@ static void __init cleanup_highmap(void) static void __init reserve_brk(void) { if (_brk_end > _brk_start) - memblock_x86_reserve_range(__pa(_brk_start), __pa(_brk_end), "BRK"); + memblock_reserve(__pa(_brk_start), + __pa(_brk_end) - __pa(_brk_start)); /* Mark brk area as locked down and no longer taking any new allocations */ @@ -337,7 +338,7 @@ static void __init relocate_initrd(void) /* Note: this includes all the lowmem currently occupied by the initrd, we rely on that fact to keep the data intact. */ - memblock_x86_reserve_range(ramdisk_here, ramdisk_here + area_size, "NEW RAMDISK"); + memblock_reserve(ramdisk_here, area_size); initrd_start = ramdisk_here + PAGE_OFFSET; initrd_end = initrd_start + ramdisk_size; printk(KERN_INFO "Allocated new RAMDISK: %08llx - %08llx\n", @@ -393,7 +394,7 @@ static void __init reserve_initrd(void) initrd_start = 0; if (ramdisk_size >= (end_of_lowmem>>1)) { - memblock_x86_free_range(ramdisk_image, ramdisk_end); + memblock_free(ramdisk_image, ramdisk_end - ramdisk_image); printk(KERN_ERR "initrd too large to handle, " "disabling initrd\n"); return; @@ -416,7 +417,7 @@ static void __init reserve_initrd(void) relocate_initrd(); - memblock_x86_free_range(ramdisk_image, ramdisk_end); + memblock_free(ramdisk_image, ramdisk_end - ramdisk_image); } #else static void __init reserve_initrd(void) @@ -490,15 +491,13 @@ static void __init memblock_x86_reserve_range_setup_data(void) { struct setup_data *data; u64 pa_data; - char buf[32]; if (boot_params.hdr.version < 0x0209) return; pa_data = boot_params.hdr.setup_data; while (pa_data) { data = early_memremap(pa_data, sizeof(*data)); - sprintf(buf, "setup data %x", data->type); - memblock_x86_reserve_range(pa_data, pa_data+sizeof(*data)+data->len, buf); + memblock_reserve(pa_data, sizeof(*data) + data->len); pa_data = data->next; early_iounmap(data, sizeof(*data)); } @@ -568,7 +567,7 @@ static void __init reserve_crashkernel(void) return; } } - memblock_x86_reserve_range(crash_base, crash_base + crash_size, "CRASH KERNEL"); + memblock_reserve(crash_base, crash_size); printk(KERN_INFO "Reserving %ldMB of memory at %ldMB " "for crashkernel (System RAM: %ldMB)\n", @@ -626,7 +625,7 @@ static __init void reserve_ibft_region(void) addr = find_ibft_region(&size); if (size) - memblock_x86_reserve_range(addr, addr + size, "* ibft"); + memblock_reserve(addr, size); } static unsigned reserve_low = CONFIG_X86_RESERVE_LOW << 10; diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c index a1f13ddb06e05..a73b61055ad63 100644 --- a/arch/x86/kernel/trampoline.c +++ b/arch/x86/kernel/trampoline.c @@ -18,7 +18,7 @@ void __init setup_trampolines(void) panic("Cannot allocate trampoline\n"); x86_trampoline_base = __va(mem); - memblock_x86_reserve_range(mem, mem + size, "TRAMPOLINE"); + memblock_reserve(mem, size); printk(KERN_DEBUG "Base memory trampoline at [%p] %llx size %zu\n", x86_trampoline_base, (unsigned long long)mem, size); diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile index 3d11327c9ab46..23d8e5fecf76b 100644 --- a/arch/x86/mm/Makefile +++ b/arch/x86/mm/Makefile @@ -27,6 +27,4 @@ obj-$(CONFIG_AMD_NUMA) += amdtopology.o obj-$(CONFIG_ACPI_NUMA) += srat.o obj-$(CONFIG_NUMA_EMU) += numa_emulation.o -obj-$(CONFIG_HAVE_MEMBLOCK) += memblock.o - obj-$(CONFIG_MEMTEST) += memtest.o diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index 13cf05a616057..0b736b99d9255 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -81,7 +81,7 @@ static void __init find_early_table_space(unsigned long end, int use_pse, void __init native_pagetable_reserve(u64 start, u64 end) { - memblock_x86_reserve_range(start, end, "PGTABLE"); + memblock_reserve(start, end - start); } struct map_range { @@ -280,8 +280,8 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, * pgt_buf_end) and free the other ones (pgt_buf_end - pgt_buf_top) * so that they can be reused for other purposes. * - * On native it just means calling memblock_x86_reserve_range, on Xen it - * also means marking RW the pagetable pages that we allocated before + * On native it just means calling memblock_reserve, on Xen it also + * means marking RW the pagetable pages that we allocated before * but that haven't been used. * * In fact on xen we mark RO the whole range pgt_buf_start - diff --git a/arch/x86/mm/memblock.c b/arch/x86/mm/memblock.c deleted file mode 100644 index 7325c5d8ace56..0000000000000 --- a/arch/x86/mm/memblock.c +++ /dev/null @@ -1,34 +0,0 @@ -#include -#include -#include -#include -#include -#include -#include -#include - -void __init memblock_x86_reserve_range(u64 start, u64 end, char *name) -{ - if (start == end) - return; - - if (WARN_ONCE(start > end, "memblock_x86_reserve_range: wrong range [%#llx, %#llx)\n", start, end)) - return; - - memblock_dbg(" memblock_x86_reserve_range: [%#010llx-%#010llx] %16s\n", start, end - 1, name); - - memblock_reserve(start, end - start); -} - -void __init memblock_x86_free_range(u64 start, u64 end) -{ - if (start == end) - return; - - if (WARN_ONCE(start > end, "memblock_x86_free_range: wrong range [%#llx, %#llx)\n", start, end)) - return; - - memblock_dbg(" memblock_x86_free_range: [%#010llx-%#010llx]\n", start, end - 1); - - memblock_free(start, end - start); -} diff --git a/arch/x86/mm/memtest.c b/arch/x86/mm/memtest.c index 46a5ff25eda40..c80b9fb95734f 100644 --- a/arch/x86/mm/memtest.c +++ b/arch/x86/mm/memtest.c @@ -34,7 +34,7 @@ static void __init reserve_bad_mem(u64 pattern, u64 start_bad, u64 end_bad) (unsigned long long) pattern, (unsigned long long) start_bad, (unsigned long long) end_bad); - memblock_x86_reserve_range(start_bad, end_bad, "BAD RAM"); + memblock_reserve(start_bad, end_bad - start_bad); } static void __init memtest(u64 pattern, u64 start_phys, u64 size) diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c index 88e562729967f..496f494593bf6 100644 --- a/arch/x86/mm/numa.c +++ b/arch/x86/mm/numa.c @@ -364,8 +364,7 @@ void __init numa_reset_distance(void) /* numa_distance could be 1LU marking allocation failure, test cnt */ if (numa_distance_cnt) - memblock_x86_free_range(__pa(numa_distance), - __pa(numa_distance) + size); + memblock_free(__pa(numa_distance), size); numa_distance_cnt = 0; numa_distance = NULL; /* enable table creation */ } @@ -394,7 +393,7 @@ static int __init numa_alloc_distance(void) numa_distance = (void *)1LU; return -ENOMEM; } - memblock_x86_reserve_range(phys, phys + size, "NUMA DIST"); + memblock_reserve(phys, size); numa_distance = __va(phys); numa_distance_cnt = cnt; diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c index 58878b536ef2a..534255a36b6ba 100644 --- a/arch/x86/mm/numa_32.c +++ b/arch/x86/mm/numa_32.c @@ -204,7 +204,7 @@ void __init init_alloc_remap(int nid, u64 start, u64 end) size, nid); return; } - memblock_x86_reserve_range(node_pa, node_pa + size, "KVA RAM"); + memblock_reserve(node_pa, size); remap_pa = memblock_find_in_range(min_low_pfn << PAGE_SHIFT, max_low_pfn << PAGE_SHIFT, @@ -212,10 +212,10 @@ void __init init_alloc_remap(int nid, u64 start, u64 end) if (!remap_pa) { pr_warning("remap_alloc: failed to allocate %lu bytes remap area for node %d\n", size, nid); - memblock_x86_free_range(node_pa, node_pa + size); + memblock_free(node_pa, size); return; } - memblock_x86_reserve_range(remap_pa, remap_pa + size, "KVA PG"); + memblock_reserve(remap_pa, size); remap_va = phys_to_virt(remap_pa); /* perform actual remap */ diff --git a/arch/x86/mm/numa_emulation.c b/arch/x86/mm/numa_emulation.c index 971fe70549b36..46db56845f180 100644 --- a/arch/x86/mm/numa_emulation.c +++ b/arch/x86/mm/numa_emulation.c @@ -361,7 +361,7 @@ void __init numa_emulation(struct numa_meminfo *numa_meminfo, int numa_dist_cnt) pr_warning("NUMA: Warning: can't allocate copy of distance table, disabling emulation\n"); goto no_emu; } - memblock_x86_reserve_range(phys, phys + phys_size, "TMP NUMA DIST"); + memblock_reserve(phys, phys_size); phys_dist = __va(phys); for (i = 0; i < numa_dist_cnt; i++) @@ -430,7 +430,7 @@ void __init numa_emulation(struct numa_meminfo *numa_meminfo, int numa_dist_cnt) /* free the copied physical distance table */ if (phys_dist) - memblock_x86_free_range(__pa(phys_dist), __pa(phys_dist) + phys_size); + memblock_free(__pa(phys_dist), phys_size); return; no_emu: diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index a4c322ca1a5d5..3b4e86bda3cb9 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -280,8 +280,7 @@ void __init efi_memblock_x86_reserve_range(void) boot_params.efi_info.efi_memdesc_size; memmap.desc_version = boot_params.efi_info.efi_memdesc_version; memmap.desc_size = boot_params.efi_info.efi_memdesc_size; - memblock_x86_reserve_range(pmap, pmap + memmap.nr_map * memmap.desc_size, - "EFI memmap"); + memblock_reserve(pmap, memmap.nr_map * memmap.desc_size); } #if EFI_DEBUG @@ -332,8 +331,7 @@ void __init efi_reserve_boot_services(void) "[0x%010llx-0x%010llx]\n", start, start+size-1); } else - memblock_x86_reserve_range(start, start+size, - "EFI Boot"); + memblock_reserve(start, size); } } diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 0ccccb67a9930..ad54fa10f8a23 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -1720,10 +1720,8 @@ pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd, __xen_write_cr3(true, __pa(pgd)); xen_mc_issue(PARAVIRT_LAZY_CPU); - memblock_x86_reserve_range(__pa(xen_start_info->pt_base), - __pa(xen_start_info->pt_base + - xen_start_info->nr_pt_frames * PAGE_SIZE), - "XEN PAGETABLES"); + memblock_reserve(__pa(xen_start_info->pt_base), + xen_start_info->nr_pt_frames * PAGE_SIZE); return pgd; } @@ -1799,10 +1797,8 @@ pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd, PFN_DOWN(__pa(initial_page_table))); xen_write_cr3(__pa(initial_page_table)); - memblock_x86_reserve_range(__pa(xen_start_info->pt_base), - __pa(xen_start_info->pt_base + - xen_start_info->nr_pt_frames * PAGE_SIZE), - "XEN PAGETABLES"); + memblock_reserve(__pa(xen_start_info->pt_base), + xen_start_info->nr_pt_frames * PAGE_SIZE)); return initial_page_table; } diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index 60aeeb56948f7..73daaf75801aa 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -63,7 +63,7 @@ static void __init xen_add_extra_mem(unsigned long pages) e820_add_region(extra_start, size, E820_RAM); sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); - memblock_x86_reserve_range(extra_start, extra_start + size, "XEN EXTRA"); + memblock_reserve(extra_start, size); xen_extra_mem_size += size; @@ -287,9 +287,8 @@ char * __init xen_memory_setup(void) * - xen_start_info * See comment above "struct start_info" in */ - memblock_x86_reserve_range(__pa(xen_start_info->mfn_list), - __pa(xen_start_info->pt_base), - "XEN START INFO"); + memblock_reserve(__pa(xen_start_info->mfn_list), + xen_start_info->pt_base - xen_start_info->mfn_list); sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); diff --git a/include/linux/memblock.h b/include/linux/memblock.h index 2491355bb6e48..90746318cec4e 100644 --- a/include/linux/memblock.h +++ b/include/linux/memblock.h @@ -17,8 +17,6 @@ #include #include -#include - #define INIT_MEMBLOCK_REGIONS 128 struct memblock_region { diff --git a/mm/memblock.c b/mm/memblock.c index ebc6119f12804..0cb4da657b9d5 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -449,6 +449,9 @@ long __init_memblock memblock_remove(phys_addr_t base, phys_addr_t size) long __init_memblock memblock_free(phys_addr_t base, phys_addr_t size) { + memblock_dbg(" memblock_free: [%#016llx-%#016llx] %pF\n", + base, base + size, (void *)_RET_IP_); + return __memblock_remove(&memblock.reserved, base, size); } @@ -456,6 +459,8 @@ long __init_memblock memblock_reserve(phys_addr_t base, phys_addr_t size) { struct memblock_type *_rgn = &memblock.reserved; + memblock_dbg("memblock_reserve: [%#016llx-%#016llx] %pF\n", + base, base + size, (void *)_RET_IP_); BUG_ON(0 == size); return memblock_add_region(_rgn, base, size); diff --git a/mm/nobootmem.c b/mm/nobootmem.c index 7075bc00fa84c..29d948ce6d0f7 100644 --- a/mm/nobootmem.c +++ b/mm/nobootmem.c @@ -47,7 +47,7 @@ static void * __init __alloc_memory_core_early(int nid, u64 size, u64 align, ptr = phys_to_virt(addr); memset(ptr, 0, size); - memblock_x86_reserve_range(addr, addr + size, "BOOTMEM"); + memblock_reserve(addr, size); /* * The min_count is set to 0 so that bootmem allocated blocks * are never reported as leaks. @@ -175,7 +175,7 @@ void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, unsigned long size) { kmemleak_free_part(__va(physaddr), size); - memblock_x86_free_range(physaddr, physaddr + size); + memblock_free(physaddr, size); } /** @@ -190,7 +190,7 @@ void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr, void __init free_bootmem(unsigned long addr, unsigned long size) { kmemleak_free_part(__va(addr), size); - memblock_x86_free_range(addr, addr + size); + memblock_free(addr, size); } static void * __init ___alloc_bootmem_nopanic(unsigned long size, -- GitLab From a150439c4a97db379f0ed6faa46fbbb6e7bf3cb2 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Thu, 14 Jul 2011 11:57:10 -0700 Subject: [PATCH 0033/3089] memblock: Cast phys_addr_t to unsigned long long for printf use phys_addr_t is not necessarily the same thing as unsigned long long. It is, however, easier to cast it to unsigned long long for printf purposes than it is to deal with differnent printf formats. Signed-off-by: H. Peter Anvin Cc: Tejun Heo Link: http://lkml.kernel.org/r/4E1F4D2C.3000507@zytor.com --- mm/memblock.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/mm/memblock.c b/mm/memblock.c index 0cb4da657b9d5..a75723d626311 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -450,7 +450,9 @@ long __init_memblock memblock_remove(phys_addr_t base, phys_addr_t size) long __init_memblock memblock_free(phys_addr_t base, phys_addr_t size) { memblock_dbg(" memblock_free: [%#016llx-%#016llx] %pF\n", - base, base + size, (void *)_RET_IP_); + (unsigned long long)base, + (unsigned long long)base + size, + (void *)_RET_IP_); return __memblock_remove(&memblock.reserved, base, size); } @@ -460,7 +462,9 @@ long __init_memblock memblock_reserve(phys_addr_t base, phys_addr_t size) struct memblock_type *_rgn = &memblock.reserved; memblock_dbg("memblock_reserve: [%#016llx-%#016llx] %pF\n", - base, base + size, (void *)_RET_IP_); + (unsigned long long)base, + (unsigned long long)base + size, + (void *)_RET_IP_); BUG_ON(0 == size); return memblock_add_region(_rgn, base, size); -- GitLab From d2eece376648d2f7ba0a7d78f3c4d0421e608ac2 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:21:54 -0700 Subject: [PATCH 0034/3089] ocfs2/cluster: Abort heartbeat start on hard-ro devices Currently if the heartbeat device is hard-ro, the o2hb thread keeps chugging along and dumping errors along the way. The user needs to manually stop the heartbeat. The patch addresses this shortcoming by adding a limit to the number of times the hb thread will iterate in an unsteady state. If the hb thread does not ready steady state in that many interation, the start is aborted. Signed-off-by: Sunil Mushran --- fs/ocfs2/cluster/heartbeat.c | 185 ++++++++++++++++++++++------------- 1 file changed, 116 insertions(+), 69 deletions(-) diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 9a3e6bbff27bd..4728a746281df 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -216,6 +216,7 @@ struct o2hb_region { struct list_head hr_all_item; unsigned hr_unclean_stop:1, + hr_aborted_start:1, hr_item_pinned:1, hr_item_dropped:1; @@ -254,6 +255,10 @@ struct o2hb_region { * a more complete api that doesn't lead to this sort of fragility. */ atomic_t hr_steady_iterations; + /* terminate o2hb thread if it does not reach steady state + * (hr_steady_iterations == 0) within hr_unsteady_iterations */ + atomic_t hr_unsteady_iterations; + char hr_dev_name[BDEVNAME_SIZE]; unsigned int hr_timeout_ms; @@ -324,6 +329,10 @@ static void o2hb_write_timeout(struct work_struct *work) static void o2hb_arm_write_timeout(struct o2hb_region *reg) { + /* Arm writeout only after thread reaches steady state */ + if (atomic_read(®->hr_steady_iterations) != 0) + return; + mlog(ML_HEARTBEAT, "Queue write timeout for %u ms\n", O2HB_MAX_WRITE_TIMEOUT_MS); @@ -537,9 +546,14 @@ static int o2hb_verify_crc(struct o2hb_region *reg, return read == computed; } -/* We want to make sure that nobody is heartbeating on top of us -- - * this will help detect an invalid configuration. */ -static void o2hb_check_last_timestamp(struct o2hb_region *reg) +/* + * Compare the slot data with what we wrote in the last iteration. + * If the match fails, print an appropriate error message. This is to + * detect errors like... another node hearting on the same slot, + * flaky device that is losing writes, etc. + * Returns 1 if check succeeds, 0 otherwise. + */ +static int o2hb_check_own_slot(struct o2hb_region *reg) { struct o2hb_disk_slot *slot; struct o2hb_disk_heartbeat_block *hb_block; @@ -548,13 +562,13 @@ static void o2hb_check_last_timestamp(struct o2hb_region *reg) slot = ®->hr_slots[o2nm_this_node()]; /* Don't check on our 1st timestamp */ if (!slot->ds_last_time) - return; + return 0; hb_block = slot->ds_raw_block; if (le64_to_cpu(hb_block->hb_seq) == slot->ds_last_time && le64_to_cpu(hb_block->hb_generation) == slot->ds_last_generation && hb_block->hb_node == slot->ds_node_num) - return; + return 1; #define ERRSTR1 "Another node is heartbeating on device" #define ERRSTR2 "Heartbeat generation mismatch on device" @@ -574,6 +588,8 @@ static void o2hb_check_last_timestamp(struct o2hb_region *reg) (unsigned long long)slot->ds_last_time, hb_block->hb_node, (unsigned long long)le64_to_cpu(hb_block->hb_generation), (unsigned long long)le64_to_cpu(hb_block->hb_seq)); + + return 0; } static inline void o2hb_prepare_block(struct o2hb_region *reg, @@ -719,17 +735,24 @@ static void o2hb_shutdown_slot(struct o2hb_disk_slot *slot) o2nm_node_put(node); } -static void o2hb_set_quorum_device(struct o2hb_region *reg, - struct o2hb_disk_slot *slot) +static void o2hb_set_quorum_device(struct o2hb_region *reg) { - assert_spin_locked(&o2hb_live_lock); - if (!o2hb_global_heartbeat_active()) return; - if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap)) + /* Prevent race with o2hb_heartbeat_group_drop_item() */ + if (kthread_should_stop()) + return; + + /* Tag region as quorum only after thread reaches steady state */ + if (atomic_read(®->hr_steady_iterations) != 0) return; + spin_lock(&o2hb_live_lock); + + if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap)) + goto unlock; + /* * A region can be added to the quorum only when it sees all * live nodes heartbeat on it. In other words, the region has been @@ -737,13 +760,10 @@ static void o2hb_set_quorum_device(struct o2hb_region *reg, */ if (memcmp(reg->hr_live_node_bitmap, o2hb_live_node_bitmap, sizeof(o2hb_live_node_bitmap))) - return; - - if (slot->ds_changed_samples < O2HB_LIVE_THRESHOLD) - return; + goto unlock; - printk(KERN_NOTICE "o2hb: Region %s is now a quorum device\n", - config_item_name(®->hr_item)); + printk(KERN_NOTICE "o2hb: Region %s (%s) is now a quorum device\n", + config_item_name(®->hr_item), reg->hr_dev_name); set_bit(reg->hr_region_num, o2hb_quorum_region_bitmap); @@ -754,6 +774,8 @@ static void o2hb_set_quorum_device(struct o2hb_region *reg, if (o2hb_pop_count(&o2hb_quorum_region_bitmap, O2NM_MAX_REGIONS) > O2HB_PIN_CUT_OFF) o2hb_region_unpin(NULL); +unlock: + spin_unlock(&o2hb_live_lock); } static int o2hb_check_slot(struct o2hb_region *reg, @@ -925,8 +947,6 @@ fire_callbacks: slot->ds_equal_samples = 0; } out: - o2hb_set_quorum_device(reg, slot); - spin_unlock(&o2hb_live_lock); o2hb_run_event_list(&event); @@ -957,7 +977,8 @@ static int o2hb_highest_node(unsigned long *nodes, static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) { - int i, ret, highest_node, change = 0; + int i, ret, highest_node; + int membership_change = 0, own_slot_ok = 0; unsigned long configured_nodes[BITS_TO_LONGS(O2NM_MAX_NODES)]; unsigned long live_node_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)]; struct o2hb_bio_wait_ctxt write_wc; @@ -966,7 +987,7 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) sizeof(configured_nodes)); if (ret) { mlog_errno(ret); - return ret; + goto bail; } /* @@ -982,8 +1003,9 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) highest_node = o2hb_highest_node(configured_nodes, O2NM_MAX_NODES); if (highest_node >= O2NM_MAX_NODES) { - mlog(ML_NOTICE, "ocfs2_heartbeat: no configured nodes found!\n"); - return -EINVAL; + mlog(ML_NOTICE, "o2hb: No configured nodes found!\n"); + ret = -EINVAL; + goto bail; } /* No sense in reading the slots of nodes that don't exist @@ -993,29 +1015,27 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) ret = o2hb_read_slots(reg, highest_node + 1); if (ret < 0) { mlog_errno(ret); - return ret; + goto bail; } /* With an up to date view of the slots, we can check that no * other node has been improperly configured to heartbeat in * our slot. */ - o2hb_check_last_timestamp(reg); + own_slot_ok = o2hb_check_own_slot(reg); /* fill in the proper info for our next heartbeat */ o2hb_prepare_block(reg, reg->hr_generation); - /* And fire off the write. Note that we don't wait on this I/O - * until later. */ ret = o2hb_issue_node_write(reg, &write_wc); if (ret < 0) { mlog_errno(ret); - return ret; + goto bail; } i = -1; while((i = find_next_bit(configured_nodes, O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) { - change |= o2hb_check_slot(reg, ®->hr_slots[i]); + membership_change |= o2hb_check_slot(reg, ®->hr_slots[i]); } /* @@ -1030,18 +1050,39 @@ static int o2hb_do_disk_heartbeat(struct o2hb_region *reg) * disk */ mlog(ML_ERROR, "Write error %d on device \"%s\"\n", write_wc.wc_error, reg->hr_dev_name); - return write_wc.wc_error; + ret = write_wc.wc_error; + goto bail; } - o2hb_arm_write_timeout(reg); + /* Skip disarming the timeout if own slot has stale/bad data */ + if (own_slot_ok) { + o2hb_set_quorum_device(reg); + o2hb_arm_write_timeout(reg); + } +bail: /* let the person who launched us know when things are steady */ - if (!change && (atomic_read(®->hr_steady_iterations) != 0)) { - if (atomic_dec_and_test(®->hr_steady_iterations)) + if (atomic_read(®->hr_steady_iterations) != 0) { + if (!ret && own_slot_ok && !membership_change) { + if (atomic_dec_and_test(®->hr_steady_iterations)) + wake_up(&o2hb_steady_queue); + } + } + + if (atomic_read(®->hr_steady_iterations) != 0) { + if (atomic_dec_and_test(®->hr_unsteady_iterations)) { + printk(KERN_NOTICE "o2hb: Unable to stabilize " + "heartbeart on region %s (%s)\n", + config_item_name(®->hr_item), + reg->hr_dev_name); + atomic_set(®->hr_steady_iterations, 0); + reg->hr_aborted_start = 1; wake_up(&o2hb_steady_queue); + ret = -EIO; + } } - return 0; + return ret; } /* Subtract b from a, storing the result in a. a *must* have a larger @@ -1095,7 +1136,8 @@ static int o2hb_thread(void *data) /* Pin node */ o2nm_depend_this_node(); - while (!kthread_should_stop() && !reg->hr_unclean_stop) { + while (!kthread_should_stop() && + !reg->hr_unclean_stop && !reg->hr_aborted_start) { /* We track the time spent inside * o2hb_do_disk_heartbeat so that we avoid more than * hr_timeout_ms between disk writes. On busy systems @@ -1103,10 +1145,7 @@ static int o2hb_thread(void *data) * likely to time itself out. */ do_gettimeofday(&before_hb); - i = 0; - do { - ret = o2hb_do_disk_heartbeat(reg); - } while (ret && ++i < 2); + ret = o2hb_do_disk_heartbeat(reg); do_gettimeofday(&after_hb); elapsed_msec = o2hb_elapsed_msecs(&before_hb, &after_hb); @@ -1117,7 +1156,8 @@ static int o2hb_thread(void *data) after_hb.tv_sec, (unsigned long) after_hb.tv_usec, elapsed_msec); - if (elapsed_msec < reg->hr_timeout_ms) { + if (!kthread_should_stop() && + elapsed_msec < reg->hr_timeout_ms) { /* the kthread api has blocked signals for us so no * need to record the return value. */ msleep_interruptible(reg->hr_timeout_ms - elapsed_msec); @@ -1134,20 +1174,20 @@ static int o2hb_thread(void *data) * to timeout on this region when we could just as easily * write a clear generation - thus indicating to them that * this node has left this region. - * - * XXX: Should we skip this on unclean_stop? */ - o2hb_prepare_block(reg, 0); - ret = o2hb_issue_node_write(reg, &write_wc); - if (ret == 0) { - o2hb_wait_on_io(reg, &write_wc); - } else { - mlog_errno(ret); + */ + if (!reg->hr_unclean_stop && !reg->hr_aborted_start) { + o2hb_prepare_block(reg, 0); + ret = o2hb_issue_node_write(reg, &write_wc); + if (ret == 0) + o2hb_wait_on_io(reg, &write_wc); + else + mlog_errno(ret); } /* Unpin node */ o2nm_undepend_this_node(); - mlog(ML_HEARTBEAT|ML_KTHREAD, "hb thread exiting\n"); + mlog(ML_HEARTBEAT|ML_KTHREAD, "o2hb thread exiting\n"); return 0; } @@ -1426,6 +1466,8 @@ static void o2hb_region_release(struct config_item *item) struct page *page; struct o2hb_region *reg = to_o2hb_region(item); + mlog(ML_HEARTBEAT, "hb region release (%s)\n", reg->hr_dev_name); + if (reg->hr_tmp_block) kfree(reg->hr_tmp_block); @@ -1792,7 +1834,10 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, live_threshold <<= 1; spin_unlock(&o2hb_live_lock); } - atomic_set(®->hr_steady_iterations, live_threshold + 1); + ++live_threshold; + atomic_set(®->hr_steady_iterations, live_threshold); + /* unsteady_iterations is double the steady_iterations */ + atomic_set(®->hr_unsteady_iterations, (live_threshold << 1)); hb_task = kthread_run(o2hb_thread, reg, "o2hb-%s", reg->hr_item.ci_name); @@ -1809,14 +1854,12 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, ret = wait_event_interruptible(o2hb_steady_queue, atomic_read(®->hr_steady_iterations) == 0); if (ret) { - /* We got interrupted (hello ptrace!). Clean up */ - spin_lock(&o2hb_live_lock); - hb_task = reg->hr_task; - reg->hr_task = NULL; - spin_unlock(&o2hb_live_lock); + atomic_set(®->hr_steady_iterations, 0); + reg->hr_aborted_start = 1; + } - if (hb_task) - kthread_stop(hb_task); + if (reg->hr_aborted_start) { + ret = -EIO; goto out; } @@ -1833,8 +1876,8 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, ret = -EIO; if (hb_task && o2hb_global_heartbeat_active()) - printk(KERN_NOTICE "o2hb: Heartbeat started on region %s\n", - config_item_name(®->hr_item)); + printk(KERN_NOTICE "o2hb: Heartbeat started on region %s (%s)\n", + config_item_name(®->hr_item), reg->hr_dev_name); out: if (filp) @@ -2092,13 +2135,6 @@ static void o2hb_heartbeat_group_drop_item(struct config_group *group, /* stop the thread when the user removes the region dir */ spin_lock(&o2hb_live_lock); - if (o2hb_global_heartbeat_active()) { - clear_bit(reg->hr_region_num, o2hb_region_bitmap); - clear_bit(reg->hr_region_num, o2hb_live_region_bitmap); - if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap)) - quorum_region = 1; - clear_bit(reg->hr_region_num, o2hb_quorum_region_bitmap); - } hb_task = reg->hr_task; reg->hr_task = NULL; reg->hr_item_dropped = 1; @@ -2107,19 +2143,30 @@ static void o2hb_heartbeat_group_drop_item(struct config_group *group, if (hb_task) kthread_stop(hb_task); + if (o2hb_global_heartbeat_active()) { + spin_lock(&o2hb_live_lock); + clear_bit(reg->hr_region_num, o2hb_region_bitmap); + clear_bit(reg->hr_region_num, o2hb_live_region_bitmap); + if (test_bit(reg->hr_region_num, o2hb_quorum_region_bitmap)) + quorum_region = 1; + clear_bit(reg->hr_region_num, o2hb_quorum_region_bitmap); + spin_unlock(&o2hb_live_lock); + printk(KERN_NOTICE "o2hb: Heartbeat %s on region %s (%s)\n", + ((atomic_read(®->hr_steady_iterations) == 0) ? + "stopped" : "start aborted"), config_item_name(item), + reg->hr_dev_name); + } + /* * If we're racing a dev_write(), we need to wake them. They will * check reg->hr_task */ if (atomic_read(®->hr_steady_iterations) != 0) { + reg->hr_aborted_start = 1; atomic_set(®->hr_steady_iterations, 0); wake_up(&o2hb_steady_queue); } - if (o2hb_global_heartbeat_active()) - printk(KERN_NOTICE "o2hb: Heartbeat stopped on region %s\n", - config_item_name(®->hr_item)); - config_item_put(item); if (!o2hb_global_heartbeat_active() || !quorum_region) -- GitLab From 1dfecf810e0eacb35987905082f23e5c2cd26e91 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:22:54 -0700 Subject: [PATCH 0035/3089] ocfs2/cluster: Clean up messages in o2net o2net messages needed a facelift. Signed-off-by: Sunil Mushran --- fs/ocfs2/cluster/tcp.c | 119 ++++++++++++++++++----------------------- 1 file changed, 53 insertions(+), 66 deletions(-) diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index db5ee4b4f47aa..6e97895ecf24a 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -545,7 +545,7 @@ static void o2net_set_nn_state(struct o2net_node *nn, } if (was_valid && !valid) { - printk(KERN_NOTICE "o2net: no longer connected to " + printk(KERN_NOTICE "o2net: No longer connected to " SC_NODEF_FMT "\n", SC_NODEF_ARGS(old_sc)); o2net_complete_nodes_nsw(nn); } @@ -555,7 +555,7 @@ static void o2net_set_nn_state(struct o2net_node *nn, cancel_delayed_work(&nn->nn_connect_expired); printk(KERN_NOTICE "o2net: %s " SC_NODEF_FMT "\n", o2nm_this_node() > sc->sc_node->nd_num ? - "connected to" : "accepted connection from", + "Connected to" : "Accepted connection from", SC_NODEF_ARGS(sc)); } @@ -643,7 +643,7 @@ static void o2net_state_change(struct sock *sk) o2net_sc_queue_work(sc, &sc->sc_connect_work); break; default: - printk(KERN_INFO "o2net: connection to " SC_NODEF_FMT + printk(KERN_INFO "o2net: Connection to " SC_NODEF_FMT " shutdown, state %d\n", SC_NODEF_ARGS(sc), sk->sk_state); o2net_sc_queue_work(sc, &sc->sc_shutdown_work); @@ -1284,11 +1284,11 @@ static int o2net_check_handshake(struct o2net_sock_container *sc) struct o2net_node *nn = o2net_nn_from_num(sc->sc_node->nd_num); if (hand->protocol_version != cpu_to_be64(O2NET_PROTOCOL_VERSION)) { - mlog(ML_NOTICE, SC_NODEF_FMT " advertised net protocol " - "version %llu but %llu is required, disconnecting\n", - SC_NODEF_ARGS(sc), - (unsigned long long)be64_to_cpu(hand->protocol_version), - O2NET_PROTOCOL_VERSION); + printk(KERN_NOTICE "o2net: " SC_NODEF_FMT " Advertised net " + "protocol version %llu but %llu is required. " + "Disconnecting.\n", SC_NODEF_ARGS(sc), + (unsigned long long)be64_to_cpu(hand->protocol_version), + O2NET_PROTOCOL_VERSION); /* don't bother reconnecting if its the wrong version. */ o2net_ensure_shutdown(nn, sc, -ENOTCONN); @@ -1302,33 +1302,33 @@ static int o2net_check_handshake(struct o2net_sock_container *sc) */ if (be32_to_cpu(hand->o2net_idle_timeout_ms) != o2net_idle_timeout()) { - mlog(ML_NOTICE, SC_NODEF_FMT " uses a network idle timeout of " - "%u ms, but we use %u ms locally. disconnecting\n", - SC_NODEF_ARGS(sc), - be32_to_cpu(hand->o2net_idle_timeout_ms), - o2net_idle_timeout()); + printk(KERN_NOTICE "o2net: " SC_NODEF_FMT " uses a network " + "idle timeout of %u ms, but we use %u ms locally. " + "Disconnecting.\n", SC_NODEF_ARGS(sc), + be32_to_cpu(hand->o2net_idle_timeout_ms), + o2net_idle_timeout()); o2net_ensure_shutdown(nn, sc, -ENOTCONN); return -1; } if (be32_to_cpu(hand->o2net_keepalive_delay_ms) != o2net_keepalive_delay()) { - mlog(ML_NOTICE, SC_NODEF_FMT " uses a keepalive delay of " - "%u ms, but we use %u ms locally. disconnecting\n", - SC_NODEF_ARGS(sc), - be32_to_cpu(hand->o2net_keepalive_delay_ms), - o2net_keepalive_delay()); + printk(KERN_NOTICE "o2net: " SC_NODEF_FMT " uses a keepalive " + "delay of %u ms, but we use %u ms locally. " + "Disconnecting.\n", SC_NODEF_ARGS(sc), + be32_to_cpu(hand->o2net_keepalive_delay_ms), + o2net_keepalive_delay()); o2net_ensure_shutdown(nn, sc, -ENOTCONN); return -1; } if (be32_to_cpu(hand->o2hb_heartbeat_timeout_ms) != O2HB_MAX_WRITE_TIMEOUT_MS) { - mlog(ML_NOTICE, SC_NODEF_FMT " uses a heartbeat timeout of " - "%u ms, but we use %u ms locally. disconnecting\n", - SC_NODEF_ARGS(sc), - be32_to_cpu(hand->o2hb_heartbeat_timeout_ms), - O2HB_MAX_WRITE_TIMEOUT_MS); + printk(KERN_NOTICE "o2net: " SC_NODEF_FMT " uses a heartbeat " + "timeout of %u ms, but we use %u ms locally. " + "Disconnecting.\n", SC_NODEF_ARGS(sc), + be32_to_cpu(hand->o2hb_heartbeat_timeout_ms), + O2HB_MAX_WRITE_TIMEOUT_MS); o2net_ensure_shutdown(nn, sc, -ENOTCONN); return -1; } @@ -1539,28 +1539,16 @@ static void o2net_idle_timer(unsigned long data) { struct o2net_sock_container *sc = (struct o2net_sock_container *)data; struct o2net_node *nn = o2net_nn_from_num(sc->sc_node->nd_num); - #ifdef CONFIG_DEBUG_FS - ktime_t now = ktime_get(); + unsigned long msecs = ktime_to_ms(ktime_get()) - + ktime_to_ms(sc->sc_tv_timer); +#else + unsigned long msecs = o2net_idle_timeout(); #endif - printk(KERN_NOTICE "o2net: connection to " SC_NODEF_FMT " has been idle for %u.%u " - "seconds, shutting it down.\n", SC_NODEF_ARGS(sc), - o2net_idle_timeout() / 1000, - o2net_idle_timeout() % 1000); - -#ifdef CONFIG_DEBUG_FS - mlog(ML_NOTICE, "Here are some times that might help debug the " - "situation: (Timer: %lld, Now %lld, DataReady %lld, Advance %lld-%lld, " - "Key 0x%08x, Func %u, FuncTime %lld-%lld)\n", - (long long)ktime_to_us(sc->sc_tv_timer), (long long)ktime_to_us(now), - (long long)ktime_to_us(sc->sc_tv_data_ready), - (long long)ktime_to_us(sc->sc_tv_advance_start), - (long long)ktime_to_us(sc->sc_tv_advance_stop), - sc->sc_msg_key, sc->sc_msg_type, - (long long)ktime_to_us(sc->sc_tv_func_start), - (long long)ktime_to_us(sc->sc_tv_func_stop)); -#endif + printk(KERN_NOTICE "o2net: Connection to " SC_NODEF_FMT " has been " + "idle for %lu.%lu secs, shutting it down.\n", SC_NODEF_ARGS(sc), + msecs / 1000, msecs % 1000); /* * Initialize the nn_timeout so that the next connection attempt @@ -1693,8 +1681,8 @@ static void o2net_start_connect(struct work_struct *work) out: if (ret) { - mlog(ML_NOTICE, "connect attempt to " SC_NODEF_FMT " failed " - "with errno %d\n", SC_NODEF_ARGS(sc), ret); + printk(KERN_NOTICE "o2net: Connect attempt to " SC_NODEF_FMT + " failed with errno %d\n", SC_NODEF_ARGS(sc), ret); /* 0 err so that another will be queued and attempted * from set_nn_state */ if (sc) @@ -1717,8 +1705,8 @@ static void o2net_connect_expired(struct work_struct *work) spin_lock(&nn->nn_lock); if (!nn->nn_sc_valid) { - mlog(ML_ERROR, "no connection established with node %u after " - "%u.%u seconds, giving up and returning errors.\n", + printk(KERN_NOTICE "o2net: No connection established with " + "node %u after %u.%u seconds, giving up.\n", o2net_num_from_nn(nn), o2net_idle_timeout() / 1000, o2net_idle_timeout() % 1000); @@ -1861,21 +1849,21 @@ static int o2net_accept_one(struct socket *sock) node = o2nm_get_node_by_ip(sin.sin_addr.s_addr); if (node == NULL) { - mlog(ML_NOTICE, "attempt to connect from unknown node at %pI4:%d\n", - &sin.sin_addr.s_addr, ntohs(sin.sin_port)); + printk(KERN_NOTICE "o2net: Attempt to connect from unknown " + "node at %pI4:%d\n", &sin.sin_addr.s_addr, + ntohs(sin.sin_port)); ret = -EINVAL; goto out; } if (o2nm_this_node() >= node->nd_num) { local_node = o2nm_get_node_by_num(o2nm_this_node()); - mlog(ML_NOTICE, "unexpected connect attempt seen at node '%s' (" - "%u, %pI4:%d) from node '%s' (%u, %pI4:%d)\n", - local_node->nd_name, local_node->nd_num, - &(local_node->nd_ipv4_address), - ntohs(local_node->nd_ipv4_port), - node->nd_name, node->nd_num, &sin.sin_addr.s_addr, - ntohs(sin.sin_port)); + printk(KERN_NOTICE "o2net: Unexpected connect attempt seen " + "at node '%s' (%u, %pI4:%d) from node '%s' (%u, " + "%pI4:%d)\n", local_node->nd_name, local_node->nd_num, + &(local_node->nd_ipv4_address), + ntohs(local_node->nd_ipv4_port), node->nd_name, + node->nd_num, &sin.sin_addr.s_addr, ntohs(sin.sin_port)); ret = -EINVAL; goto out; } @@ -1900,10 +1888,10 @@ static int o2net_accept_one(struct socket *sock) ret = 0; spin_unlock(&nn->nn_lock); if (ret) { - mlog(ML_NOTICE, "attempt to connect from node '%s' at " - "%pI4:%d but it already has an open connection\n", - node->nd_name, &sin.sin_addr.s_addr, - ntohs(sin.sin_port)); + printk(KERN_NOTICE "o2net: Attempt to connect from node '%s' " + "at %pI4:%d but it already has an open connection\n", + node->nd_name, &sin.sin_addr.s_addr, + ntohs(sin.sin_port)); goto out; } @@ -1983,7 +1971,7 @@ static int o2net_open_listening_sock(__be32 addr, __be16 port) ret = sock_create(PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock); if (ret < 0) { - mlog(ML_ERROR, "unable to create socket, ret=%d\n", ret); + printk(KERN_ERR "o2net: Error %d while creating socket\n", ret); goto out; } @@ -2000,16 +1988,15 @@ static int o2net_open_listening_sock(__be32 addr, __be16 port) sock->sk->sk_reuse = 1; ret = sock->ops->bind(sock, (struct sockaddr *)&sin, sizeof(sin)); if (ret < 0) { - mlog(ML_ERROR, "unable to bind socket at %pI4:%u, " - "ret=%d\n", &addr, ntohs(port), ret); + printk(KERN_ERR "o2net: Error %d while binding socket at " + "%pI4:%u\n", ret, &addr, ntohs(port)); goto out; } ret = sock->ops->listen(sock, 64); - if (ret < 0) { - mlog(ML_ERROR, "unable to listen on %pI4:%u, ret=%d\n", - &addr, ntohs(port), ret); - } + if (ret < 0) + printk(KERN_ERR "o2net: Error %d while listening on %pI4:%u\n", + ret, &addr, ntohs(port)); out: if (ret) { -- GitLab From 8decab3c8dadcdf4f54ffb30df6e6f67b398b6e0 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:23:54 -0700 Subject: [PATCH 0036/3089] ocfs2/dlm: Clean up messages in o2dlm o2dlm messages needed a facelift. Signed-off-by: Sunil Mushran --- fs/ocfs2/dlm/dlmdomain.c | 18 +++++++------ fs/ocfs2/dlm/dlmlock.c | 22 +++++++--------- fs/ocfs2/dlm/dlmmaster.c | 46 +++++++++++++++------------------ fs/ocfs2/dlm/dlmrecovery.c | 53 ++++++++++++++++++++++---------------- fs/ocfs2/dlm/dlmthread.c | 2 -- 5 files changed, 71 insertions(+), 70 deletions(-) diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 6ed6b95dcf935..ce225fb0cea71 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -539,17 +539,17 @@ again: static void __dlm_print_nodes(struct dlm_ctxt *dlm) { - int node = -1; + int node = -1, num = 0; assert_spin_locked(&dlm->spinlock); - printk(KERN_NOTICE "o2dlm: Nodes in domain %s: ", dlm->name); - + printk("( "); while ((node = find_next_bit(dlm->domain_map, O2NM_MAX_NODES, node + 1)) < O2NM_MAX_NODES) { printk("%d ", node); + ++num; } - printk("\n"); + printk(") %u nodes\n", num); } static int dlm_exit_domain_handler(struct o2net_msg *msg, u32 len, void *data, @@ -566,11 +566,10 @@ static int dlm_exit_domain_handler(struct o2net_msg *msg, u32 len, void *data, node = exit_msg->node_idx; - printk(KERN_NOTICE "o2dlm: Node %u leaves domain %s\n", node, dlm->name); - spin_lock(&dlm->spinlock); clear_bit(node, dlm->domain_map); clear_bit(node, dlm->exit_domain_map); + printk(KERN_NOTICE "o2dlm: Node %u leaves domain %s ", node, dlm->name); __dlm_print_nodes(dlm); /* notify anything attached to the heartbeat events */ @@ -755,6 +754,7 @@ void dlm_unregister_domain(struct dlm_ctxt *dlm) dlm_mark_domain_leaving(dlm); dlm_leave_domain(dlm); + printk(KERN_NOTICE "o2dlm: Leaving domain %s\n", dlm->name); dlm_force_free_mles(dlm); dlm_complete_dlm_shutdown(dlm); } @@ -970,7 +970,7 @@ static int dlm_assert_joined_handler(struct o2net_msg *msg, u32 len, void *data, clear_bit(assert->node_idx, dlm->exit_domain_map); __dlm_set_joining_node(dlm, DLM_LOCK_RES_OWNER_UNKNOWN); - printk(KERN_NOTICE "o2dlm: Node %u joins domain %s\n", + printk(KERN_NOTICE "o2dlm: Node %u joins domain %s ", assert->node_idx, dlm->name); __dlm_print_nodes(dlm); @@ -1701,8 +1701,10 @@ static int dlm_try_to_join_domain(struct dlm_ctxt *dlm) bail: spin_lock(&dlm->spinlock); __dlm_set_joining_node(dlm, DLM_LOCK_RES_OWNER_UNKNOWN); - if (!status) + if (!status) { + printk(KERN_NOTICE "o2dlm: Joining domain %s ", dlm->name); __dlm_print_nodes(dlm); + } spin_unlock(&dlm->spinlock); if (ctxt) { diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c index 8d39e0fd66f73..c7f3e22bda1ea 100644 --- a/fs/ocfs2/dlm/dlmlock.c +++ b/fs/ocfs2/dlm/dlmlock.c @@ -319,27 +319,23 @@ static enum dlm_status dlm_send_remote_lock_request(struct dlm_ctxt *dlm, tmpret = o2net_send_message(DLM_CREATE_LOCK_MSG, dlm->key, &create, sizeof(create), res->owner, &status); if (tmpret >= 0) { - // successfully sent and received - ret = status; // this is already a dlm_status + ret = status; if (ret == DLM_REJECTED) { - mlog(ML_ERROR, "%s:%.*s: BUG. this is a stale lockres " - "no longer owned by %u. that node is coming back " - "up currently.\n", dlm->name, create.namelen, + mlog(ML_ERROR, "%s: res %.*s, Stale lockres no longer " + "owned by node %u. That node is coming back up " + "currently.\n", dlm->name, create.namelen, create.name, res->owner); dlm_print_one_lock_resource(res); BUG(); } } else { - mlog(ML_ERROR, "Error %d when sending message %u (key 0x%x) to " - "node %u\n", tmpret, DLM_CREATE_LOCK_MSG, dlm->key, - res->owner); - if (dlm_is_host_down(tmpret)) { + mlog(ML_ERROR, "%s: res %.*s, Error %d send CREATE LOCK to " + "node %u\n", dlm->name, create.namelen, create.name, + tmpret, res->owner); + if (dlm_is_host_down(tmpret)) ret = DLM_RECOVERING; - mlog(0, "node %u died so returning DLM_RECOVERING " - "from lock message!\n", res->owner); - } else { + else ret = dlm_err_to_dlm_status(tmpret); - } } return ret; diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index 11eefb8c12e98..9f3b093b629ac 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -829,8 +829,8 @@ lookup: * but they might own this lockres. wait on them. */ bit = find_next_bit(dlm->recovery_map, O2NM_MAX_NODES, 0); if (bit < O2NM_MAX_NODES) { - mlog(ML_NOTICE, "%s:%.*s: at least one node (%d) to " - "recover before lock mastery can begin\n", + mlog(0, "%s: res %.*s, At least one node (%d) " + "to recover before lock mastery can begin\n", dlm->name, namelen, (char *)lockid, bit); wait_on_recovery = 1; } @@ -864,8 +864,8 @@ redo_request: * dlm spinlock would be detectable be a change on the mle, * so we only need to clear out the recovery map once. */ if (dlm_is_recovery_lock(lockid, namelen)) { - mlog(ML_NOTICE, "%s: recovery map is not empty, but " - "must master $RECOVERY lock now\n", dlm->name); + mlog(0, "%s: Recovery map is not empty, but must " + "master $RECOVERY lock now\n", dlm->name); if (!dlm_pre_master_reco_lockres(dlm, res)) wait_on_recovery = 0; else { @@ -883,8 +883,8 @@ redo_request: spin_lock(&dlm->spinlock); bit = find_next_bit(dlm->recovery_map, O2NM_MAX_NODES, 0); if (bit < O2NM_MAX_NODES) { - mlog(ML_NOTICE, "%s:%.*s: at least one node (%d) to " - "recover before lock mastery can begin\n", + mlog(0, "%s: res %.*s, At least one node (%d) " + "to recover before lock mastery can begin\n", dlm->name, namelen, (char *)lockid, bit); wait_on_recovery = 1; } else @@ -913,8 +913,8 @@ redo_request: * yet, keep going until it does. this is how the * master will know that asserts are needed back to * the lower nodes. */ - mlog(0, "%s:%.*s: requests only up to %u but master " - "is %u, keep going\n", dlm->name, namelen, + mlog(0, "%s: res %.*s, Requests only up to %u but " + "master is %u, keep going\n", dlm->name, namelen, lockid, nodenum, mle->master); } } @@ -924,13 +924,12 @@ wait: ret = dlm_wait_for_lock_mastery(dlm, res, mle, &blocked); if (ret < 0) { wait_on_recovery = 1; - mlog(0, "%s:%.*s: node map changed, redo the " - "master request now, blocked=%d\n", - dlm->name, res->lockname.len, + mlog(0, "%s: res %.*s, Node map changed, redo the master " + "request now, blocked=%d\n", dlm->name, res->lockname.len, res->lockname.name, blocked); if (++tries > 20) { - mlog(ML_ERROR, "%s:%.*s: spinning on " - "dlm_wait_for_lock_mastery, blocked=%d\n", + mlog(ML_ERROR, "%s: res %.*s, Spinning on " + "dlm_wait_for_lock_mastery, blocked = %d\n", dlm->name, res->lockname.len, res->lockname.name, blocked); dlm_print_one_lock_resource(res); @@ -940,7 +939,8 @@ wait: goto redo_request; } - mlog(0, "lockres mastered by %u\n", res->owner); + mlog(0, "%s: res %.*s, Mastered by %u\n", dlm->name, res->lockname.len, + res->lockname.name, res->owner); /* make sure we never continue without this */ BUG_ON(res->owner == O2NM_MAX_NODES); @@ -2187,8 +2187,6 @@ int dlm_drop_lockres_ref(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) namelen = res->lockname.len; BUG_ON(namelen > O2NM_MAX_NAME_LEN); - mlog(0, "%s:%.*s: sending deref to %d\n", - dlm->name, namelen, lockname, res->owner); memset(&deref, 0, sizeof(deref)); deref.node_idx = dlm->node_num; deref.namelen = namelen; @@ -2197,14 +2195,12 @@ int dlm_drop_lockres_ref(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) ret = o2net_send_message(DLM_DEREF_LOCKRES_MSG, dlm->key, &deref, sizeof(deref), res->owner, &r); if (ret < 0) - mlog(ML_ERROR, "Error %d when sending message %u (key 0x%x) to " - "node %u\n", ret, DLM_DEREF_LOCKRES_MSG, dlm->key, - res->owner); + mlog(ML_ERROR, "%s: res %.*s, error %d send DEREF to node %u\n", + dlm->name, namelen, lockname, ret, res->owner); else if (r < 0) { /* BAD. other node says I did not have a ref. */ - mlog(ML_ERROR,"while dropping ref on %s:%.*s " - "(master=%u) got %d.\n", dlm->name, namelen, - lockname, res->owner, r); + mlog(ML_ERROR, "%s: res %.*s, DEREF to node %u got %d\n", + dlm->name, namelen, lockname, res->owner, r); dlm_print_one_lock_resource(res); BUG(); } @@ -2916,9 +2912,9 @@ static int dlm_do_migrate_request(struct dlm_ctxt *dlm, &migrate, sizeof(migrate), nodenum, &status); if (ret < 0) { - mlog(ML_ERROR, "Error %d when sending message %u (key " - "0x%x) to node %u\n", ret, DLM_MIGRATE_REQUEST_MSG, - dlm->key, nodenum); + mlog(ML_ERROR, "%s: res %.*s, Error %d send " + "MIGRATE_REQUEST to node %u\n", dlm->name, + migrate.namelen, migrate.name, ret, nodenum); if (!dlm_is_host_down(ret)) { mlog(ML_ERROR, "unhandled error=%d!\n", ret); BUG(); diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index 7efab6d28a21b..a3c312c43b90e 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -430,6 +430,8 @@ static void dlm_begin_recovery(struct dlm_ctxt *dlm) { spin_lock(&dlm->spinlock); BUG_ON(dlm->reco.state & DLM_RECO_STATE_ACTIVE); + printk(KERN_NOTICE "o2dlm: Begin recovery on domain %s for node %u\n", + dlm->name, dlm->reco.dead_node); dlm->reco.state |= DLM_RECO_STATE_ACTIVE; spin_unlock(&dlm->spinlock); } @@ -440,9 +442,18 @@ static void dlm_end_recovery(struct dlm_ctxt *dlm) BUG_ON(!(dlm->reco.state & DLM_RECO_STATE_ACTIVE)); dlm->reco.state &= ~DLM_RECO_STATE_ACTIVE; spin_unlock(&dlm->spinlock); + printk(KERN_NOTICE "o2dlm: End recovery on domain %s\n", dlm->name); wake_up(&dlm->reco.event); } +static void dlm_print_recovery_master(struct dlm_ctxt *dlm) +{ + printk(KERN_NOTICE "o2dlm: Node %u (%s) is the Recovery Master for the " + "dead node %u in domain %s\n", dlm->reco.new_master, + (dlm->node_num == dlm->reco.new_master ? "me" : "he"), + dlm->reco.dead_node, dlm->name); +} + static int dlm_do_recovery(struct dlm_ctxt *dlm) { int status = 0; @@ -505,9 +516,8 @@ static int dlm_do_recovery(struct dlm_ctxt *dlm) } mlog(0, "another node will master this recovery session.\n"); } - mlog(0, "dlm=%s (%d), new_master=%u, this node=%u, dead_node=%u\n", - dlm->name, task_pid_nr(dlm->dlm_reco_thread_task), dlm->reco.new_master, - dlm->node_num, dlm->reco.dead_node); + + dlm_print_recovery_master(dlm); /* it is safe to start everything back up here * because all of the dead node's lock resources @@ -518,15 +528,13 @@ static int dlm_do_recovery(struct dlm_ctxt *dlm) return 0; master_here: - mlog(ML_NOTICE, "(%d) Node %u is the Recovery Master for the Dead Node " - "%u for Domain %s\n", task_pid_nr(dlm->dlm_reco_thread_task), - dlm->node_num, dlm->reco.dead_node, dlm->name); + dlm_print_recovery_master(dlm); status = dlm_remaster_locks(dlm, dlm->reco.dead_node); if (status < 0) { /* we should never hit this anymore */ - mlog(ML_ERROR, "error %d remastering locks for node %u, " - "retrying.\n", status, dlm->reco.dead_node); + mlog(ML_ERROR, "%s: Error %d remastering locks for node %u, " + "retrying.\n", dlm->name, status, dlm->reco.dead_node); /* yield a bit to allow any final network messages * to get handled on remaining nodes */ msleep(100); @@ -567,7 +575,7 @@ static int dlm_remaster_locks(struct dlm_ctxt *dlm, u8 dead_node) BUG_ON(ndata->state != DLM_RECO_NODE_DATA_INIT); ndata->state = DLM_RECO_NODE_DATA_REQUESTING; - mlog(0, "requesting lock info from node %u\n", + mlog(0, "%s: Requesting lock info from node %u\n", dlm->name, ndata->node_num); if (ndata->node_num == dlm->node_num) { @@ -640,7 +648,7 @@ static int dlm_remaster_locks(struct dlm_ctxt *dlm, u8 dead_node) spin_unlock(&dlm_reco_state_lock); } - mlog(0, "done requesting all lock info\n"); + mlog(0, "%s: Done requesting all lock info\n", dlm->name); /* nodes should be sending reco data now * just need to wait */ @@ -802,10 +810,9 @@ static int dlm_request_all_locks(struct dlm_ctxt *dlm, u8 request_from, /* negative status is handled by caller */ if (ret < 0) - mlog(ML_ERROR, "Error %d when sending message %u (key " - "0x%x) to node %u\n", ret, DLM_LOCK_REQUEST_MSG, - dlm->key, request_from); - + mlog(ML_ERROR, "%s: Error %d send LOCK_REQUEST to node %u " + "to recover dead node %u\n", dlm->name, ret, + request_from, dead_node); // return from here, then // sleep until all received or error return ret; @@ -956,9 +963,9 @@ static int dlm_send_all_done_msg(struct dlm_ctxt *dlm, u8 dead_node, u8 send_to) ret = o2net_send_message(DLM_RECO_DATA_DONE_MSG, dlm->key, &done_msg, sizeof(done_msg), send_to, &tmpret); if (ret < 0) { - mlog(ML_ERROR, "Error %d when sending message %u (key " - "0x%x) to node %u\n", ret, DLM_RECO_DATA_DONE_MSG, - dlm->key, send_to); + mlog(ML_ERROR, "%s: Error %d send RECO_DATA_DONE to node %u " + "to recover dead node %u\n", dlm->name, ret, send_to, + dead_node); if (!dlm_is_host_down(ret)) { BUG(); } @@ -1127,9 +1134,11 @@ static int dlm_send_mig_lockres_msg(struct dlm_ctxt *dlm, if (ret < 0) { /* XXX: negative status is not handled. * this will end up killing this node. */ - mlog(ML_ERROR, "Error %d when sending message %u (key " - "0x%x) to node %u\n", ret, DLM_MIG_LOCKRES_MSG, - dlm->key, send_to); + mlog(ML_ERROR, "%s: res %.*s, Error %d send MIG_LOCKRES to " + "node %u (%s)\n", dlm->name, mres->lockname_len, + mres->lockname, ret, send_to, + (orig_flags & DLM_MRES_MIGRATION ? + "migration" : "recovery")); } else { /* might get an -ENOMEM back here */ ret = status; @@ -2324,9 +2333,9 @@ static void dlm_do_local_recovery_cleanup(struct dlm_ctxt *dlm, u8 dead_node) dlm_revalidate_lvb(dlm, res, dead_node); if (res->owner == dead_node) { if (res->state & DLM_LOCK_RES_DROPPING_REF) { - mlog(ML_NOTICE, "Ignore %.*s for " + mlog(ML_NOTICE, "%s: res %.*s, Skip " "recovery as it is being freed\n", - res->lockname.len, + dlm->name, res->lockname.len, res->lockname.name); } else dlm_move_lockres_to_recovery_list(dlm, diff --git a/fs/ocfs2/dlm/dlmthread.c b/fs/ocfs2/dlm/dlmthread.c index 1d6d1d22c4715..46faac20f16c9 100644 --- a/fs/ocfs2/dlm/dlmthread.c +++ b/fs/ocfs2/dlm/dlmthread.c @@ -185,8 +185,6 @@ static void dlm_purge_lockres(struct dlm_ctxt *dlm, /* clear our bit from the master's refmap, ignore errors */ ret = dlm_drop_lockres_ref(dlm, res); if (ret < 0) { - mlog(ML_ERROR, "%s: deref %.*s failed %d\n", dlm->name, - res->lockname.len, res->lockname.name, ret); if (!dlm_is_host_down(ret)) BUG(); } -- GitLab From 394eb3d38a3ecc549cc34a3040103a9164be516b Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:24:54 -0700 Subject: [PATCH 0037/3089] ocfs2: Clean up messages in stack_o2cb.c o2cb messages needed a facelift. Signed-off-by: Sunil Mushran --- fs/ocfs2/stack_o2cb.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/ocfs2/stack_o2cb.c b/fs/ocfs2/stack_o2cb.c index 19965b00c43ca..be935d89a4e88 100644 --- a/fs/ocfs2/stack_o2cb.c +++ b/fs/ocfs2/stack_o2cb.c @@ -263,8 +263,8 @@ static void o2dlm_eviction_cb(int node_num, void *data) { struct ocfs2_cluster_connection *conn = data; - mlog(ML_NOTICE, "o2dlm has evicted node %d from group %.*s\n", - node_num, conn->cc_namelen, conn->cc_name); + printk(KERN_NOTICE "o2cb: o2dlm has evicted node %d from domain %.*s\n", + node_num, conn->cc_namelen, conn->cc_name); conn->cc_recovery_handler(node_num, conn->cc_recovery_data); } -- GitLab From 0afbba13226fcdbd4327f6b13a42f6efbb8c9caf Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:25:54 -0700 Subject: [PATCH 0038/3089] ocfs2/dlm: Cleanup up dlm_finish_local_lockres_recovery() dlm_finish_local_lockres_recovery() needed a facelift. Signed-off-by: Sunil Mushran --- fs/ocfs2/dlm/dlmrecovery.c | 57 +++++++++++++++++--------------------- 1 file changed, 25 insertions(+), 32 deletions(-) diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index a3c312c43b90e..41e74f0e2d35c 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -2093,6 +2093,9 @@ static void dlm_finish_local_lockres_recovery(struct dlm_ctxt *dlm, list_for_each_entry_safe(res, next, &dlm->reco.resources, recovering) { if (res->owner == dead_node) { + mlog(0, "%s: res %.*s, Changing owner from %u to %u\n", + dlm->name, res->lockname.len, res->lockname.name, + res->owner, new_master); list_del_init(&res->recovering); spin_lock(&res->spinlock); /* new_master has our reference from @@ -2114,40 +2117,30 @@ static void dlm_finish_local_lockres_recovery(struct dlm_ctxt *dlm, for (i = 0; i < DLM_HASH_BUCKETS; i++) { bucket = dlm_lockres_hash(dlm, i); hlist_for_each_entry(res, hash_iter, bucket, hash_node) { - if (res->state & DLM_LOCK_RES_RECOVERING) { - if (res->owner == dead_node) { - mlog(0, "(this=%u) res %.*s owner=%u " - "was not on recovering list, but " - "clearing state anyway\n", - dlm->node_num, res->lockname.len, - res->lockname.name, new_master); - } else if (res->owner == dlm->node_num) { - mlog(0, "(this=%u) res %.*s owner=%u " - "was not on recovering list, " - "owner is THIS node, clearing\n", - dlm->node_num, res->lockname.len, - res->lockname.name, new_master); - } else - continue; + if (!(res->state & DLM_LOCK_RES_RECOVERING)) + continue; - if (!list_empty(&res->recovering)) { - mlog(0, "%s:%.*s: lockres was " - "marked RECOVERING, owner=%u\n", - dlm->name, res->lockname.len, - res->lockname.name, res->owner); - list_del_init(&res->recovering); - dlm_lockres_put(res); - } - spin_lock(&res->spinlock); - /* new_master has our reference from - * the lock state sent during recovery */ - dlm_change_lockres_owner(dlm, res, new_master); - res->state &= ~DLM_LOCK_RES_RECOVERING; - if (__dlm_lockres_has_locks(res)) - __dlm_dirty_lockres(dlm, res); - spin_unlock(&res->spinlock); - wake_up(&res->wq); + if (res->owner != dead_node && + res->owner != dlm->node_num) + continue; + + if (!list_empty(&res->recovering)) { + list_del_init(&res->recovering); + dlm_lockres_put(res); } + + /* new_master has our reference from + * the lock state sent during recovery */ + mlog(0, "%s: res %.*s, Changing owner from %u to %u\n", + dlm->name, res->lockname.len, res->lockname.name, + res->owner, new_master); + spin_lock(&res->spinlock); + dlm_change_lockres_owner(dlm, res, new_master); + res->state &= ~DLM_LOCK_RES_RECOVERING; + if (__dlm_lockres_has_locks(res)) + __dlm_dirty_lockres(dlm, res); + spin_unlock(&res->spinlock); + wake_up(&res->wq); } } } -- GitLab From 8d400b81cc83b171ff872587723a37eb7fae9abe Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:26:54 -0700 Subject: [PATCH 0039/3089] ocfs2/dlm: Clean up refmap helpers Patch cleans up helpers that set/clear refmap bits and grab/drop inflight lock ref counts. Signed-off-by: Sunil Mushran --- fs/ocfs2/dlm/dlmcommon.h | 49 ++++----------------- fs/ocfs2/dlm/dlmmaster.c | 88 +++++++++++++++++++++++--------------- fs/ocfs2/dlm/dlmrecovery.c | 8 ++-- 3 files changed, 66 insertions(+), 79 deletions(-) diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h index d602abb51b610..4a5d5804b1309 100644 --- a/fs/ocfs2/dlm/dlmcommon.h +++ b/fs/ocfs2/dlm/dlmcommon.h @@ -902,46 +902,15 @@ struct dlm_lock_resource *dlm_new_lockres(struct dlm_ctxt *dlm, const char *name, unsigned int namelen); -#define dlm_lockres_set_refmap_bit(bit,res) \ - __dlm_lockres_set_refmap_bit(bit,res,__FILE__,__LINE__) -#define dlm_lockres_clear_refmap_bit(bit,res) \ - __dlm_lockres_clear_refmap_bit(bit,res,__FILE__,__LINE__) - -static inline void __dlm_lockres_set_refmap_bit(int bit, - struct dlm_lock_resource *res, - const char *file, - int line) -{ - //printk("%s:%d:%.*s: setting bit %d\n", file, line, - // res->lockname.len, res->lockname.name, bit); - set_bit(bit, res->refmap); -} - -static inline void __dlm_lockres_clear_refmap_bit(int bit, - struct dlm_lock_resource *res, - const char *file, - int line) -{ - //printk("%s:%d:%.*s: clearing bit %d\n", file, line, - // res->lockname.len, res->lockname.name, bit); - clear_bit(bit, res->refmap); -} - -void __dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res, - const char *file, - int line); -void __dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res, - int new_lockres, - const char *file, - int line); -#define dlm_lockres_drop_inflight_ref(d,r) \ - __dlm_lockres_drop_inflight_ref(d,r,__FILE__,__LINE__) -#define dlm_lockres_grab_inflight_ref(d,r) \ - __dlm_lockres_grab_inflight_ref(d,r,0,__FILE__,__LINE__) -#define dlm_lockres_grab_inflight_ref_new(d,r) \ - __dlm_lockres_grab_inflight_ref(d,r,1,__FILE__,__LINE__) +void dlm_lockres_set_refmap_bit(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res, int bit); +void dlm_lockres_clear_refmap_bit(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res, int bit); + +void dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res); +void dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res); void dlm_queue_ast(struct dlm_ctxt *dlm, struct dlm_lock *lock); void dlm_queue_bast(struct dlm_ctxt *dlm, struct dlm_lock *lock); diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index 9f3b093b629ac..11e446ffb67a9 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -631,39 +631,58 @@ error: return NULL; } -void __dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res, - int new_lockres, - const char *file, - int line) +void dlm_lockres_set_refmap_bit(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res, int bit) { - if (!new_lockres) - assert_spin_locked(&res->spinlock); + assert_spin_locked(&res->spinlock); + + mlog(0, "res %.*s, set node %u, %ps()\n", res->lockname.len, + res->lockname.name, bit, __builtin_return_address(0)); + + set_bit(bit, res->refmap); +} + +void dlm_lockres_clear_refmap_bit(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res, int bit) +{ + assert_spin_locked(&res->spinlock); + + mlog(0, "res %.*s, clr node %u, %ps()\n", res->lockname.len, + res->lockname.name, bit, __builtin_return_address(0)); + + clear_bit(bit, res->refmap); +} + + +void dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res) +{ + assert_spin_locked(&res->spinlock); if (!test_bit(dlm->node_num, res->refmap)) { BUG_ON(res->inflight_locks != 0); - dlm_lockres_set_refmap_bit(dlm->node_num, res); + dlm_lockres_set_refmap_bit(dlm, res, dlm->node_num); } res->inflight_locks++; - mlog(0, "%s:%.*s: inflight++: now %u\n", - dlm->name, res->lockname.len, res->lockname.name, - res->inflight_locks); + mlog(0, "%s: res %.*s, inflight++: now %u, %ps()\n", dlm->name, + res->lockname.len, res->lockname.name, res->inflight_locks, + __builtin_return_address(0)); } -void __dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res, - const char *file, - int line) +void dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm, + struct dlm_lock_resource *res) { assert_spin_locked(&res->spinlock); BUG_ON(res->inflight_locks == 0); + res->inflight_locks--; - mlog(0, "%s:%.*s: inflight--: now %u\n", - dlm->name, res->lockname.len, res->lockname.name, - res->inflight_locks); + mlog(0, "%s: res %.*s, inflight--: now %u, %ps()\n", dlm->name, + res->lockname.len, res->lockname.name, res->inflight_locks, + __builtin_return_address(0)); + if (res->inflight_locks == 0) - dlm_lockres_clear_refmap_bit(dlm->node_num, res); + dlm_lockres_clear_refmap_bit(dlm, res, dlm->node_num); wake_up(&res->wq); } @@ -843,8 +862,10 @@ lookup: /* finally add the lockres to its hash bucket */ __dlm_insert_lockres(dlm, res); - /* since this lockres is new it doesn't not require the spinlock */ - dlm_lockres_grab_inflight_ref_new(dlm, res); + + spin_lock(&res->spinlock); + dlm_lockres_grab_inflight_ref(dlm, res); + spin_unlock(&res->spinlock); /* if this node does not become the master make sure to drop * this inflight reference below */ @@ -1426,9 +1447,7 @@ way_up_top: } if (res->owner == dlm->node_num) { - mlog(0, "%s:%.*s: setting bit %u in refmap\n", - dlm->name, namelen, name, request->node_idx); - dlm_lockres_set_refmap_bit(request->node_idx, res); + dlm_lockres_set_refmap_bit(dlm, res, request->node_idx); spin_unlock(&res->spinlock); response = DLM_MASTER_RESP_YES; if (mle) @@ -1493,10 +1512,8 @@ way_up_top: * go back and clean the mles on any * other nodes */ dispatch_assert = 1; - dlm_lockres_set_refmap_bit(request->node_idx, res); - mlog(0, "%s:%.*s: setting bit %u in refmap\n", - dlm->name, namelen, name, - request->node_idx); + dlm_lockres_set_refmap_bit(dlm, res, + request->node_idx); } else response = DLM_MASTER_RESP_NO; } else { @@ -1702,7 +1719,7 @@ again: "lockres, set the bit in the refmap\n", namelen, lockname, to); spin_lock(&res->spinlock); - dlm_lockres_set_refmap_bit(to, res); + dlm_lockres_set_refmap_bit(dlm, res, to); spin_unlock(&res->spinlock); } } @@ -2256,7 +2273,7 @@ int dlm_deref_lockres_handler(struct o2net_msg *msg, u32 len, void *data, else { BUG_ON(res->state & DLM_LOCK_RES_DROPPING_REF); if (test_bit(node, res->refmap)) { - dlm_lockres_clear_refmap_bit(node, res); + dlm_lockres_clear_refmap_bit(dlm, res, node); cleared = 1; } } @@ -2316,7 +2333,7 @@ static void dlm_deref_lockres_worker(struct dlm_work_item *item, void *data) BUG_ON(res->state & DLM_LOCK_RES_DROPPING_REF); if (test_bit(node, res->refmap)) { __dlm_wait_on_lockres_flags(res, DLM_LOCK_RES_SETREF_INPROG); - dlm_lockres_clear_refmap_bit(node, res); + dlm_lockres_clear_refmap_bit(dlm, res, node); cleared = 1; } spin_unlock(&res->spinlock); @@ -2798,7 +2815,8 @@ static void dlm_remove_nonlocal_locks(struct dlm_ctxt *dlm, BUG_ON(!list_empty(&lock->bast_list)); BUG_ON(lock->ast_pending); BUG_ON(lock->bast_pending); - dlm_lockres_clear_refmap_bit(lock->ml.node, res); + dlm_lockres_clear_refmap_bit(dlm, res, + lock->ml.node); list_del_init(&lock->list); dlm_lock_put(lock); /* In a normal unlock, we would have added a @@ -2819,7 +2837,7 @@ static void dlm_remove_nonlocal_locks(struct dlm_ctxt *dlm, mlog(0, "%s:%.*s: node %u had a ref to this " "migrating lockres, clearing\n", dlm->name, res->lockname.len, res->lockname.name, bit); - dlm_lockres_clear_refmap_bit(bit, res); + dlm_lockres_clear_refmap_bit(dlm, res, bit); } bit++; } @@ -2933,7 +2951,7 @@ static int dlm_do_migrate_request(struct dlm_ctxt *dlm, dlm->name, res->lockname.len, res->lockname.name, nodenum); spin_lock(&res->spinlock); - dlm_lockres_set_refmap_bit(nodenum, res); + dlm_lockres_set_refmap_bit(dlm, res, nodenum); spin_unlock(&res->spinlock); } } @@ -3267,7 +3285,7 @@ int dlm_finish_migration(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, * mastery reference here since old_master will briefly have * a reference after the migration completes */ spin_lock(&res->spinlock); - dlm_lockres_set_refmap_bit(old_master, res); + dlm_lockres_set_refmap_bit(dlm, res, old_master); spin_unlock(&res->spinlock); mlog(0, "now time to do a migrate request to other nodes\n"); diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index 41e74f0e2d35c..f1a3aa84ad898 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -1776,7 +1776,7 @@ static int dlm_process_recovery_data(struct dlm_ctxt *dlm, dlm->name, mres->lockname_len, mres->lockname, from); spin_lock(&res->spinlock); - dlm_lockres_set_refmap_bit(from, res); + dlm_lockres_set_refmap_bit(dlm, res, from); spin_unlock(&res->spinlock); added++; break; @@ -1974,7 +1974,7 @@ skip_lvb: mlog(0, "%s:%.*s: added lock for node %u, " "setting refmap bit\n", dlm->name, res->lockname.len, res->lockname.name, ml->node); - dlm_lockres_set_refmap_bit(ml->node, res); + dlm_lockres_set_refmap_bit(dlm, res, ml->node); added++; } spin_unlock(&res->spinlock); @@ -2254,12 +2254,12 @@ static void dlm_free_dead_locks(struct dlm_ctxt *dlm, res->lockname.len, res->lockname.name, freed, dead_node); __dlm_print_one_lock_resource(res); } - dlm_lockres_clear_refmap_bit(dead_node, res); + dlm_lockres_clear_refmap_bit(dlm, res, dead_node); } else if (test_bit(dead_node, res->refmap)) { mlog(0, "%s:%.*s: dead node %u had a ref, but had " "no locks and had not purged before dying\n", dlm->name, res->lockname.len, res->lockname.name, dead_node); - dlm_lockres_clear_refmap_bit(dead_node, res); + dlm_lockres_clear_refmap_bit(dlm, res, dead_node); } /* do not kick thread yet */ -- GitLab From e9f0b6a6233105c188064328c3f15e21942d0241 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:27:54 -0700 Subject: [PATCH 0040/3089] ocfs2/dlm: Trace insert/remove of resource to/from hash Add mlog to trace adding and removing the resource from/to the hash table. Signed-off-by: Sunil Mushran --- fs/ocfs2/dlm/dlmcommon.h | 5 ++--- fs/ocfs2/dlm/dlmdomain.c | 19 ++++++++++++------- fs/ocfs2/dlm/dlmthread.c | 2 +- 3 files changed, 15 insertions(+), 11 deletions(-) diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h index 4a5d5804b1309..b1748ac747518 100644 --- a/fs/ocfs2/dlm/dlmcommon.h +++ b/fs/ocfs2/dlm/dlmcommon.h @@ -877,9 +877,8 @@ static inline void dlm_lockres_get(struct dlm_lock_resource *res) kref_get(&res->refs); } void dlm_lockres_put(struct dlm_lock_resource *res); -void __dlm_unhash_lockres(struct dlm_lock_resource *res); -void __dlm_insert_lockres(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res); +void __dlm_unhash_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res); +void __dlm_insert_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res); struct dlm_lock_resource * __dlm_lookup_lockres_full(struct dlm_ctxt *dlm, const char *name, unsigned int len, diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index ce225fb0cea71..200d9888ffee6 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -157,16 +157,18 @@ static int dlm_protocol_compare(struct dlm_protocol_version *existing, static void dlm_unregister_domain_handlers(struct dlm_ctxt *dlm); -void __dlm_unhash_lockres(struct dlm_lock_resource *lockres) +void __dlm_unhash_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { - if (!hlist_unhashed(&lockres->hash_node)) { - hlist_del_init(&lockres->hash_node); - dlm_lockres_put(lockres); - } + if (hlist_unhashed(&res->hash_node)) + return; + + mlog(0, "%s: Unhash res %.*s\n", dlm->name, res->lockname.len, + res->lockname.name); + hlist_del_init(&res->hash_node); + dlm_lockres_put(res); } -void __dlm_insert_lockres(struct dlm_ctxt *dlm, - struct dlm_lock_resource *res) +void __dlm_insert_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { struct hlist_head *bucket; struct qstr *q; @@ -180,6 +182,9 @@ void __dlm_insert_lockres(struct dlm_ctxt *dlm, dlm_lockres_get(res); hlist_add_head(&res->hash_node, bucket); + + mlog(0, "%s: Hash res %.*s\n", dlm->name, res->lockname.len, + res->lockname.name); } struct dlm_lock_resource * __dlm_lookup_lockres_full(struct dlm_ctxt *dlm, diff --git a/fs/ocfs2/dlm/dlmthread.c b/fs/ocfs2/dlm/dlmthread.c index 46faac20f16c9..4eff65e7e59e0 100644 --- a/fs/ocfs2/dlm/dlmthread.c +++ b/fs/ocfs2/dlm/dlmthread.c @@ -207,7 +207,7 @@ static void dlm_purge_lockres(struct dlm_ctxt *dlm, BUG(); } - __dlm_unhash_lockres(res); + __dlm_unhash_lockres(dlm, res); /* lockres is not in the hash now. drop the flag and wake up * any processes waiting in dlm_get_lock_resource. */ -- GitLab From ed8625c6fb93d750ed022db571a8a7b7a6724b3b Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:28:54 -0700 Subject: [PATCH 0041/3089] ocfs2/dlm: Cleanup dlm_wait_for_node_death() and dlm_wait_for_node_recovery() dlm_wait_for_node_death() and dlm_wait_for_node_recovery() needed a facelift. Signed-off-by: Sunil Mushran --- fs/ocfs2/dlm/dlmcommon.h | 4 ++-- fs/ocfs2/dlm/dlmrecovery.c | 46 ++++++++++++++++++-------------------- 2 files changed, 24 insertions(+), 26 deletions(-) diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h index b1748ac747518..a5952ceecba5a 100644 --- a/fs/ocfs2/dlm/dlmcommon.h +++ b/fs/ocfs2/dlm/dlmcommon.h @@ -859,8 +859,8 @@ void dlm_complete_recovery_thread(struct dlm_ctxt *dlm); void dlm_wait_for_recovery(struct dlm_ctxt *dlm); void dlm_kick_recovery_thread(struct dlm_ctxt *dlm); int dlm_is_node_dead(struct dlm_ctxt *dlm, u8 node); -int dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout); -int dlm_wait_for_node_recovery(struct dlm_ctxt *dlm, u8 node, int timeout); +void dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout); +void dlm_wait_for_node_recovery(struct dlm_ctxt *dlm, u8 node, int timeout); void dlm_put(struct dlm_ctxt *dlm); struct dlm_ctxt *dlm_grab(struct dlm_ctxt *dlm); diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index f1a3aa84ad898..01ebfd0bdad72 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -362,40 +362,38 @@ static int dlm_is_node_recovered(struct dlm_ctxt *dlm, u8 node) } -int dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout) +void dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout) { - if (timeout) { - mlog(ML_NOTICE, "%s: waiting %dms for notification of " - "death of node %u\n", dlm->name, timeout, node); + if (dlm_is_node_dead(dlm, node)) + return; + + printk(KERN_NOTICE "o2dlm: Waiting on the death of node %u in " + "domain %s\n", node, dlm->name); + + if (timeout) wait_event_timeout(dlm->dlm_reco_thread_wq, - dlm_is_node_dead(dlm, node), - msecs_to_jiffies(timeout)); - } else { - mlog(ML_NOTICE, "%s: waiting indefinitely for notification " - "of death of node %u\n", dlm->name, node); + dlm_is_node_dead(dlm, node), + msecs_to_jiffies(timeout)); + else wait_event(dlm->dlm_reco_thread_wq, dlm_is_node_dead(dlm, node)); - } - /* for now, return 0 */ - return 0; } -int dlm_wait_for_node_recovery(struct dlm_ctxt *dlm, u8 node, int timeout) +void dlm_wait_for_node_recovery(struct dlm_ctxt *dlm, u8 node, int timeout) { - if (timeout) { - mlog(0, "%s: waiting %dms for notification of " - "recovery of node %u\n", dlm->name, timeout, node); + if (dlm_is_node_recovered(dlm, node)) + return; + + printk(KERN_NOTICE "o2dlm: Waiting on the recovery of node %u in " + "domain %s\n", node, dlm->name); + + if (timeout) wait_event_timeout(dlm->dlm_reco_thread_wq, - dlm_is_node_recovered(dlm, node), - msecs_to_jiffies(timeout)); - } else { - mlog(0, "%s: waiting indefinitely for notification " - "of recovery of node %u\n", dlm->name, node); + dlm_is_node_recovered(dlm, node), + msecs_to_jiffies(timeout)); + else wait_event(dlm->dlm_reco_thread_wq, dlm_is_node_recovered(dlm, node)); - } - /* for now, return 0 */ - return 0; } /* callers of the top-level api calls (dlmlock/dlmunlock) should -- GitLab From ff0a522e7db79625aa27a433467eb94c5e255718 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:29:54 -0700 Subject: [PATCH 0042/3089] ocfs2/dlm: Take inflight reference count for remotely mastered resources too The inflight reference count, in the lock resource, is taken to pin the resource in memory. We take it when a new resource is created and release it after a lock is attached to it. We do this to prevent the resource from getting purged prematurely. Earlier this reference count was being taken for locally mastered resources only. This patch extends the same functionality for remotely mastered ones. We are doing this because the same premature purging could occur for remotely mastered resources if the remote node were to die before completion of the create lock. Fix for Oracle bug#12405575. Signed-off-by: Sunil Mushran --- fs/ocfs2/dlm/dlmlock.c | 12 ++++++---- fs/ocfs2/dlm/dlmmaster.c | 47 +++++++++++++++------------------------- fs/ocfs2/dlm/dlmthread.c | 12 +++++----- 3 files changed, 32 insertions(+), 39 deletions(-) diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c index c7f3e22bda1ea..3ef2c1adfb8ff 100644 --- a/fs/ocfs2/dlm/dlmlock.c +++ b/fs/ocfs2/dlm/dlmlock.c @@ -183,10 +183,6 @@ static enum dlm_status dlmlock_master(struct dlm_ctxt *dlm, kick_thread = 1; } } - /* reduce the inflight count, this may result in the lockres - * being purged below during calc_usage */ - if (lock->ml.node == dlm->node_num) - dlm_lockres_drop_inflight_ref(dlm, res); spin_unlock(&res->spinlock); wake_up(&res->wq); @@ -737,6 +733,14 @@ retry_lock: } } + /* Inflight taken in dlm_get_lock_resource() is dropped here */ + spin_lock(&res->spinlock); + dlm_lockres_drop_inflight_ref(dlm, res); + spin_unlock(&res->spinlock); + + dlm_lockres_calc_usage(dlm, res); + dlm_kick_thread(dlm, res); + if (status != DLM_NORMAL) { lock->lksb->flags &= ~DLM_LKSB_GET_LVB; if (status != DLM_NOTQUEUED) diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index 11e446ffb67a9..005261c333b09 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -659,11 +659,8 @@ void dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm, { assert_spin_locked(&res->spinlock); - if (!test_bit(dlm->node_num, res->refmap)) { - BUG_ON(res->inflight_locks != 0); - dlm_lockres_set_refmap_bit(dlm, res, dlm->node_num); - } res->inflight_locks++; + mlog(0, "%s: res %.*s, inflight++: now %u, %ps()\n", dlm->name, res->lockname.len, res->lockname.name, res->inflight_locks, __builtin_return_address(0)); @@ -677,12 +674,11 @@ void dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm, BUG_ON(res->inflight_locks == 0); res->inflight_locks--; + mlog(0, "%s: res %.*s, inflight--: now %u, %ps()\n", dlm->name, res->lockname.len, res->lockname.name, res->inflight_locks, __builtin_return_address(0)); - if (res->inflight_locks == 0) - dlm_lockres_clear_refmap_bit(dlm, res, dlm->node_num); wake_up(&res->wq); } @@ -716,7 +712,6 @@ struct dlm_lock_resource * dlm_get_lock_resource(struct dlm_ctxt *dlm, unsigned int hash; int tries = 0; int bit, wait_on_recovery = 0; - int drop_inflight_if_nonlocal = 0; BUG_ON(!lockid); @@ -728,36 +723,33 @@ lookup: spin_lock(&dlm->spinlock); tmpres = __dlm_lookup_lockres_full(dlm, lockid, namelen, hash); if (tmpres) { - int dropping_ref = 0; - spin_unlock(&dlm->spinlock); - spin_lock(&tmpres->spinlock); - /* We wait for the other thread that is mastering the resource */ + /* Wait on the thread that is mastering the resource */ if (tmpres->owner == DLM_LOCK_RES_OWNER_UNKNOWN) { __dlm_wait_on_lockres(tmpres); BUG_ON(tmpres->owner == DLM_LOCK_RES_OWNER_UNKNOWN); + spin_unlock(&tmpres->spinlock); + dlm_lockres_put(tmpres); + tmpres = NULL; + goto lookup; } - if (tmpres->owner == dlm->node_num) { - BUG_ON(tmpres->state & DLM_LOCK_RES_DROPPING_REF); - dlm_lockres_grab_inflight_ref(dlm, tmpres); - } else if (tmpres->state & DLM_LOCK_RES_DROPPING_REF) - dropping_ref = 1; - spin_unlock(&tmpres->spinlock); - - /* wait until done messaging the master, drop our ref to allow - * the lockres to be purged, start over. */ - if (dropping_ref) { - spin_lock(&tmpres->spinlock); - __dlm_wait_on_lockres_flags(tmpres, DLM_LOCK_RES_DROPPING_REF); + /* Wait on the resource purge to complete before continuing */ + if (tmpres->state & DLM_LOCK_RES_DROPPING_REF) { + BUG_ON(tmpres->owner == dlm->node_num); + __dlm_wait_on_lockres_flags(tmpres, + DLM_LOCK_RES_DROPPING_REF); spin_unlock(&tmpres->spinlock); dlm_lockres_put(tmpres); tmpres = NULL; goto lookup; } - mlog(0, "found in hash!\n"); + /* Grab inflight ref to pin the resource */ + dlm_lockres_grab_inflight_ref(dlm, tmpres); + + spin_unlock(&tmpres->spinlock); if (res) dlm_lockres_put(res); res = tmpres; @@ -863,14 +855,11 @@ lookup: /* finally add the lockres to its hash bucket */ __dlm_insert_lockres(dlm, res); + /* Grab inflight ref to pin the resource */ spin_lock(&res->spinlock); dlm_lockres_grab_inflight_ref(dlm, res); spin_unlock(&res->spinlock); - /* if this node does not become the master make sure to drop - * this inflight reference below */ - drop_inflight_if_nonlocal = 1; - /* get an extra ref on the mle in case this is a BLOCK * if so, the creator of the BLOCK may try to put the last * ref at this time in the assert master handler, so we @@ -973,8 +962,6 @@ wait: wake_waiters: spin_lock(&res->spinlock); - if (res->owner != dlm->node_num && drop_inflight_if_nonlocal) - dlm_lockres_drop_inflight_ref(dlm, res); res->state &= ~DLM_LOCK_RES_IN_PROGRESS; spin_unlock(&res->spinlock); wake_up(&res->wq); diff --git a/fs/ocfs2/dlm/dlmthread.c b/fs/ocfs2/dlm/dlmthread.c index 4eff65e7e59e0..e73c833fc2a1a 100644 --- a/fs/ocfs2/dlm/dlmthread.c +++ b/fs/ocfs2/dlm/dlmthread.c @@ -94,24 +94,26 @@ int __dlm_lockres_unused(struct dlm_lock_resource *res) { int bit; + assert_spin_locked(&res->spinlock); + if (__dlm_lockres_has_locks(res)) return 0; + /* Locks are in the process of being created */ + if (res->inflight_locks) + return 0; + if (!list_empty(&res->dirty) || res->state & DLM_LOCK_RES_DIRTY) return 0; if (res->state & DLM_LOCK_RES_RECOVERING) return 0; + /* Another node has this resource with this node as the master */ bit = find_next_bit(res->refmap, O2NM_MAX_NODES, 0); if (bit < O2NM_MAX_NODES) return 0; - /* - * since the bit for dlm->node_num is not set, inflight_locks better - * be zero - */ - BUG_ON(res->inflight_locks != 0); return 1; } -- GitLab From a2c0cc1579176bd0808ef7deea456767dfa80217 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:30:54 -0700 Subject: [PATCH 0043/3089] ocfs2/dlm: dlmlock_remote() needs to account for remastery In dlmlock_remote(), we wait for the resource to stop being active before setting the inprogress flag. Active includes recovery, migration, etc. The problem here is that if the resource was being recovered or migrated, the new owner could very well be that node itself (and thus not a remote node). This problem was observed in Oracle bug#12583620. The error messages observed were as follows: dlm_send_remote_lock_request:337 ERROR: Error -40 (ELOOP) when sending message 503 (key 0xd6d8c7) to node 2 dlmlock_remote:271 ERROR: dlm status = DLM_BADARGS dlmlock:751 ERROR: dlm status = DLM_BADARGS Signed-off-by: Sunil Mushran --- fs/ocfs2/dlm/dlmlock.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-) diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c index 3ef2c1adfb8ff..f32fcba04923d 100644 --- a/fs/ocfs2/dlm/dlmlock.c +++ b/fs/ocfs2/dlm/dlmlock.c @@ -227,10 +227,16 @@ static enum dlm_status dlmlock_remote(struct dlm_ctxt *dlm, lock->ml.type, res->lockname.len, res->lockname.name, flags); + /* + * Wait if resource is getting recovered, remastered, etc. + * If the resource was remastered and new owner is self, then exit. + */ spin_lock(&res->spinlock); - - /* will exit this call with spinlock held */ __dlm_wait_on_lockres(res); + if (res->owner == dlm->node_num) { + spin_unlock(&res->spinlock); + return DLM_RECOVERING; + } res->state |= DLM_LOCK_RES_IN_PROGRESS; /* add lock to local (secondary) queue */ @@ -710,18 +716,10 @@ retry_lock: if (status == DLM_RECOVERING || status == DLM_MIGRATING || status == DLM_FORWARD) { - mlog(0, "retrying lock with migration/" - "recovery/in progress\n"); msleep(100); - /* no waiting for dlm_reco_thread */ if (recovery) { if (status != DLM_RECOVERING) goto retry_lock; - - mlog(0, "%s: got RECOVERING " - "for $RECOVERY lock, master " - "was %u\n", dlm->name, - res->owner); /* wait to see the node go down, then * drop down and allow the lockres to * get cleaned up. need to remaster. */ -- GitLab From bb570a5d9e74f71d32751823052db4a97d6a5e7c Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:31:54 -0700 Subject: [PATCH 0044/3089] ocfs2/cluster: Fix output in file elapsed_time_in_ms The o2hb debugfs file, elapsed_time_in_ms, should return values only after the timer is armed atleast once. Signed-off-by: Sunil Mushran --- fs/ocfs2/cluster/heartbeat.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 4728a746281df..a4e855e3690e6 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -1198,6 +1198,7 @@ static int o2hb_debug_open(struct inode *inode, struct file *file) struct o2hb_debug_buf *db = inode->i_private; struct o2hb_region *reg; unsigned long map[BITS_TO_LONGS(O2NM_MAX_NODES)]; + unsigned long lts; char *buf = NULL; int i = -1; int out = 0; @@ -1234,9 +1235,11 @@ static int o2hb_debug_open(struct inode *inode, struct file *file) case O2HB_DB_TYPE_REGION_ELAPSED_TIME: reg = (struct o2hb_region *)db->db_data; - out += snprintf(buf + out, PAGE_SIZE - out, "%u\n", - jiffies_to_msecs(jiffies - - reg->hr_last_timeout_start)); + lts = reg->hr_last_timeout_start; + /* If 0, it has never been set before */ + if (lts) + lts = jiffies_to_msecs(jiffies - lts); + out += snprintf(buf + out, PAGE_SIZE - out, "%lu\n", lts); goto done; case O2HB_DB_TYPE_REGION_PINNED: -- GitLab From 3ba169ccec1c5ad0f678e04fd29b990197fdfe79 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:32:54 -0700 Subject: [PATCH 0045/3089] ocfs2/cluster: Add new function o2net_fill_node_map() Patch adds function o2net_fill_node_map() to return the bitmap of nodes that it is connected to. This bitmap is also accessible by the user via the debugfs file, /sys/kernel/debug/o2net/connected_nodes. Signed-off-by: Sunil Mushran --- fs/ocfs2/cluster/netdebug.c | 102 ++++++++++++++++++++++++------------ fs/ocfs2/cluster/tcp.c | 19 +++++++ fs/ocfs2/cluster/tcp.h | 2 + 3 files changed, 90 insertions(+), 33 deletions(-) diff --git a/fs/ocfs2/cluster/netdebug.c b/fs/ocfs2/cluster/netdebug.c index 3a5835904b3db..dc45deb19e688 100644 --- a/fs/ocfs2/cluster/netdebug.c +++ b/fs/ocfs2/cluster/netdebug.c @@ -47,6 +47,7 @@ #define SC_DEBUG_NAME "sock_containers" #define NST_DEBUG_NAME "send_tracking" #define STATS_DEBUG_NAME "stats" +#define NODES_DEBUG_NAME "connected_nodes" #define SHOW_SOCK_CONTAINERS 0 #define SHOW_SOCK_STATS 1 @@ -55,6 +56,7 @@ static struct dentry *o2net_dentry; static struct dentry *sc_dentry; static struct dentry *nst_dentry; static struct dentry *stats_dentry; +static struct dentry *nodes_dentry; static DEFINE_SPINLOCK(o2net_debug_lock); @@ -491,53 +493,87 @@ static const struct file_operations sc_seq_fops = { .release = sc_fop_release, }; -int o2net_debugfs_init(void) +static int o2net_fill_bitmap(char *buf, int len) { - o2net_dentry = debugfs_create_dir(O2NET_DEBUG_DIR, NULL); - if (!o2net_dentry) { - mlog_errno(-ENOMEM); - goto bail; - } + unsigned long map[BITS_TO_LONGS(O2NM_MAX_NODES)]; + int i = -1, out = 0; - nst_dentry = debugfs_create_file(NST_DEBUG_NAME, S_IFREG|S_IRUSR, - o2net_dentry, NULL, - &nst_seq_fops); - if (!nst_dentry) { - mlog_errno(-ENOMEM); - goto bail; - } + o2net_fill_node_map(map, sizeof(map)); - sc_dentry = debugfs_create_file(SC_DEBUG_NAME, S_IFREG|S_IRUSR, - o2net_dentry, NULL, - &sc_seq_fops); - if (!sc_dentry) { - mlog_errno(-ENOMEM); - goto bail; - } + while ((i = find_next_bit(map, O2NM_MAX_NODES, i + 1)) < O2NM_MAX_NODES) + out += snprintf(buf + out, PAGE_SIZE - out, "%d ", i); + out += snprintf(buf + out, PAGE_SIZE - out, "\n"); - stats_dentry = debugfs_create_file(STATS_DEBUG_NAME, S_IFREG|S_IRUSR, - o2net_dentry, NULL, - &stats_seq_fops); - if (!stats_dentry) { - mlog_errno(-ENOMEM); - goto bail; - } + return out; +} + +static int nodes_fop_open(struct inode *inode, struct file *file) +{ + char *buf; + + buf = kmalloc(PAGE_SIZE, GFP_KERNEL); + if (!buf) + return -ENOMEM; + + i_size_write(inode, o2net_fill_bitmap(buf, PAGE_SIZE)); + + file->private_data = buf; return 0; -bail: - debugfs_remove(stats_dentry); - debugfs_remove(sc_dentry); - debugfs_remove(nst_dentry); - debugfs_remove(o2net_dentry); - return -ENOMEM; } +static int o2net_debug_release(struct inode *inode, struct file *file) +{ + kfree(file->private_data); + return 0; +} + +static ssize_t o2net_debug_read(struct file *file, char __user *buf, + size_t nbytes, loff_t *ppos) +{ + return simple_read_from_buffer(buf, nbytes, ppos, file->private_data, + i_size_read(file->f_mapping->host)); +} + +static const struct file_operations nodes_fops = { + .open = nodes_fop_open, + .release = o2net_debug_release, + .read = o2net_debug_read, + .llseek = generic_file_llseek, +}; + void o2net_debugfs_exit(void) { + debugfs_remove(nodes_dentry); debugfs_remove(stats_dentry); debugfs_remove(sc_dentry); debugfs_remove(nst_dentry); debugfs_remove(o2net_dentry); } +int o2net_debugfs_init(void) +{ + mode_t mode = S_IFREG|S_IRUSR; + + o2net_dentry = debugfs_create_dir(O2NET_DEBUG_DIR, NULL); + if (o2net_dentry) + nst_dentry = debugfs_create_file(NST_DEBUG_NAME, mode, + o2net_dentry, NULL, &nst_seq_fops); + if (nst_dentry) + sc_dentry = debugfs_create_file(SC_DEBUG_NAME, mode, + o2net_dentry, NULL, &sc_seq_fops); + if (sc_dentry) + stats_dentry = debugfs_create_file(STATS_DEBUG_NAME, mode, + o2net_dentry, NULL, &stats_seq_fops); + if (stats_dentry) + nodes_dentry = debugfs_create_file(NODES_DEBUG_NAME, mode, + o2net_dentry, NULL, &nodes_fops); + if (nodes_dentry) + return 0; + + o2net_debugfs_exit(); + mlog_errno(-ENOMEM); + return -ENOMEM; +} + #endif /* CONFIG_DEBUG_FS */ diff --git a/fs/ocfs2/cluster/tcp.c b/fs/ocfs2/cluster/tcp.c index 6e97895ecf24a..ae13d5ca7908c 100644 --- a/fs/ocfs2/cluster/tcp.c +++ b/fs/ocfs2/cluster/tcp.c @@ -1034,6 +1034,25 @@ static int o2net_tx_can_proceed(struct o2net_node *nn, return ret; } +/* Get a map of all nodes to which this node is currently connected to */ +void o2net_fill_node_map(unsigned long *map, unsigned bytes) +{ + struct o2net_sock_container *sc; + int node, ret; + + BUG_ON(bytes < (BITS_TO_LONGS(O2NM_MAX_NODES) * sizeof(unsigned long))); + + memset(map, 0, bytes); + for (node = 0; node < O2NM_MAX_NODES; ++node) { + o2net_tx_can_proceed(o2net_nn_from_num(node), &sc, &ret); + if (!ret) { + set_bit(node, map); + sc_put(sc); + } + } +} +EXPORT_SYMBOL_GPL(o2net_fill_node_map); + int o2net_send_message_vec(u32 msg_type, u32 key, struct kvec *caller_vec, size_t caller_veclen, u8 target_node, int *status) { diff --git a/fs/ocfs2/cluster/tcp.h b/fs/ocfs2/cluster/tcp.h index fd6179eb26d4c..5bada2a69b503 100644 --- a/fs/ocfs2/cluster/tcp.h +++ b/fs/ocfs2/cluster/tcp.h @@ -106,6 +106,8 @@ int o2net_register_handler(u32 msg_type, u32 key, u32 max_len, struct list_head *unreg_list); void o2net_unregister_handler_list(struct list_head *list); +void o2net_fill_node_map(unsigned long *map, unsigned bytes); + struct o2nm_node; int o2net_register_hb_callbacks(void); void o2net_unregister_hb_callbacks(void); -- GitLab From 6b27f62fc750d85bc6fc3718b3b38ec60edc2d74 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:33:54 -0700 Subject: [PATCH 0046/3089] ocfs2/cluster: Cluster up now includes network connections too The cluster up check only checks to see if the node is heartbeating or not. If yes it continues assuming that the node is connected to all the nodes. But if that is not the case, the cluster join aborts with a stack of errors that are not easy to comprehend. This patch adds the network connect check upfront and prints the nodes that the node is not yet connected to, before aborting. Signed-off-by: Sunil Mushran --- fs/ocfs2/dlm/dlmdomain.c | 7 ----- fs/ocfs2/stack_o2cb.c | 67 ++++++++++++++++++++++++++++++++++++---- 2 files changed, 61 insertions(+), 13 deletions(-) diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 200d9888ffee6..92f2ead0fab6d 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -2138,13 +2138,6 @@ struct dlm_ctxt * dlm_register_domain(const char *domain, goto leave; } - if (!o2hb_check_local_node_heartbeating()) { - mlog(ML_ERROR, "the local node has not been configured, or is " - "not heartbeating\n"); - ret = -EPROTO; - goto leave; - } - mlog(0, "register called for domain \"%s\"\n", domain); retry: diff --git a/fs/ocfs2/stack_o2cb.c b/fs/ocfs2/stack_o2cb.c index be935d89a4e88..94368017edb37 100644 --- a/fs/ocfs2/stack_o2cb.c +++ b/fs/ocfs2/stack_o2cb.c @@ -28,6 +28,7 @@ #include "cluster/masklog.h" #include "cluster/nodemanager.h" #include "cluster/heartbeat.h" +#include "cluster/tcp.h" #include "stackglue.h" @@ -255,6 +256,61 @@ static void o2cb_dump_lksb(struct ocfs2_dlm_lksb *lksb) dlm_print_one_lock(lksb->lksb_o2dlm.lockid); } +/* + * Check if this node is heartbeating and is connected to all other + * heartbeating nodes. + */ +static int o2cb_cluster_check(void) +{ + u8 node_num; + int i; + unsigned long hbmap[BITS_TO_LONGS(O2NM_MAX_NODES)]; + unsigned long netmap[BITS_TO_LONGS(O2NM_MAX_NODES)]; + + node_num = o2nm_this_node(); + if (node_num == O2NM_MAX_NODES) { + printk(KERN_ERR "o2cb: This node has not been configured.\n"); + return -EINVAL; + } + + /* + * o2dlm expects o2net sockets to be created. If not, then + * dlm_join_domain() fails with a stack of errors which are both cryptic + * and incomplete. The idea here is to detect upfront whether we have + * managed to connect to all nodes or not. If not, then list the nodes + * to allow the user to check the configuration (incorrect IP, firewall, + * etc.) Yes, this is racy. But its not the end of the world. + */ +#define O2CB_MAP_STABILIZE_COUNT 60 + for (i = 0; i < O2CB_MAP_STABILIZE_COUNT; ++i) { + o2hb_fill_node_map(hbmap, sizeof(hbmap)); + if (!test_bit(node_num, hbmap)) { + printk(KERN_ERR "o2cb: %s heartbeat has not been " + "started.\n", (o2hb_global_heartbeat_active() ? + "Global" : "Local")); + return -EINVAL; + } + o2net_fill_node_map(netmap, sizeof(netmap)); + /* Force set the current node to allow easy compare */ + set_bit(node_num, netmap); + if (!memcmp(hbmap, netmap, sizeof(hbmap))) + return 0; + if (i < O2CB_MAP_STABILIZE_COUNT) + msleep(1000); + } + + printk(KERN_ERR "o2cb: This node could not connect to nodes:"); + i = -1; + while ((i = find_next_bit(hbmap, O2NM_MAX_NODES, + i + 1)) < O2NM_MAX_NODES) { + if (!test_bit(i, netmap)) + printk(" %u", i); + } + printk(".\n"); + + return -ENOTCONN; +} + /* * Called from the dlm when it's about to evict a node. This is how the * classic stack signals node death. @@ -280,12 +336,11 @@ static int o2cb_cluster_connect(struct ocfs2_cluster_connection *conn) BUG_ON(conn == NULL); BUG_ON(conn->cc_proto == NULL); - /* for now we only have one cluster/node, make sure we see it - * in the heartbeat universe */ - if (!o2hb_check_local_node_heartbeating()) { - if (o2hb_global_heartbeat_active()) - mlog(ML_ERROR, "Global heartbeat not started\n"); - rc = -EINVAL; + /* Ensure cluster stack is up and all nodes are connected */ + rc = o2cb_cluster_check(); + if (rc) { + printk(KERN_ERR "o2cb: Cluster check failed. Fix errors " + "before retrying.\n"); goto out; } -- GitLab From 619c200de144b44f5e405305241bcd7edbb8c6cf Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:34:54 -0700 Subject: [PATCH 0047/3089] ocfs2: Clean up messages in the fs Convert useful messages from ML_NOTICE to KERN_NOTICE to improve readability. Signed-off-by: Sunil Mushran --- fs/ocfs2/journal.c | 9 ++++++--- fs/ocfs2/quota_local.c | 13 +++++++++---- fs/ocfs2/slot_map.c | 4 ++-- fs/ocfs2/super.c | 10 +++++----- 4 files changed, 22 insertions(+), 14 deletions(-) diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index 295d56454e8b2..8eaaa78544096 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -1544,9 +1544,9 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb, /* we need to run complete recovery for offline orphan slots */ ocfs2_replay_map_set_state(osb, REPLAY_NEEDED); - mlog(ML_NOTICE, "Recovering node %d from slot %d on device (%u,%u)\n", - node_num, slot_num, - MAJOR(osb->sb->s_dev), MINOR(osb->sb->s_dev)); + printk(KERN_NOTICE "ocfs2: Begin replay journal (node %d, slot %d) on "\ + "device (%u,%u)\n", node_num, slot_num, MAJOR(osb->sb->s_dev), + MINOR(osb->sb->s_dev)); OCFS2_I(inode)->ip_clusters = le32_to_cpu(fe->i_clusters); @@ -1601,6 +1601,9 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb, jbd2_journal_destroy(journal); + printk(KERN_NOTICE "ocfs2: End replay journal (node %d, slot %d) on "\ + "device (%u,%u)\n", node_num, slot_num, MAJOR(osb->sb->s_dev), + MINOR(osb->sb->s_dev)); done: /* drop the lock on this nodes journal */ if (got_lock) diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c index dc8007fc92471..942fd65bdad3f 100644 --- a/fs/ocfs2/quota_local.c +++ b/fs/ocfs2/quota_local.c @@ -404,7 +404,9 @@ struct ocfs2_quota_recovery *ocfs2_begin_quota_recovery( int status = 0; struct ocfs2_quota_recovery *rec; - mlog(ML_NOTICE, "Beginning quota recovery in slot %u\n", slot_num); + printk(KERN_NOTICE "ocfs2: Beginning quota recovery on device (%s) for " + "slot %u\n", osb->dev_str, slot_num); + rec = ocfs2_alloc_quota_recovery(); if (!rec) return ERR_PTR(-ENOMEM); @@ -596,7 +598,9 @@ int ocfs2_finish_quota_recovery(struct ocfs2_super *osb, struct inode *lqinode; unsigned int flags; - mlog(ML_NOTICE, "Finishing quota recovery in slot %u\n", slot_num); + printk(KERN_NOTICE "ocfs2: Finishing quota recovery on device (%s) for " + "slot %u\n", osb->dev_str, slot_num); + mutex_lock(&sb_dqopt(sb)->dqonoff_mutex); for (type = 0; type < MAXQUOTAS; type++) { if (list_empty(&(rec->r_list[type]))) @@ -612,8 +616,9 @@ int ocfs2_finish_quota_recovery(struct ocfs2_super *osb, /* Someone else is holding the lock? Then he must be * doing the recovery. Just skip the file... */ if (status == -EAGAIN) { - mlog(ML_NOTICE, "skipping quota recovery for slot %d " - "because quota file is locked.\n", slot_num); + printk(KERN_NOTICE "ocfs2: Skipping quota recovery on " + "device (%s) for slot %d because quota file is " + "locked.\n", osb->dev_str, slot_num); status = 0; goto out_put; } else if (status < 0) { diff --git a/fs/ocfs2/slot_map.c b/fs/ocfs2/slot_map.c index 26fc0014d5093..1424c151cccce 100644 --- a/fs/ocfs2/slot_map.c +++ b/fs/ocfs2/slot_map.c @@ -493,8 +493,8 @@ int ocfs2_find_slot(struct ocfs2_super *osb) goto bail; } } else - mlog(ML_NOTICE, "slot %d is already allocated to this node!\n", - slot); + printk(KERN_INFO "ocfs2: Slot %d on device (%s) was already " + "allocated to this node!\n", slot, osb->dev_str); ocfs2_set_slot(si, slot, osb->node_num); osb->slot_num = slot; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 56f61027236b6..3e7850380d2b6 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1107,9 +1107,9 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent) ocfs2_set_ro_flag(osb, 1); - printk(KERN_NOTICE "Readonly device detected. No cluster " - "services will be utilized for this mount. Recovery " - "will be skipped.\n"); + printk(KERN_NOTICE "ocfs2: Readonly device (%s) detected. " + "Cluster services will not be used for this mount. " + "Recovery will be skipped.\n", osb->dev_str); } if (!ocfs2_is_hard_readonly(osb)) { @@ -2462,8 +2462,8 @@ static int ocfs2_check_volume(struct ocfs2_super *osb) goto finally; } } else { - mlog(ML_NOTICE, "File system was not unmounted cleanly, " - "recovering volume.\n"); + printk(KERN_NOTICE "ocfs2: File system on device (%s) was not " + "unmounted cleanly, recovering it.\n", osb->dev_str); } local = ocfs2_mount_local(osb); -- GitLab From a035bff6b82aca89c1223e2c614adc2d17ec8aa2 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Sun, 24 Jul 2011 10:35:54 -0700 Subject: [PATCH 0048/3089] ocfs2: Add comment about orphan scanning Add a comment that explains the reason as to why orphan scan scans all the slots. Signed-off-by: Sunil Mushran --- fs/ocfs2/journal.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index 8eaaa78544096..0a42ae96dca7d 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -1811,6 +1811,20 @@ static inline unsigned long ocfs2_orphan_scan_timeout(void) * every slot, queuing a recovery of the slot on the ocfs2_wq thread. This * is done to catch any orphans that are left over in orphan directories. * + * It scans all slots, even ones that are in use. It does so to handle the + * case described below: + * + * Node 1 has an inode it was using. The dentry went away due to memory + * pressure. Node 1 closes the inode, but it's on the free list. The node + * has the open lock. + * Node 2 unlinks the inode. It grabs the dentry lock to notify others, + * but node 1 has no dentry and doesn't get the message. It trylocks the + * open lock, sees that another node has a PR, and does nothing. + * Later node 2 runs its orphan dir. It igets the inode, trylocks the + * open lock, sees the PR still, and does nothing. + * Basically, we have to trigger an orphan iput on node 1. The only way + * for this to happen is if node 1 runs node 2's orphan dir. + * * ocfs2_queue_orphan_scan gets called every ORPHAN_SCAN_SCHEDULE_TIMEOUT * seconds. It gets an EX lock on os_lockres and checks sequence number * stored in LVB. If the sequence number has changed, it means some other -- GitLab From 5cffff9e29866a3de98c2c25135b3199491f93b0 Mon Sep 17 00:00:00 2001 From: Wengang Wang Date: Sun, 24 Jul 2011 10:36:54 -0700 Subject: [PATCH 0049/3089] ocfs2: Fix ocfs2_page_mkwrite() This patch address two shortcomings in ocfs2_page_mkwrite(): 1. Makes the function return better VM_FAULT_* errors. 2. It handles a error that is triggered when a page is dropped from the mapping due to memory pressure. This patch locks the page to prevent that. [Patch was cleaned up by Sunil Mushran.] Signed-off-by: Wengang Wang Signed-off-by: Sunil Mushran --- fs/ocfs2/aops.c | 51 +++++++++++++++++++++++++++++++++++++++-------- fs/ocfs2/mmap.c | 53 ++++++++++++++++++++++--------------------------- 2 files changed, 67 insertions(+), 37 deletions(-) diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index c1efe939c774e..ff98c169b631e 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -862,6 +862,12 @@ struct ocfs2_write_ctxt { struct page *w_pages[OCFS2_MAX_CTXT_PAGES]; struct page *w_target_page; + /* + * w_target_locked is used for page_mkwrite path indicating no unlocking + * against w_target_page in ocfs2_write_end_nolock. + */ + unsigned int w_target_locked:1; + /* * ocfs2_write_end() uses this to know what the real range to * write in the target should be. @@ -895,6 +901,24 @@ void ocfs2_unlock_and_free_pages(struct page **pages, int num_pages) static void ocfs2_free_write_ctxt(struct ocfs2_write_ctxt *wc) { + int i; + + /* + * w_target_locked is only set to true in the page_mkwrite() case. + * The intent is to allow us to lock the target page from write_begin() + * to write_end(). The caller must hold a ref on w_target_page. + */ + if (wc->w_target_locked) { + BUG_ON(!wc->w_target_page); + for (i = 0; i < wc->w_num_pages; i++) { + if (wc->w_target_page == wc->w_pages[i]) { + wc->w_pages[i] = NULL; + break; + } + } + mark_page_accessed(wc->w_target_page); + page_cache_release(wc->w_target_page); + } ocfs2_unlock_and_free_pages(wc->w_pages, wc->w_num_pages); brelse(wc->w_di_bh); @@ -1132,20 +1156,17 @@ static int ocfs2_grab_pages_for_write(struct address_space *mapping, */ lock_page(mmap_page); + /* Exit and let the caller retry */ if (mmap_page->mapping != mapping) { + WARN_ON(mmap_page->mapping); unlock_page(mmap_page); - /* - * Sanity check - the locking in - * ocfs2_pagemkwrite() should ensure - * that this code doesn't trigger. - */ - ret = -EINVAL; - mlog_errno(ret); + ret = -EAGAIN; goto out; } page_cache_get(mmap_page); wc->w_pages[i] = mmap_page; + wc->w_target_locked = true; } else { wc->w_pages[i] = find_or_create_page(mapping, index, GFP_NOFS); @@ -1160,6 +1181,8 @@ static int ocfs2_grab_pages_for_write(struct address_space *mapping, wc->w_target_page = wc->w_pages[i]; } out: + if (ret) + wc->w_target_locked = false; return ret; } @@ -1817,11 +1840,23 @@ try_again: */ ret = ocfs2_grab_pages_for_write(mapping, wc, wc->w_cpos, pos, len, cluster_of_pages, mmap_page); - if (ret) { + if (ret && ret != -EAGAIN) { mlog_errno(ret); goto out_quota; } + /* + * ocfs2_grab_pages_for_write() returns -EAGAIN if it could not lock + * the target page. In this case, we exit with no error and no target + * page. This will trigger the caller, page_mkwrite(), to re-try + * the operation. + */ + if (ret == -EAGAIN) { + BUG_ON(wc->w_target_page); + ret = 0; + goto out_quota; + } + ret = ocfs2_write_cluster_by_desc(mapping, data_ac, meta_ac, wc, pos, len); if (ret) { diff --git a/fs/ocfs2/mmap.c b/fs/ocfs2/mmap.c index 3e9393ca39ebd..9cd41083e9912 100644 --- a/fs/ocfs2/mmap.c +++ b/fs/ocfs2/mmap.c @@ -61,7 +61,7 @@ static int ocfs2_fault(struct vm_area_struct *area, struct vm_fault *vmf) static int __ocfs2_page_mkwrite(struct file *file, struct buffer_head *di_bh, struct page *page) { - int ret; + int ret = VM_FAULT_NOPAGE; struct inode *inode = file->f_path.dentry->d_inode; struct address_space *mapping = inode->i_mapping; loff_t pos = page_offset(page); @@ -71,32 +71,25 @@ static int __ocfs2_page_mkwrite(struct file *file, struct buffer_head *di_bh, void *fsdata; loff_t size = i_size_read(inode); - /* - * Another node might have truncated while we were waiting on - * cluster locks. - * We don't check size == 0 before the shift. This is borrowed - * from do_generic_file_read. - */ last_index = (size - 1) >> PAGE_CACHE_SHIFT; - if (unlikely(!size || page->index > last_index)) { - ret = -EINVAL; - goto out; - } /* - * The i_size check above doesn't catch the case where nodes - * truncated and then re-extended the file. We'll re-check the - * page mapping after taking the page lock inside of - * ocfs2_write_begin_nolock(). + * There are cases that lead to the page no longer bebongs to the + * mapping. + * 1) pagecache truncates locally due to memory pressure. + * 2) pagecache truncates when another is taking EX lock against + * inode lock. see ocfs2_data_convert_worker. + * + * The i_size check doesn't catch the case where nodes truncated and + * then re-extended the file. We'll re-check the page mapping after + * taking the page lock inside of ocfs2_write_begin_nolock(). + * + * Let VM retry with these cases. */ - if (!PageUptodate(page) || page->mapping != inode->i_mapping) { - /* - * the page has been umapped in ocfs2_data_downconvert_worker. - * So return 0 here and let VFS retry. - */ - ret = 0; + if ((page->mapping != inode->i_mapping) || + (!PageUptodate(page)) || + (page_offset(page) >= size)) goto out; - } /* * Call ocfs2_write_begin() and ocfs2_write_end() to take @@ -116,17 +109,21 @@ static int __ocfs2_page_mkwrite(struct file *file, struct buffer_head *di_bh, if (ret) { if (ret != -ENOSPC) mlog_errno(ret); + if (ret == -ENOMEM) + ret = VM_FAULT_OOM; + else + ret = VM_FAULT_SIGBUS; goto out; } - ret = ocfs2_write_end_nolock(mapping, pos, len, len, locked_page, - fsdata); - if (ret < 0) { - mlog_errno(ret); + if (!locked_page) { + ret = VM_FAULT_NOPAGE; goto out; } + ret = ocfs2_write_end_nolock(mapping, pos, len, len, locked_page, + fsdata); BUG_ON(ret != len); - ret = 0; + ret = VM_FAULT_LOCKED; out: return ret; } @@ -168,8 +165,6 @@ static int ocfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) out: ocfs2_unblock_signals(&oldset); - if (ret) - ret = VM_FAULT_SIGBUS; return ret; } -- GitLab From 93862d5e1ab875664c6cc95254fc365028a48bb1 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Mon, 25 Jul 2011 14:58:15 -0700 Subject: [PATCH 0050/3089] ocfs2: Implement llseek() ocfs2 implements its own llseek() to provide the SEEK_HOLE/SEEK_DATA functionality. SEEK_HOLE sets the file pointer to the start of either a hole or an unwritten (preallocated) extent, that is greater than or equal to the supplied offset. SEEK_DATA sets the file pointer to the start of an allocated extent (not unwritten) that is greater than or equal to the supplied offset. If the supplied offset is on a desired region, then the file pointer is set to it. Offsets greater than or equal to the file size return -ENXIO. Unwritten (preallocated) extents are considered holes because the file system treats reads to such regions in the same way as it does to holes. Signed-off-by: Sunil Mushran --- fs/ocfs2/extent_map.c | 96 +++++++++++++++++++++++++++++++++++++++++++ fs/ocfs2/extent_map.h | 2 + fs/ocfs2/file.c | 55 ++++++++++++++++++++++++- 3 files changed, 151 insertions(+), 2 deletions(-) diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c index 23457b491e8ce..2f5b92ef0e533 100644 --- a/fs/ocfs2/extent_map.c +++ b/fs/ocfs2/extent_map.c @@ -832,6 +832,102 @@ out: return ret; } +int ocfs2_seek_data_hole_offset(struct file *file, loff_t *offset, int origin) +{ + struct inode *inode = file->f_mapping->host; + int ret; + unsigned int is_last = 0, is_data = 0; + u16 cs_bits = OCFS2_SB(inode->i_sb)->s_clustersize_bits; + u32 cpos, cend, clen, hole_size; + u64 extoff, extlen; + struct buffer_head *di_bh = NULL; + struct ocfs2_extent_rec rec; + + BUG_ON(origin != SEEK_DATA && origin != SEEK_HOLE); + + ret = ocfs2_inode_lock(inode, &di_bh, 0); + if (ret) { + mlog_errno(ret); + goto out; + } + + down_read(&OCFS2_I(inode)->ip_alloc_sem); + + if (*offset >= inode->i_size) { + ret = -ENXIO; + goto out_unlock; + } + + if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) { + if (origin == SEEK_HOLE) + *offset = inode->i_size; + goto out_unlock; + } + + clen = 0; + cpos = *offset >> cs_bits; + cend = ocfs2_clusters_for_bytes(inode->i_sb, inode->i_size); + + while (cpos < cend && !is_last) { + ret = ocfs2_get_clusters_nocache(inode, di_bh, cpos, &hole_size, + &rec, &is_last); + if (ret) { + mlog_errno(ret); + goto out_unlock; + } + + extoff = cpos; + extoff <<= cs_bits; + + if (rec.e_blkno == 0ULL) { + clen = hole_size; + is_data = 0; + } else { + clen = le16_to_cpu(rec.e_leaf_clusters) - + (cpos - le32_to_cpu(rec.e_cpos)); + is_data = (rec.e_flags & OCFS2_EXT_UNWRITTEN) ? 0 : 1; + } + + if ((!is_data && origin == SEEK_HOLE) || + (is_data && origin == SEEK_DATA)) { + if (extoff > *offset) + *offset = extoff; + goto out_unlock; + } + + if (!is_last) + cpos += clen; + } + + if (origin == SEEK_HOLE) { + extoff = cpos; + extoff <<= cs_bits; + extlen = clen; + extlen <<= cs_bits; + + if ((extoff + extlen) > inode->i_size) + extlen = inode->i_size - extoff; + extoff += extlen; + if (extoff > *offset) + *offset = extoff; + goto out_unlock; + } + + ret = -ENXIO; + +out_unlock: + + brelse(di_bh); + + up_read(&OCFS2_I(inode)->ip_alloc_sem); + + ocfs2_inode_unlock(inode, 0); +out: + if (ret && ret != -ENXIO) + ret = -ENXIO; + return ret; +} + int ocfs2_read_virt_blocks(struct inode *inode, u64 v_block, int nr, struct buffer_head *bhs[], int flags, int (*validate)(struct super_block *sb, diff --git a/fs/ocfs2/extent_map.h b/fs/ocfs2/extent_map.h index e79d41c2c9097..67ea57d2fd594 100644 --- a/fs/ocfs2/extent_map.h +++ b/fs/ocfs2/extent_map.h @@ -53,6 +53,8 @@ int ocfs2_extent_map_get_blocks(struct inode *inode, u64 v_blkno, u64 *p_blkno, int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, u64 map_start, u64 map_len); +int ocfs2_seek_data_hole_offset(struct file *file, loff_t *offset, int origin); + int ocfs2_xattr_get_clusters(struct inode *inode, u32 v_cluster, u32 *p_cluster, u32 *num_clusters, struct ocfs2_extent_list *el, diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 0fc2bd34039da..c0f015e11c28f 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2591,6 +2591,57 @@ bail: return ret; } +/* Refer generic_file_llseek_unlocked() */ +static loff_t ocfs2_file_llseek(struct file *file, loff_t offset, int origin) +{ + struct inode *inode = file->f_mapping->host; + int ret = 0; + + mutex_lock(&inode->i_mutex); + + switch (origin) { + case SEEK_SET: + break; + case SEEK_END: + offset += inode->i_size; + break; + case SEEK_CUR: + if (offset == 0) { + offset = file->f_pos; + goto out; + } + offset += file->f_pos; + break; + case SEEK_DATA: + case SEEK_HOLE: + ret = ocfs2_seek_data_hole_offset(file, &offset, origin); + if (ret) + goto out; + break; + default: + ret = -EINVAL; + goto out; + } + + if (offset < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) + ret = -EINVAL; + if (!ret && offset > inode->i_sb->s_maxbytes) + ret = -EINVAL; + if (ret) + goto out; + + if (offset != file->f_pos) { + file->f_pos = offset; + file->f_version = 0; + } + +out: + mutex_unlock(&inode->i_mutex); + if (ret) + return ret; + return offset; +} + const struct inode_operations ocfs2_file_iops = { .setattr = ocfs2_setattr, .getattr = ocfs2_getattr, @@ -2615,7 +2666,7 @@ const struct inode_operations ocfs2_special_file_iops = { * ocfs2_fops_no_plocks and ocfs2_dops_no_plocks! */ const struct file_operations ocfs2_fops = { - .llseek = generic_file_llseek, + .llseek = ocfs2_file_llseek, .read = do_sync_read, .write = do_sync_write, .mmap = ocfs2_mmap, @@ -2663,7 +2714,7 @@ const struct file_operations ocfs2_dops = { * the cluster. */ const struct file_operations ocfs2_fops_no_plocks = { - .llseek = generic_file_llseek, + .llseek = ocfs2_file_llseek, .read = do_sync_read, .write = do_sync_write, .mmap = ocfs2_mmap, -- GitLab From a11f7e63c59810a81494d4c4b028af707d4c7ca4 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Wed, 22 Jun 2011 14:23:38 -0700 Subject: [PATCH 0051/3089] ocfs2: serialize unaligned aio Fix a corruption that can happen when we have (two or more) outstanding aio's to an overlapping unaligned region. Ext4 (e9e3bcecf44c04b9e6b505fd8e2eb9cea58fb94d) and xfs recently had to fix similar issues. In our case what happens is that we can have an outstanding aio on a region and if a write comes in with some bytes overlapping the original aio we may decide to read that region into a page before continuing (typically because of buffered-io fallback). Since we have no ordering guarantees with the aio, we can read stale or bad data into the page and then write it back out. If the i/o is page and block aligned, then we avoid this issue as there won't be any need to read data from disk. I took the same approach as Eric in the ext4 patch and introduced some serialization of unaligned async direct i/o. I don't expect this to have an effect on the most common cases of AIO. Unaligned aio will be slower though, but that's far more acceptable than data corruption. Signed-off-by: Mark Fasheh Signed-off-by: Joel Becker --- fs/ocfs2/aops.c | 10 ++++++++++ fs/ocfs2/aops.h | 14 ++++++++++++++ fs/ocfs2/file.c | 38 ++++++++++++++++++++++++++++++++++++++ fs/ocfs2/inode.h | 3 +++ fs/ocfs2/super.c | 10 ++++++++-- 5 files changed, 73 insertions(+), 2 deletions(-) diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index ac97bca282d25..4c1ec8f2d8c10 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -564,6 +564,7 @@ static void ocfs2_dio_end_io(struct kiocb *iocb, { struct inode *inode = iocb->ki_filp->f_path.dentry->d_inode; int level; + wait_queue_head_t *wq = ocfs2_ioend_wq(inode); /* this io's submitter should not have unlocked this before we could */ BUG_ON(!ocfs2_iocb_is_rw_locked(iocb)); @@ -573,6 +574,15 @@ static void ocfs2_dio_end_io(struct kiocb *iocb, ocfs2_iocb_clear_sem_locked(iocb); } + if (ocfs2_iocb_is_unaligned_aio(iocb)) { + ocfs2_iocb_clear_unaligned_aio(iocb); + + if (atomic_dec_and_test(&OCFS2_I(inode)->ip_unaligned_aio) && + waitqueue_active(wq)) { + wake_up_all(wq); + } + } + ocfs2_iocb_clear_rw_locked(iocb); level = ocfs2_iocb_rw_locked_level(iocb); diff --git a/fs/ocfs2/aops.h b/fs/ocfs2/aops.h index 75cf3ad987a66..ffb2da370a99d 100644 --- a/fs/ocfs2/aops.h +++ b/fs/ocfs2/aops.h @@ -78,6 +78,7 @@ enum ocfs2_iocb_lock_bits { OCFS2_IOCB_RW_LOCK = 0, OCFS2_IOCB_RW_LOCK_LEVEL, OCFS2_IOCB_SEM, + OCFS2_IOCB_UNALIGNED_IO, OCFS2_IOCB_NUM_LOCKS }; @@ -91,4 +92,17 @@ enum ocfs2_iocb_lock_bits { clear_bit(OCFS2_IOCB_SEM, (unsigned long *)&iocb->private) #define ocfs2_iocb_is_sem_locked(iocb) \ test_bit(OCFS2_IOCB_SEM, (unsigned long *)&iocb->private) + +#define ocfs2_iocb_set_unaligned_aio(iocb) \ + set_bit(OCFS2_IOCB_UNALIGNED_IO, (unsigned long *)&iocb->private) +#define ocfs2_iocb_clear_unaligned_aio(iocb) \ + clear_bit(OCFS2_IOCB_UNALIGNED_IO, (unsigned long *)&iocb->private) +#define ocfs2_iocb_is_unaligned_aio(iocb) \ + test_bit(OCFS2_IOCB_UNALIGNED_IO, (unsigned long *)&iocb->private) + +#define OCFS2_IOEND_WQ_HASH_SZ 37 +#define ocfs2_ioend_wq(v) (&ocfs2__ioend_wq[((unsigned long)(v)) %\ + OCFS2_IOEND_WQ_HASH_SZ]) +extern wait_queue_head_t ocfs2__ioend_wq[OCFS2_IOEND_WQ_HASH_SZ]; + #endif /* OCFS2_FILE_H */ diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index b1e35a392ca52..145f4533a9364 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2038,6 +2038,23 @@ out: return ret; } +static void ocfs2_aiodio_wait(struct inode *inode) +{ + wait_queue_head_t *wq = ocfs2_ioend_wq(inode); + + wait_event(*wq, (atomic_read(&OCFS2_I(inode)->ip_unaligned_aio) == 0)); +} + +static int ocfs2_is_io_unaligned(struct inode *inode, size_t count, loff_t pos) +{ + int blockmask = inode->i_sb->s_blocksize - 1; + loff_t final_size = pos + count; + + if ((pos & blockmask) || (final_size & blockmask)) + return 1; + return 0; +} + static int ocfs2_prepare_inode_for_refcount(struct inode *inode, struct file *file, loff_t pos, size_t count, @@ -2216,6 +2233,7 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb, struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); int full_coherency = !(osb->s_mount_opt & OCFS2_MOUNT_COHERENCY_BUFFERED); + int unaligned_dio = 0; trace_ocfs2_file_aio_write(inode, file, file->f_path.dentry, (unsigned long long)OCFS2_I(inode)->ip_blkno, @@ -2284,6 +2302,10 @@ relock: goto out; } + if (direct_io && !is_sync_kiocb(iocb)) + unaligned_dio = ocfs2_is_io_unaligned(inode, iocb->ki_left, + *ppos); + /* * We can't complete the direct I/O as requested, fall back to * buffered I/O. @@ -2299,6 +2321,18 @@ relock: goto relock; } + if (unaligned_dio) { + /* + * Wait on previous unaligned aio to complete before + * proceeding. + */ + ocfs2_aiodio_wait(inode); + + /* Mark the iocb as needing a decrement in ocfs2_dio_end_io */ + atomic_inc(&OCFS2_I(inode)->ip_unaligned_aio); + ocfs2_iocb_set_unaligned_aio(iocb); + } + /* * To later detect whether a journal commit for sync writes is * necessary, we sample i_size, and cluster count here. @@ -2371,8 +2405,12 @@ out_dio: if ((ret == -EIOCBQUEUED) || (!ocfs2_iocb_is_rw_locked(iocb))) { rw_level = -1; have_alloc_sem = 0; + unaligned_dio = 0; } + if (unaligned_dio) + atomic_dec(&OCFS2_I(inode)->ip_unaligned_aio); + out: if (rw_level != -1) ocfs2_rw_unlock(inode, rw_level); diff --git a/fs/ocfs2/inode.h b/fs/ocfs2/inode.h index 1c508b149b3ac..88924a3133fae 100644 --- a/fs/ocfs2/inode.h +++ b/fs/ocfs2/inode.h @@ -43,6 +43,9 @@ struct ocfs2_inode_info /* protects extended attribute changes on this inode */ struct rw_semaphore ip_xattr_sem; + /* Number of outstanding AIO's which are not page aligned */ + atomic_t ip_unaligned_aio; + /* These fields are protected by ip_lock */ spinlock_t ip_lock; u32 ip_open_count; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 029c4cd8a6915..603f5fe9f8160 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -54,6 +54,7 @@ #include "ocfs1_fs_compat.h" #include "alloc.h" +#include "aops.h" #include "blockcheck.h" #include "dlmglue.h" #include "export.h" @@ -1616,12 +1617,17 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt) return 0; } +wait_queue_head_t ocfs2__ioend_wq[OCFS2_IOEND_WQ_HASH_SZ]; + static int __init ocfs2_init(void) { - int status; + int status, i; ocfs2_print_version(); + for (i = 0; i < OCFS2_IOEND_WQ_HASH_SZ; i++) + init_waitqueue_head(&ocfs2__ioend_wq[i]); + status = init_ocfs2_uptodate_cache(); if (status < 0) { mlog_errno(status); @@ -1760,7 +1766,7 @@ static void ocfs2_inode_init_once(void *data) ocfs2_extent_map_init(&oi->vfs_inode); INIT_LIST_HEAD(&oi->ip_io_markers); oi->ip_dir_start_lookup = 0; - + atomic_set(&oi->ip_unaligned_aio, 0); init_rwsem(&oi->ip_alloc_sem); init_rwsem(&oi->ip_xattr_sem); mutex_init(&oi->ip_io_mutex); -- GitLab From c7e25e6e0b0486492c5faaf6312b37413642c48e Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 23 Jun 2011 22:51:47 +0200 Subject: [PATCH 0052/3089] ocfs2: Avoid livelock in ocfs2_readpage() When someone writes to an inode, readers accessing the same inode via ocfs2_readpage() just busyloop trying to get ip_alloc_sem because do_generic_file_read() looks up the page again and retries ->readpage() when previous attempt failed with AOP_TRUNCATED_PAGE. When there are enough readers, they can occupy all CPUs and in non-preempt kernel the system is deadlocked because writer holding ip_alloc_sem is never run to release the semaphore. Fix the problem by making reader block on ip_alloc_sem to break the busy loop. Signed-off-by: Jan Kara Signed-off-by: Joel Becker --- fs/ocfs2/aops.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index 4c1ec8f2d8c10..ba3ca1e63b51b 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -290,7 +290,15 @@ static int ocfs2_readpage(struct file *file, struct page *page) } if (down_read_trylock(&oi->ip_alloc_sem) == 0) { + /* + * Unlock the page and cycle ip_alloc_sem so that we don't + * busyloop waiting for ip_alloc_sem to unlock + */ ret = AOP_TRUNCATED_PAGE; + unlock_page(page); + unlock = 0; + down_read(&oi->ip_alloc_sem); + up_read(&oi->ip_alloc_sem); goto out_inode_unlock; } -- GitLab From dc9c0beecfeb4866fdbc245c295cf30d340b702c Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Tue, 11 Oct 2011 15:19:43 -0400 Subject: [PATCH 0053/3089] piix: ICH7 MWDMA1 errata Based on libata commit c611bed7. Signed-off-by: Bartlomiej Zolnierkiewicz Signed-off-by: David S. Miller --- drivers/ide/piix.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/drivers/ide/piix.c b/drivers/ide/piix.c index b59d04c720517..1892e81fb00f6 100644 --- a/drivers/ide/piix.c +++ b/drivers/ide/piix.c @@ -331,7 +331,7 @@ static const struct ide_port_ops ich_port_ops = { .udma_mask = udma, \ } -#define DECLARE_ICH_DEV(udma) \ +#define DECLARE_ICH_DEV(mwdma, udma) \ { \ .name = DRV_NAME, \ .init_chipset = init_chipset_ich, \ @@ -340,7 +340,7 @@ static const struct ide_port_ops ich_port_ops = { .port_ops = &ich_port_ops, \ .pio_mask = ATA_PIO4, \ .swdma_mask = ATA_SWDMA2_ONLY, \ - .mwdma_mask = ATA_MWDMA12_ONLY, \ + .mwdma_mask = mwdma, \ .udma_mask = udma, \ } @@ -362,13 +362,15 @@ static const struct ide_port_info piix_pci_info[] __devinitdata = { /* 2: PIIX4 */ DECLARE_PIIX_DEV(ATA_UDMA2), /* 3: ICH0 */ - DECLARE_ICH_DEV(ATA_UDMA2), + DECLARE_ICH_DEV(ATA_MWDMA12_ONLY, ATA_UDMA2), /* 4: ICH */ - DECLARE_ICH_DEV(ATA_UDMA4), + DECLARE_ICH_DEV(ATA_MWDMA12_ONLY, ATA_UDMA4), /* 5: PIIX4 */ DECLARE_PIIX_DEV(ATA_UDMA4), - /* 6: ICH[2-7]/ICH[2-3]M/C-ICH/ICH5-SATA/ESB2/ICH8M */ - DECLARE_ICH_DEV(ATA_UDMA5), + /* 6: ICH[2-6]/ICH[2-3]M/C-ICH/ICH5-SATA/ESB2/ICH8M */ + DECLARE_ICH_DEV(ATA_MWDMA12_ONLY, ATA_UDMA5), + /* 7: ICH7/7-R, no MWDMA1 */ + DECLARE_ICH_DEV(ATA_MWDMA2_ONLY, ATA_UDMA5), }; /** @@ -438,9 +440,9 @@ static const struct pci_device_id piix_pci_tbl[] = { #endif { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ESB_2), 6 }, { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH6_19), 6 }, - { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH7_21), 6 }, + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH7_21), 7 }, { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801DB_1), 6 }, - { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ESB2_18), 6 }, + { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ESB2_18), 7 }, { PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH8_6), 6 }, { 0, }, }; -- GitLab From 839e7306e5d1ebd57cf12e83d9910b8b2dbe79ba Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Wed, 12 Oct 2011 04:31:51 +0000 Subject: [PATCH 0054/3089] IDE: Don't powerdown Compaq Triflex IDE device on suspend Don't powerdown Compaq Triflex IDE device on suspend This fixes APM suspend on Compaq Armada 7400. APM BIOS doesn't suspend if IDE is powered down when suspending. The Triflex controller is found only on old Compaq boards, so this patch will hopefully have no side effects. This patch fixes a suspend regression introduced in feb22b7f8e62b1b987a3a1dbad95af767a1df832 ("ide: add proper PCI PM support (v2)"). Signed-off-by: Mikulas Patocka [bart: add commit's summary in parens] Signed-off-by: Bartlomiej Zolnierkiewicz Signed-off-by: David S. Miller --- drivers/ide/triflex.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/drivers/ide/triflex.c b/drivers/ide/triflex.c index e53a1b78378b0..281c914263459 100644 --- a/drivers/ide/triflex.c +++ b/drivers/ide/triflex.c @@ -113,12 +113,26 @@ static const struct pci_device_id triflex_pci_tbl[] = { }; MODULE_DEVICE_TABLE(pci, triflex_pci_tbl); +#ifdef CONFIG_PM +static int triflex_ide_pci_suspend(struct pci_dev *dev, pm_message_t state) +{ + /* + * We must not disable or powerdown the device. + * APM bios refuses to suspend if IDE is not accessible. + */ + pci_save_state(dev); + return 0; +} +#else +#define triflex_ide_pci_suspend NULL +#endif + static struct pci_driver triflex_pci_driver = { .name = "TRIFLEX_IDE", .id_table = triflex_pci_tbl, .probe = triflex_init_one, .remove = ide_pci_remove, - .suspend = ide_pci_suspend, + .suspend = triflex_ide_pci_suspend, .resume = ide_pci_resume, }; -- GitLab From acc8dbe7f44f1bab6fcf21f2d5efb32ea92e19fd Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Wed, 12 Oct 2011 04:45:34 +0000 Subject: [PATCH 0055/3089] icside: DMA support fix Fix problem introduced by commit 5e37bdc ("ide: add struct ide_dma_ops (take 3)"): d.dma_ops shouldn't be cleared if we are going to use DMA. Signed-off-by: Bartlomiej Zolnierkiewicz Signed-off-by: David S. Miller --- drivers/ide/icside.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ide/icside.c b/drivers/ide/icside.c index 4a697a238e280..8716066a2f2b7 100644 --- a/drivers/ide/icside.c +++ b/drivers/ide/icside.c @@ -521,8 +521,8 @@ icside_register_v6(struct icside_state *state, struct expansion_card *ec) if (ec->dma != NO_DMA && !request_dma(ec->dma, DRV_NAME)) { d.init_dma = icside_dma_init; d.port_ops = &icside_v6_port_ops; + } else d.dma_ops = NULL; - } ret = ide_host_register(host, &d, hws); if (ret) -- GitLab From 0ab3d8b3213c8bb55370b11fcc5321ee4f2c5e92 Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Thu, 13 Oct 2011 00:28:54 +0000 Subject: [PATCH 0056/3089] cy82c693: fix PCI device selection Wrong PCI device may be selected by cy82c693_set_pio_mode() if modular IDE host drivers are used and there are additional IDE PCI devices installed in the system. Fix it. Signed-off-by: Bartlomiej Zolnierkiewicz Signed-off-by: David S. Miller --- drivers/ide/cy82c693.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/ide/cy82c693.c b/drivers/ide/cy82c693.c index 67cbcfa351225..847553fd8b963 100644 --- a/drivers/ide/cy82c693.c +++ b/drivers/ide/cy82c693.c @@ -1,7 +1,7 @@ /* * Copyright (C) 1998-2000 Andreas S. Krebs (akrebs@altavista.net), Maintainer * Copyright (C) 1998-2002 Andre Hedrick , Integrator - * Copyright (C) 2007-2010 Bartlomiej Zolnierkiewicz + * Copyright (C) 2007-2011 Bartlomiej Zolnierkiewicz * * CYPRESS CY82C693 chipset IDE controller * @@ -90,7 +90,7 @@ static void cy82c693_set_pio_mode(ide_hwif_t *hwif, ide_drive_t *drive) u8 time_16, time_8; /* select primary or secondary channel */ - if (hwif->index > 0) { /* drive is on the secondary channel */ + if (drive->dn > 1) { /* drive is on the secondary channel */ dev = pci_get_slot(dev->bus, dev->devfn+1); if (!dev) { printk(KERN_ERR "%s: tune_drive: " @@ -141,7 +141,7 @@ static void cy82c693_set_pio_mode(ide_hwif_t *hwif, ide_drive_t *drive) pci_write_config_byte(dev, CY82_IDE_SLAVE_IOW, time_16); pci_write_config_byte(dev, CY82_IDE_SLAVE_8BIT, time_8); } - if (hwif->index > 0) + if (drive->dn > 1) pci_dev_put(dev); } -- GitLab From c2bc11113c50449f23c40b724fe410fc2380a8e9 Mon Sep 17 00:00:00 2001 From: John Stultz Date: Thu, 27 Oct 2011 18:12:42 -0700 Subject: [PATCH 0057/3089] time: Improve documentation of timekeeeping_adjust() After getting a number of questions in private emails about the math around admittedly very complex timekeeping_adjust() and timekeeping_big_adjust(), I figure the code needs some better comments. Hopefully the explanations are clear enough and don't muddy the water any worse. Still needs documentation for ntp_error, but I couldn't recall exactly the full explanation behind the code that's there (although I do recall once working it out when Roman first proposed it). Given a bit more time I can probably work it out, but I don't want to hold back this documentation until then. Signed-off-by: John Stultz Cc: Chen Jie Cc: Steven Rostedt Link: http://lkml.kernel.org/r/1319764362-32367-1-git-send-email-john.stultz@linaro.org Signed-off-by: Ingo Molnar --- kernel/time/timekeeping.c | 81 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 80 insertions(+), 1 deletion(-) diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 2b021b0e8507e..025e136f3881d 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -802,14 +802,44 @@ static void timekeeping_adjust(s64 offset) s64 error, interval = timekeeper.cycle_interval; int adj; + /* + * The point of this is to check if the error is greater then half + * an interval. + * + * First we shift it down from NTP_SHIFT to clocksource->shifted nsecs. + * + * Note we subtract one in the shift, so that error is really error*2. + * This "saves" dividing(shifting) intererval twice, but keeps the + * (error > interval) comparision as still measuring if error is + * larger then half an interval. + * + * Note: It does not "save" on aggrivation when reading the code. + */ error = timekeeper.ntp_error >> (timekeeper.ntp_error_shift - 1); if (error > interval) { + /* + * We now divide error by 4(via shift), which checks if + * the error is greater then twice the interval. + * If it is greater, we need a bigadjust, if its smaller, + * we can adjust by 1. + */ error >>= 2; + /* + * XXX - In update_wall_time, we round up to the next + * nanosecond, and store the amount rounded up into + * the error. This causes the likely below to be unlikely. + * + * The properfix is to avoid rounding up by using + * the high precision timekeeper.xtime_nsec instead of + * xtime.tv_nsec everywhere. Fixing this will take some + * time. + */ if (likely(error <= interval)) adj = 1; else adj = timekeeping_bigadjust(error, &interval, &offset); } else if (error < -interval) { + /* See comment above, this is just switched for the negative */ error >>= 2; if (likely(error >= -interval)) { adj = -1; @@ -817,9 +847,58 @@ static void timekeeping_adjust(s64 offset) offset = -offset; } else adj = timekeeping_bigadjust(error, &interval, &offset); - } else + } else /* No adjustment needed */ return; + /* + * So the following can be confusing. + * + * To keep things simple, lets assume adj == 1 for now. + * + * When adj != 1, remember that the interval and offset values + * have been appropriately scaled so the math is the same. + * + * The basic idea here is that we're increasing the multiplier + * by one, this causes the xtime_interval to be incremented by + * one cycle_interval. This is because: + * xtime_interval = cycle_interval * mult + * So if mult is being incremented by one: + * xtime_interval = cycle_interval * (mult + 1) + * Its the same as: + * xtime_interval = (cycle_interval * mult) + cycle_interval + * Which can be shortened to: + * xtime_interval += cycle_interval + * + * So offset stores the non-accumulated cycles. Thus the current + * time (in shifted nanoseconds) is: + * now = (offset * adj) + xtime_nsec + * Now, even though we're adjusting the clock frequency, we have + * to keep time consistent. In other words, we can't jump back + * in time, and we also want to avoid jumping forward in time. + * + * So given the same offset value, we need the time to be the same + * both before and after the freq adjustment. + * now = (offset * adj_1) + xtime_nsec_1 + * now = (offset * adj_2) + xtime_nsec_2 + * So: + * (offset * adj_1) + xtime_nsec_1 = + * (offset * adj_2) + xtime_nsec_2 + * And we know: + * adj_2 = adj_1 + 1 + * So: + * (offset * adj_1) + xtime_nsec_1 = + * (offset * (adj_1+1)) + xtime_nsec_2 + * (offset * adj_1) + xtime_nsec_1 = + * (offset * adj_1) + offset + xtime_nsec_2 + * Canceling the sides: + * xtime_nsec_1 = offset + xtime_nsec_2 + * Which gives us: + * xtime_nsec_2 = xtime_nsec_1 - offset + * Which simplfies to: + * xtime_nsec -= offset + * + * XXX - TODO: Doc ntp_error calculation. + */ timekeeper.mult += adj; timekeeper.xtime_interval += interval; timekeeper.xtime_nsec -= offset; -- GitLab From f74974c763c872d65da4062a14f4e73c77a0b577 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Tue, 11 Oct 2011 17:27:51 +0200 Subject: [PATCH 0058/3089] drm/i915: disable temporal dithering on the internal panel MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit i'm getting tempted to just disable temporal Approved. apparently it makes the screen look pulse-y which is worse than the disease. References: http://lists.freedesktop.org/archives/intel-gfx/2011-October/012545.html Tested-by: Олег Герман Reviewed-by: Adam Jackson Signed-off-by: Daniel Vetter Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_display.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 981b1f1c04d8d..b75bd93cf59b0 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -5671,7 +5671,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, pipeconf &= ~PIPECONF_DITHER_TYPE_MASK; if ((is_lvds && dev_priv->lvds_dither) || dither) { pipeconf |= PIPECONF_DITHER_EN; - pipeconf |= PIPECONF_DITHER_TYPE_ST1; + pipeconf |= PIPECONF_DITHER_TYPE_SP; } if (is_dp || intel_encoder_is_pch_edp(&has_edp_encoder->base)) { intel_dp_set_m_n(crtc, mode, adjusted_mode); -- GitLab From 80a2901d2a59946d098c4479d650d0c3c82c3d85 Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Tue, 11 Oct 2011 10:59:05 +0200 Subject: [PATCH 0059/3089] drm/i915: only match on PCI_BASE_CLASS_DISPLAY ... not DISPLAY_VGA, because we ignore the VGA subclass with our class_mask. It confused me until Chris Wilson clued me up. Signed-off-by: Daniel Vetter Reviewed-by: Adam Jackson Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 4c8d681c2151e..548e04bade3c3 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -106,7 +106,7 @@ static struct drm_driver driver; extern int intel_agp_enabled; #define INTEL_VGA_DEVICE(id, info) { \ - .class = PCI_CLASS_DISPLAY_VGA << 8, \ + .class = PCI_BASE_CLASS_DISPLAY << 16, \ .class_mask = 0xff0000, \ .vendor = 0x8086, \ .device = id, \ -- GitLab From 828204903945002be837d938401e242978b7b505 Mon Sep 17 00:00:00 2001 From: Adam Jackson Date: Mon, 10 Oct 2011 16:33:34 -0400 Subject: [PATCH 0060/3089] drm/i915: intel_choose_pipe_bpp_dither messages should be DRM_DEBUG_KMS Shouldn't hide these behind _DRIVER, they're all KMS-related. Signed-off-by: Adam Jackson Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_display.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index b75bd93cf59b0..9fa342e89454b 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -4711,7 +4711,7 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc, lvds_bpc = 6; if (lvds_bpc < display_bpc) { - DRM_DEBUG_DRIVER("clamping display bpc (was %d) to LVDS (%d)\n", display_bpc, lvds_bpc); + DRM_DEBUG_KMS("clamping display bpc (was %d) to LVDS (%d)\n", display_bpc, lvds_bpc); display_bpc = lvds_bpc; } continue; @@ -4722,7 +4722,7 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc, unsigned int edp_bpc = dev_priv->edp.bpp / 3; if (edp_bpc < display_bpc) { - DRM_DEBUG_DRIVER("clamping display bpc (was %d) to eDP (%d)\n", display_bpc, edp_bpc); + DRM_DEBUG_KMS("clamping display bpc (was %d) to eDP (%d)\n", display_bpc, edp_bpc); display_bpc = edp_bpc; } continue; @@ -4737,7 +4737,7 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc, /* Don't use an invalid EDID bpc value */ if (connector->display_info.bpc && connector->display_info.bpc < display_bpc) { - DRM_DEBUG_DRIVER("clamping display bpc (was %d) to EDID reported max of %d\n", display_bpc, connector->display_info.bpc); + DRM_DEBUG_KMS("clamping display bpc (was %d) to EDID reported max of %d\n", display_bpc, connector->display_info.bpc); display_bpc = connector->display_info.bpc; } } @@ -4748,10 +4748,10 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc, */ if (intel_encoder->type == INTEL_OUTPUT_HDMI) { if (display_bpc > 8 && display_bpc < 12) { - DRM_DEBUG_DRIVER("forcing bpc to 12 for HDMI\n"); + DRM_DEBUG_KMS("forcing bpc to 12 for HDMI\n"); display_bpc = 12; } else { - DRM_DEBUG_DRIVER("forcing bpc to 8 for HDMI\n"); + DRM_DEBUG_KMS("forcing bpc to 8 for HDMI\n"); display_bpc = 8; } } @@ -4789,8 +4789,8 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc, display_bpc = min(display_bpc, bpc); - DRM_DEBUG_DRIVER("setting pipe bpc to %d (max display bpc %d)\n", - bpc, display_bpc); + DRM_DEBUG_KMS("setting pipe bpc to %d (max display bpc %d)\n", + bpc, display_bpc); *pipe_bpp = display_bpc * 3; -- GitLab From 3fca806cf343155d0bb0dfbdbc2e51ef47b65750 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 28 Oct 2011 14:42:41 +0300 Subject: [PATCH 0061/3089] drm/i915: fix if statement (bogus semi-colon) The semi-colon is a typo here and it makes the if statement unconditional. Signed-off-by: Dan Carpenter Signed-off-by: Keith Packard --- drivers/char/agp/intel-gtt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c index 66cd0b8096ca4..3a8d44886010d 100644 --- a/drivers/char/agp/intel-gtt.c +++ b/drivers/char/agp/intel-gtt.c @@ -1236,7 +1236,7 @@ static int i9xx_setup(void) intel_private.gtt_bus_addr = reg_addr + gtt_offset; } - if (needs_idle_maps()); + if (needs_idle_maps()) intel_private.base.do_idle_maps = 1; intel_i9xx_setup_flush(); -- GitLab From c5e62bdd73da80e1017058c923336a5a150af9f2 Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Fri, 28 Oct 2011 10:28:00 -0700 Subject: [PATCH 0062/3089] agp: iommu_gfx_mapped only available if CONFIG_INTEL_IOMMU is set Kernels with no iommu support cannot ever need the Ironlake work-around, so never enable it in that case. Might be better to completely remove the work-around from the kernel in this case? Signed-off-by: Keith Packard Reviewed-by: Ben Widawsky --- drivers/char/agp/intel-gtt.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c index 3a8d44886010d..c92424ca1a553 100644 --- a/drivers/char/agp/intel-gtt.c +++ b/drivers/char/agp/intel-gtt.c @@ -1186,10 +1186,11 @@ static void gen6_cleanup(void) /* Certain Gen5 chipsets require require idling the GPU before * unmapping anything from the GTT when VT-d is enabled. */ -extern int intel_iommu_gfx_mapped; static inline int needs_idle_maps(void) { +#ifdef CONFIG_INTEL_IOMMU const unsigned short gpu_devid = intel_private.pcidev->device; + extern int intel_iommu_gfx_mapped; /* Query intel_iommu to see if we need the workaround. Presumably that * was loaded first. @@ -1198,7 +1199,7 @@ static inline int needs_idle_maps(void) gpu_devid == PCI_DEVICE_ID_INTEL_IRONLAKE_M_IG) && intel_iommu_gfx_mapped) return 1; - +#endif return 0; } -- GitLab From 0d52f54e2ef64c189dedc332e680b2eb4a34590a Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 22 Oct 2011 00:43:38 +0200 Subject: [PATCH 0063/3089] PCI / ACPI: Make acpiphp ignore root bridges using PCIe native hotplug If the kernel has requested control of the PCIe native hotplug feature for a given root complex, the acpiphp driver should not try to handle that root complex and it should leave it to pciehp. Failing to do so causes problems to happen if acpiphp is loaded before pciehp on such systems. To address this issue make find_root_bridges() ignore PCIe root complexes with PCIe native hotplug enabled and make add_bridge() return error code if PCIe native hotplug is enabled for the given root port. This causes acpiphp to refuse to load if PCIe native hotplug is enabled for all complexes and to refuse binding to the root complexes with PCIe native hotplug is enabled. Acked-by: Kenji Kaneshige Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes --- drivers/pci/hotplug/acpiphp_glue.c | 29 ++++++++++++++++++++++++----- 1 file changed, 24 insertions(+), 5 deletions(-) diff --git a/drivers/pci/hotplug/acpiphp_glue.c b/drivers/pci/hotplug/acpiphp_glue.c index 596172b4ae955..fce1c54a0c8d8 100644 --- a/drivers/pci/hotplug/acpiphp_glue.c +++ b/drivers/pci/hotplug/acpiphp_glue.c @@ -459,8 +459,17 @@ static int add_bridge(acpi_handle handle) { acpi_status status; unsigned long long tmp; + struct acpi_pci_root *root; acpi_handle dummy_handle; + /* + * We shouldn't use this bridge if PCIe native hotplug control has been + * granted by the BIOS for it. + */ + root = acpi_pci_find_root(handle); + if (root && (root->osc_control_set & OSC_PCI_EXPRESS_NATIVE_HP_CONTROL)) + return -ENODEV; + /* if the bridge doesn't have _STA, we assume it is always there */ status = acpi_get_handle(handle, "_STA", &dummy_handle); if (ACPI_SUCCESS(status)) { @@ -1376,13 +1385,23 @@ static void handle_hotplug_event_func(acpi_handle handle, u32 type, static acpi_status find_root_bridges(acpi_handle handle, u32 lvl, void *context, void **rv) { + struct acpi_pci_root *root; int *count = (int *)context; - if (acpi_is_root_bridge(handle)) { - acpi_install_notify_handler(handle, ACPI_SYSTEM_NOTIFY, - handle_hotplug_event_bridge, NULL); - (*count)++; - } + if (!acpi_is_root_bridge(handle)) + return AE_OK; + + root = acpi_pci_find_root(handle); + if (!root) + return AE_OK; + + if (root->osc_control_set & OSC_PCI_EXPRESS_NATIVE_HP_CONTROL) + return AE_OK; + + (*count)++; + acpi_install_notify_handler(handle, ACPI_SYSTEM_NOTIFY, + handle_hotplug_event_bridge, NULL); + return AE_OK ; } -- GitLab From 466b3ddfbcf4f5ce402a77397630a0fa9ea9ce6b Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Sun, 30 Oct 2011 16:35:08 +0100 Subject: [PATCH 0064/3089] PCI: Fix compile errors with PCI_ATS and !PCI_IOV The ats and sroiv members of 'struct pci_dev' are required for the ATS code already, even without IOV support compiled in. So depend on ATS here. This is fine with PCI_IOV too because it selects PCI_ATS. Also the prototypes for ATS need to be available for PCI_ATS. Reported-by: Randy Dunlap Signed-off-by: Joerg Roedel Signed-off-by: Jesse Barnes --- include/linux/pci-ats.h | 6 +++--- include/linux/pci.h | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/include/linux/pci-ats.h b/include/linux/pci-ats.h index e3d0b38902491..7ef68724f0f05 100644 --- a/include/linux/pci-ats.h +++ b/include/linux/pci-ats.h @@ -12,7 +12,7 @@ struct pci_ats { unsigned int is_enabled:1; /* Enable bit is set */ }; -#ifdef CONFIG_PCI_IOV +#ifdef CONFIG_PCI_ATS extern int pci_enable_ats(struct pci_dev *dev, int ps); extern void pci_disable_ats(struct pci_dev *dev); @@ -29,7 +29,7 @@ static inline int pci_ats_enabled(struct pci_dev *dev) return dev->ats && dev->ats->is_enabled; } -#else /* CONFIG_PCI_IOV */ +#else /* CONFIG_PCI_ATS */ static inline int pci_enable_ats(struct pci_dev *dev, int ps) { @@ -50,7 +50,7 @@ static inline int pci_ats_enabled(struct pci_dev *dev) return 0; } -#endif /* CONFIG_PCI_IOV */ +#endif /* CONFIG_PCI_ATS */ #ifdef CONFIG_PCI_PRI diff --git a/include/linux/pci.h b/include/linux/pci.h index 337df0d5d5f7e..7cda65b5f7980 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -338,7 +338,7 @@ struct pci_dev { struct list_head msi_list; #endif struct pci_vpd *vpd; -#ifdef CONFIG_PCI_IOV +#ifdef CONFIG_PCI_ATS union { struct pci_sriov *sriov; /* SR-IOV capability related */ struct pci_dev *physfn; /* the PF this VF is associated with */ -- GitLab From c54420d3302d30999060e2669d0800b85e5b5a20 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Sun, 30 Oct 2011 16:35:07 +0100 Subject: [PATCH 0065/3089] PCI: Let PCI_PRI depend on PCI This avoids the PCI_PRI question in 'make config' when PCI is not selected. Reported-by: Geert Uytterhoeven Signed-off-by: Joerg Roedel Signed-off-by: Jesse Barnes --- drivers/pci/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig index cec66064ee4b5..8252fc3430bdb 100644 --- a/drivers/pci/Kconfig +++ b/drivers/pci/Kconfig @@ -87,6 +87,7 @@ config PCI_IOV config PCI_PRI bool "PCI PRI support" + depends on PCI select PCI_ATS help PRI is the PCI Page Request Interface. It allows PCI devices that are -- GitLab From d631097577f6fe027f4903f62eabced6445d19bf Mon Sep 17 00:00:00 2001 From: Ilya Dryomov Date: Mon, 31 Oct 2011 11:07:42 +0200 Subject: [PATCH 0066/3089] tracing: fix event_subsystem ref counting Fix a bug introduced by e9dbfae5, which prevents event_subsystem from ever being released. Ref_count was added to keep track of subsystem users, not for counting events. Subsystem is created with ref_count = 1, so there is no need to increment it for every event, we have nr_events for that. Fix this by touching ref_count only when we actually have a new user - subsystem_open(). Cc: stable@vger.kernel.org Signed-off-by: Ilya Dryomov Link: http://lkml.kernel.org/r/1320052062-7846-1-git-send-email-idryomov@gmail.com Signed-off-by: Steven Rostedt --- kernel/trace/trace_events.c | 1 - 1 file changed, 1 deletion(-) diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index 581876f9f3872..c212a7f934ec4 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -1078,7 +1078,6 @@ event_subsystem_dir(const char *name, struct dentry *d_events) /* First see if we did not already create this dir */ list_for_each_entry(system, &event_subsystems, list) { if (strcmp(system->name, name) == 0) { - __get_system(system); system->nr_events++; return system->entry; } -- GitLab From 627f7675f0f530ea555d76543dc4e469d70a1532 Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Mon, 31 Oct 2011 11:30:10 -0700 Subject: [PATCH 0067/3089] drm/i915: Use mode_config.mutex in ironlake_panel_vdd_work Use of the struct_mutex is not correct for locking in mode setting paths. Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/intel_dp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index fc1a0832af4fe..7259034b33d1d 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -971,9 +971,9 @@ static void ironlake_panel_vdd_work(struct work_struct *__work) struct intel_dp, panel_vdd_work); struct drm_device *dev = intel_dp->base.base.dev; - mutex_lock(&dev->struct_mutex); + mutex_lock(&dev->mode_config.mutex); ironlake_panel_vdd_off_sync(intel_dp); - mutex_unlock(&dev->struct_mutex); + mutex_unlock(&dev->mode_config.mutex); } static void ironlake_edp_panel_vdd_off(struct intel_dp *intel_dp, bool sync) -- GitLab From ff56b0bc84c01fb2c5d2ec6438becfe0f5f2f78f Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Mon, 31 Oct 2011 23:16:21 -0700 Subject: [PATCH 0068/3089] drm/i915: Fix object refcount leak on mmappable size limit error path. I've been seeing memory leaks on my system in the form of large (300-400MB) GEM objects created by now-dead processes laying around clogging up memory. I usually notice when it gets to about 1.2GB of them. Hopefully this clears up the issue, but I just found this bug by inspection. Signed-off-by: Eric Anholt Cc: stable@kernel.org Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_gem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 6651c36b6e8a1..d18b07adcffa3 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1396,7 +1396,7 @@ i915_gem_mmap_gtt(struct drm_file *file, if (obj->base.size > dev_priv->mm.gtt_mappable_end) { ret = -E2BIG; - goto unlock; + goto out; } if (obj->madv != I915_MADV_WILLNEED) { -- GitLab From ed0449af5373abd766c79fbf83254bebc996bd23 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 1 Nov 2011 09:09:35 +0800 Subject: [PATCH 0069/3089] tracing: Restore system filter behavior Though not all events have field 'prev_pid', it was allowed to do this: # echo 'prev_pid == 100' > events/sched/filter but commit 75b8e98263fdb0bfbdeba60d4db463259f1fe8a2 (tracing/filter: Swap entire filter of events) broke it without any reason. Link: http://lkml.kernel.org/r/4EAF46CF.8040408@cn.fujitsu.com Signed-off-by: Li Zefan Signed-off-by: Steven Rostedt --- include/linux/ftrace_event.h | 2 ++ kernel/trace/trace_events_filter.c | 7 ++++++- 2 files changed, 8 insertions(+), 1 deletion(-) diff --git a/include/linux/ftrace_event.h b/include/linux/ftrace_event.h index 96efa6794ea52..c3da42dd22baf 100644 --- a/include/linux/ftrace_event.h +++ b/include/linux/ftrace_event.h @@ -172,6 +172,7 @@ enum { TRACE_EVENT_FL_FILTERED_BIT, TRACE_EVENT_FL_RECORDED_CMD_BIT, TRACE_EVENT_FL_CAP_ANY_BIT, + TRACE_EVENT_FL_NO_SET_FILTER_BIT, }; enum { @@ -179,6 +180,7 @@ enum { TRACE_EVENT_FL_FILTERED = (1 << TRACE_EVENT_FL_FILTERED_BIT), TRACE_EVENT_FL_RECORDED_CMD = (1 << TRACE_EVENT_FL_RECORDED_CMD_BIT), TRACE_EVENT_FL_CAP_ANY = (1 << TRACE_EVENT_FL_CAP_ANY_BIT), + TRACE_EVENT_FL_NO_SET_FILTER = (1 << TRACE_EVENT_FL_NO_SET_FILTER_BIT), }; struct ftrace_event_call { diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 816d3d0749793..86040d9c1accf 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -1649,7 +1649,9 @@ static int replace_system_preds(struct event_subsystem *system, */ err = replace_preds(call, NULL, ps, filter_string, true); if (err) - goto fail; + call->flags |= TRACE_EVENT_FL_NO_SET_FILTER; + else + call->flags &= ~TRACE_EVENT_FL_NO_SET_FILTER; } list_for_each_entry(call, &ftrace_events, list) { @@ -1658,6 +1660,9 @@ static int replace_system_preds(struct event_subsystem *system, if (strcmp(call->class->system, system->name) != 0) continue; + if (call->flags & TRACE_EVENT_FL_NO_SET_FILTER) + continue; + filter_item = kzalloc(sizeof(*filter_item), GFP_KERNEL); if (!filter_item) goto fail_mem; -- GitLab From f43d623164022dcbf6750ef220b7a1133a1183eb Mon Sep 17 00:00:00 2001 From: Don Zickus Date: Thu, 20 Oct 2011 23:52:14 -0400 Subject: [PATCH 0070/3089] usb, xhci: fix lockdep warning on endpoint timeout While debugging a usb3 problem, I stumbled upon this lockdep warning. Oct 18 21:41:17 dhcp47-74 kernel: ================================= Oct 18 21:41:17 dhcp47-74 kernel: [ INFO: inconsistent lock state ] Oct 18 21:41:17 dhcp47-74 kernel: 3.1.0-rc4nmi+ #456 Oct 18 21:41:17 dhcp47-74 kernel: --------------------------------- Oct 18 21:41:17 dhcp47-74 kernel: inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage. Oct 18 21:41:17 dhcp47-74 kernel: swapper/0 [HC0[0]:SC1[1]:HE1:SE0] takes: Oct 18 21:41:17 dhcp47-74 kernel: (&(&xhci->lock)->rlock){?.-...}, at: [] xhci_stop_endpoint_command_watchdog+0x30/0x340 [xhci_hcd] Oct 18 21:41:17 dhcp47-74 kernel: {IN-HARDIRQ-W} state was registered at: Oct 18 21:41:17 dhcp47-74 kernel: [] __lock_acquire+0x781/0x1660 Oct 18 21:41:17 dhcp47-74 kernel: [] lock_acquire+0x97/0x170 Oct 18 21:41:17 dhcp47-74 kernel: [] _raw_spin_lock+0x46/0x80 Oct 18 21:41:17 dhcp47-74 kernel: [] xhci_irq+0x3a/0x1960 [xhci_hcd] Oct 18 21:41:17 dhcp47-74 kernel: [] xhci_msi_irq+0x31/0x40 [xhci_hcd] Oct 18 21:41:17 dhcp47-74 kernel: [] handle_irq_event_percpu+0x85/0x320 Oct 18 21:41:17 dhcp47-74 kernel: [] handle_irq_event+0x48/0x70 Oct 18 21:41:17 dhcp47-74 kernel: [] handle_edge_irq+0x6d/0x130 Oct 18 21:41:17 dhcp47-74 kernel: [] handle_irq+0x49/0xa0 Oct 18 21:41:17 dhcp47-74 kernel: [] do_IRQ+0x5d/0xe0 Oct 18 21:41:17 dhcp47-74 kernel: [] ret_from_intr+0x0/0x13 Oct 18 21:41:17 dhcp47-74 kernel: [] usb_set_device_state+0x8a/0x180 Oct 18 21:41:17 dhcp47-74 kernel: [] usb_add_hcd+0x2b8/0x730 Oct 18 21:41:17 dhcp47-74 kernel: [] xhci_pci_probe+0x9e/0xd4 [xhci_hcd] Oct 18 21:41:17 dhcp47-74 kernel: [] local_pci_probe+0x5f/0xd0 Oct 18 21:41:17 dhcp47-74 kernel: [] pci_device_probe+0x119/0x120 Oct 18 21:41:17 dhcp47-74 kernel: [] driver_probe_device+0xa3/0x2c0 Oct 18 21:41:17 dhcp47-74 kernel: [] __driver_attach+0xab/0xb0 Oct 18 21:41:17 dhcp47-74 kernel: [] bus_for_each_dev+0x6c/0xa0 Oct 18 21:41:17 dhcp47-74 kernel: [] driver_attach+0x1e/0x20 Oct 18 21:41:17 dhcp47-74 kernel: [] bus_add_driver+0x1f8/0x2b0 Oct 18 21:41:17 dhcp47-74 kernel: [] driver_register+0x76/0x140 Oct 18 21:41:17 dhcp47-74 kernel: [] __pci_register_driver+0x66/0xe0 Oct 18 21:41:17 dhcp47-74 kernel: [] snd_timer_find+0x4a/0x70 [snd_timer] Oct 18 21:41:17 dhcp47-74 kernel: [] snd_timer_find+0xe/0x70 [snd_timer] Oct 18 21:41:17 dhcp47-74 kernel: [] do_one_initcall+0x43/0x180 Oct 18 21:41:17 dhcp47-74 kernel: [] sys_init_module+0x92/0x1f0 Oct 18 21:41:17 dhcp47-74 kernel: [] system_call_fastpath+0x16/0x1b Oct 18 21:41:17 dhcp47-74 kernel: irq event stamp: 631984 Oct 18 21:41:17 dhcp47-74 kernel: hardirqs last enabled at (631984): [] _raw_spin_unlock_irq+0x30/0x50 Oct 18 21:41:17 dhcp47-74 kernel: hardirqs last disabled at (631983): [] _raw_spin_lock_irq+0x19/0x90 Oct 18 21:41:17 dhcp47-74 kernel: softirqs last enabled at (631980): [] _local_bh_enable+0x13/0x20 Oct 18 21:41:17 dhcp47-74 kernel: softirqs last disabled at (631981): [] call_softirq+0x1c/0x30 Oct 18 21:41:17 dhcp47-74 kernel: Oct 18 21:41:17 dhcp47-74 kernel: other info that might help us debug this: Oct 18 21:41:17 dhcp47-74 kernel: Possible unsafe locking scenario: Oct 18 21:41:17 dhcp47-74 kernel: Oct 18 21:41:17 dhcp47-74 kernel: CPU0 Oct 18 21:41:17 dhcp47-74 kernel: ---- Oct 18 21:41:17 dhcp47-74 kernel: lock(&(&xhci->lock)->rlock); Oct 18 21:41:17 dhcp47-74 kernel: Oct 18 21:41:17 dhcp47-74 kernel: lock(&(&xhci->lock)->rlock); Oct 18 21:41:17 dhcp47-74 kernel: Oct 18 21:41:17 dhcp47-74 kernel: *** DEADLOCK *** Oct 18 21:41:17 dhcp47-74 kernel: Oct 18 21:41:17 dhcp47-74 kernel: 1 lock held by swapper/0: Oct 18 21:41:17 dhcp47-74 kernel: #0: (&ep->stop_cmd_timer){+.-...}, at: [] run_timer_softirq+0x162/0x570 Oct 18 21:41:17 dhcp47-74 kernel: Oct 18 21:41:17 dhcp47-74 kernel: stack backtrace: Oct 18 21:41:17 dhcp47-74 kernel: Pid: 0, comm: swapper Tainted: G W 3.1.0-rc4nmi+ #456 Oct 18 21:41:17 dhcp47-74 kernel: Call Trace: Oct 18 21:41:17 dhcp47-74 kernel: [] print_usage_bug+0x227/0x270 Oct 18 21:41:17 dhcp47-74 kernel: [] mark_lock+0x346/0x410 Oct 18 21:41:17 dhcp47-74 kernel: [] __lock_acquire+0x61e/0x1660 Oct 18 21:41:17 dhcp47-74 kernel: [] ? mark_lock+0x213/0x410 Oct 18 21:41:17 dhcp47-74 kernel: [] lock_acquire+0x97/0x170 Oct 18 21:41:17 dhcp47-74 kernel: [] ? xhci_stop_endpoint_command_watchdog+0x30/0x340 [xhci_hcd] Oct 18 21:41:17 dhcp47-74 kernel: [] _raw_spin_lock+0x46/0x80 Oct 18 21:41:17 dhcp47-74 kernel: [] ? xhci_stop_endpoint_command_watchdog+0x30/0x340 [xhci_hcd] Oct 18 21:41:17 dhcp47-74 kernel: [] xhci_stop_endpoint_command_watchdog+0x30/0x340 [xhci_hcd] Oct 18 21:41:17 dhcp47-74 kernel: [] ? run_timer_softirq+0x162/0x570 Oct 18 21:41:17 dhcp47-74 kernel: [] run_timer_softirq+0x20d/0x570 Oct 18 21:41:17 dhcp47-74 kernel: [] ? run_timer_softirq+0x162/0x570 Oct 18 21:41:17 dhcp47-74 kernel: [] ? xhci_queue_isoc_tx_prepare+0x8e0/0x8e0 [xhci_hcd] Oct 18 21:41:17 dhcp47-74 kernel: [] __do_softirq+0xf2/0x3f0 Oct 18 21:41:17 dhcp47-74 kernel: [] ? lapic_next_event+0x1d/0x30 Oct 18 21:41:17 dhcp47-74 kernel: [] ? clockevents_program_event+0x5e/0x90 Oct 18 21:41:17 dhcp47-74 kernel: [] call_softirq+0x1c/0x30 Oct 18 21:41:17 dhcp47-74 kernel: [] do_softirq+0x8d/0xc0 Oct 18 21:41:17 dhcp47-74 kernel: [] irq_exit+0xe5/0x100 Oct 18 21:41:17 dhcp47-74 kernel: [] smp_apic_timer_interrupt+0x6e/0x99 Oct 18 21:41:17 dhcp47-74 kernel: [] apic_timer_interrupt+0x70/0x80 Oct 18 21:41:17 dhcp47-74 kernel: [] ? trace_hardirqs_off+0xd/0x10 Oct 18 21:41:17 dhcp47-74 kernel: [] ? acpi_idle_enter_bm+0x227/0x25b Oct 18 21:41:17 dhcp47-74 kernel: [] ? acpi_idle_enter_bm+0x222/0x25b Oct 18 21:41:17 dhcp47-74 kernel: [] cpuidle_idle_call+0x103/0x290 Oct 18 21:41:17 dhcp47-74 kernel: [] cpu_idle+0xe5/0x160 Oct 18 21:41:17 dhcp47-74 kernel: [] rest_init+0xe0/0xf0 Oct 18 21:41:17 dhcp47-74 kernel: [] ? csum_partial_copy_generic+0x170/0x170 Oct 18 21:41:17 dhcp47-74 kernel: [] start_kernel+0x3fc/0x407 Oct 18 21:41:17 dhcp47-74 kernel: [] x86_64_start_reservations+0x131/0x135 Oct 18 21:41:17 dhcp47-74 kernel: [] x86_64_start_kernel+0xed/0xf4 Oct 18 21:41:17 dhcp47-74 kernel: xhci_hcd 0000:00:14.0: xHCI host not responding to stop endpoint command. Oct 18 21:41:17 dhcp47-74 kernel: xhci_hcd 0000:00:14.0: Assuming host is dying, halting host. Oct 18 21:41:17 dhcp47-74 kernel: xhci_hcd 0000:00:14.0: HC died; cleaning up Oct 18 21:41:17 dhcp47-74 kernel: usb 3-4: device descriptor read/8, error -110 Oct 18 21:41:17 dhcp47-74 kernel: usb 3-4: device descriptor read/8, error -22 Oct 18 21:41:17 dhcp47-74 kernel: hub 3-0:1.0: cannot disable port 4 (err = -19) Basically what is happening is in xhci_stop_endpoint_command_watchdog() the xhci->lock is grabbed with just spin_lock. What lockdep deduces is that if an interrupt occurred while in this function it would deadlock with xhci_irq because that function also grabs the xhci->lock. Fixing it is trivial by using spin_lock_irqsave instead. This should be queued to stable kernels as far back as 2.6.33. Signed-off-by: Don Zickus Signed-off-by: Sarah Sharp Cc: stable@kernel.org --- drivers/usb/host/xhci-ring.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/usb/host/xhci-ring.c b/drivers/usb/host/xhci-ring.c index 940321b3ec68b..9f1d4b15d8185 100644 --- a/drivers/usb/host/xhci-ring.c +++ b/drivers/usb/host/xhci-ring.c @@ -816,23 +816,24 @@ void xhci_stop_endpoint_command_watchdog(unsigned long arg) struct xhci_ring *ring; struct xhci_td *cur_td; int ret, i, j; + unsigned long flags; ep = (struct xhci_virt_ep *) arg; xhci = ep->xhci; - spin_lock(&xhci->lock); + spin_lock_irqsave(&xhci->lock, flags); ep->stop_cmds_pending--; if (xhci->xhc_state & XHCI_STATE_DYING) { xhci_dbg(xhci, "Stop EP timer ran, but another timer marked " "xHCI as DYING, exiting.\n"); - spin_unlock(&xhci->lock); + spin_unlock_irqrestore(&xhci->lock, flags); return; } if (!(ep->stop_cmds_pending == 0 && (ep->ep_state & EP_HALT_PENDING))) { xhci_dbg(xhci, "Stop EP timer ran, but no command pending, " "exiting.\n"); - spin_unlock(&xhci->lock); + spin_unlock_irqrestore(&xhci->lock, flags); return; } @@ -844,11 +845,11 @@ void xhci_stop_endpoint_command_watchdog(unsigned long arg) xhci->xhc_state |= XHCI_STATE_DYING; /* Disable interrupts from the host controller and start halting it */ xhci_quiesce(xhci); - spin_unlock(&xhci->lock); + spin_unlock_irqrestore(&xhci->lock, flags); ret = xhci_halt(xhci); - spin_lock(&xhci->lock); + spin_lock_irqsave(&xhci->lock, flags); if (ret < 0) { /* This is bad; the host is not responding to commands and it's * not allowing itself to be halted. At least interrupts are @@ -896,7 +897,7 @@ void xhci_stop_endpoint_command_watchdog(unsigned long arg) } } } - spin_unlock(&xhci->lock); + spin_unlock_irqrestore(&xhci->lock, flags); xhci_dbg(xhci, "Calling usb_hc_died()\n"); usb_hc_died(xhci_to_hcd(xhci)->primary_hcd); xhci_dbg(xhci, "xHCI host controller is dead.\n"); -- GitLab From c75d720fca8a91ce99196d33adea383621027bf2 Mon Sep 17 00:00:00 2001 From: Edward Donovan Date: Tue, 1 Nov 2011 15:29:44 -0400 Subject: [PATCH 0071/3089] genirq: Fix irqfixup, irqpoll regression commit d05c65fff0 ("genirq: spurious: Run only one poller at a time") introduced a regression, leaving the boot options 'irqfixup' and 'irqpoll' non-functional. The patch placed tests in each function, to exit if the function is already running. The test in 'misrouted_irq' exited when it should have proceeded, effectively disabling 'misrouted_irq' and 'poll_spurious_irqs'. The check for an already running poller needs to be "!= 1" not "== 1" as "1" is the value when the first poller starts running. Signed-off-by: Edward Donovan Cc: maciej.rutecki@gmail.com Link: http://lkml.kernel.org/r/1320175784-6745-1-git-send-email-edward.donovan@numble.net Cc: stable@vger.kernel.org # >= 2.6.39 Signed-off-by: Thomas Gleixner --- kernel/irq/spurious.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c index aa57d5da18c1d..b5f4742693c01 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -115,7 +115,7 @@ static int misrouted_irq(int irq) struct irq_desc *desc; int i, ok = 0; - if (atomic_inc_return(&irq_poll_active) == 1) + if (atomic_inc_return(&irq_poll_active) != 1) goto out; irq_poll_cpu = smp_processor_id(); -- GitLab From 616fdb5afb12b4d5bb9e36df40d039e86aaaefc2 Mon Sep 17 00:00:00 2001 From: Ben Widawsky Date: Wed, 5 Oct 2011 11:44:54 -0700 Subject: [PATCH 0072/3089] drm/i915: forcewake warning fixes in debugfs Some more unsafe debugfs access are fixed with this patch. I tested all reads, but didn't thoroughly test the writes. Cc: "Nicolas Kalkhof" Signed-off-by: Ben Widawsky Reviewed-by: Daniel Vetter Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_debugfs.c | 57 ++++++++++++++++++++++++++--- 1 file changed, 51 insertions(+), 6 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 8e95d66800b09..f2e02075d5100 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -635,11 +635,16 @@ static int i915_ringbuffer_info(struct seq_file *m, void *data) struct drm_device *dev = node->minor->dev; drm_i915_private_t *dev_priv = dev->dev_private; struct intel_ring_buffer *ring; + int ret; ring = &dev_priv->ring[(uintptr_t)node->info_ent->data]; if (ring->size == 0) return 0; + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; + seq_printf(m, "Ring %s:\n", ring->name); seq_printf(m, " Head : %08x\n", I915_READ_HEAD(ring) & HEAD_ADDR); seq_printf(m, " Tail : %08x\n", I915_READ_TAIL(ring) & TAIL_ADDR); @@ -653,6 +658,8 @@ static int i915_ringbuffer_info(struct seq_file *m, void *data) seq_printf(m, " Control : %08x\n", I915_READ_CTL(ring)); seq_printf(m, " Start : %08x\n", I915_READ_START(ring)); + mutex_unlock(&dev->struct_mutex); + return 0; } @@ -841,7 +848,16 @@ static int i915_rstdby_delays(struct seq_file *m, void *unused) struct drm_info_node *node = (struct drm_info_node *) m->private; struct drm_device *dev = node->minor->dev; drm_i915_private_t *dev_priv = dev->dev_private; - u16 crstanddelay = I915_READ16(CRSTANDVID); + u16 crstanddelay; + int ret; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; + + crstanddelay = I915_READ16(CRSTANDVID); + + mutex_unlock(&dev->struct_mutex); seq_printf(m, "w/ctx: %d, w/o ctx: %d\n", (crstanddelay >> 8) & 0x3f, (crstanddelay & 0x3f)); @@ -939,7 +955,11 @@ static int i915_delayfreq_table(struct seq_file *m, void *unused) struct drm_device *dev = node->minor->dev; drm_i915_private_t *dev_priv = dev->dev_private; u32 delayfreq; - int i; + int ret, i; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; for (i = 0; i < 16; i++) { delayfreq = I915_READ(PXVFREQ_BASE + i * 4); @@ -947,6 +967,8 @@ static int i915_delayfreq_table(struct seq_file *m, void *unused) (delayfreq & PXVFREQ_PX_MASK) >> PXVFREQ_PX_SHIFT); } + mutex_unlock(&dev->struct_mutex); + return 0; } @@ -961,13 +983,19 @@ static int i915_inttoext_table(struct seq_file *m, void *unused) struct drm_device *dev = node->minor->dev; drm_i915_private_t *dev_priv = dev->dev_private; u32 inttoext; - int i; + int ret, i; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; for (i = 1; i <= 32; i++) { inttoext = I915_READ(INTTOEXT_BASE_ILK + i * 4); seq_printf(m, "INTTOEXT%02d: 0x%08x\n", i, inttoext); } + mutex_unlock(&dev->struct_mutex); + return 0; } @@ -976,9 +1004,19 @@ static int i915_drpc_info(struct seq_file *m, void *unused) struct drm_info_node *node = (struct drm_info_node *) m->private; struct drm_device *dev = node->minor->dev; drm_i915_private_t *dev_priv = dev->dev_private; - u32 rgvmodectl = I915_READ(MEMMODECTL); - u32 rstdbyctl = I915_READ(RSTDBYCTL); - u16 crstandvid = I915_READ16(CRSTANDVID); + u32 rgvmodectl, rstdbyctl; + u16 crstandvid; + int ret; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; + + rgvmodectl = I915_READ(MEMMODECTL); + rstdbyctl = I915_READ(RSTDBYCTL); + crstandvid = I915_READ16(CRSTANDVID); + + mutex_unlock(&dev->struct_mutex); seq_printf(m, "HD boost: %s\n", (rgvmodectl & MEMMODE_BOOST_EN) ? "yes" : "no"); @@ -1166,9 +1204,16 @@ static int i915_gfxec(struct seq_file *m, void *unused) struct drm_info_node *node = (struct drm_info_node *) m->private; struct drm_device *dev = node->minor->dev; drm_i915_private_t *dev_priv = dev->dev_private; + int ret; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; seq_printf(m, "GFXEC: %ld\n", (unsigned long)I915_READ(0x112f4)); + mutex_unlock(&dev->struct_mutex); + return 0; } -- GitLab From 775d17b6ca4357048f36c22151335addfe15db4b Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Sun, 9 Oct 2011 21:52:01 +0200 Subject: [PATCH 0073/3089] drm/i915: Ivybridge still has fences! So don't forget to restore them on resume and dump them into the error state. Cc: stable@kernel.org Signed-off-by: Daniel Vetter Reviewed-by: Chris Wilson Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_irq.c | 1 + drivers/gpu/drm/i915/i915_suspend.c | 2 ++ 2 files changed, 3 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 9ee2729fe5c65..b40004b559771 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -824,6 +824,7 @@ static void i915_gem_record_fences(struct drm_device *dev, /* Fences */ switch (INTEL_INFO(dev)->gen) { + case 7: case 6: for (i = 0; i < 16; i++) error->fence[i] = I915_READ64(FENCE_REG_SANDYBRIDGE_0 + (i * 8)); diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c index f8f602d766501..7886e4fb60e3e 100644 --- a/drivers/gpu/drm/i915/i915_suspend.c +++ b/drivers/gpu/drm/i915/i915_suspend.c @@ -370,6 +370,7 @@ static void i915_save_modeset_reg(struct drm_device *dev) /* Fences */ switch (INTEL_INFO(dev)->gen) { + case 7: case 6: for (i = 0; i < 16; i++) dev_priv->saveFENCE[i] = I915_READ64(FENCE_REG_SANDYBRIDGE_0 + (i * 8)); @@ -404,6 +405,7 @@ static void i915_restore_modeset_reg(struct drm_device *dev) /* Fences */ switch (INTEL_INFO(dev)->gen) { + case 7: case 6: for (i = 0; i < 16; i++) I915_WRITE64(FENCE_REG_SANDYBRIDGE_0 + (i * 8), dev_priv->saveFENCE[i]); -- GitLab From 4b9de737fad5bd8993e6070530802de22f32744d Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Sun, 9 Oct 2011 21:52:02 +0200 Subject: [PATCH 0074/3089] drm/i915: add constants to size fence arrays and fields In preparation of to support 32 fences on Ivybdrigde. Signed-Off-by: Daniel Vetter Reviewed-by: Chris Wilson Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_drv.h | 15 ++++++++------- drivers/gpu/drm/i915/i915_gem.c | 4 ++-- 2 files changed, 10 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 06a37f4fd74b1..d2da91f902528 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -126,6 +126,9 @@ struct drm_i915_master_private { struct _drm_i915_sarea *sarea_priv; }; #define I915_FENCE_REG_NONE -1 +#define I915_MAX_NUM_FENCES 16 +/* 16 fences + sign bit for FENCE_REG_NONE */ +#define I915_MAX_NUM_FENCE_BITS 5 struct drm_i915_fence_reg { struct list_head lru_list; @@ -168,7 +171,7 @@ struct drm_i915_error_state { u32 instdone1; u32 seqno; u64 bbaddr; - u64 fence[16]; + u64 fence[I915_MAX_NUM_FENCES]; struct timeval time; struct drm_i915_error_object { int page_count; @@ -182,7 +185,7 @@ struct drm_i915_error_state { u32 gtt_offset; u32 read_domains; u32 write_domain; - s32 fence_reg:5; + s32 fence_reg:I915_MAX_NUM_FENCE_BITS; s32 pinned:2; u32 tiling:2; u32 dirty:1; @@ -375,7 +378,7 @@ typedef struct drm_i915_private { struct notifier_block lid_notifier; int crt_ddc_pin; - struct drm_i915_fence_reg fence_regs[16]; /* assume 965 */ + struct drm_i915_fence_reg fence_regs[I915_MAX_NUM_FENCES]; /* assume 965 */ int fence_reg_start; /* 4 if userland hasn't ioctl'd us yet */ int num_fence_regs; /* 8 on pre-965, 16 otherwise */ @@ -506,7 +509,7 @@ typedef struct drm_i915_private { u8 saveAR[21]; u8 saveDACMASK; u8 saveCR[37]; - uint64_t saveFENCE[16]; + uint64_t saveFENCE[I915_MAX_NUM_FENCES]; u32 saveCURACNTR; u32 saveCURAPOS; u32 saveCURABASE; @@ -777,10 +780,8 @@ struct drm_i915_gem_object { * Fence register bits (if any) for this object. Will be set * as needed when mapped into the GTT. * Protected by dev->struct_mutex. - * - * Size: 4 bits for 16 fences + sign (for FENCE_REG_NONE) */ - signed int fence_reg:5; + signed int fence_reg:I915_MAX_NUM_FENCE_BITS; /** * Advice: are the backing pages purgeable? diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index d18b07adcffa3..a83859767d48c 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1745,7 +1745,7 @@ static void i915_gem_reset_fences(struct drm_device *dev) struct drm_i915_private *dev_priv = dev->dev_private; int i; - for (i = 0; i < 16; i++) { + for (i = 0; i < dev_priv->num_fence_regs; i++) { struct drm_i915_fence_reg *reg = &dev_priv->fence_regs[i]; struct drm_i915_gem_object *obj = reg->obj; @@ -3877,7 +3877,7 @@ i915_gem_load(struct drm_device *dev) INIT_LIST_HEAD(&dev_priv->mm.gtt_list); for (i = 0; i < I915_NUM_RINGS; i++) init_ring_lists(&dev_priv->ring[i]); - for (i = 0; i < 16; i++) + for (i = 0; i < I915_MAX_NUM_FENCES; i++) INIT_LIST_HEAD(&dev_priv->fence_regs[i].lru_list); INIT_DELAYED_WORK(&dev_priv->mm.retire_work, i915_gem_retire_work_handler); -- GitLab From a9866a0975881ba5e819b0a02324ba683b5cdb81 Mon Sep 17 00:00:00 2001 From: Shawn Guo Date: Fri, 21 Oct 2011 11:53:34 +0800 Subject: [PATCH 0075/3089] MAINTAINERS: add ARM/FREESCALE MXS entry As suggested by Pengutronix, they want to get Sascha relieved from the burden of maintaining MXS sub-architecture. Since I brought the most of MXS core code to mainline (with the great help from Pengutronix people, thanks!), I would like to step up for maintaining MXS. Signed-off-by: Shawn Guo Acked-by: Sascha Hauer --- MAINTAINERS | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index ab7767001286b..8fe3f01849902 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -804,6 +804,13 @@ S: Maintained T: git git://git.linaro.org/people/shawnguo/linux-2.6.git F: arch/arm/mach-imx/*imx6* +ARM/FREESCALE MXS ARM ARCHITECTURE +M: Shawn Guo +L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) +S: Maintained +T: git git://git.linaro.org/people/shawnguo/linux-2.6.git +F: arch/arm/mach-mxs/ + ARM/GLOMATION GESBC9312SX MACHINE SUPPORT M: Lennert Buytenhek L: linux-arm-kernel@lists.infradead.org (moderated for non-subscribers) -- GitLab From 680da876f44a644aee891e1d0df5a560cfa4720e Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Thu, 3 Nov 2011 14:15:13 -0700 Subject: [PATCH 0076/3089] drm/i915: enable cacheable objects on Ivybridge IVB supports these bits as well. Signed-off-by: Jesse Barnes Reviewed-by: Daniel Vetter Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_gem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index a83859767d48c..ed0b68fdb9703 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -3613,7 +3613,7 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev, obj->base.write_domain = I915_GEM_DOMAIN_CPU; obj->base.read_domains = I915_GEM_DOMAIN_CPU; - if (IS_GEN6(dev)) { + if (IS_GEN6(dev) || IS_GEN7(dev)) { /* On Gen6, we can have the GPU use the LLC (the CPU * cache) for about a 10% performance improvement * compared to uncached. Graphics requests other than -- GitLab From 87121ca504fd1d963a66b3fb0c72054b0fd9a177 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Fri, 7 Oct 2011 16:31:46 +0200 Subject: [PATCH 0077/3089] oprofile: Fix crash when unloading module (hr timer mode) Oprofile may crash in a KVM guest while unlaoding modules. This happens if oprofile_arch_init() fails and oprofile switches to the hr timer mode as a fallback. In this case oprofile_arch_exit() is called, but it never was initialized properly which causes the crash. This patch fixes this. oprofile: using timer interrupt. BUG: unable to handle kernel NULL pointer dereference at 0000000000000008 IP: [] unregister_syscore_ops+0x41/0x58 PGD 41da3f067 PUD 41d80e067 PMD 0 Oops: 0002 [#1] PREEMPT SMP CPU 5 Modules linked in: oprofile(-) Pid: 2382, comm: modprobe Not tainted 3.1.0-rc7-00018-g709a39d #18 Advanced Micro Device Anaheim/Anaheim RIP: 0010:[] [] unregister_syscore_ops+0x41/0x58 RSP: 0018:ffff88041de1de98 EFLAGS: 00010296 RAX: 0000000000000000 RBX: ffffffffa00060e0 RCX: dead000000200200 RDX: 0000000000000000 RSI: dead000000100100 RDI: ffffffff8178c620 RBP: ffff88041de1dea8 R08: 0000000000000001 R09: 0000000000000082 R10: 0000000000000000 R11: ffff88041de1dde8 R12: 0000000000000080 R13: fffffffffffffff5 R14: 0000000000000001 R15: 0000000000610210 FS: 00007f9ae5bef700(0000) GS:ffff88042fd40000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000000008 CR3: 000000041ca44000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process modprobe (pid: 2382, threadinfo ffff88041de1c000, task ffff88042db6d040) Stack: ffff88041de1deb8 ffffffffa0006770 ffff88041de1deb8 ffffffffa000251e ffff88041de1dec8 ffffffffa00022c2 ffff88041de1ded8 ffffffffa0004993 ffff88041de1df78 ffffffff81073115 656c69666f72706f 0000000000610200 Call Trace: [] op_nmi_exit+0x15/0x17 [oprofile] [] oprofile_arch_exit+0xe/0x10 [oprofile] [] oprofile_exit+0x13/0x15 [oprofile] [] sys_delete_module+0x1c3/0x22f [] ? trace_hardirqs_on_thunk+0x3a/0x3f [] system_call_fastpath+0x16/0x1b Code: 20 c6 78 81 e8 c5 cc 23 00 48 8b 13 48 8b 43 08 48 be 00 01 10 00 00 00 ad de 48 b9 00 02 20 00 00 00 ad de 48 c7 c7 20 c6 78 81 89 42 08 48 89 10 48 89 33 48 89 4b 08 e8 a6 c0 23 00 5a 5b RIP [] unregister_syscore_ops+0x41/0x58 RSP CR2: 0000000000000008 ---[ end trace 06d4e95b6aa3b437 ]--- CC: stable@kernel.org # 2.6.37+ Signed-off-by: Robert Richter --- drivers/oprofile/oprof.c | 29 ++++++++++++++++++++++++----- drivers/oprofile/timer_int.c | 1 + 2 files changed, 25 insertions(+), 5 deletions(-) diff --git a/drivers/oprofile/oprof.c b/drivers/oprofile/oprof.c index dccd8636095cb..f8c752e408a66 100644 --- a/drivers/oprofile/oprof.c +++ b/drivers/oprofile/oprof.c @@ -239,26 +239,45 @@ int oprofile_set_ulong(unsigned long *addr, unsigned long val) return err; } +static int timer_mode; + static int __init oprofile_init(void) { int err; + /* always init architecture to setup backtrace support */ err = oprofile_arch_init(&oprofile_ops); - if (err < 0 || timer) { - printk(KERN_INFO "oprofile: using timer interrupt.\n"); + + timer_mode = err || timer; /* fall back to timer mode on errors */ + if (timer_mode) { + if (!err) + oprofile_arch_exit(); err = oprofile_timer_init(&oprofile_ops); if (err) return err; } - return oprofilefs_register(); + + err = oprofilefs_register(); + if (!err) + return 0; + + /* failed */ + if (timer_mode) + oprofile_timer_exit(); + else + oprofile_arch_exit(); + + return err; } static void __exit oprofile_exit(void) { - oprofile_timer_exit(); oprofilefs_unregister(); - oprofile_arch_exit(); + if (timer_mode) + oprofile_timer_exit(); + else + oprofile_arch_exit(); } diff --git a/drivers/oprofile/timer_int.c b/drivers/oprofile/timer_int.c index 3ef44624f5103..878fba1265829 100644 --- a/drivers/oprofile/timer_int.c +++ b/drivers/oprofile/timer_int.c @@ -110,6 +110,7 @@ int oprofile_timer_init(struct oprofile_operations *ops) ops->start = oprofile_hrtimer_start; ops->stop = oprofile_hrtimer_stop; ops->cpu_type = "timer"; + printk(KERN_INFO "oprofile: using timer interrupt.\n"); return 0; } -- GitLab From 97f7f8189fe54e3cfe324ef9ad35064f3d2d3bff Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Mon, 10 Oct 2011 16:21:10 +0200 Subject: [PATCH 0078/3089] oprofile, x86: Fix crash when unloading module (nmi timer mode) If oprofile uses the nmi timer interrupt there is a crash while unloading the module. The bug can be triggered with oprofile build as module and kernel parameter nolapic set. This patch fixes this. oprofile: using NMI timer interrupt. BUG: unable to handle kernel NULL pointer dereference at 0000000000000008 IP: [] unregister_syscore_ops+0x41/0x58 PGD 42dbca067 PUD 41da6a067 PMD 0 Oops: 0002 [#1] PREEMPT SMP CPU 5 Modules linked in: oprofile(-) [last unloaded: oprofile] Pid: 2518, comm: modprobe Not tainted 3.1.0-rc7-00019-gb2fb49d #19 Advanced Micro Device Anaheim/Anaheim RIP: 0010:[] [] unregister_syscore_ops+0x41/0x58 RSP: 0018:ffff88041ef71e98 EFLAGS: 00010296 RAX: 0000000000000000 RBX: ffffffffa0017100 RCX: dead000000200200 RDX: 0000000000000000 RSI: dead000000100100 RDI: ffffffff8178c620 RBP: ffff88041ef71ea8 R08: 0000000000000001 R09: 0000000000000082 R10: 0000000000000000 R11: ffff88041ef71de8 R12: 0000000000000080 R13: fffffffffffffff5 R14: 0000000000000001 R15: 0000000000610210 FS: 00007fc902f20700(0000) GS:ffff88042fd40000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000000008 CR3: 000000041cdb6000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process modprobe (pid: 2518, threadinfo ffff88041ef70000, task ffff88041d348040) Stack: ffff88041ef71eb8 ffffffffa0017790 ffff88041ef71eb8 ffffffffa0013532 ffff88041ef71ec8 ffffffffa00132d6 ffff88041ef71ed8 ffffffffa00159b2 ffff88041ef71f78 ffffffff81073115 656c69666f72706f 0000000000610200 Call Trace: [] op_nmi_exit+0x15/0x17 [oprofile] [] oprofile_arch_exit+0xe/0x10 [oprofile] [] oprofile_exit+0x1e/0x20 [oprofile] [] sys_delete_module+0x1c3/0x22f [] ? trace_hardirqs_on_thunk+0x3a/0x3f [] system_call_fastpath+0x16/0x1b Code: 20 c6 78 81 e8 c5 cc 23 00 48 8b 13 48 8b 43 08 48 be 00 01 10 00 00 00 ad de 48 b9 00 02 20 00 00 00 ad de 48 c7 c7 20 c6 78 81 89 42 08 48 89 10 48 89 33 48 89 4b 08 e8 a6 c0 23 00 5a 5b RIP [] unregister_syscore_ops+0x41/0x58 RSP CR2: 0000000000000008 ---[ end trace 43a541a52956b7b0 ]--- CC: stable@kernel.org # 2.6.37+ Signed-off-by: Robert Richter --- arch/x86/oprofile/init.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/arch/x86/oprofile/init.c b/arch/x86/oprofile/init.c index cdfe4c54decac..f148cf6526783 100644 --- a/arch/x86/oprofile/init.c +++ b/arch/x86/oprofile/init.c @@ -21,6 +21,7 @@ extern int op_nmi_timer_init(struct oprofile_operations *ops); extern void op_nmi_exit(void); extern void x86_backtrace(struct pt_regs * const regs, unsigned int depth); +static int nmi_timer; int __init oprofile_arch_init(struct oprofile_operations *ops) { @@ -31,8 +32,9 @@ int __init oprofile_arch_init(struct oprofile_operations *ops) #ifdef CONFIG_X86_LOCAL_APIC ret = op_nmi_init(ops); #endif + nmi_timer = (ret != 0); #ifdef CONFIG_X86_IO_APIC - if (ret < 0) + if (nmi_timer) ret = op_nmi_timer_init(ops); #endif ops->backtrace = x86_backtrace; @@ -44,6 +46,7 @@ int __init oprofile_arch_init(struct oprofile_operations *ops) void oprofile_arch_exit(void) { #ifdef CONFIG_X86_LOCAL_APIC - op_nmi_exit(); + if (!nmi_timer) + op_nmi_exit(); #endif } -- GitLab From 159a80b2142df709416ab369113de7d511c48331 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Tue, 11 Oct 2011 19:39:16 +0200 Subject: [PATCH 0079/3089] oprofile, x86: Add kernel parameter oprofile.cpu_type=timer We need this to better test x86 NMI timer mode. Otherwise it is very hard to setup systems with NMI timer enabled, but we have this e.g. in virtual machine environments. Signed-off-by: Robert Richter --- Documentation/kernel-parameters.txt | 3 +++ arch/x86/oprofile/nmi_int.c | 27 +++++++++++++++++++++------ 2 files changed, 24 insertions(+), 6 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 854ed5ca7e3f1..f7735a125f4b2 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1848,6 +1848,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted. arch_perfmon: [X86] Force use of architectural perfmon on Intel CPUs instead of the CPU specific event set. + timer: [X86] Force use of architectural NMI + timer mode (see also oprofile.timer + for generic hr timer mode) oops=panic Always panic on oopses. Default is to just kill the process, but there is a small probability of diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c index 68894fdc034bc..7ca4d43e8988d 100644 --- a/arch/x86/oprofile/nmi_int.c +++ b/arch/x86/oprofile/nmi_int.c @@ -613,24 +613,36 @@ static int __init p4_init(char **cpu_type) return 0; } -static int force_arch_perfmon; -static int force_cpu_type(const char *str, struct kernel_param *kp) +enum __force_cpu_type { + reserved = 0, /* do not force */ + timer, + arch_perfmon, +}; + +static int force_cpu_type; + +static int set_cpu_type(const char *str, struct kernel_param *kp) { - if (!strcmp(str, "arch_perfmon")) { - force_arch_perfmon = 1; + if (!strcmp(str, "timer")) { + force_cpu_type = timer; + printk(KERN_INFO "oprofile: forcing NMI timer mode\n"); + } else if (!strcmp(str, "arch_perfmon")) { + force_cpu_type = arch_perfmon; printk(KERN_INFO "oprofile: forcing architectural perfmon\n"); + } else { + force_cpu_type = 0; } return 0; } -module_param_call(cpu_type, force_cpu_type, NULL, NULL, 0); +module_param_call(cpu_type, set_cpu_type, NULL, NULL, 0); static int __init ppro_init(char **cpu_type) { __u8 cpu_model = boot_cpu_data.x86_model; struct op_x86_model_spec *spec = &op_ppro_spec; /* default */ - if (force_arch_perfmon && cpu_has_arch_perfmon) + if (force_cpu_type == arch_perfmon && cpu_has_arch_perfmon) return 0; /* @@ -697,6 +709,9 @@ int __init op_nmi_init(struct oprofile_operations *ops) if (!cpu_has_apic) return -ENODEV; + if (force_cpu_type == timer) + return -ENODEV; + switch (vendor) { case X86_VENDOR_AMD: /* Needs to be at least an Athlon (or hammer in 32bit mode) */ -- GitLab From 75c43a20b220f885c39ffa7cdbbb1191e257a9a9 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Fri, 14 Oct 2011 15:46:10 +0200 Subject: [PATCH 0080/3089] oprofile: Remove exit function for timer mode Remove exit functions by moving init/exit code to oprofile's setup/ shutdown functions. Doing so the oprofile module exit code will be easier and less error-prone. Signed-off-by: Robert Richter --- drivers/oprofile/oprof.c | 8 ++------ drivers/oprofile/timer_int.c | 30 +++++++++++++++--------------- 2 files changed, 17 insertions(+), 21 deletions(-) diff --git a/drivers/oprofile/oprof.c b/drivers/oprofile/oprof.c index f8c752e408a66..f7cd06967aedd 100644 --- a/drivers/oprofile/oprof.c +++ b/drivers/oprofile/oprof.c @@ -262,9 +262,7 @@ static int __init oprofile_init(void) return 0; /* failed */ - if (timer_mode) - oprofile_timer_exit(); - else + if (!timer_mode) oprofile_arch_exit(); return err; @@ -274,9 +272,7 @@ static int __init oprofile_init(void) static void __exit oprofile_exit(void) { oprofilefs_unregister(); - if (timer_mode) - oprofile_timer_exit(); - else + if (!timer_mode) oprofile_arch_exit(); } diff --git a/drivers/oprofile/timer_int.c b/drivers/oprofile/timer_int.c index 878fba1265829..93404f72dfa81 100644 --- a/drivers/oprofile/timer_int.c +++ b/drivers/oprofile/timer_int.c @@ -97,24 +97,24 @@ static struct notifier_block __refdata oprofile_cpu_notifier = { .notifier_call = oprofile_cpu_notify, }; -int oprofile_timer_init(struct oprofile_operations *ops) +static int oprofile_hrtimer_setup(void) { - int rc; - - rc = register_hotcpu_notifier(&oprofile_cpu_notifier); - if (rc) - return rc; - ops->create_files = NULL; - ops->setup = NULL; - ops->shutdown = NULL; - ops->start = oprofile_hrtimer_start; - ops->stop = oprofile_hrtimer_stop; - ops->cpu_type = "timer"; - printk(KERN_INFO "oprofile: using timer interrupt.\n"); - return 0; + return register_hotcpu_notifier(&oprofile_cpu_notifier); } -void oprofile_timer_exit(void) +static void oprofile_hrtimer_shutdown(void) { unregister_hotcpu_notifier(&oprofile_cpu_notifier); } + +int oprofile_timer_init(struct oprofile_operations *ops) +{ + ops->create_files = NULL; + ops->setup = oprofile_hrtimer_setup; + ops->shutdown = oprofile_hrtimer_shutdown; + ops->start = oprofile_hrtimer_start; + ops->stop = oprofile_hrtimer_stop; + ops->cpu_type = "timer"; + printk(KERN_INFO "oprofile: using timer interrupt.\n"); + return 0; +} -- GitLab From dcfce4a095932e6e95d83ad982be3609947963bc Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Tue, 11 Oct 2011 17:11:08 +0200 Subject: [PATCH 0081/3089] oprofile, x86: Reimplement nmi timer mode using perf event The legacy x86 nmi watchdog code was removed with the implementation of the perf based nmi watchdog. This broke Oprofile's nmi timer mode. To run nmi timer mode we relied on a continuous ticking nmi source which the nmi watchdog provided. The nmi tick was no longer available and current watchdog can not be used anymore since it runs with very long periods in the range of seconds. This patch reimplements the nmi timer mode using a perf counter nmi source. V2: * removing pr_info() * fix undefined reference to `__udivdi3' for 32 bit build * fix section mismatch of .cpuinit.data:nmi_timer_cpu_nb * removed nmi timer setup in arch/x86 * implemented function stubs for op_nmi_init/exit() * made code more readable in oprofile_init() V3: * fix architectural initialization in oprofile_init() * fix CONFIG_OPROFILE_NMI_TIMER dependencies Acked-by: Peter Zijlstra Signed-off-by: Robert Richter --- arch/Kconfig | 4 + arch/x86/oprofile/Makefile | 3 +- arch/x86/oprofile/init.c | 30 ++---- arch/x86/oprofile/nmi_timer_int.c | 66 ------------ drivers/oprofile/nmi_timer_int.c | 173 ++++++++++++++++++++++++++++++ drivers/oprofile/oprof.c | 24 ++--- drivers/oprofile/oprof.h | 9 ++ kernel/events/core.c | 2 + 8 files changed, 208 insertions(+), 103 deletions(-) delete mode 100644 arch/x86/oprofile/nmi_timer_int.c create mode 100644 drivers/oprofile/nmi_timer_int.c diff --git a/arch/Kconfig b/arch/Kconfig index 4b0669cbb3b01..2505740b81d2e 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -30,6 +30,10 @@ config OPROFILE_EVENT_MULTIPLEX config HAVE_OPROFILE bool +config OPROFILE_NMI_TIMER + def_bool y + depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI + config KPROBES bool "Kprobes" depends on MODULES diff --git a/arch/x86/oprofile/Makefile b/arch/x86/oprofile/Makefile index 446902b2a6b6a..1599f568f0e2f 100644 --- a/arch/x86/oprofile/Makefile +++ b/arch/x86/oprofile/Makefile @@ -4,9 +4,8 @@ DRIVER_OBJS = $(addprefix ../../../drivers/oprofile/, \ oprof.o cpu_buffer.o buffer_sync.o \ event_buffer.o oprofile_files.o \ oprofilefs.o oprofile_stats.o \ - timer_int.o ) + timer_int.o nmi_timer_int.o ) oprofile-y := $(DRIVER_OBJS) init.o backtrace.o oprofile-$(CONFIG_X86_LOCAL_APIC) += nmi_int.o op_model_amd.o \ op_model_ppro.o op_model_p4.o -oprofile-$(CONFIG_X86_IO_APIC) += nmi_timer_int.o diff --git a/arch/x86/oprofile/init.c b/arch/x86/oprofile/init.c index f148cf6526783..9e138d00ad36d 100644 --- a/arch/x86/oprofile/init.c +++ b/arch/x86/oprofile/init.c @@ -16,37 +16,23 @@ * with the NMI mode driver. */ +#ifdef CONFIG_X86_LOCAL_APIC extern int op_nmi_init(struct oprofile_operations *ops); -extern int op_nmi_timer_init(struct oprofile_operations *ops); extern void op_nmi_exit(void); -extern void x86_backtrace(struct pt_regs * const regs, unsigned int depth); +#else +static int op_nmi_init(struct oprofile_operations *ops) { return -ENODEV; } +static void op_nmi_exit(void) { } +#endif -static int nmi_timer; +extern void x86_backtrace(struct pt_regs * const regs, unsigned int depth); int __init oprofile_arch_init(struct oprofile_operations *ops) { - int ret; - - ret = -ENODEV; - -#ifdef CONFIG_X86_LOCAL_APIC - ret = op_nmi_init(ops); -#endif - nmi_timer = (ret != 0); -#ifdef CONFIG_X86_IO_APIC - if (nmi_timer) - ret = op_nmi_timer_init(ops); -#endif ops->backtrace = x86_backtrace; - - return ret; + return op_nmi_init(ops); } - void oprofile_arch_exit(void) { -#ifdef CONFIG_X86_LOCAL_APIC - if (!nmi_timer) - op_nmi_exit(); -#endif + op_nmi_exit(); } diff --git a/arch/x86/oprofile/nmi_timer_int.c b/arch/x86/oprofile/nmi_timer_int.c deleted file mode 100644 index 720bf5a53c513..0000000000000 --- a/arch/x86/oprofile/nmi_timer_int.c +++ /dev/null @@ -1,66 +0,0 @@ -/** - * @file nmi_timer_int.c - * - * @remark Copyright 2003 OProfile authors - * @remark Read the file COPYING - * - * @author Zwane Mwaikambo - */ - -#include -#include -#include -#include -#include -#include - -#include -#include -#include - -static int profile_timer_exceptions_notify(struct notifier_block *self, - unsigned long val, void *data) -{ - struct die_args *args = (struct die_args *)data; - int ret = NOTIFY_DONE; - - switch (val) { - case DIE_NMI: - oprofile_add_sample(args->regs, 0); - ret = NOTIFY_STOP; - break; - default: - break; - } - return ret; -} - -static struct notifier_block profile_timer_exceptions_nb = { - .notifier_call = profile_timer_exceptions_notify, - .next = NULL, - .priority = NMI_LOW_PRIOR, -}; - -static int timer_start(void) -{ - if (register_die_notifier(&profile_timer_exceptions_nb)) - return 1; - return 0; -} - - -static void timer_stop(void) -{ - unregister_die_notifier(&profile_timer_exceptions_nb); - synchronize_sched(); /* Allow already-started NMIs to complete. */ -} - - -int __init op_nmi_timer_init(struct oprofile_operations *ops) -{ - ops->start = timer_start; - ops->stop = timer_stop; - ops->cpu_type = "timer"; - printk(KERN_INFO "oprofile: using NMI timer interrupt.\n"); - return 0; -} diff --git a/drivers/oprofile/nmi_timer_int.c b/drivers/oprofile/nmi_timer_int.c new file mode 100644 index 0000000000000..76f1c9357f39d --- /dev/null +++ b/drivers/oprofile/nmi_timer_int.c @@ -0,0 +1,173 @@ +/** + * @file nmi_timer_int.c + * + * @remark Copyright 2011 Advanced Micro Devices, Inc. + * + * @author Robert Richter + */ + +#include +#include +#include +#include +#include + +#ifdef CONFIG_OPROFILE_NMI_TIMER + +static DEFINE_PER_CPU(struct perf_event *, nmi_timer_events); +static int ctr_running; + +static struct perf_event_attr nmi_timer_attr = { + .type = PERF_TYPE_HARDWARE, + .config = PERF_COUNT_HW_CPU_CYCLES, + .size = sizeof(struct perf_event_attr), + .pinned = 1, + .disabled = 1, +}; + +static void nmi_timer_callback(struct perf_event *event, + struct perf_sample_data *data, + struct pt_regs *regs) +{ + event->hw.interrupts = 0; /* don't throttle interrupts */ + oprofile_add_sample(regs, 0); +} + +static int nmi_timer_start_cpu(int cpu) +{ + struct perf_event *event = per_cpu(nmi_timer_events, cpu); + + if (!event) { + event = perf_event_create_kernel_counter(&nmi_timer_attr, cpu, NULL, + nmi_timer_callback, NULL); + if (IS_ERR(event)) + return PTR_ERR(event); + per_cpu(nmi_timer_events, cpu) = event; + } + + if (event && ctr_running) + perf_event_enable(event); + + return 0; +} + +static void nmi_timer_stop_cpu(int cpu) +{ + struct perf_event *event = per_cpu(nmi_timer_events, cpu); + + if (event && ctr_running) + perf_event_disable(event); +} + +static int nmi_timer_cpu_notifier(struct notifier_block *b, unsigned long action, + void *data) +{ + int cpu = (unsigned long)data; + switch (action) { + case CPU_DOWN_FAILED: + case CPU_ONLINE: + nmi_timer_start_cpu(cpu); + break; + case CPU_DOWN_PREPARE: + nmi_timer_stop_cpu(cpu); + break; + } + return NOTIFY_DONE; +} + +static struct notifier_block nmi_timer_cpu_nb = { + .notifier_call = nmi_timer_cpu_notifier +}; + +static int nmi_timer_start(void) +{ + int cpu; + + get_online_cpus(); + ctr_running = 1; + for_each_online_cpu(cpu) + nmi_timer_start_cpu(cpu); + put_online_cpus(); + + return 0; +} + +static void nmi_timer_stop(void) +{ + int cpu; + + get_online_cpus(); + for_each_online_cpu(cpu) + nmi_timer_stop_cpu(cpu); + ctr_running = 0; + put_online_cpus(); +} + +static void nmi_timer_shutdown(void) +{ + struct perf_event *event; + int cpu; + + get_online_cpus(); + unregister_cpu_notifier(&nmi_timer_cpu_nb); + for_each_possible_cpu(cpu) { + event = per_cpu(nmi_timer_events, cpu); + if (!event) + continue; + perf_event_disable(event); + per_cpu(nmi_timer_events, cpu) = NULL; + perf_event_release_kernel(event); + } + + put_online_cpus(); +} + +static int nmi_timer_setup(void) +{ + int cpu, err; + u64 period; + + /* clock cycles per tick: */ + period = (u64)cpu_khz * 1000; + do_div(period, HZ); + nmi_timer_attr.sample_period = period; + + get_online_cpus(); + err = register_cpu_notifier(&nmi_timer_cpu_nb); + if (err) + goto out; + /* can't attach events to offline cpus: */ + for_each_online_cpu(cpu) { + err = nmi_timer_start_cpu(cpu); + if (err) + break; + } + if (err) + nmi_timer_shutdown(); +out: + put_online_cpus(); + return err; +} + +int __init op_nmi_timer_init(struct oprofile_operations *ops) +{ + int err = 0; + + err = nmi_timer_setup(); + if (err) + return err; + nmi_timer_shutdown(); /* only check, don't alloc */ + + ops->create_files = NULL; + ops->setup = nmi_timer_setup; + ops->shutdown = nmi_timer_shutdown; + ops->start = nmi_timer_start; + ops->stop = nmi_timer_stop; + ops->cpu_type = "timer"; + + printk(KERN_INFO "oprofile: using NMI timer interrupt.\n"); + + return 0; +} + +#endif diff --git a/drivers/oprofile/oprof.c b/drivers/oprofile/oprof.c index f7cd06967aedd..ed2c3ec07024d 100644 --- a/drivers/oprofile/oprof.c +++ b/drivers/oprofile/oprof.c @@ -246,26 +246,24 @@ static int __init oprofile_init(void) int err; /* always init architecture to setup backtrace support */ + timer_mode = 0; err = oprofile_arch_init(&oprofile_ops); + if (!err) { + if (!timer && !oprofilefs_register()) + return 0; + oprofile_arch_exit(); + } - timer_mode = err || timer; /* fall back to timer mode on errors */ - if (timer_mode) { - if (!err) - oprofile_arch_exit(); + /* setup timer mode: */ + timer_mode = 1; + /* no nmi timer mode if oprofile.timer is set */ + if (timer || op_nmi_timer_init(&oprofile_ops)) { err = oprofile_timer_init(&oprofile_ops); if (err) return err; } - err = oprofilefs_register(); - if (!err) - return 0; - - /* failed */ - if (!timer_mode) - oprofile_arch_exit(); - - return err; + return oprofilefs_register(); } diff --git a/drivers/oprofile/oprof.h b/drivers/oprofile/oprof.h index 177b73de5e5f1..769fb0fcac444 100644 --- a/drivers/oprofile/oprof.h +++ b/drivers/oprofile/oprof.h @@ -36,6 +36,15 @@ struct dentry; void oprofile_create_files(struct super_block *sb, struct dentry *root); int oprofile_timer_init(struct oprofile_operations *ops); void oprofile_timer_exit(void); +#ifdef CONFIG_OPROFILE_NMI_TIMER +int op_nmi_timer_init(struct oprofile_operations *ops); +#else +static inline int op_nmi_timer_init(struct oprofile_operations *ops) +{ + return -ENODEV; +} +#endif + int oprofile_set_ulong(unsigned long *addr, unsigned long val); int oprofile_set_timeout(unsigned long time); diff --git a/kernel/events/core.c b/kernel/events/core.c index d1a1bee35228c..d2e28bdd523a9 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1322,6 +1322,7 @@ retry: } raw_spin_unlock_irq(&ctx->lock); } +EXPORT_SYMBOL_GPL(perf_event_disable); static void perf_set_shadow_time(struct perf_event *event, struct perf_event_context *ctx, @@ -1806,6 +1807,7 @@ retry: out: raw_spin_unlock_irq(&ctx->lock); } +EXPORT_SYMBOL_GPL(perf_event_enable); int perf_event_refresh(struct perf_event *event, int refresh) { -- GitLab From d31c285b3a71cf9056e6a060de41f37780b0af86 Mon Sep 17 00:00:00 2001 From: Sarah Sharp Date: Thu, 3 Nov 2011 13:06:08 -0700 Subject: [PATCH 0082/3089] xhci: Set slot and ep0 flags for address command. Matt's AsMedia xHCI host controller was responding with a Context Error to an address device command after a configured device reset. Some sequence of events leads both the slot and endpoint zero add flags cleared to zero, which the AsMedia host doesn't like: [ 223.701839] xhci_hcd 0000:03:00.0: Slot ID 1 Input Context: [ 223.701841] xhci_hcd 0000:03:00.0: @ffff880137b25000 (virt) @ffffc000 (dma) 0x000000 - drop flags [ 223.701843] xhci_hcd 0000:03:00.0: @ffff880137b25004 (virt) @ffffc004 (dma) 0x000000 - add flags [ 223.701846] xhci_hcd 0000:03:00.0: @ffff880137b25008 (virt) @ffffc008 (dma) 0x000000 - rsvd2[0] [ 223.701848] xhci_hcd 0000:03:00.0: @ffff880137b2500c (virt) @ffffc00c (dma) 0x000000 - rsvd2[1] [ 223.701850] xhci_hcd 0000:03:00.0: @ffff880137b25010 (virt) @ffffc010 (dma) 0x000000 - rsvd2[2] [ 223.701852] xhci_hcd 0000:03:00.0: @ffff880137b25014 (virt) @ffffc014 (dma) 0x000000 - rsvd2[3] [ 223.701854] xhci_hcd 0000:03:00.0: @ffff880137b25018 (virt) @ffffc018 (dma) 0x000000 - rsvd2[4] [ 223.701857] xhci_hcd 0000:03:00.0: @ffff880137b2501c (virt) @ffffc01c (dma) 0x000000 - rsvd2[5] [ 223.701858] xhci_hcd 0000:03:00.0: Slot Context: [ 223.701860] xhci_hcd 0000:03:00.0: @ffff880137b25020 (virt) @ffffc020 (dma) 0x8400000 - dev_info [ 223.701862] xhci_hcd 0000:03:00.0: @ffff880137b25024 (virt) @ffffc024 (dma) 0x010000 - dev_info2 [ 223.701864] xhci_hcd 0000:03:00.0: @ffff880137b25028 (virt) @ffffc028 (dma) 0x000000 - tt_info [ 223.701866] xhci_hcd 0000:03:00.0: @ffff880137b2502c (virt) @ffffc02c (dma) 0x000000 - dev_state [ 223.701869] xhci_hcd 0000:03:00.0: @ffff880137b25030 (virt) @ffffc030 (dma) 0x000000 - rsvd[0] [ 223.701871] xhci_hcd 0000:03:00.0: @ffff880137b25034 (virt) @ffffc034 (dma) 0x000000 - rsvd[1] [ 223.701873] xhci_hcd 0000:03:00.0: @ffff880137b25038 (virt) @ffffc038 (dma) 0x000000 - rsvd[2] [ 223.701875] xhci_hcd 0000:03:00.0: @ffff880137b2503c (virt) @ffffc03c (dma) 0x000000 - rsvd[3] [ 223.701877] xhci_hcd 0000:03:00.0: Endpoint 00 Context: [ 223.701879] xhci_hcd 0000:03:00.0: @ffff880137b25040 (virt) @ffffc040 (dma) 0x000000 - ep_info [ 223.701881] xhci_hcd 0000:03:00.0: @ffff880137b25044 (virt) @ffffc044 (dma) 0x2000026 - ep_info2 [ 223.701883] xhci_hcd 0000:03:00.0: @ffff880137b25048 (virt) @ffffc048 (dma) 0xffffe8e0 - deq [ 223.701885] xhci_hcd 0000:03:00.0: @ffff880137b25050 (virt) @ffffc050 (dma) 0x000000 - tx_info [ 223.701887] xhci_hcd 0000:03:00.0: @ffff880137b25054 (virt) @ffffc054 (dma) 0x000000 - rsvd[0] [ 223.701889] xhci_hcd 0000:03:00.0: @ffff880137b25058 (virt) @ffffc058 (dma) 0x000000 - rsvd[1] [ 223.701892] xhci_hcd 0000:03:00.0: @ffff880137b2505c (virt) @ffffc05c (dma) 0x000000 - rsvd[2] ... [ 223.701927] xhci_hcd 0000:03:00.0: // Ding dong! [ 223.701992] xhci_hcd 0000:03:00.0: Setup ERROR: address device command for slot 1. The xHCI spec says that both flags must be set to one for the Address Device command. When the device is first enumerated, xhci_setup_addressable_virt_dev() does set those flags. However, when the device is addressed after it has been reset in the configured state, xhci_setup_addressable_virt_dev() is not called, and xhci_copy_ep0_dequeue_into_input_ctx() is called instead. That function relies on the flags being set up by previous commands, which apparently isn't a good assumption. Move the setting of the flags into the common parent function. This should be queued for stable kernels as old as 2.6.35, since that was the first introduction of xhci_copy_ep0_dequeue_into_input_ctx. Signed-off-by: Sarah Sharp Tested-by: Matt Cc: stable@vger.kernel.org --- drivers/usb/host/xhci-mem.c | 5 ----- drivers/usb/host/xhci.c | 5 ++++- 2 files changed, 4 insertions(+), 6 deletions(-) diff --git a/drivers/usb/host/xhci-mem.c b/drivers/usb/host/xhci-mem.c index 42a22b8e69226..0e4b25fa3bcd2 100644 --- a/drivers/usb/host/xhci-mem.c +++ b/drivers/usb/host/xhci-mem.c @@ -982,7 +982,6 @@ int xhci_setup_addressable_virt_dev(struct xhci_hcd *xhci, struct usb_device *ud struct xhci_virt_device *dev; struct xhci_ep_ctx *ep0_ctx; struct xhci_slot_ctx *slot_ctx; - struct xhci_input_control_ctx *ctrl_ctx; u32 port_num; struct usb_device *top_dev; @@ -994,12 +993,8 @@ int xhci_setup_addressable_virt_dev(struct xhci_hcd *xhci, struct usb_device *ud return -EINVAL; } ep0_ctx = xhci_get_ep_ctx(xhci, dev->in_ctx, 0); - ctrl_ctx = xhci_get_input_control_ctx(xhci, dev->in_ctx); slot_ctx = xhci_get_slot_ctx(xhci, dev->in_ctx); - /* 2) New slot context and endpoint 0 context are valid*/ - ctrl_ctx->add_flags = cpu_to_le32(SLOT_FLAG | EP0_FLAG); - /* 3) Only the control endpoint is valid - one endpoint context */ slot_ctx->dev_info |= cpu_to_le32(LAST_CTX(1) | udev->route); switch (udev->speed) { diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 1ff95a0df5767..747c5ead922bb 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -3504,6 +3504,10 @@ int xhci_address_device(struct usb_hcd *hcd, struct usb_device *udev) /* Otherwise, update the control endpoint ring enqueue pointer. */ else xhci_copy_ep0_dequeue_into_input_ctx(xhci, udev); + ctrl_ctx = xhci_get_input_control_ctx(xhci, virt_dev->in_ctx); + ctrl_ctx->add_flags = cpu_to_le32(SLOT_FLAG | EP0_FLAG); + ctrl_ctx->drop_flags = 0; + xhci_dbg(xhci, "Slot ID %d Input Context:\n", udev->slot_id); xhci_dbg_ctx(xhci, virt_dev->in_ctx, 2); @@ -3585,7 +3589,6 @@ int xhci_address_device(struct usb_hcd *hcd, struct usb_device *udev) virt_dev->address = (le32_to_cpu(slot_ctx->dev_state) & DEV_ADDR_MASK) + 1; /* Zero the input context control for later use */ - ctrl_ctx = xhci_get_input_control_ctx(xhci, virt_dev->in_ctx); ctrl_ctx->add_flags = 0; ctrl_ctx->drop_flags = 0; -- GitLab From 79c3dd8150fd5236d95766a9e662e3e932b462c9 Mon Sep 17 00:00:00 2001 From: Don Zickus Date: Thu, 3 Nov 2011 09:07:18 -0400 Subject: [PATCH 0083/3089] usb, xhci: Clear warm reset change event during init I noticed on my Panther Point system that I wasn't getting hotplug events for my usb3.0 disk on a usb3 port. I tracked it down to the fact that the system had the warm reset change bit still set. This seemed to block future events from being received, including a hotplug event. Clearing this bit during initialization allowed the hotplug event to be received and the disk to be recognized correctly. This patch should be backported to kernels as old as 2.6.39. Signed-off-by: Don Zickus Signed-off-by: Sarah Sharp Cc: stable@vger.kernel.org --- drivers/usb/core/hub.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c index 96f05b29c9ad2..79781461eec97 100644 --- a/drivers/usb/core/hub.c +++ b/drivers/usb/core/hub.c @@ -813,6 +813,12 @@ static void hub_activate(struct usb_hub *hub, enum hub_activation_type type) USB_PORT_FEAT_C_PORT_LINK_STATE); } + if ((portchange & USB_PORT_STAT_C_BH_RESET) && + hub_is_superspeed(hub->hdev)) { + need_debounce_delay = true; + clear_port_feature(hub->hdev, port1, + USB_PORT_FEAT_C_BH_PORT_RESET); + } /* We can forget about a "removed" device when there's a * physical disconnect or the connect status changes. */ -- GitLab From 1788ea6e3b2a58cf4fb00206e362d9caff8d86a7 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Fri, 4 Nov 2011 13:31:21 -0400 Subject: [PATCH 0084/3089] nfs: when attempting to open a directory, fall back on normal lookup (try #5) commit d953126 changed how nfs_atomic_lookup handles an -EISDIR return from an OPEN call. Prior to that patch, that caused the client to fall back to doing a normal lookup. When that patch went in, the code began returning that error to userspace. The d_revalidate codepath however never had the corresponding change, so it was still possible to end up with a NULL ctx->state pointer after that. That patch caused a regression. When we attempt to open a directory that does not have a cached dentry, that open now errors out with EISDIR. If you attempt the same open with a cached dentry, it will succeed. Fix this by reverting the change in nfs_atomic_lookup and allowing attempts to open directories to fall back to a normal lookup Also, add a NFSv4-specific f_ops->open routine that just returns -ENOTDIR. This should never be called if things are working properly, but if it ever is, then the dprintk may help in debugging. To facilitate this, a new file_operations field is also added to the nfs_rpc_ops struct. Cc: stable@kernel.org Signed-off-by: Jeff Layton Signed-off-by: Trond Myklebust --- fs/nfs/dir.c | 2 +- fs/nfs/file.c | 32 ++++++++++++++++++++++++++++++++ fs/nfs/inode.c | 2 +- fs/nfs/nfs3proc.c | 1 + fs/nfs/nfs4proc.c | 1 + fs/nfs/proc.c | 1 + include/linux/nfs_fs.h | 3 +++ include/linux/nfs_xdr.h | 1 + 8 files changed, 41 insertions(+), 2 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index b238d95ac48c7..ac28990981476 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1468,12 +1468,12 @@ static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry res = NULL; goto out; /* This turned out not to be a regular file */ + case -EISDIR: case -ENOTDIR: goto no_open; case -ELOOP: if (!(nd->intent.open.flags & O_NOFOLLOW)) goto no_open; - /* case -EISDIR: */ /* case -EINVAL: */ default: res = ERR_CAST(inode); diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 0a1f8312b4dcf..6d93e0754b5ed 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -886,3 +886,35 @@ static int nfs_setlease(struct file *file, long arg, struct file_lock **fl) file->f_path.dentry->d_name.name, arg); return -EINVAL; } + +#ifdef CONFIG_NFS_V4 +static int +nfs4_file_open(struct inode *inode, struct file *filp) +{ + /* + * NFSv4 opens are handled in d_lookup and d_revalidate. If we get to + * this point, then something is very wrong + */ + dprintk("NFS: %s called! inode=%p filp=%p\n", __func__, inode, filp); + return -ENOTDIR; +} + +const struct file_operations nfs4_file_operations = { + .llseek = nfs_file_llseek, + .read = do_sync_read, + .write = do_sync_write, + .aio_read = nfs_file_read, + .aio_write = nfs_file_write, + .mmap = nfs_file_mmap, + .open = nfs4_file_open, + .flush = nfs_file_flush, + .release = nfs_file_release, + .fsync = nfs_file_fsync, + .lock = nfs_lock, + .flock = nfs_flock, + .splice_read = nfs_file_splice_read, + .splice_write = nfs_file_splice_write, + .check_flags = nfs_check_flags, + .setlease = nfs_setlease, +}; +#endif /* CONFIG_NFS_V4 */ diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index c07a55aec8386..50a15fa8cf985 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -291,7 +291,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) */ inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->file_inode_ops; if (S_ISREG(inode->i_mode)) { - inode->i_fop = &nfs_file_operations; + inode->i_fop = NFS_SB(sb)->nfs_client->rpc_ops->file_ops; inode->i_data.a_ops = &nfs_file_aops; inode->i_data.backing_dev_info = &NFS_SB(sb)->backing_dev_info; } else if (S_ISDIR(inode->i_mode)) { diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 85f1690ca08c1..d4bc9ed917484 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -853,6 +853,7 @@ const struct nfs_rpc_ops nfs_v3_clientops = { .dentry_ops = &nfs_dentry_operations, .dir_inode_ops = &nfs3_dir_inode_operations, .file_inode_ops = &nfs3_file_inode_operations, + .file_ops = &nfs_file_operations, .getroot = nfs3_proc_get_root, .getattr = nfs3_proc_getattr, .setattr = nfs3_proc_setattr, diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index b60fddf606f7d..069cb8094d43c 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -6253,6 +6253,7 @@ const struct nfs_rpc_ops nfs_v4_clientops = { .dentry_ops = &nfs4_dentry_operations, .dir_inode_ops = &nfs4_dir_inode_operations, .file_inode_ops = &nfs4_file_inode_operations, + .file_ops = &nfs4_file_operations, .getroot = nfs4_proc_get_root, .getattr = nfs4_proc_getattr, .setattr = nfs4_proc_setattr, diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index ac40b8535d7e0..f48125da198a2 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -710,6 +710,7 @@ const struct nfs_rpc_ops nfs_v2_clientops = { .dentry_ops = &nfs_dentry_operations, .dir_inode_ops = &nfs_dir_inode_operations, .file_inode_ops = &nfs_file_inode_operations, + .file_ops = &nfs_file_operations, .getroot = nfs_proc_get_root, .getattr = nfs_proc_getattr, .setattr = nfs_proc_setattr, diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index ab2c6343361ae..92ecf5585facf 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -410,6 +410,9 @@ extern const struct inode_operations nfs_file_inode_operations; extern const struct inode_operations nfs3_file_inode_operations; #endif /* CONFIG_NFS_V3 */ extern const struct file_operations nfs_file_operations; +#ifdef CONFIG_NFS_V4 +extern const struct file_operations nfs4_file_operations; +#endif /* CONFIG_NFS_V4 */ extern const struct address_space_operations nfs_file_aops; extern const struct address_space_operations nfs_dir_aops; diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index c74595ba70947..2a7c533be5dd8 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -1192,6 +1192,7 @@ struct nfs_rpc_ops { const struct dentry_operations *dentry_ops; const struct inode_operations *dir_inode_ops; const struct inode_operations *file_inode_ops; + const struct file_operations *file_ops; int (*getroot) (struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *); -- GitLab From 0486958f57a496212e3c1e3d9194deebba3dc3d4 Mon Sep 17 00:00:00 2001 From: Jeff Layton Date: Fri, 4 Nov 2011 13:31:22 -0400 Subject: [PATCH 0085/3089] nfs: move nfs_file_operations declaration to bottom of file.c (try #2) ...a remove a set of forward declarations. Signed-off-by: Jeff Layton Signed-off-by: Trond Myklebust --- fs/nfs/file.c | 59 +++++++++++++++++---------------------------------- 1 file changed, 19 insertions(+), 40 deletions(-) diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 6d93e0754b5ed..eca56d4b39c0a 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -40,48 +40,8 @@ #define NFSDBG_FACILITY NFSDBG_FILE -static int nfs_file_open(struct inode *, struct file *); -static int nfs_file_release(struct inode *, struct file *); -static loff_t nfs_file_llseek(struct file *file, loff_t offset, int origin); -static int nfs_file_mmap(struct file *, struct vm_area_struct *); -static ssize_t nfs_file_splice_read(struct file *filp, loff_t *ppos, - struct pipe_inode_info *pipe, - size_t count, unsigned int flags); -static ssize_t nfs_file_read(struct kiocb *, const struct iovec *iov, - unsigned long nr_segs, loff_t pos); -static ssize_t nfs_file_splice_write(struct pipe_inode_info *pipe, - struct file *filp, loff_t *ppos, - size_t count, unsigned int flags); -static ssize_t nfs_file_write(struct kiocb *, const struct iovec *iov, - unsigned long nr_segs, loff_t pos); -static int nfs_file_flush(struct file *, fl_owner_t id); -static int nfs_file_fsync(struct file *, loff_t, loff_t, int datasync); -static int nfs_check_flags(int flags); -static int nfs_lock(struct file *filp, int cmd, struct file_lock *fl); -static int nfs_flock(struct file *filp, int cmd, struct file_lock *fl); -static int nfs_setlease(struct file *file, long arg, struct file_lock **fl); - static const struct vm_operations_struct nfs_file_vm_ops; -const struct file_operations nfs_file_operations = { - .llseek = nfs_file_llseek, - .read = do_sync_read, - .write = do_sync_write, - .aio_read = nfs_file_read, - .aio_write = nfs_file_write, - .mmap = nfs_file_mmap, - .open = nfs_file_open, - .flush = nfs_file_flush, - .release = nfs_file_release, - .fsync = nfs_file_fsync, - .lock = nfs_lock, - .flock = nfs_flock, - .splice_read = nfs_file_splice_read, - .splice_write = nfs_file_splice_write, - .check_flags = nfs_check_flags, - .setlease = nfs_setlease, -}; - const struct inode_operations nfs_file_inode_operations = { .permission = nfs_permission, .getattr = nfs_getattr, @@ -887,6 +847,25 @@ static int nfs_setlease(struct file *file, long arg, struct file_lock **fl) return -EINVAL; } +const struct file_operations nfs_file_operations = { + .llseek = nfs_file_llseek, + .read = do_sync_read, + .write = do_sync_write, + .aio_read = nfs_file_read, + .aio_write = nfs_file_write, + .mmap = nfs_file_mmap, + .open = nfs_file_open, + .flush = nfs_file_flush, + .release = nfs_file_release, + .fsync = nfs_file_fsync, + .lock = nfs_lock, + .flock = nfs_flock, + .splice_read = nfs_file_splice_read, + .splice_write = nfs_file_splice_write, + .check_flags = nfs_check_flags, + .setlease = nfs_setlease, +}; + #ifdef CONFIG_NFS_V4 static int nfs4_file_open(struct inode *inode, struct file *filp) -- GitLab From 49aa29513ec995f201664cf6eee36e5326ed38bf Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 2 Nov 2011 16:46:46 -0400 Subject: [PATCH 0086/3089] tracing: Add boiler plate for subsystem filter The system filter can be used to set multiple event filters that exist within the system. But currently it displays the last filter written that does not necessarily correspond to the filters within the system. The system filter itself is not used to filter any events. The system filter is just a means to set filters of the events within it. Because this causes an ambiguous state when the system filter reads a filter string but the events within the system have different strings it is best to just show a boiler plate: ### global filter ### # Use this to set filters for multiple events. # Only events with the given fields will be affected. # If no events are modified, an error message will be displayed here. If an error occurs while writing to the system filter, the system filter will replace the boiler plate with the error message as it currently does. Signed-off-by: Steven Rostedt --- kernel/trace/trace_events_filter.c | 26 +++++++++++++++++++------- 1 file changed, 19 insertions(+), 7 deletions(-) diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c index 86040d9c1accf..fdc6d22d406b6 100644 --- a/kernel/trace/trace_events_filter.c +++ b/kernel/trace/trace_events_filter.c @@ -27,6 +27,12 @@ #include "trace.h" #include "trace_output.h" +#define DEFAULT_SYS_FILTER_MESSAGE \ + "### global filter ###\n" \ + "# Use this to set filters for multiple events.\n" \ + "# Only events with the given fields will be affected.\n" \ + "# If no events are modified, an error message will be displayed here" + enum filter_op_ids { OP_OR, @@ -646,7 +652,7 @@ void print_subsystem_event_filter(struct event_subsystem *system, if (filter && filter->filter_string) trace_seq_printf(s, "%s\n", filter->filter_string); else - trace_seq_printf(s, "none\n"); + trace_seq_printf(s, DEFAULT_SYS_FILTER_MESSAGE "\n"); mutex_unlock(&event_mutex); } @@ -1838,7 +1844,10 @@ int apply_subsystem_event_filter(struct event_subsystem *system, if (!filter) goto out; - replace_filter_string(filter, filter_string); + /* System filters just show a default message */ + kfree(filter->filter_string); + filter->filter_string = NULL; + /* * No event actually uses the system filter * we can free it without synchronize_sched(). @@ -1848,14 +1857,12 @@ int apply_subsystem_event_filter(struct event_subsystem *system, parse_init(ps, filter_ops, filter_string); err = filter_parse(ps); - if (err) { - append_filter_err(ps, system->filter); - goto out; - } + if (err) + goto err_filter; err = replace_system_preds(system, ps, filter_string); if (err) - append_filter_err(ps, system->filter); + goto err_filter; out: filter_opstack_clear(ps); @@ -1865,6 +1872,11 @@ out_unlock: mutex_unlock(&event_mutex); return err; + +err_filter: + replace_filter_string(filter, filter_string); + append_filter_err(ps, system->filter); + goto out; } #ifdef CONFIG_PERF_EVENTS -- GitLab From 3890c136357284cb0656f9dd0e62286995ad32e9 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Fri, 28 Oct 2011 15:48:36 +0800 Subject: [PATCH 0087/3089] tracing: update Documentation on max preds limit Preds is no longer stored in a fixed-size stack. The max preds now is 2^14, and is way more than enough, so don't bother to document it. Link: http://lkml.kernel.org/r/4EAA5E54.1080102@cn.fujitsu.com Signed-off-by: Li Zefan Signed-off-by: Steven Rostedt --- Documentation/trace/events.txt | 2 -- 1 file changed, 2 deletions(-) diff --git a/Documentation/trace/events.txt b/Documentation/trace/events.txt index b510564aac7eb..bb24c2a0e870d 100644 --- a/Documentation/trace/events.txt +++ b/Documentation/trace/events.txt @@ -191,8 +191,6 @@ And for string fields they are: Currently, only exact string matches are supported. -Currently, the maximum number of predicates in a filter is 16. - 5.2 Setting filters ------------------- -- GitLab From b42c909743595208f7987e331f0921b73af2ce25 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Tomas=20Miljenovi=C4=87?= Date: Fri, 4 Nov 2011 23:59:49 -0700 Subject: [PATCH 0088/3089] Input: i8042 - add HP Pavilion dv4s to 'notimeout' and 'nomux' blacklists Touchpad input doesn't work with newer HP Pavilion dv4 laptops due to bad i8042 timeout data. Booting with i8042.notimeout and i8042.nomux successfully works around the problem. This patch adds the devices to the i8042 notimeout and nomux blacklists. Signed-off-by: Tomas Miljenovic Signed-off-by: Dmitry Torokhov --- drivers/input/serio/i8042-x86ia64io.h | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/drivers/input/serio/i8042-x86ia64io.h b/drivers/input/serio/i8042-x86ia64io.h index bb9f5d31f0d06..b4cfc6c8be89d 100644 --- a/drivers/input/serio/i8042-x86ia64io.h +++ b/drivers/input/serio/i8042-x86ia64io.h @@ -431,6 +431,13 @@ static const struct dmi_system_id __initconst i8042_dmi_nomux_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "Vostro V13"), }, }, + { + /* Newer HP Pavilion dv4 models */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"), + DMI_MATCH(DMI_PRODUCT_NAME, "HP Pavilion dv4 Notebook PC"), + }, + }, { } }; @@ -560,6 +567,13 @@ static const struct dmi_system_id __initconst i8042_dmi_notimeout_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "Vostro V13"), }, }, + { + /* Newer HP Pavilion dv4 models */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"), + DMI_MATCH(DMI_PRODUCT_NAME, "HP Pavilion dv4 Notebook PC"), + }, + }, { } }; -- GitLab From 15a2015fbc692e1c97d7ce12d96e077f5ae7ea6d Mon Sep 17 00:00:00 2001 From: Sage Weil Date: Sat, 5 Nov 2011 22:06:31 -0700 Subject: [PATCH 0089/3089] ceph: fix iput race when queueing inode work If we queue a work item that calls iput(), make sure we ihold() before attempting to queue work. Otherwise our queued work might miraculously run before we notice the queue_work() succeeded and call ihold(), allowing the inode to be destroyed. That is, instead of if (queue_work(...)) ihold(); we need to do ihold(); if (!queue_work(...)) iput(); Reported-by: Amon Ott Signed-off-by: Sage Weil --- fs/ceph/inode.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index e392bfce84a3c..116f36502f178 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -1328,12 +1328,13 @@ int ceph_inode_set_size(struct inode *inode, loff_t size) */ void ceph_queue_writeback(struct inode *inode) { + ihold(inode); if (queue_work(ceph_inode_to_client(inode)->wb_wq, &ceph_inode(inode)->i_wb_work)) { dout("ceph_queue_writeback %p\n", inode); - ihold(inode); } else { dout("ceph_queue_writeback %p failed\n", inode); + iput(inode); } } @@ -1353,12 +1354,13 @@ static void ceph_writeback_work(struct work_struct *work) */ void ceph_queue_invalidate(struct inode *inode) { + ihold(inode); if (queue_work(ceph_inode_to_client(inode)->pg_inv_wq, &ceph_inode(inode)->i_pg_inv_work)) { dout("ceph_queue_invalidate %p\n", inode); - ihold(inode); } else { dout("ceph_queue_invalidate %p failed\n", inode); + iput(inode); } } @@ -1434,13 +1436,14 @@ void ceph_queue_vmtruncate(struct inode *inode) { struct ceph_inode_info *ci = ceph_inode(inode); + ihold(inode); if (queue_work(ceph_sb_to_client(inode->i_sb)->trunc_wq, &ci->i_vmtruncate_work)) { dout("ceph_queue_vmtruncate %p\n", inode); - ihold(inode); } else { dout("ceph_queue_vmtruncate %p failed, pending=%d\n", inode, ci->i_truncate_pending); + iput(inode); } } -- GitLab From eb513689c97e3e73bb9b4459d490a8e894b4a546 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Mon, 7 Nov 2011 10:47:42 -0500 Subject: [PATCH 0090/3089] ext4: ignore journalled data options on remount if fs has no journal This avoids a confusing failure in the init scripts when the /etc/fstab has data=writeback or data=journal but the file system does not have a journal. So check for this case explicitly, and warn the user that we are ignoring the (pointless, since they have no journal) data=* mount option. Signed-off-by: "Theodore Ts'o" --- fs/ext4/super.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 9953d80145ad0..0435013dbd467 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1683,7 +1683,9 @@ static int parse_options(char *options, struct super_block *sb, data_opt = EXT4_MOUNT_WRITEBACK_DATA; datacheck: if (is_remount) { - if (test_opt(sb, DATA_FLAGS) != data_opt) { + if (!sbi->s_journal) + ext4_msg(sb, KERN_WARNING, "Remounting file system with no journal so ignoring journalled data option"); + else if (test_opt(sb, DATA_FLAGS) != data_opt) { ext4_msg(sb, KERN_ERR, "Cannot change data mode on remount"); return 0; -- GitLab From 2397256d6218e7bf5147a3b01dcc6aec20fd3916 Mon Sep 17 00:00:00 2001 From: Richard Weinberger Date: Mon, 7 Nov 2011 10:50:09 -0500 Subject: [PATCH 0091/3089] ext4: Remove kernel_lock annotations The BKL is gone, these annotations are useless. Signed-off-by: Richard Weinberger Signed-off-by: "Theodore Ts'o" --- fs/ext4/super.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 0435013dbd467..3858767ec672e 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -3101,8 +3101,6 @@ static void ext4_destroy_lazyinit_thread(void) } static int ext4_fill_super(struct super_block *sb, void *data, int silent) - __releases(kernel_lock) - __acquires(kernel_lock) { char *orig_data = kstrdup(data, GFP_KERNEL); struct buffer_head *bh; -- GitLab From 3c1fcb2c24519febd7ca9cf292fa5bdf513b601f Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Mon, 7 Nov 2011 11:01:13 -0500 Subject: [PATCH 0092/3089] ext4: add blk_finish_plug in error case of writepages. blk_finish_plug is needed in error case of writepages. Signed-off-by: Namjae Jeon Signed-off-by: "Theodore Ts'o" --- fs/ext4/inode.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index cc5a6da030a14..18b262bb6a4a1 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2270,6 +2270,7 @@ retry: ext4_msg(inode->i_sb, KERN_CRIT, "%s: jbd2_start: " "%ld pages, ino %lu; err %d", __func__, wbc->nr_to_write, inode->i_ino, ret); + blk_finish_plug(&plug); goto out_writepages; } -- GitLab From e5e78d08f3ab3094783b8df08a5b6d1d1a56a58f Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 2 Nov 2011 20:24:16 -0400 Subject: [PATCH 0093/3089] lockdep: Show subclass in pretty print of lockdep output The pretty print of the lockdep debug splat uses just the lock name to show how the locking scenario happens. But when it comes to nesting locks, the output becomes confusing which takes away the point of the pretty printing of the lock scenario. Without displaying the subclass info, we get the following output: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(slock-AF_INET); lock(slock-AF_INET); lock(slock-AF_INET); lock(slock-AF_INET); *** DEADLOCK *** The above looks more of a A->A locking bug than a A->B B->A. By adding the subclass to the output, we can see what really happened: other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(slock-AF_INET); lock(slock-AF_INET/1); lock(slock-AF_INET); lock(slock-AF_INET/1); *** DEADLOCK *** This bug was discovered while tracking down a real bug caught by lockdep. Link: http://lkml.kernel.org/r/20111025202049.GB25043@hostway.ca Cc: Peter Zijlstra Reported-by: Thomas Gleixner Tested-by: Simon Kirby Signed-off-by: Steven Rostedt --- kernel/lockdep.c | 30 +++++++++++++----------------- 1 file changed, 13 insertions(+), 17 deletions(-) diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 91d67ce3a8d52..6bd915df5fd34 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -490,36 +490,32 @@ void get_usage_chars(struct lock_class *class, char usage[LOCK_USAGE_CHARS]) usage[i] = '\0'; } -static int __print_lock_name(struct lock_class *class) +static void __print_lock_name(struct lock_class *class) { char str[KSYM_NAME_LEN]; const char *name; - name = class->name; - if (!name) - name = __get_key_name(class->key, str); - - return printk("%s", name); -} - -static void print_lock_name(struct lock_class *class) -{ - char str[KSYM_NAME_LEN], usage[LOCK_USAGE_CHARS]; - const char *name; - - get_usage_chars(class, usage); - name = class->name; if (!name) { name = __get_key_name(class->key, str); - printk(" (%s", name); + printk("%s", name); } else { - printk(" (%s", name); + printk("%s", name); if (class->name_version > 1) printk("#%d", class->name_version); if (class->subclass) printk("/%d", class->subclass); } +} + +static void print_lock_name(struct lock_class *class) +{ + char usage[LOCK_USAGE_CHARS]; + + get_usage_chars(class, usage); + + printk(" ("); + __print_lock_name(class); printk("){%s}", usage); } -- GitLab From 8ee3c92b7f2751c392be2d8fc360a410480b8757 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 4 Nov 2011 10:45:23 -0400 Subject: [PATCH 0094/3089] ftrace: Remove force undef config value left for testing A forced undef of a config value was used for testing and was accidently left in during the final commit. This causes x86 to run slower than needed while running function tracing as well as causes the function graph selftest to fail when DYNMAIC_FTRACE is not set. This is because the code in MCOUNT expects the ftrace code to be processed with the config value set that happened to be forced not set. The forced config option was left in by: commit 6331c28c962561aee59e5a493b7556a4bb585957 ftrace: Fix dynamic selftest failure on some archs Link: http://lkml.kernel.org/r/20111102150255.GA6973@debian Cc: stable@vger.kernel.org Reported-by: Rabin Vincent Signed-off-by: Steven Rostedt --- kernel/trace/ftrace.c | 1 - 1 file changed, 1 deletion(-) diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 077d853879088..0fcc6caead1cf 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -151,7 +151,6 @@ void clear_ftrace_function(void) ftrace_pid_function = ftrace_stub; } -#undef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST #ifndef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST /* * For those archs that do not test ftrace_trace_stop in their -- GitLab From c8452afb7426f7e21388492f40227582e3e83879 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Tue, 18 Oct 2011 19:55:51 +0200 Subject: [PATCH 0095/3089] jump_label: jump_label_inc may return before the code is patched If cpu A calls jump_label_inc() just after atomic_add_return() is called by cpu B, atomic_inc_not_zero() will return value greater then zero and jump_label_inc() will return to a caller before jump_label_update() finishes its job on cpu B. Link: http://lkml.kernel.org/r/20111018175551.GH17571@redhat.com Cc: stable@vger.kernel.org Cc: Peter Zijlstra Acked-by: Jason Baron Signed-off-by: Gleb Natapov Signed-off-by: Steven Rostedt --- kernel/jump_label.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/jump_label.c b/kernel/jump_label.c index a8ce45097f3d2..e6f1f24ad5778 100644 --- a/kernel/jump_label.c +++ b/kernel/jump_label.c @@ -66,8 +66,9 @@ void jump_label_inc(struct jump_label_key *key) return; jump_label_lock(); - if (atomic_add_return(1, &key->enabled) == 1) + if (atomic_read(&key->enabled) == 0) jump_label_update(key, JUMP_LABEL_ENABLE); + atomic_inc(&key->enabled); jump_label_unlock(); } -- GitLab From 49908a1b25d448d68fd26faca260e1850201575f Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 4 Nov 2011 16:32:25 -0400 Subject: [PATCH 0096/3089] perf: Fix parsing of __print_flags() in TP_printk() A update is made to the sched:sched_switch event that adds some logic to the first parameter of the __print_flags() that shows the state of tasks. This change cause perf to fail parsing the flags. A simple fix is needed to have the parser be able to process ops within the argument. Cc: stable@vger.kernel.org Reported-by: Andrew Vagin Signed-off-by: Steven Rostedt --- tools/perf/util/trace-event-parse.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tools/perf/util/trace-event-parse.c b/tools/perf/util/trace-event-parse.c index 0a7ed5b5e281c..6c164dc9ee957 100644 --- a/tools/perf/util/trace-event-parse.c +++ b/tools/perf/util/trace-event-parse.c @@ -1537,6 +1537,8 @@ process_flags(struct event *event, struct print_arg *arg, char **tok) field = malloc_or_die(sizeof(*field)); type = process_arg(event, field, &token); + while (type == EVENT_OP) + type = process_op(event, field, &token); if (test_type_token(type, token, EVENT_DELIM, ",")) goto out_free; -- GitLab From fdbd3ce9efb3a045266f2f6b2f1b6047882ff092 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Mon, 7 Nov 2011 07:53:23 -0800 Subject: [PATCH 0097/3089] PCI: pciehp: Retrieve link speed after link is trained During hot plug, board_added will call pciehp_power_on_slot(). But link speed is updated in pciehp_power_on_slot(). We should not update link speed there, because that is too early. So move the link speed update to pciehp_check_link_status() after making sure the link has been trained. -v2: fix compile warning that Kenji found. Signed-off-by: Yinghai Lu Reviewed-by: Kenji Kaneshige Tested-by: Kenji Kaneshige Signed-off-by: Jesse Barnes --- drivers/pci/hotplug/pciehp_hpc.c | 11 ++--------- 1 file changed, 2 insertions(+), 9 deletions(-) diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c index 96dc4734e4aff..6692832c9c81d 100644 --- a/drivers/pci/hotplug/pciehp_hpc.c +++ b/drivers/pci/hotplug/pciehp_hpc.c @@ -294,6 +294,8 @@ int pciehp_check_link_status(struct controller *ctrl) return retval; } + pcie_update_link_speed(ctrl->pcie->port->subordinate, lnk_status); + return retval; } @@ -484,7 +486,6 @@ int pciehp_power_on_slot(struct slot * slot) u16 slot_cmd; u16 cmd_mask; u16 slot_status; - u16 lnk_status; int retval = 0; /* Clear sticky power-fault bit from previous power failures */ @@ -516,14 +517,6 @@ int pciehp_power_on_slot(struct slot * slot) ctrl_dbg(ctrl, "%s: SLOTCTRL %x write cmd %x\n", __func__, pci_pcie_cap(ctrl->pcie->port) + PCI_EXP_SLTCTL, slot_cmd); - retval = pciehp_readw(ctrl, PCI_EXP_LNKSTA, &lnk_status); - if (retval) { - ctrl_err(ctrl, "%s: Cannot read LNKSTA register\n", - __func__); - return retval; - } - pcie_update_link_speed(ctrl->pcie->port->subordinate, lnk_status); - return retval; } -- GitLab From d4d34b981a5327eec956c6cb4cce397ce6f57279 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 4 Nov 2011 20:32:39 -0400 Subject: [PATCH 0098/3089] ftrace: Fix hash record accounting bug If the set_ftrace_filter is cleared by writing just whitespace to it, then the filter hash refcounts will be decremented but not updated. This causes two bugs: 1) No functions will be enabled for tracing when they all should be 2) If the users clears the set_ftrace_filter twice, it will crash ftrace: ------------[ cut here ]------------ WARNING: at /home/rostedt/work/git/linux-trace.git/kernel/trace/ftrace.c:1384 __ftrace_hash_rec_update.part.27+0x157/0x1a7() Modules linked in: Pid: 2330, comm: bash Not tainted 3.1.0-test+ #32 Call Trace: [] warn_slowpath_common+0x83/0x9b [] warn_slowpath_null+0x1a/0x1c [] __ftrace_hash_rec_update.part.27+0x157/0x1a7 [] ? ftrace_regex_release+0xa7/0x10f [] ? kfree+0xe5/0x115 [] ftrace_hash_move+0x2e/0x151 [] ftrace_regex_release+0xba/0x10f [] fput+0xfd/0x1c2 [] filp_close+0x6d/0x78 [] sys_dup3+0x197/0x1c1 [] sys_dup2+0x4f/0x54 [] system_call_fastpath+0x16/0x1b ---[ end trace 77a3a7ee73794a02 ]--- Link: http://lkml.kernel.org/r/20111101141420.GA4918@debian Reported-by: Rabin Vincent Signed-off-by: Steven Rostedt --- kernel/trace/ftrace.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 0fcc6caead1cf..7caa4505508da 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -1210,7 +1210,9 @@ ftrace_hash_move(struct ftrace_ops *ops, int enable, if (!src->count) { free_ftrace_hash_rcu(*dst); rcu_assign_pointer(*dst, EMPTY_HASH); - return 0; + /* still need to update the function records */ + ret = 0; + goto out; } /* -- GitLab From 7e9a49ef542610609144d1afcd516dc3fafac4d6 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Mon, 7 Nov 2011 16:08:49 +0100 Subject: [PATCH 0099/3089] tracing/latency: Fix header output for latency tracers In case the the graph tracer (CONFIG_FUNCTION_GRAPH_TRACER) or even the function tracer (CONFIG_FUNCTION_TRACER) are not set, the latency tracers do not display proper latency header. The involved/fixed latency tracers are: wakeup_rt wakeup preemptirqsoff preemptoff irqsoff The patch adds proper handling of tracer configuration options for latency tracers, and displaying correct header info accordingly. * The current output (for wakeup tracer) with both graph and function tracers disabled is: # tracer: wakeup # -0 0d.h5 1us+: 0:120:R + [000] 7: 0:R watchdog/0 -0 0d.h5 3us+: ttwu_do_activate.clone.1 <-try_to_wake_up ... * The fixed output is: # tracer: wakeup # # wakeup latency trace v1.1.5 on 3.1.0-tip+ # -------------------------------------------------------------------- # latency: 55 us, #4/4, CPU#0 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:2) # ----------------- # | task: migration/0-6 (uid:0 nice:0 policy:1 rt_prio:99) # ----------------- # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / cat-1129 0d..4 1us : 1129:120:R + [000] 6: 0:R migration/0 cat-1129 0d..4 2us+: ttwu_do_activate.clone.1 <-try_to_wake_up * The current output (for wakeup tracer) with only function tracer enabled is: # tracer: wakeup # cat-1140 0d..4 1us+: 1140:120:R + [000] 6: 0:R migration/0 cat-1140 0d..4 2us : ttwu_do_activate.clone.1 <-try_to_wake_up * The fixed output is: # tracer: wakeup # # wakeup latency trace v1.1.5 on 3.1.0-tip+ # -------------------------------------------------------------------- # latency: 207 us, #109/109, CPU#1 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:2) # ----------------- # | task: watchdog/1-12 (uid:0 nice:0 policy:1 rt_prio:99) # ----------------- # # _------=> CPU# # / _-----=> irqs-off # | / _----=> need-resched # || / _---=> hardirq/softirq # ||| / _--=> preempt-depth # |||| / delay # cmd pid ||||| time | caller # \ / ||||| \ | / -0 1d.h5 1us+: 0:120:R + [001] 12: 0:R watchdog/1 -0 1d.h5 3us : ttwu_do_activate.clone.1 <-try_to_wake_up Link: http://lkml.kernel.org/r/20111107150849.GE1807@m.brq.redhat.com Cc: Frederic Weisbecker Cc: Ingo Molnar Signed-off-by: Jiri Olsa Signed-off-by: Steven Rostedt --- kernel/trace/trace.c | 15 +++++++++++++++ kernel/trace/trace.h | 1 + kernel/trace/trace_irqsoff.c | 13 ++++++++++++- kernel/trace/trace_sched_wakeup.c | 13 ++++++++++++- 4 files changed, 40 insertions(+), 2 deletions(-) diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index b24a72d350082..b296186eb93a0 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -2140,6 +2140,21 @@ enum print_line_t print_trace_line(struct trace_iterator *iter) return print_trace_fmt(iter); } +void trace_latency_header(struct seq_file *m) +{ + struct trace_iterator *iter = m->private; + + /* print nothing if the buffers are empty */ + if (trace_empty(iter)) + return; + + if (iter->iter_flags & TRACE_FILE_LAT_FMT) + print_trace_header(m, iter); + + if (!(trace_flags & TRACE_ITER_VERBOSE)) + print_lat_help_header(m); +} + void trace_default_header(struct seq_file *m) { struct trace_iterator *iter = m->private; diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 092e1f8d18dc8..f8ec2291b522a 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -370,6 +370,7 @@ void trace_graph_function(struct trace_array *tr, unsigned long ip, unsigned long parent_ip, unsigned long flags, int pc); +void trace_latency_header(struct seq_file *m); void trace_default_header(struct seq_file *m); void print_trace_header(struct seq_file *m, struct trace_iterator *iter); int trace_empty(struct trace_iterator *iter); diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index a1a3359996a7c..a248c686b2b8e 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -280,9 +280,20 @@ static enum print_line_t irqsoff_print_line(struct trace_iterator *iter) } static void irqsoff_graph_return(struct ftrace_graph_ret *trace) { } -static void irqsoff_print_header(struct seq_file *s) { } static void irqsoff_trace_open(struct trace_iterator *iter) { } static void irqsoff_trace_close(struct trace_iterator *iter) { } + +#ifdef CONFIG_FUNCTION_TRACER +static void irqsoff_print_header(struct seq_file *s) +{ + trace_default_header(s); +} +#else +static void irqsoff_print_header(struct seq_file *s) +{ + trace_latency_header(s); +} +#endif /* CONFIG_FUNCTION_TRACER */ #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ /* diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index e4a70c0c71b64..ff791ea48b570 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -280,9 +280,20 @@ static enum print_line_t wakeup_print_line(struct trace_iterator *iter) } static void wakeup_graph_return(struct ftrace_graph_ret *trace) { } -static void wakeup_print_header(struct seq_file *s) { } static void wakeup_trace_open(struct trace_iterator *iter) { } static void wakeup_trace_close(struct trace_iterator *iter) { } + +#ifdef CONFIG_FUNCTION_TRACER +static void wakeup_print_header(struct seq_file *s) +{ + trace_default_header(s); +} +#else +static void wakeup_print_header(struct seq_file *s) +{ + trace_latency_header(s); +} +#endif /* CONFIG_FUNCTION_TRACER */ #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ /* -- GitLab From abf015f04614a3a7da43d35b55edc90189af4e6a Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 7 Nov 2011 19:04:36 +0000 Subject: [PATCH 0100/3089] Revert "ARM: 7098/1: kdump: copy kernel relocation code at the kexec prepare stage" This reverts commit 2b034922af2caa19df86f0c502e76cb6f6e910f9. Will Deacon reports: This is causing kexec to fail. The symptoms are that the .init.text section is not loaded as part of the new kernel image, so when we try to do the SMP/UP fixups we hit a whole sea of poison left there by the previous kernel. So my guess is that machine_kexec_prepare *is* too early for preparing the reboot_code_buffer and, unless anybody has a good reason not to, I'd like to revert the patch causing these problems. Reported-by: Will Deacon --- arch/arm/kernel/machine_kexec.c | 35 ++++++++++++++++----------------- 1 file changed, 17 insertions(+), 18 deletions(-) diff --git a/arch/arm/kernel/machine_kexec.c b/arch/arm/kernel/machine_kexec.c index c1b4463dcc839..e59bbd496c391 100644 --- a/arch/arm/kernel/machine_kexec.c +++ b/arch/arm/kernel/machine_kexec.c @@ -32,24 +32,6 @@ static atomic_t waiting_for_crash_ipi; int machine_kexec_prepare(struct kimage *image) { - unsigned long page_list; - void *reboot_code_buffer; - page_list = image->head & PAGE_MASK; - - reboot_code_buffer = page_address(image->control_code_page); - - /* Prepare parameters for reboot_code_buffer*/ - kexec_start_address = image->start; - kexec_indirection_page = page_list; - kexec_mach_type = machine_arch_type; - kexec_boot_atags = image->start - KEXEC_ARM_ZIMAGE_OFFSET + KEXEC_ARM_ATAGS_OFFSET; - - /* copy our kernel relocation code to the control code page */ - memcpy(reboot_code_buffer, - relocate_new_kernel, relocate_new_kernel_size); - - flush_icache_range((unsigned long) reboot_code_buffer, - (unsigned long) reboot_code_buffer + KEXEC_CONTROL_PAGE_SIZE); return 0; } @@ -100,14 +82,31 @@ void (*kexec_reinit)(void); void machine_kexec(struct kimage *image) { + unsigned long page_list; unsigned long reboot_code_buffer_phys; void *reboot_code_buffer; + + page_list = image->head & PAGE_MASK; + /* we need both effective and real address here */ reboot_code_buffer_phys = page_to_pfn(image->control_code_page) << PAGE_SHIFT; reboot_code_buffer = page_address(image->control_code_page); + /* Prepare parameters for reboot_code_buffer*/ + kexec_start_address = image->start; + kexec_indirection_page = page_list; + kexec_mach_type = machine_arch_type; + kexec_boot_atags = image->start - KEXEC_ARM_ZIMAGE_OFFSET + KEXEC_ARM_ATAGS_OFFSET; + + /* copy our kernel relocation code to the control code page */ + memcpy(reboot_code_buffer, + relocate_new_kernel, relocate_new_kernel_size); + + + flush_icache_range((unsigned long) reboot_code_buffer, + (unsigned long) reboot_code_buffer + KEXEC_CONTROL_PAGE_SIZE); printk(KERN_INFO "Bye!\n"); if (kexec_reinit) -- GitLab From a6f498a891c730327645a7afa10c5ae977de6fd8 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 7 Nov 2011 15:50:32 -0500 Subject: [PATCH 0101/3089] NFS: Fix a regression in the referral code Fix a regression that was introduced by commit 0c2e53f11a6dae9e3af5f50f5ad0382e7c3e0cfa (NFS: Remove the unused "lookupfh()" version of nfs4_proc_lookup()). In the case where the lookup gets an NFS4ERR_MOVED, we want to return the result of nfs4_get_referral(). Instead, that value is getting clobbered by the call to nfs4_handle_exception()... Reported-by: Tigran Mkrtchyan Tested-by: Tigran Mkrtchyan Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 069cb8094d43c..be2bbac13817c 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2464,8 +2464,7 @@ static int nfs4_proc_lookup(struct rpc_clnt *clnt, struct inode *dir, struct qst case -NFS4ERR_BADNAME: return -ENOENT; case -NFS4ERR_MOVED: - err = nfs4_get_referral(dir, name, fattr, fhandle); - break; + return nfs4_get_referral(dir, name, fattr, fhandle); case -NFS4ERR_WRONGSEC: nfs_fixup_secinfo_attributes(fattr, fhandle); } -- GitLab From 5698eb4eda657fd4a74612a65cd5d583f35a5bdb Mon Sep 17 00:00:00 2001 From: Kevin Hilman Date: Mon, 7 Nov 2011 15:58:40 -0800 Subject: [PATCH 0102/3089] ARM: OMAP3: CPUidle: include The CPUidle use THIS_MODULE, so needs Signed-off-by: Kevin Hilman Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/cpuidle34xx.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/mach-omap2/cpuidle34xx.c b/arch/arm/mach-omap2/cpuidle34xx.c index 4bf6e6e8b1001..2e0fa5a4aaeb3 100644 --- a/arch/arm/mach-omap2/cpuidle34xx.c +++ b/arch/arm/mach-omap2/cpuidle34xx.c @@ -24,6 +24,7 @@ #include #include +#include #include #include -- GitLab From 406478dc911e16677fbd9c84d1d50cdffbc031ab Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Mon, 7 Nov 2011 16:07:04 -0800 Subject: [PATCH 0103/3089] drm/i915: Turn on a required 3D clock gating bit on Sandybridge. Fixes rendering failures in Unigine Tropics and Sanctuary and the mesa "fire" demo. Signed-off-by: Eric Anholt Cc: stable@kernel.org Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_reg.h | 3 +++ drivers/gpu/drm/i915/intel_display.c | 9 +++++++++ 2 files changed, 12 insertions(+) diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 5a09416e611f5..b807275ea7393 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -3444,6 +3444,9 @@ #define GT_FIFO_FREE_ENTRIES 0x120008 #define GT_FIFO_NUM_RESERVED_ENTRIES 20 +#define GEN6_UCGCTL2 0x9404 +# define GEN6_RCPBUNIT_CLOCK_GATE_DISABLE (1 << 12) + #define GEN6_RPNSWREQ 0xA008 #define GEN6_TURBO_DISABLE (1<<31) #define GEN6_FREQUENCY(x) ((x)<<25) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 9fa342e89454b..2b2a7645cd0c4 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -8148,6 +8148,15 @@ static void gen6_init_clock_gating(struct drm_device *dev) I915_WRITE(WM2_LP_ILK, 0); I915_WRITE(WM1_LP_ILK, 0); + /* According to the BSpec vol1g, bit 12 (RCPBUNIT) clock + * gating disable must be set. Failure to set it results in + * flickering pixels due to Z write ordering failures after + * some amount of runtime in the Mesa "fire" demo, and Unigine + * Sanctuary and Tropics, and apparently anything else with + * alpha test or pixel discard. + */ + I915_WRITE(GEN6_UCGCTL2, GEN6_RCPBUNIT_CLOCK_GATE_DISABLE); + /* * According to the spec the following bits should be * set in order to enable memory self-refresh and fbc: -- GitLab From 9ca1d10d748e56964de95e3ed80211b192f56cf4 Mon Sep 17 00:00:00 2001 From: Eric Anholt Date: Mon, 7 Nov 2011 16:07:05 -0800 Subject: [PATCH 0104/3089] drm/i915: Turn on another required clock gating bit on gen6. Unlike the previous one, I don't have known testcases it fixes. I'd rather not go through the same debug cycle on whatever testcases those might be. Signed-off-by: Eric Anholt Cc: stable@kernel.org Signed-off-by: Keith Packard --- drivers/gpu/drm/i915/i915_reg.h | 1 + drivers/gpu/drm/i915/intel_display.c | 7 ++++++- 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index b807275ea7393..a34e86630f263 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -3446,6 +3446,7 @@ #define GEN6_UCGCTL2 0x9404 # define GEN6_RCPBUNIT_CLOCK_GATE_DISABLE (1 << 12) +# define GEN6_RCCUNIT_CLOCK_GATE_DISABLE (1 << 11) #define GEN6_RPNSWREQ 0xA008 #define GEN6_TURBO_DISABLE (1<<31) diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 2b2a7645cd0c4..591eb0ed31103 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -8154,8 +8154,13 @@ static void gen6_init_clock_gating(struct drm_device *dev) * some amount of runtime in the Mesa "fire" demo, and Unigine * Sanctuary and Tropics, and apparently anything else with * alpha test or pixel discard. + * + * According to the spec, bit 11 (RCCUNIT) must also be set, + * but we didn't debug actual testcases to find it out. */ - I915_WRITE(GEN6_UCGCTL2, GEN6_RCPBUNIT_CLOCK_GATE_DISABLE); + I915_WRITE(GEN6_UCGCTL2, + GEN6_RCPBUNIT_CLOCK_GATE_DISABLE | + GEN6_RCCUNIT_CLOCK_GATE_DISABLE); /* * According to the spec the following bits should be -- GitLab From 1258ea596fe0422272525d9ec8a25b703de4623f Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Tue, 8 Nov 2011 03:16:09 -0700 Subject: [PATCH 0105/3089] ARM: OMAP2xxx: HWMOD: Fix DSS reset DSS needs all DSS clocks to be enabled to be able to finish reset properly. Before v3.1-rc1 the omapdss driver was managing clocks and resets correctly. However, when omapdss started using runtime PM at v3.1-rc1, the responsibility for the reset moved to HWMOD framework. HWMOD framework does not currently enable all the DSS clocks when resetting the DSS hardware. This hasn't caused any problems so far, but we may just have been lucky. This patch sets HWMOD_CONTROL_OPT_CLKS_IN_RESET for dss_core in OMAP2xxx HWMOD data, fixing the issue. Signed-off-by: Tomi Valkeinen [paul@pwsan.com: merged duplicate .flags fields] Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/omap_hwmod_2420_data.c | 6 +++++- arch/arm/mach-omap2/omap_hwmod_2430_data.c | 6 +++++- 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-omap2/omap_hwmod_2420_data.c b/arch/arm/mach-omap2/omap_hwmod_2420_data.c index 6d7206213525d..5160e3509d5df 100644 --- a/arch/arm/mach-omap2/omap_hwmod_2420_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_2420_data.c @@ -875,6 +875,10 @@ static struct omap_hwmod_ocp_if *omap2420_dss_slaves[] = { }; static struct omap_hwmod_opt_clk dss_opt_clks[] = { + /* + * The DSS HW needs all DSS clocks enabled during reset. The dss_core + * driver does not use these clocks. + */ { .role = "tv_clk", .clk = "dss_54m_fck" }, { .role = "sys_clk", .clk = "dss2_fck" }, }; @@ -899,7 +903,7 @@ static struct omap_hwmod omap2420_dss_core_hwmod = { .slaves_cnt = ARRAY_SIZE(omap2420_dss_slaves), .masters = omap2420_dss_masters, .masters_cnt = ARRAY_SIZE(omap2420_dss_masters), - .flags = HWMOD_NO_IDLEST, + .flags = HWMOD_NO_IDLEST | HWMOD_CONTROL_OPT_CLKS_IN_RESET, }; /* l4_core -> dss_dispc */ diff --git a/arch/arm/mach-omap2/omap_hwmod_2430_data.c b/arch/arm/mach-omap2/omap_hwmod_2430_data.c index a2580d01c3ff9..d02a44ea7f9b5 100644 --- a/arch/arm/mach-omap2/omap_hwmod_2430_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_2430_data.c @@ -942,6 +942,10 @@ static struct omap_hwmod_ocp_if *omap2430_dss_slaves[] = { }; static struct omap_hwmod_opt_clk dss_opt_clks[] = { + /* + * The DSS HW needs all DSS clocks enabled during reset. The dss_core + * driver does not use these clocks. + */ { .role = "tv_clk", .clk = "dss_54m_fck" }, { .role = "sys_clk", .clk = "dss2_fck" }, }; @@ -966,7 +970,7 @@ static struct omap_hwmod omap2430_dss_core_hwmod = { .slaves_cnt = ARRAY_SIZE(omap2430_dss_slaves), .masters = omap2430_dss_masters, .masters_cnt = ARRAY_SIZE(omap2430_dss_masters), - .flags = HWMOD_NO_IDLEST, + .flags = HWMOD_NO_IDLEST | HWMOD_CONTROL_OPT_CLKS_IN_RESET, }; /* l4_core -> dss_dispc */ -- GitLab From b8ac10d8b75843c9ddd6aadf70a0cdf8aa783659 Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Tue, 8 Nov 2011 03:16:09 -0700 Subject: [PATCH 0106/3089] ARM: OMAP2xxx: HWMOD: fix DSS clock data The OMAP2xxx HWMOD data currently contains two errors with DSS clocks: - dss_rfbi is missing ick opt-clock, which is needed for RFBI to calculate timings - dss_venc's interface and main clocks are wrong, causing VENC to fail to start These problems were temporarily fixed with a DSS patch 9ede365aa6f74428a1f69c21ca1cf21213167576 ("HACK: OMAP: DSS2: clk hack for OMAP2/3"), which can be reverted after this patch (and the similar patches for other OMAPs). Signed-off-by: Tomi Valkeinen Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/omap_hwmod_2420_data.c | 10 ++++++++-- arch/arm/mach-omap2/omap_hwmod_2430_data.c | 10 ++++++++-- 2 files changed, 16 insertions(+), 4 deletions(-) diff --git a/arch/arm/mach-omap2/omap_hwmod_2420_data.c b/arch/arm/mach-omap2/omap_hwmod_2420_data.c index 5160e3509d5df..74037a5807afb 100644 --- a/arch/arm/mach-omap2/omap_hwmod_2420_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_2420_data.c @@ -965,6 +965,10 @@ static struct omap_hwmod_ocp_if *omap2420_dss_rfbi_slaves[] = { &omap2420_l4_core__dss_rfbi, }; +static struct omap_hwmod_opt_clk dss_rfbi_opt_clks[] = { + { .role = "ick", .clk = "dss_ick" }, +}; + static struct omap_hwmod omap2420_dss_rfbi_hwmod = { .name = "dss_rfbi", .class = &omap2_rfbi_hwmod_class, @@ -976,6 +980,8 @@ static struct omap_hwmod omap2420_dss_rfbi_hwmod = { .module_offs = CORE_MOD, }, }, + .opt_clks = dss_rfbi_opt_clks, + .opt_clks_cnt = ARRAY_SIZE(dss_rfbi_opt_clks), .slaves = omap2420_dss_rfbi_slaves, .slaves_cnt = ARRAY_SIZE(omap2420_dss_rfbi_slaves), .flags = HWMOD_NO_IDLEST, @@ -985,7 +991,7 @@ static struct omap_hwmod omap2420_dss_rfbi_hwmod = { static struct omap_hwmod_ocp_if omap2420_l4_core__dss_venc = { .master = &omap2420_l4_core_hwmod, .slave = &omap2420_dss_venc_hwmod, - .clk = "dss_54m_fck", + .clk = "dss_ick", .addr = omap2_dss_venc_addrs, .fw = { .omap2 = { @@ -1005,7 +1011,7 @@ static struct omap_hwmod_ocp_if *omap2420_dss_venc_slaves[] = { static struct omap_hwmod omap2420_dss_venc_hwmod = { .name = "dss_venc", .class = &omap2_venc_hwmod_class, - .main_clk = "dss1_fck", + .main_clk = "dss_54m_fck", .prcm = { .omap2 = { .prcm_reg_id = 1, diff --git a/arch/arm/mach-omap2/omap_hwmod_2430_data.c b/arch/arm/mach-omap2/omap_hwmod_2430_data.c index d02a44ea7f9b5..6f0365388dfc5 100644 --- a/arch/arm/mach-omap2/omap_hwmod_2430_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_2430_data.c @@ -1020,6 +1020,10 @@ static struct omap_hwmod_ocp_if *omap2430_dss_rfbi_slaves[] = { &omap2430_l4_core__dss_rfbi, }; +static struct omap_hwmod_opt_clk dss_rfbi_opt_clks[] = { + { .role = "ick", .clk = "dss_ick" }, +}; + static struct omap_hwmod omap2430_dss_rfbi_hwmod = { .name = "dss_rfbi", .class = &omap2_rfbi_hwmod_class, @@ -1031,6 +1035,8 @@ static struct omap_hwmod omap2430_dss_rfbi_hwmod = { .module_offs = CORE_MOD, }, }, + .opt_clks = dss_rfbi_opt_clks, + .opt_clks_cnt = ARRAY_SIZE(dss_rfbi_opt_clks), .slaves = omap2430_dss_rfbi_slaves, .slaves_cnt = ARRAY_SIZE(omap2430_dss_rfbi_slaves), .flags = HWMOD_NO_IDLEST, @@ -1040,7 +1046,7 @@ static struct omap_hwmod omap2430_dss_rfbi_hwmod = { static struct omap_hwmod_ocp_if omap2430_l4_core__dss_venc = { .master = &omap2430_l4_core_hwmod, .slave = &omap2430_dss_venc_hwmod, - .clk = "dss_54m_fck", + .clk = "dss_ick", .addr = omap2_dss_venc_addrs, .flags = OCPIF_SWSUP_IDLE, .user = OCP_USER_MPU | OCP_USER_SDMA, @@ -1054,7 +1060,7 @@ static struct omap_hwmod_ocp_if *omap2430_dss_venc_slaves[] = { static struct omap_hwmod omap2430_dss_venc_hwmod = { .name = "dss_venc", .class = &omap2_venc_hwmod_class, - .main_clk = "dss1_fck", + .main_clk = "dss_54m_fck", .prcm = { .omap2 = { .prcm_reg_id = 1, -- GitLab From 8c3105ca1a42783b13930fbd375484ec8c72e608 Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Tue, 8 Nov 2011 03:16:10 -0700 Subject: [PATCH 0107/3089] ARM: OMAP3: HWMOD: Fix DSS reset DSS needs all DSS clocks to be enabled to be able to finish reset properly. Before v3.1-rc1 the omapdss driver was managing clocks and resets correctly. However, when omapdss started using runtime PM at v3.1-rc1, the responsibility for the reset moved to HWMOD framework. HWMOD framework does not currently enable all the DSS clocks when resetting the DSS hardware. This hasn't caused any problems so far, but we may just have been lucky. dss_core's opt-clocks is also missing dss_96m_fck, which is a DSS clock present only on OMAP3430, and thus required on OMAP3430 to finish the reset. This patch sets HWMOD_CONTROL_OPT_CLKS_IN_RESET and adds the dss_96m_fck opt-clock for dss_core in OMAP3 HWMOD data, fixing the issue. Signed-off-by: Tomi Valkeinen [paul@pwsan.com: merged duplicate .flags fields] Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/omap_hwmod_3xxx_data.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c index bc9035ec87fc5..45ccd4f07cffa 100644 --- a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c @@ -1369,9 +1369,14 @@ static struct omap_hwmod_ocp_if *omap3xxx_dss_slaves[] = { }; static struct omap_hwmod_opt_clk dss_opt_clks[] = { - { .role = "tv_clk", .clk = "dss_tv_fck" }, - { .role = "video_clk", .clk = "dss_96m_fck" }, + /* + * The DSS HW needs all DSS clocks enabled during reset. The dss_core + * driver does not use these clocks. + */ { .role = "sys_clk", .clk = "dss2_alwon_fck" }, + { .role = "tv_clk", .clk = "dss_tv_fck" }, + /* required only on OMAP3430 */ + { .role = "tv_dac_clk", .clk = "dss_96m_fck" }, }; static struct omap_hwmod omap3430es1_dss_core_hwmod = { @@ -1394,11 +1399,12 @@ static struct omap_hwmod omap3430es1_dss_core_hwmod = { .slaves_cnt = ARRAY_SIZE(omap3430es1_dss_slaves), .masters = omap3xxx_dss_masters, .masters_cnt = ARRAY_SIZE(omap3xxx_dss_masters), - .flags = HWMOD_NO_IDLEST, + .flags = HWMOD_NO_IDLEST | HWMOD_CONTROL_OPT_CLKS_IN_RESET, }; static struct omap_hwmod omap3xxx_dss_core_hwmod = { .name = "dss_core", + .flags = HWMOD_CONTROL_OPT_CLKS_IN_RESET, .class = &omap2_dss_hwmod_class, .main_clk = "dss1_alwon_fck", /* instead of dss_fck */ .sdma_reqs = omap3xxx_dss_sdma_chs, -- GitLab From 6c3d7e34d68ffae888d33bf364f447e0b415591f Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Tue, 8 Nov 2011 03:16:10 -0700 Subject: [PATCH 0108/3089] ARM: OMAP3: HWMOD: fix DSS clock data The OMAP3 HWMOD data currently contains these errors with DSS clocks: - dss_rfbi is missing ick opt-clock, which is needed for RFBI to calculate timings - dss_dsi is missing ick and sys_clk - dss_venc is missing dss_96m_fck opt-clock, which is required on OMAP3430 - dss_venc's interface and main clocks are wrong, causing VENC to fail to start These problems were temporarily fixed with a DSS patch 9ede365aa6f74428a1f69c21ca1cf21213167576 ("HACK: OMAP: DSS2: clk hack for OMAP2/3"), which can be reverted after this patch (and the similar patches for other OMAPs). Signed-off-by: Tomi Valkeinen Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/omap_hwmod_3xxx_data.c | 24 ++++++++++++++++++++-- 1 file changed, 22 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c index 45ccd4f07cffa..4aaf0caaf3cc3 100644 --- a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c @@ -1492,6 +1492,7 @@ static struct omap_hwmod_addr_space omap3xxx_dss_dsi1_addrs[] = { static struct omap_hwmod_ocp_if omap3xxx_l4_core__dss_dsi1 = { .master = &omap3xxx_l4_core_hwmod, .slave = &omap3xxx_dss_dsi1_hwmod, + .clk = "dss_ick", .addr = omap3xxx_dss_dsi1_addrs, .fw = { .omap2 = { @@ -1508,6 +1509,10 @@ static struct omap_hwmod_ocp_if *omap3xxx_dss_dsi1_slaves[] = { &omap3xxx_l4_core__dss_dsi1, }; +static struct omap_hwmod_opt_clk dss_dsi1_opt_clks[] = { + { .role = "sys_clk", .clk = "dss2_alwon_fck" }, +}; + static struct omap_hwmod omap3xxx_dss_dsi1_hwmod = { .name = "dss_dsi1", .class = &omap3xxx_dsi_hwmod_class, @@ -1520,6 +1525,8 @@ static struct omap_hwmod omap3xxx_dss_dsi1_hwmod = { .module_offs = OMAP3430_DSS_MOD, }, }, + .opt_clks = dss_dsi1_opt_clks, + .opt_clks_cnt = ARRAY_SIZE(dss_dsi1_opt_clks), .slaves = omap3xxx_dss_dsi1_slaves, .slaves_cnt = ARRAY_SIZE(omap3xxx_dss_dsi1_slaves), .flags = HWMOD_NO_IDLEST, @@ -1546,6 +1553,10 @@ static struct omap_hwmod_ocp_if *omap3xxx_dss_rfbi_slaves[] = { &omap3xxx_l4_core__dss_rfbi, }; +static struct omap_hwmod_opt_clk dss_rfbi_opt_clks[] = { + { .role = "ick", .clk = "dss_ick" }, +}; + static struct omap_hwmod omap3xxx_dss_rfbi_hwmod = { .name = "dss_rfbi", .class = &omap2_rfbi_hwmod_class, @@ -1557,6 +1568,8 @@ static struct omap_hwmod omap3xxx_dss_rfbi_hwmod = { .module_offs = OMAP3430_DSS_MOD, }, }, + .opt_clks = dss_rfbi_opt_clks, + .opt_clks_cnt = ARRAY_SIZE(dss_rfbi_opt_clks), .slaves = omap3xxx_dss_rfbi_slaves, .slaves_cnt = ARRAY_SIZE(omap3xxx_dss_rfbi_slaves), .flags = HWMOD_NO_IDLEST, @@ -1566,7 +1579,7 @@ static struct omap_hwmod omap3xxx_dss_rfbi_hwmod = { static struct omap_hwmod_ocp_if omap3xxx_l4_core__dss_venc = { .master = &omap3xxx_l4_core_hwmod, .slave = &omap3xxx_dss_venc_hwmod, - .clk = "dss_tv_fck", + .clk = "dss_ick", .addr = omap2_dss_venc_addrs, .fw = { .omap2 = { @@ -1584,10 +1597,15 @@ static struct omap_hwmod_ocp_if *omap3xxx_dss_venc_slaves[] = { &omap3xxx_l4_core__dss_venc, }; +static struct omap_hwmod_opt_clk dss_venc_opt_clks[] = { + /* required only on OMAP3430 */ + { .role = "tv_dac_clk", .clk = "dss_96m_fck" }, +}; + static struct omap_hwmod omap3xxx_dss_venc_hwmod = { .name = "dss_venc", .class = &omap2_venc_hwmod_class, - .main_clk = "dss1_alwon_fck", + .main_clk = "dss_tv_fck", .prcm = { .omap2 = { .prcm_reg_id = 1, @@ -1595,6 +1613,8 @@ static struct omap_hwmod omap3xxx_dss_venc_hwmod = { .module_offs = OMAP3430_DSS_MOD, }, }, + .opt_clks = dss_venc_opt_clks, + .opt_clks_cnt = ARRAY_SIZE(dss_venc_opt_clks), .slaves = omap3xxx_dss_venc_slaves, .slaves_cnt = ARRAY_SIZE(omap3xxx_dss_venc_slaves), .flags = HWMOD_NO_IDLEST, -- GitLab From 4247878efca76143aa015ed1142652056052abb8 Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Tue, 8 Nov 2011 03:16:11 -0700 Subject: [PATCH 0109/3089] ARM: OMAP4: HWMOD: remove extra clocks Remove the dss_dss_clk from dss_core's opt-clocks. dss_dss_clk already defined as the dss main_clk, and thus is not needed as an opt-clock. Remove opt-clocks for dss_dispc, as dispc only uses the main_clk. Signed-off-by: Tomi Valkeinen Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/omap_hwmod_44xx_data.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c index 7695e5d433166..3b9f2d9f14710 100644 --- a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c @@ -1240,7 +1240,6 @@ static struct omap_hwmod_ocp_if *omap44xx_dss_slaves[] = { static struct omap_hwmod_opt_clk dss_opt_clks[] = { { .role = "sys_clk", .clk = "dss_sys_clk" }, { .role = "tv_clk", .clk = "dss_tv_clk" }, - { .role = "dss_clk", .clk = "dss_dss_clk" }, { .role = "video_clk", .clk = "dss_48mhz_clk" }, }; @@ -1340,12 +1339,6 @@ static struct omap_hwmod_ocp_if *omap44xx_dss_dispc_slaves[] = { &omap44xx_l4_per__dss_dispc, }; -static struct omap_hwmod_opt_clk dss_dispc_opt_clks[] = { - { .role = "sys_clk", .clk = "dss_sys_clk" }, - { .role = "tv_clk", .clk = "dss_tv_clk" }, - { .role = "hdmi_clk", .clk = "dss_48mhz_clk" }, -}; - static struct omap_hwmod omap44xx_dss_dispc_hwmod = { .name = "dss_dispc", .class = &omap44xx_dispc_hwmod_class, @@ -1359,8 +1352,6 @@ static struct omap_hwmod omap44xx_dss_dispc_hwmod = { .context_offs = OMAP4_RM_DSS_DSS_CONTEXT_OFFSET, }, }, - .opt_clks = dss_dispc_opt_clks, - .opt_clks_cnt = ARRAY_SIZE(dss_dispc_opt_clks), .slaves = omap44xx_dss_dispc_slaves, .slaves_cnt = ARRAY_SIZE(omap44xx_dss_dispc_slaves), }; -- GitLab From 37ad085527860397438f27c81a2cb13272378458 Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Tue, 8 Nov 2011 03:16:11 -0700 Subject: [PATCH 0110/3089] ARM: OMAP4: HWMOD: Add HWMOD_CONTROL_OPT_CLKS_IN_RESET for dss_core DSS needs all DSS clocks to be enabled to be able to finish reset properly. Before v3.1-rc1 the omapdss driver was managing clocks and resets correctly. However, when omapdss started using runtime PM at v3.1-rc1, the responsibility for the reset moved to HWMOD framework. HWMOD framework does not currently enable all the DSS clocks when resetting the DSS hardware. This causes the HWMOD frameworks boot-time reset to fail, possibly leaving the DSS hardware in undefined state. This patch sets HWMOD_CONTROL_OPT_CLKS_IN_RESET for dss_core. The flag is actually not used on OMAP4, because dss_core hardware does not have soft-reset functionality and thus the HWMOD framework never resets nor waits for the reset to finish. However, while the flag is not strictly needed currently, I think it represents the HW correctly: all the DSS clocks should be enabled after power-on to allow DSS hardware to finish its reset. A custom reset function will be added in the following patches which manages this correctly for OMAP4. Signed-off-by: Tomi Valkeinen Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/omap_hwmod_44xx_data.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c index 3b9f2d9f14710..b82720abf2104 100644 --- a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c @@ -1245,6 +1245,7 @@ static struct omap_hwmod_opt_clk dss_opt_clks[] = { static struct omap_hwmod omap44xx_dss_hwmod = { .name = "dss_core", + .flags = HWMOD_CONTROL_OPT_CLKS_IN_RESET, .class = &omap44xx_dss_hwmod_class, .clkdm_name = "l3_dss_clkdm", .main_clk = "dss_dss_clk", -- GitLab From 4d0698d98f501b86d3bef38c46a54a702e0469a0 Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Tue, 8 Nov 2011 03:16:12 -0700 Subject: [PATCH 0111/3089] ARM: OMAP4: HWMOD: fix DSS clock data The OMAP4 HWMOD data currently contains errors with DSS clocks: dss_hdmi and dss_venc have their main_clks wrong. The clocks should be dss_48mhz_clk and dss_tv_clk, respectively. These problems were temporarily fixed with the DSS patches 9ede365aa6f74428a1f69c21ca1cf21213167576 ("HACK: OMAP: DSS2: clk hack for OMAP2/3"), and df5d3ed23cf73ee0763a8963003bda9b69d9620f ("OMAP: DSS2: HDMI: fix hdmi clock name"), which can be reverted after this patch. Signed-off-by: Tomi Valkeinen Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/omap_hwmod_44xx_data.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c index b82720abf2104..cadf0bb2d3b72 100644 --- a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c @@ -1240,7 +1240,7 @@ static struct omap_hwmod_ocp_if *omap44xx_dss_slaves[] = { static struct omap_hwmod_opt_clk dss_opt_clks[] = { { .role = "sys_clk", .clk = "dss_sys_clk" }, { .role = "tv_clk", .clk = "dss_tv_clk" }, - { .role = "video_clk", .clk = "dss_48mhz_clk" }, + { .role = "hdmi_clk", .clk = "dss_48mhz_clk" }, }; static struct omap_hwmod omap44xx_dss_hwmod = { @@ -1616,7 +1616,7 @@ static struct omap_hwmod omap44xx_dss_hdmi_hwmod = { .clkdm_name = "l3_dss_clkdm", .mpu_irqs = omap44xx_dss_hdmi_irqs, .sdma_reqs = omap44xx_dss_hdmi_sdma_reqs, - .main_clk = "dss_dss_clk", + .main_clk = "dss_48mhz_clk", .prcm = { .omap4 = { .clkctrl_offs = OMAP4_CM_DSS_DSS_CLKCTRL_OFFSET, @@ -1777,7 +1777,7 @@ static struct omap_hwmod omap44xx_dss_venc_hwmod = { .name = "dss_venc", .class = &omap44xx_venc_hwmod_class, .clkdm_name = "l3_dss_clkdm", - .main_clk = "dss_dss_clk", + .main_clk = "dss_tv_clk", .prcm = { .omap4 = { .clkctrl_offs = OMAP4_CM_DSS_DSS_CLKCTRL_OFFSET, -- GitLab From 3ce32676bb355420ceeda57b73dd84df0ff5ad6f Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Tue, 8 Nov 2011 03:16:12 -0700 Subject: [PATCH 0112/3089] ARM: OMAP2/3: HWMOD: Add SYSS_HAS_RESET_STATUS for dss OMAP2/3 dss_core has a reset status flag in sysstatus register. Add SYSS_HAS_RESET_STATUS flag to HWMOD data so it can be used. Signed-off-by: Tomi Valkeinen Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c b/arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c index c451729d289ad..d78c1324ae597 100644 --- a/arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c @@ -43,7 +43,8 @@ static struct omap_hwmod_class_sysconfig omap2_dss_sysc = { .rev_offs = 0x0000, .sysc_offs = 0x0010, .syss_offs = 0x0014, - .sysc_flags = (SYSC_HAS_SOFTRESET | SYSC_HAS_AUTOIDLE), + .sysc_flags = (SYSC_HAS_SOFTRESET | SYSC_HAS_AUTOIDLE | + SYSS_HAS_RESET_STATUS), .sysc_fields = &omap_hwmod_sysc_type1, }; -- GitLab From 13662dc5b177d68885695ef513dd4ae0e4d2a099 Mon Sep 17 00:00:00 2001 From: Tomi Valkeinen Date: Tue, 8 Nov 2011 03:16:13 -0700 Subject: [PATCH 0113/3089] ARM: OMAP: HWMOD: Unify DSS resets for OMAPs This patch adds a custom DSS reset function used on OMAPs from OMAP2 forward. The function doesn't actually do a reset, it only waits for the reset to complete. The reason for this is that on OMAP4 there is no possibility to do a SW reset, and on OMAP2/3 doing a SW reset for dss_core resets all the other DSS modules also, thus breaking the HWMOD model where every DSS module is handled independently. This fixes the problem with DSS reset on OMAP4, caused by the fact that because there's no SW reset for dss_core on OMAP4, the HWMOD framework doesn't try to reset dss_core and thus the DSS clocks were never enabled at the same time. This causes causes the HWMOD reset to fail for dss_dispc and dss_rfbi. The common reset function will also allow us to fix another problem in the future: before doing a reset we need to disable DSS outputs, which are in some cases enabled by the bootloader, as otherwise DSS HW seems to get more or less stuck, requiring a power reset to recover. Signed-off-by: Tomi Valkeinen [paul@pwsan.com: modified to build arch/arm/mach-omap2/display.o unconditionally to avoid an error when !CONFIG_OMAP2_DSS] Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/Makefile | 5 +-- arch/arm/mach-omap2/display.c | 35 +++++++++++++++++++ .../omap_hwmod_2xxx_3xxx_ipblock_data.c | 2 ++ arch/arm/mach-omap2/omap_hwmod_44xx_data.c | 2 ++ arch/arm/plat-omap/include/plat/common.h | 3 ++ include/video/omapdss.h | 7 ---- 6 files changed, 43 insertions(+), 11 deletions(-) diff --git a/arch/arm/mach-omap2/Makefile b/arch/arm/mach-omap2/Makefile index 69ab1c069134c..b009f17dee560 100644 --- a/arch/arm/mach-omap2/Makefile +++ b/arch/arm/mach-omap2/Makefile @@ -4,7 +4,7 @@ # Common support obj-y := id.o io.o control.o mux.o devices.o serial.o gpmc.o timer.o pm.o \ - common.o gpio.o dma.o wd_timer.o + common.o gpio.o dma.o wd_timer.o display.o omap-2-3-common = irq.o sdrc.o hwmod-common = omap_hwmod.o \ @@ -264,7 +264,4 @@ smsc911x-$(CONFIG_SMSC911X) := gpmc-smsc911x.o obj-y += $(smsc911x-m) $(smsc911x-y) obj-$(CONFIG_ARCH_OMAP4) += hwspinlock.o -disp-$(CONFIG_OMAP2_DSS) := display.o -obj-y += $(disp-m) $(disp-y) - obj-y += common-board-devices.o twl-common.o diff --git a/arch/arm/mach-omap2/display.c b/arch/arm/mach-omap2/display.c index adb2756e242f1..941b5459707fc 100644 --- a/arch/arm/mach-omap2/display.c +++ b/arch/arm/mach-omap2/display.c @@ -27,6 +27,7 @@ #include #include #include +#include #include "control.h" @@ -172,3 +173,37 @@ int __init omap_display_init(struct omap_dss_board_info *board_data) return r; } + +#define MAX_MODULE_SOFTRESET_WAIT 10000 +int omap_dss_reset(struct omap_hwmod *oh) +{ + struct omap_hwmod_opt_clk *oc; + int c = 0; + int i, r; + + if (!(oh->class->sysc->sysc_flags & SYSS_HAS_RESET_STATUS)) { + pr_err("dss_core: hwmod data doesn't contain reset data\n"); + return -EINVAL; + } + + for (i = oh->opt_clks_cnt, oc = oh->opt_clks; i > 0; i--, oc++) + if (oc->_clk) + clk_enable(oc->_clk); + + omap_test_timeout((omap_hwmod_read(oh, oh->class->sysc->syss_offs) + & SYSS_RESETDONE_MASK), + MAX_MODULE_SOFTRESET_WAIT, c); + + if (c == MAX_MODULE_SOFTRESET_WAIT) + pr_warning("dss_core: waiting for reset to finish failed\n"); + else + pr_debug("dss_core: softreset done\n"); + + for (i = oh->opt_clks_cnt, oc = oh->opt_clks; i > 0; i--, oc++) + if (oc->_clk) + clk_disable(oc->_clk); + + r = (c == MAX_MODULE_SOFTRESET_WAIT) ? -ETIMEDOUT : 0; + + return r; +} diff --git a/arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c b/arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c index d78c1324ae597..c11273da5dcc3 100644 --- a/arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_2xxx_3xxx_ipblock_data.c @@ -11,6 +11,7 @@ #include #include #include +#include #include @@ -51,6 +52,7 @@ static struct omap_hwmod_class_sysconfig omap2_dss_sysc = { struct omap_hwmod_class omap2_dss_hwmod_class = { .name = "dss", .sysc = &omap2_dss_sysc, + .reset = omap_dss_reset, }; /* diff --git a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c index cadf0bb2d3b72..3b04d63316f58 100644 --- a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c @@ -30,6 +30,7 @@ #include #include #include +#include #include "omap_hwmod_common_data.h" @@ -1187,6 +1188,7 @@ static struct omap_hwmod_class_sysconfig omap44xx_dss_sysc = { static struct omap_hwmod_class omap44xx_dss_hwmod_class = { .name = "dss", .sysc = &omap44xx_dss_sysc, + .reset = omap_dss_reset, }; /* dss */ diff --git a/arch/arm/plat-omap/include/plat/common.h b/arch/arm/plat-omap/include/plat/common.h index c50df4814f6f4..3ff3e36580f26 100644 --- a/arch/arm/plat-omap/include/plat/common.h +++ b/arch/arm/plat-omap/include/plat/common.h @@ -30,6 +30,7 @@ #include #include +#include struct sys_timer; @@ -55,6 +56,8 @@ void am35xx_init_early(void); void ti816x_init_early(void); void omap4430_init_early(void); +extern int omap_dss_reset(struct omap_hwmod *); + void omap_sram_init(void); /* diff --git a/include/video/omapdss.h b/include/video/omapdss.h index b66ebb2032c6d..378c7ed6760be 100644 --- a/include/video/omapdss.h +++ b/include/video/omapdss.h @@ -307,15 +307,8 @@ struct omap_dss_board_info { void (*dsi_disable_pads)(int dsi_id, unsigned lane_mask); }; -#if defined(CONFIG_OMAP2_DSS_MODULE) || defined(CONFIG_OMAP2_DSS) /* Init with the board info */ extern int omap_display_init(struct omap_dss_board_info *board_data); -#else -static inline int omap_display_init(struct omap_dss_board_info *board_data) -{ - return 0; -} -#endif struct omap_display_platform_data { struct omap_dss_board_info *board_data; -- GitLab From b923d40dd4211c4ef7d4efa2bd81b7ca1d8744c1 Mon Sep 17 00:00:00 2001 From: Archit Taneja Date: Thu, 6 Oct 2011 18:04:08 -0600 Subject: [PATCH 0114/3089] ARM: OMAP2PLUS: DSS: Ensure DSS works correctly if display is enabled in bootloader Resetting DISPC when a DISPC output is enabled causes the DSS to go into an inconsistent state. Thus if the bootloader has enabled a display, the hwmod code cannot reset the DISPC module just like that, but the outputs need to be disabled first. Add function dispc_disable_outputs() which disables all active overlay manager and ensure all frame transfers are completed. Modify omap_dss_reset() to call this function and clear DSS_CONTROL, DSS_SDI_CONTROL and DSS_PLL_CONTROL so that DSS is in a clean state when the DSS2 driver starts. This resolves the hang issue(caused by a L3 error during boot) seen on the beagle board C3, which has a factory bootloader that enables display. The issue is resolved with this patch. Thanks to Tomi and Sricharan for some additional testing. Acked-by: Tomi Valkeinen Tested-by: R, Sricharan Signed-off-by: Archit Taneja [paul@pwsan.com: restructured code, removed omap_{read,write}l(), removed cpu_is_omap*() calls and converted to dev_attr] Signed-off-by: Paul Walmsley --- arch/arm/mach-omap2/display.c | 124 +++++++++++++++++++ arch/arm/mach-omap2/display.h | 29 +++++ arch/arm/mach-omap2/omap_hwmod_2420_data.c | 1 + arch/arm/mach-omap2/omap_hwmod_2430_data.c | 1 + arch/arm/mach-omap2/omap_hwmod_3xxx_data.c | 1 + arch/arm/mach-omap2/omap_hwmod_44xx_data.c | 6 + arch/arm/mach-omap2/omap_hwmod_common_data.c | 4 + arch/arm/mach-omap2/omap_hwmod_common_data.h | 4 + 8 files changed, 170 insertions(+) create mode 100644 arch/arm/mach-omap2/display.h diff --git a/arch/arm/mach-omap2/display.c b/arch/arm/mach-omap2/display.c index 941b5459707fc..dce9905d64bb6 100644 --- a/arch/arm/mach-omap2/display.c +++ b/arch/arm/mach-omap2/display.c @@ -30,6 +30,32 @@ #include #include "control.h" +#include "display.h" + +#define DISPC_CONTROL 0x0040 +#define DISPC_CONTROL2 0x0238 +#define DISPC_IRQSTATUS 0x0018 + +#define DSS_SYSCONFIG 0x10 +#define DSS_SYSSTATUS 0x14 +#define DSS_CONTROL 0x40 +#define DSS_SDI_CONTROL 0x44 +#define DSS_PLL_CONTROL 0x48 + +#define LCD_EN_MASK (0x1 << 0) +#define DIGIT_EN_MASK (0x1 << 1) + +#define FRAMEDONE_IRQ_SHIFT 0 +#define EVSYNC_EVEN_IRQ_SHIFT 2 +#define EVSYNC_ODD_IRQ_SHIFT 3 +#define FRAMEDONE2_IRQ_SHIFT 22 +#define FRAMEDONETV_IRQ_SHIFT 24 + +/* + * FRAMEDONE_IRQ_TIMEOUT: how long (in milliseconds) to wait during DISPC + * reset before deciding that something has gone wrong + */ +#define FRAMEDONE_IRQ_TIMEOUT 100 static struct platform_device omap_display_device = { .name = "omapdss", @@ -174,6 +200,90 @@ int __init omap_display_init(struct omap_dss_board_info *board_data) return r; } +static void dispc_disable_outputs(void) +{ + u32 v, irq_mask = 0; + bool lcd_en, digit_en, lcd2_en = false; + int i; + struct omap_dss_dispc_dev_attr *da; + struct omap_hwmod *oh; + + oh = omap_hwmod_lookup("dss_dispc"); + if (!oh) { + WARN(1, "display: could not disable outputs during reset - could not find dss_dispc hwmod\n"); + return; + } + + if (!oh->dev_attr) { + pr_err("display: could not disable outputs during reset due to missing dev_attr\n"); + return; + } + + da = (struct omap_dss_dispc_dev_attr *)oh->dev_attr; + + /* store value of LCDENABLE and DIGITENABLE bits */ + v = omap_hwmod_read(oh, DISPC_CONTROL); + lcd_en = v & LCD_EN_MASK; + digit_en = v & DIGIT_EN_MASK; + + /* store value of LCDENABLE for LCD2 */ + if (da->manager_count > 2) { + v = omap_hwmod_read(oh, DISPC_CONTROL2); + lcd2_en = v & LCD_EN_MASK; + } + + if (!(lcd_en | digit_en | lcd2_en)) + return; /* no managers currently enabled */ + + /* + * If any manager was enabled, we need to disable it before + * DSS clocks are disabled or DISPC module is reset + */ + if (lcd_en) + irq_mask |= 1 << FRAMEDONE_IRQ_SHIFT; + + if (digit_en) { + if (da->has_framedonetv_irq) { + irq_mask |= 1 << FRAMEDONETV_IRQ_SHIFT; + } else { + irq_mask |= 1 << EVSYNC_EVEN_IRQ_SHIFT | + 1 << EVSYNC_ODD_IRQ_SHIFT; + } + } + + if (lcd2_en) + irq_mask |= 1 << FRAMEDONE2_IRQ_SHIFT; + + /* + * clear any previous FRAMEDONE, FRAMEDONETV, + * EVSYNC_EVEN/ODD or FRAMEDONE2 interrupts + */ + omap_hwmod_write(irq_mask, oh, DISPC_IRQSTATUS); + + /* disable LCD and TV managers */ + v = omap_hwmod_read(oh, DISPC_CONTROL); + v &= ~(LCD_EN_MASK | DIGIT_EN_MASK); + omap_hwmod_write(v, oh, DISPC_CONTROL); + + /* disable LCD2 manager */ + if (da->manager_count > 2) { + v = omap_hwmod_read(oh, DISPC_CONTROL2); + v &= ~LCD_EN_MASK; + omap_hwmod_write(v, oh, DISPC_CONTROL2); + } + + i = 0; + while ((omap_hwmod_read(oh, DISPC_IRQSTATUS) & irq_mask) != + irq_mask) { + i++; + if (i > FRAMEDONE_IRQ_TIMEOUT) { + pr_err("didn't get FRAMEDONE1/2 or TV interrupt\n"); + break; + } + mdelay(1); + } +} + #define MAX_MODULE_SOFTRESET_WAIT 10000 int omap_dss_reset(struct omap_hwmod *oh) { @@ -190,6 +300,20 @@ int omap_dss_reset(struct omap_hwmod *oh) if (oc->_clk) clk_enable(oc->_clk); + dispc_disable_outputs(); + + /* clear SDI registers */ + if (cpu_is_omap3430()) { + omap_hwmod_write(0x0, oh, DSS_SDI_CONTROL); + omap_hwmod_write(0x0, oh, DSS_PLL_CONTROL); + } + + /* + * clear DSS_CONTROL register to switch DSS clock sources to + * PRCM clock, if any + */ + omap_hwmod_write(0x0, oh, DSS_CONTROL); + omap_test_timeout((omap_hwmod_read(oh, oh->class->sysc->syss_offs) & SYSS_RESETDONE_MASK), MAX_MODULE_SOFTRESET_WAIT, c); diff --git a/arch/arm/mach-omap2/display.h b/arch/arm/mach-omap2/display.h new file mode 100644 index 0000000000000..b871b017b3522 --- /dev/null +++ b/arch/arm/mach-omap2/display.h @@ -0,0 +1,29 @@ +/* + * display.h - OMAP2+ integration-specific DSS header + * + * Copyright (C) 2011 Texas Instruments, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2 as published by + * the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see . + */ + +#ifndef __ARCH_ARM_MACH_OMAP2_DISPLAY_H +#define __ARCH_ARM_MACH_OMAP2_DISPLAY_H + +#include + +struct omap_dss_dispc_dev_attr { + u8 manager_count; + bool has_framedonetv_irq; +}; + +#endif diff --git a/arch/arm/mach-omap2/omap_hwmod_2420_data.c b/arch/arm/mach-omap2/omap_hwmod_2420_data.c index 74037a5807afb..a5409ce3f3233 100644 --- a/arch/arm/mach-omap2/omap_hwmod_2420_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_2420_data.c @@ -943,6 +943,7 @@ static struct omap_hwmod omap2420_dss_dispc_hwmod = { .slaves = omap2420_dss_dispc_slaves, .slaves_cnt = ARRAY_SIZE(omap2420_dss_dispc_slaves), .flags = HWMOD_NO_IDLEST, + .dev_attr = &omap2_3_dss_dispc_dev_attr }; /* l4_core -> dss_rfbi */ diff --git a/arch/arm/mach-omap2/omap_hwmod_2430_data.c b/arch/arm/mach-omap2/omap_hwmod_2430_data.c index 6f0365388dfc5..c4f56cb60d7d6 100644 --- a/arch/arm/mach-omap2/omap_hwmod_2430_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_2430_data.c @@ -1004,6 +1004,7 @@ static struct omap_hwmod omap2430_dss_dispc_hwmod = { .slaves = omap2430_dss_dispc_slaves, .slaves_cnt = ARRAY_SIZE(omap2430_dss_dispc_slaves), .flags = HWMOD_NO_IDLEST, + .dev_attr = &omap2_3_dss_dispc_dev_attr }; /* l4_core -> dss_rfbi */ diff --git a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c index 4aaf0caaf3cc3..7f8915ad50990 100644 --- a/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_3xxx_data.c @@ -1462,6 +1462,7 @@ static struct omap_hwmod omap3xxx_dss_dispc_hwmod = { .slaves = omap3xxx_dss_dispc_slaves, .slaves_cnt = ARRAY_SIZE(omap3xxx_dss_dispc_slaves), .flags = HWMOD_NO_IDLEST, + .dev_attr = &omap2_3_dss_dispc_dev_attr }; /* diff --git a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c index 3b04d63316f58..daaf165af696f 100644 --- a/arch/arm/mach-omap2/omap_hwmod_44xx_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_44xx_data.c @@ -1327,6 +1327,11 @@ static struct omap_hwmod_addr_space omap44xx_dss_dispc_addrs[] = { { } }; +static struct omap_dss_dispc_dev_attr omap44xx_dss_dispc_dev_attr = { + .manager_count = 3, + .has_framedonetv_irq = 1 +}; + /* l4_per -> dss_dispc */ static struct omap_hwmod_ocp_if omap44xx_l4_per__dss_dispc = { .master = &omap44xx_l4_per_hwmod, @@ -1357,6 +1362,7 @@ static struct omap_hwmod omap44xx_dss_dispc_hwmod = { }, .slaves = omap44xx_dss_dispc_slaves, .slaves_cnt = ARRAY_SIZE(omap44xx_dss_dispc_slaves), + .dev_attr = &omap44xx_dss_dispc_dev_attr }; /* diff --git a/arch/arm/mach-omap2/omap_hwmod_common_data.c b/arch/arm/mach-omap2/omap_hwmod_common_data.c index de832ebc93a98..51e5418899fb4 100644 --- a/arch/arm/mach-omap2/omap_hwmod_common_data.c +++ b/arch/arm/mach-omap2/omap_hwmod_common_data.c @@ -49,3 +49,7 @@ struct omap_hwmod_sysc_fields omap_hwmod_sysc_type2 = { .srst_shift = SYSC_TYPE2_SOFTRESET_SHIFT, }; +struct omap_dss_dispc_dev_attr omap2_3_dss_dispc_dev_attr = { + .manager_count = 2, + .has_framedonetv_irq = 0 +}; diff --git a/arch/arm/mach-omap2/omap_hwmod_common_data.h b/arch/arm/mach-omap2/omap_hwmod_common_data.h index 39a7c37f45870..ad5d8f04c0b88 100644 --- a/arch/arm/mach-omap2/omap_hwmod_common_data.h +++ b/arch/arm/mach-omap2/omap_hwmod_common_data.h @@ -16,6 +16,8 @@ #include +#include "display.h" + /* Common address space across OMAP2xxx */ extern struct omap_hwmod_addr_space omap2xxx_uart1_addr_space[]; extern struct omap_hwmod_addr_space omap2xxx_uart2_addr_space[]; @@ -111,4 +113,6 @@ extern struct omap_hwmod_class omap2xxx_dma_hwmod_class; extern struct omap_hwmod_class omap2xxx_mailbox_hwmod_class; extern struct omap_hwmod_class omap2xxx_mcspi_class; +extern struct omap_dss_dispc_dev_attr omap2_3_dss_dispc_dev_attr; + #endif -- GitLab From 617d1017545c27a56229483cc1d689eff6c38561 Mon Sep 17 00:00:00 2001 From: Thomas Jarosch Date: Sat, 29 Oct 2011 15:45:56 +0200 Subject: [PATCH 0115/3089] MIPS: Yosemite, Emma: Fix off-by-two in arcs_cmdline buffer size check Cause is a misplaced bracket. The code strlen(buf+1) will be two bytes less than strlen(buf)+1 The +1 is in this code to reserve space for an additional space character. [ralf@linux-mips.org: Thomas' original patch fixed the issue only for Yosemite but the same bug exists also in Emma.] Signed-off-by: Thomas Jarosch Patchwork: https://patchwork.linux-mips.org/patch/2861/ Signed-off-by: Ralf Baechle --- arch/mips/emma/common/prom.c | 2 +- arch/mips/pmc-sierra/yosemite/prom.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/mips/emma/common/prom.c b/arch/mips/emma/common/prom.c index 708f08761406f..cae42259d6da9 100644 --- a/arch/mips/emma/common/prom.c +++ b/arch/mips/emma/common/prom.c @@ -50,7 +50,7 @@ void __init prom_init(void) /* arg[0] is "g", the rest is boot parameters */ for (i = 1; i < argc; i++) { - if (strlen(arcs_cmdline) + strlen(arg[i] + 1) + if (strlen(arcs_cmdline) + strlen(arg[i]) + 1 >= sizeof(arcs_cmdline)) break; strcat(arcs_cmdline, arg[i]); diff --git a/arch/mips/pmc-sierra/yosemite/prom.c b/arch/mips/pmc-sierra/yosemite/prom.c index cf4c868715aca..dcc926e06fcec 100644 --- a/arch/mips/pmc-sierra/yosemite/prom.c +++ b/arch/mips/pmc-sierra/yosemite/prom.c @@ -102,7 +102,7 @@ void __init prom_init(void) /* Get the boot parameters */ for (i = 1; i < argc; i++) { - if (strlen(arcs_cmdline) + strlen(arg[i] + 1) >= + if (strlen(arcs_cmdline) + strlen(arg[i]) + 1 >= sizeof(arcs_cmdline)) break; -- GitLab From 798580815126c899616146f286d5cc259abb5e16 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Fri, 4 Nov 2011 10:45:57 +0000 Subject: [PATCH 0116/3089] MIPS: NXP: Remove unused source files. The NXP code was moved for 2.6.36 but two files survived unused in the old location. Signed-off-by: Ralf Baechle --- arch/mips/nxp/pnx8550/common/pci.c | 134 ------------------------- arch/mips/nxp/pnx8550/common/setup.c | 143 --------------------------- 2 files changed, 277 deletions(-) delete mode 100644 arch/mips/nxp/pnx8550/common/pci.c delete mode 100644 arch/mips/nxp/pnx8550/common/setup.c diff --git a/arch/mips/nxp/pnx8550/common/pci.c b/arch/mips/nxp/pnx8550/common/pci.c deleted file mode 100644 index 98e86ddb86ccd..0000000000000 --- a/arch/mips/nxp/pnx8550/common/pci.c +++ /dev/null @@ -1,134 +0,0 @@ -/* - * - * BRIEF MODULE DESCRIPTION - * - * Author: source@mvista.com - * - * This program is free software; you can distribute it and/or modify it - * under the terms of the GNU General Public License (Version 2) as - * published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License - * for more details. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write to the Free Software Foundation, Inc., - * 59 Temple Place - Suite 330, Boston MA 02111-1307, USA. - */ -#include -#include -#include -#include - -#include -#include -#include - -static struct resource pci_io_resource = { - .start = PNX8550_PCIIO + 0x1000, /* reserve regacy I/O space */ - .end = PNX8550_PCIIO + PNX8550_PCIIO_SIZE, - .name = "pci IO space", - .flags = IORESOURCE_IO -}; - -static struct resource pci_mem_resource = { - .start = PNX8550_PCIMEM, - .end = PNX8550_PCIMEM + PNX8550_PCIMEM_SIZE - 1, - .name = "pci memory space", - .flags = IORESOURCE_MEM -}; - -extern struct pci_ops pnx8550_pci_ops; - -static struct pci_controller pnx8550_controller = { - .pci_ops = &pnx8550_pci_ops, - .io_map_base = PNX8550_PORT_BASE, - .io_resource = &pci_io_resource, - .mem_resource = &pci_mem_resource, -}; - -/* Return the total size of DRAM-memory, (RANK0 + RANK1) */ -static inline unsigned long get_system_mem_size(void) -{ - /* Read IP2031_RANK0_ADDR_LO */ - unsigned long dram_r0_lo = inl(PCI_BASE | 0x65010); - /* Read IP2031_RANK1_ADDR_HI */ - unsigned long dram_r1_hi = inl(PCI_BASE | 0x65018); - - return dram_r1_hi - dram_r0_lo + 1; -} - -static int __init pnx8550_pci_setup(void) -{ - int pci_mem_code; - int mem_size = get_system_mem_size() >> 20; - - /* Clear the Global 2 Register, PCI Inta Output Enable Registers - Bit 1:Enable DAC Powerdown - -> 0:DACs are enabled and are working normally - 1:DACs are powerdown - Bit 0:Enable of PCI inta output - -> 0 = Disable PCI inta output - 1 = Enable PCI inta output - */ - PNX8550_GLB2_ENAB_INTA_O = 0; - - /* Calc the PCI mem size code */ - if (mem_size >= 128) - pci_mem_code = SIZE_128M; - else if (mem_size >= 64) - pci_mem_code = SIZE_64M; - else if (mem_size >= 32) - pci_mem_code = SIZE_32M; - else - pci_mem_code = SIZE_16M; - - /* Set PCI_XIO registers */ - outl(pci_mem_resource.start, PCI_BASE | PCI_BASE1_LO); - outl(pci_mem_resource.end + 1, PCI_BASE | PCI_BASE1_HI); - outl(pci_io_resource.start, PCI_BASE | PCI_BASE2_LO); - outl(pci_io_resource.end, PCI_BASE | PCI_BASE2_HI); - - /* Send memory transaction via PCI_BASE2 */ - outl(0x00000001, PCI_BASE | PCI_IO); - - /* Unlock the setup register */ - outl(0xca, PCI_BASE | PCI_UNLOCKREG); - - /* - * BAR0 of PNX8550 (pci base 10) must be zero in order for ide - * to work, and in order for bus_to_baddr to work without any - * hacks. - */ - outl(0x00000000, PCI_BASE | PCI_BASE10); - - /* - *These two bars are set by default or the boot code. - * However, it's safer to set them here so we're not boot - * code dependent. - */ - outl(0x1be00000, PCI_BASE | PCI_BASE14); /* PNX MMIO */ - outl(PNX8550_NAND_BASE_ADDR, PCI_BASE | PCI_BASE18); /* XIO */ - - outl(PCI_EN_TA | - PCI_EN_PCI2MMI | - PCI_EN_XIO | - PCI_SETUP_BASE18_SIZE(SIZE_32M) | - PCI_SETUP_BASE18_EN | - PCI_SETUP_BASE14_EN | - PCI_SETUP_BASE10_PREF | - PCI_SETUP_BASE10_SIZE(pci_mem_code) | - PCI_SETUP_CFGMANAGE_EN | - PCI_SETUP_PCIARB_EN, - PCI_BASE | - PCI_SETUP); /* PCI_SETUP */ - outl(0x00000000, PCI_BASE | PCI_CTRL); /* PCI_CONTROL */ - - register_pci_controller(&pnx8550_controller); - - return 0; -} - -arch_initcall(pnx8550_pci_setup); diff --git a/arch/mips/nxp/pnx8550/common/setup.c b/arch/mips/nxp/pnx8550/common/setup.c deleted file mode 100644 index 71adac3233239..0000000000000 --- a/arch/mips/nxp/pnx8550/common/setup.c +++ /dev/null @@ -1,143 +0,0 @@ -/* - * - * 2.6 port, Embedded Alley Solutions, Inc - * - * Based on Per Hallsmark, per.hallsmark@mvista.com - * - * This program is free software; you can distribute it and/or modify it - * under the terms of the GNU General Public License (Version 2) as - * published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License - * for more details. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write to the Free Software Foundation, Inc., - * 59 Temple Place - Suite 330, Boston MA 02111-1307, USA. - */ -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include - -extern void __init board_setup(void); -extern void pnx8550_machine_restart(char *); -extern void pnx8550_machine_halt(void); -extern void pnx8550_machine_power_off(void); -extern struct resource ioport_resource; -extern struct resource iomem_resource; -extern char *prom_getcmdline(void); - -struct resource standard_io_resources[] = { - { - .start = 0x00, - .end = 0x1f, - .name = "dma1", - .flags = IORESOURCE_BUSY - }, { - .start = 0x40, - .end = 0x5f, - .name = "timer", - .flags = IORESOURCE_BUSY - }, { - .start = 0x80, - .end = 0x8f, - .name = "dma page reg", - .flags = IORESOURCE_BUSY - }, { - .start = 0xc0, - .end = 0xdf, - .name = "dma2", - .flags = IORESOURCE_BUSY - }, -}; - -#define STANDARD_IO_RESOURCES ARRAY_SIZE(standard_io_resources) - -extern struct resource pci_io_resource; -extern struct resource pci_mem_resource; - -/* Return the total size of DRAM-memory, (RANK0 + RANK1) */ -unsigned long get_system_mem_size(void) -{ - /* Read IP2031_RANK0_ADDR_LO */ - unsigned long dram_r0_lo = inl(PCI_BASE | 0x65010); - /* Read IP2031_RANK1_ADDR_HI */ - unsigned long dram_r1_hi = inl(PCI_BASE | 0x65018); - - return dram_r1_hi - dram_r0_lo + 1; -} - -int pnx8550_console_port = -1; - -void __init plat_mem_setup(void) -{ - int i; - char* argptr; - - board_setup(); /* board specific setup */ - - _machine_restart = pnx8550_machine_restart; - _machine_halt = pnx8550_machine_halt; - pm_power_off = pnx8550_machine_power_off; - - /* Clear the Global 2 Register, PCI Inta Output Enable Registers - Bit 1:Enable DAC Powerdown - -> 0:DACs are enabled and are working normally - 1:DACs are powerdown - Bit 0:Enable of PCI inta output - -> 0 = Disable PCI inta output - 1 = Enable PCI inta output - */ - PNX8550_GLB2_ENAB_INTA_O = 0; - - /* IO/MEM resources. */ - set_io_port_base(PNX8550_PORT_BASE); - ioport_resource.start = 0; - ioport_resource.end = ~0; - iomem_resource.start = 0; - iomem_resource.end = ~0; - - /* Request I/O space for devices on this board */ - for (i = 0; i < STANDARD_IO_RESOURCES; i++) - request_resource(&ioport_resource, standard_io_resources + i); - - /* Place the Mode Control bit for GPIO pin 16 in primary function */ - /* Pin 16 is used by UART1, UA1_TX */ - outl((PNX8550_GPIO_MODE_PRIMOP << PNX8550_GPIO_MC_16_BIT) | - (PNX8550_GPIO_MODE_PRIMOP << PNX8550_GPIO_MC_17_BIT), - PNX8550_GPIO_MC1); - - argptr = prom_getcmdline(); - if ((argptr = strstr(argptr, "console=ttyS")) != NULL) { - argptr += strlen("console=ttyS"); - pnx8550_console_port = *argptr == '0' ? 0 : 1; - - /* We must initialize the UART (console) before early printk */ - /* Set LCR to 8-bit and BAUD to 38400 (no 5) */ - ip3106_lcr(UART_BASE, pnx8550_console_port) = - PNX8XXX_UART_LCR_8BIT; - ip3106_baud(UART_BASE, pnx8550_console_port) = 5; - } -} -- GitLab From df55a66addf87bc21ef5678bea238f2c1b2db48a Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Fri, 4 Nov 2011 11:17:46 +0100 Subject: [PATCH 0117/3089] MIPS: BCM47xx: fix build with GENERIC_GPIO configuration Since eb9ae7f2 (gpio: fix build error in include/asm-generic/gpio.h) the generic version of gpio.h calls __gpio_{set,get}_value which we do not define. Get rid of asm-generic/gpio.h and define the missing stubs directly for BCM47xx to build. Reported-by: Ralf Baechle To: linux-mips@linux-mips.org CC: Hauke Mehrtens Signed-off-by: Florian Fainelli Patchwork: https://patchwork.linux-mips.org/patch/2885/ Signed-off-by: Ralf Baechle --- arch/mips/include/asm/mach-bcm47xx/gpio.h | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-) diff --git a/arch/mips/include/asm/mach-bcm47xx/gpio.h b/arch/mips/include/asm/mach-bcm47xx/gpio.h index 76961cabeedfe..2ef17e8df4034 100644 --- a/arch/mips/include/asm/mach-bcm47xx/gpio.h +++ b/arch/mips/include/asm/mach-bcm47xx/gpio.h @@ -36,6 +36,8 @@ static inline int gpio_get_value(unsigned gpio) return -EINVAL; } +#define gpio_get_value_cansleep gpio_get_value + static inline void gpio_set_value(unsigned gpio, int value) { switch (bcm47xx_bus_type) { @@ -54,6 +56,19 @@ static inline void gpio_set_value(unsigned gpio, int value) } } +#define gpio_set_value_cansleep gpio_set_value + +static inline int gpio_cansleep(unsigned gpio) +{ + return 0; +} + +static inline int gpio_is_valid(unsigned gpio) +{ + return gpio < (BCM47XX_EXTIF_GPIO_LINES + BCM47XX_CHIPCO_GPIO_LINES); +} + + static inline int gpio_direction_input(unsigned gpio) { switch (bcm47xx_bus_type) { @@ -137,7 +152,4 @@ static inline int gpio_polarity(unsigned gpio, int value) } -/* cansleep wrappers */ -#include - #endif /* __BCM47XX_GPIO_H */ -- GitLab From e63fb7a9dae820b3ff6754f794fd713f83e32fff Mon Sep 17 00:00:00 2001 From: Venkat Subbiah Date: Mon, 3 Oct 2011 13:31:10 -0700 Subject: [PATCH 0118/3089] MIPS: Octeon: Mark SMP-IPI interrupt as IRQF_NO_THREAD This is to exclude it from force threading to allow RT patch set to work. And while on this line * Remove IRQF_DISABLED as as this flag is NOOP * Add IRQF_PERCPU as this is a per cpu interrupt. Signed-off-by: Venkat Subbiah Acked-by: David Daney To: linux-mips@linux-mips.org To: linux-rt-users@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/2817/ Acked-by: Thomas Gleixner Signed-off-by: Ralf Baechle --- arch/mips/cavium-octeon/smp.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/mips/cavium-octeon/smp.c b/arch/mips/cavium-octeon/smp.c index 8b606423bbd7f..efcfff4d4627c 100644 --- a/arch/mips/cavium-octeon/smp.c +++ b/arch/mips/cavium-octeon/smp.c @@ -207,8 +207,9 @@ void octeon_prepare_cpus(unsigned int max_cpus) * the other bits alone. */ cvmx_write_csr(CVMX_CIU_MBOX_CLRX(cvmx_get_core_num()), 0xffff); - if (request_irq(OCTEON_IRQ_MBOX0, mailbox_interrupt, IRQF_DISABLED, - "SMP-IPI", mailbox_interrupt)) { + if (request_irq(OCTEON_IRQ_MBOX0, mailbox_interrupt, + IRQF_PERCPU | IRQF_NO_THREAD, "SMP-IPI", + mailbox_interrupt)) { panic("Cannot request_irq(OCTEON_IRQ_MBOX0)\n"); } } -- GitLab From 66f5ddf30a59f811818656cb2833c80da0340cfa Mon Sep 17 00:00:00 2001 From: "Luck, Tony" Date: Thu, 3 Nov 2011 11:46:47 -0700 Subject: [PATCH 0119/3089] x86/mce: Make mce_chrdev_ops 'static const' Arjan would like to make struct file_operations const, but mce-inject directly writes to the mce_chrdev_ops to install its write handler. In an ideal world mce-inject would have its own character device, but we have a sizable legacy of test scripts that hardwire "/dev/mcelog", so it would be painful to switch to a separate device now. Instead, this patch switches to a stub function in the mce code, with a registration helper that mce-inject can call when it is loaded. Note that this would also allow for a sane process to allow mce-inject to be unloaded again (with an unregister function, and appropriate module_{get,put}() calls), but that is left for potential future patches. Reported-by: Arjan van de Ven Signed-off-by: Tony Luck Link: http://lkml.kernel.org/r/4eb2e1971326651a3b@agluck-desktop.sc.intel.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/mce.h | 5 ++++- arch/x86/kernel/cpu/mcheck/mce-inject.c | 2 +- arch/x86/kernel/cpu/mcheck/mce.c | 25 ++++++++++++++++++++++--- 3 files changed, 27 insertions(+), 5 deletions(-) diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index c9321f34e55b3..0e8ae57d3656c 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -201,7 +201,10 @@ int mce_notify_irq(void); void mce_notify_process(void); DECLARE_PER_CPU(struct mce, injectm); -extern struct file_operations mce_chrdev_ops; + +extern void register_mce_write_callback(ssize_t (*)(struct file *filp, + const char __user *ubuf, + size_t usize, loff_t *off)); /* * Exception handler diff --git a/arch/x86/kernel/cpu/mcheck/mce-inject.c b/arch/x86/kernel/cpu/mcheck/mce-inject.c index 6199232161cff..319882ef848d3 100644 --- a/arch/x86/kernel/cpu/mcheck/mce-inject.c +++ b/arch/x86/kernel/cpu/mcheck/mce-inject.c @@ -208,7 +208,7 @@ static int inject_init(void) if (!alloc_cpumask_var(&mce_inject_cpumask, GFP_KERNEL)) return -ENOMEM; printk(KERN_INFO "Machine check injector initialized\n"); - mce_chrdev_ops.write = mce_write; + register_mce_write_callback(mce_write); register_nmi_handler(NMI_LOCAL, mce_raise_notify, 0, "mce_notify"); return 0; diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 362056aefeb47..2af127d4c3d1d 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -1634,16 +1634,35 @@ static long mce_chrdev_ioctl(struct file *f, unsigned int cmd, } } -/* Modified in mce-inject.c, so not static or const */ -struct file_operations mce_chrdev_ops = { +static ssize_t (*mce_write)(struct file *filp, const char __user *ubuf, + size_t usize, loff_t *off); + +void register_mce_write_callback(ssize_t (*fn)(struct file *filp, + const char __user *ubuf, + size_t usize, loff_t *off)) +{ + mce_write = fn; +} +EXPORT_SYMBOL_GPL(register_mce_write_callback); + +ssize_t mce_chrdev_write(struct file *filp, const char __user *ubuf, + size_t usize, loff_t *off) +{ + if (mce_write) + return mce_write(filp, ubuf, usize, off); + else + return -EINVAL; +} + +static const struct file_operations mce_chrdev_ops = { .open = mce_chrdev_open, .release = mce_chrdev_release, .read = mce_chrdev_read, + .write = mce_chrdev_write, .poll = mce_chrdev_poll, .unlocked_ioctl = mce_chrdev_ioctl, .llseek = no_llseek, }; -EXPORT_SYMBOL_GPL(mce_chrdev_ops); static struct miscdevice mce_chrdev_device = { MISC_MCELOG_MINOR, -- GitLab From f1e10354fc2a12773e5e8efcf841380aa57d4aa5 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Sat, 5 Nov 2011 14:47:19 +0800 Subject: [PATCH 0120/3089] ASoC: wm9081: Fix reading wrong register for setting VMID 2*240k VMID Divider Enable and Select is controlled by BIT[2:1] of WM9081_VMID_CONTROL register (04h). Current code reads wrong register (WM9081_BIAS_CONTROL_1) for setting VMID 2*240k. Signed-off-by: Axel Lin Signed-off-by: Mark Brown --- sound/soc/codecs/wm9081.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/codecs/wm9081.c b/sound/soc/codecs/wm9081.c index 3cd35a02c28c7..fe6561885f39e 100644 --- a/sound/soc/codecs/wm9081.c +++ b/sound/soc/codecs/wm9081.c @@ -818,7 +818,7 @@ static int wm9081_set_bias_level(struct snd_soc_codec *codec, } /* VMID 2*240k */ - reg = snd_soc_read(codec, WM9081_BIAS_CONTROL_1); + reg = snd_soc_read(codec, WM9081_VMID_CONTROL); reg &= ~WM9081_VMID_SEL_MASK; reg |= 0x04; snd_soc_write(codec, WM9081_VMID_CONTROL, reg); -- GitLab From adf463626ad8e0a2cdbe17d8bb64c1d9d0ac160d Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Sat, 5 Nov 2011 14:49:21 +0800 Subject: [PATCH 0121/3089] ASoC: wm9081: Don't write WM9081_BIAS_ENA bit to WM9081_VMID_CONTROL register WM9081_BIAS_ENA is the bit[1] of WM9081_BIAS_CONTROL_1 register (05h). Current code incorrectly write it to WM9081_VMID_CONTROL(04h) register. Signed-off-by: Axel Lin Signed-off-by: Mark Brown --- sound/soc/codecs/wm9081.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/sound/soc/codecs/wm9081.c b/sound/soc/codecs/wm9081.c index fe6561885f39e..4a398c3bfe84a 100644 --- a/sound/soc/codecs/wm9081.c +++ b/sound/soc/codecs/wm9081.c @@ -807,7 +807,6 @@ static int wm9081_set_bias_level(struct snd_soc_codec *codec, mdelay(100); /* Normal bias enable & soft start off */ - reg |= WM9081_BIAS_ENA; reg &= ~WM9081_VMID_RAMP; snd_soc_write(codec, WM9081_VMID_CONTROL, reg); @@ -830,14 +829,15 @@ static int wm9081_set_bias_level(struct snd_soc_codec *codec, break; case SND_SOC_BIAS_OFF: - /* Startup bias source */ + /* Startup bias source and disable bias */ reg = snd_soc_read(codec, WM9081_BIAS_CONTROL_1); reg |= WM9081_BIAS_SRC; + reg &= ~WM9081_BIAS_ENA; snd_soc_write(codec, WM9081_BIAS_CONTROL_1, reg); - /* Disable VMID and biases with soft ramping */ + /* Disable VMID with soft ramping */ reg = snd_soc_read(codec, WM9081_VMID_CONTROL); - reg &= ~(WM9081_VMID_SEL_MASK | WM9081_BIAS_ENA); + reg &= ~WM9081_VMID_SEL_MASK; reg |= WM9081_VMID_RAMP; snd_soc_write(codec, WM9081_VMID_CONTROL, reg); -- GitLab From 10953db8e1a278742ef7e64a3d1491802bcfa98b Mon Sep 17 00:00:00 2001 From: hayeswang Date: Mon, 7 Nov 2011 20:44:37 +0000 Subject: [PATCH 0122/3089] r8169: increase the delay parameter of pm_schedule_suspend The link down would occur when reseting PHY. And it would take about 2 ~ 5 seconds from link down to link up. If the delay of pm_schedule_suspend is not long enough, the device would enter runtime_suspend before link up. After link up, the device would wake up and reset PHY again. Then, you would find the driver keep in a loop of runtime_suspend and rumtime_resume. Signed-off-by: Hayes Wang Acked-by: Francois Romieu Signed-off-by: David S. Miller --- drivers/net/ethernet/realtek/r8169.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c index 92b45f08858fa..6f06aa10f0d72 100644 --- a/drivers/net/ethernet/realtek/r8169.c +++ b/drivers/net/ethernet/realtek/r8169.c @@ -1292,7 +1292,7 @@ static void __rtl8169_check_link_status(struct net_device *dev, netif_carrier_off(dev); netif_info(tp, ifdown, dev, "link down\n"); if (pm) - pm_schedule_suspend(&tp->pci_dev->dev, 100); + pm_schedule_suspend(&tp->pci_dev->dev, 5000); } spin_unlock_irqrestore(&tp->lock, flags); } -- GitLab From c457338d7acd3823e765b684a62294cfda9d2f55 Mon Sep 17 00:00:00 2001 From: Brian Haley Date: Tue, 8 Nov 2011 04:41:42 +0000 Subject: [PATCH 0123/3089] ipv6: drop packets when source address is multicast RFC 4291 Section 2.7 says Multicast addresses must not be used as source addresses in IPv6 packets - drop them on input so we don't process the packet further. Signed-off-by: Brian Haley Reported-and-Tested-by: Kumar Sanghvi Signed-off-by: David S. Miller --- net/ipv6/ip6_input.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/net/ipv6/ip6_input.c b/net/ipv6/ip6_input.c index 027c7ff6f1e53..a46c64eb0a660 100644 --- a/net/ipv6/ip6_input.c +++ b/net/ipv6/ip6_input.c @@ -111,6 +111,14 @@ int ipv6_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt ipv6_addr_loopback(&hdr->daddr)) goto err; + /* + * RFC4291 2.7 + * Multicast addresses must not be used as source addresses in IPv6 + * packets or appear in any Routing header. + */ + if (ipv6_addr_is_multicast(&hdr->saddr)) + goto err; + skb->transport_header = skb->network_header + sizeof(*hdr); IP6CB(skb)->nhoff = offsetof(struct ipv6hdr, nexthdr); -- GitLab From ea1649dea95fb270af94f75e842402c4cfbfc15f Mon Sep 17 00:00:00 2001 From: Nico Erfurth Date: Tue, 8 Nov 2011 07:30:40 +0000 Subject: [PATCH 0124/3089] Fix incorrect usage of NET_IP_ALIGN The driver used NET_IP_ALIGN to remove some additional padding inside of the rx_fixup function. On many architectures NET_IP_ALIGN defaults to 2 which removed the correct amount of bytes. On MCORE2-machines commit ea812ca1b06113597adcd8e70c0f84a413d97544 introduces a change which sets NET_IP_ALIGN to 0 by default. Which triggered the bug on these machines. This fix introduces a new RXW_PADDING define and uses this instead of NET_IP_ALIGN. The name was taken from the original SMSC7500 driver which is provided by SMSC. Signed-off-by: Nico Erfurth Tested-by: Phil Sutter Signed-off-by: David S. Miller --- drivers/net/usb/smsc75xx.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/net/usb/smsc75xx.c b/drivers/net/usb/smsc75xx.c index 22a7cf951e728..a5b9b12ef2681 100644 --- a/drivers/net/usb/smsc75xx.c +++ b/drivers/net/usb/smsc75xx.c @@ -51,6 +51,7 @@ #define USB_VENDOR_ID_SMSC (0x0424) #define USB_PRODUCT_ID_LAN7500 (0x7500) #define USB_PRODUCT_ID_LAN7505 (0x7505) +#define RXW_PADDING 2 #define check_warn(ret, fmt, args...) \ ({ if (ret < 0) netdev_warn(dev->net, fmt, ##args); }) @@ -1088,13 +1089,13 @@ static int smsc75xx_rx_fixup(struct usbnet *dev, struct sk_buff *skb) memcpy(&rx_cmd_b, skb->data, sizeof(rx_cmd_b)); le32_to_cpus(&rx_cmd_b); - skb_pull(skb, 4 + NET_IP_ALIGN); + skb_pull(skb, 4 + RXW_PADDING); packet = skb->data; /* get the packet length */ - size = (rx_cmd_a & RX_CMD_A_LEN) - NET_IP_ALIGN; - align_count = (4 - ((size + NET_IP_ALIGN) % 4)) % 4; + size = (rx_cmd_a & RX_CMD_A_LEN) - RXW_PADDING; + align_count = (4 - ((size + RXW_PADDING) % 4)) % 4; if (unlikely(rx_cmd_a & RX_CMD_A_RED)) { netif_dbg(dev, rx_err, dev->net, -- GitLab From 4f1a1eb530071c39fb239fd26c912a64284b1408 Mon Sep 17 00:00:00 2001 From: Al Cooper Date: Tue, 8 Nov 2011 09:59:01 -0500 Subject: [PATCH 0125/3089] MIPS: Kernel hangs occasionally during boot. The Kernel hangs occasionally during boot after "Calibrating delay loop..". This is caused by the c0_compare_int_usable() routine in cevt-r4k.c returning false which causes the system to disable the timer and hang later. The false return happens because the routine is using a series of four calls to irq_disable_hazard() as a delay while it waits for the timer changes to propagate to the cp0 cause register. On newer MIPS cores, like the 74K, the series of irq_disable_hazard() calls turn into ehb instructions and can take as little as a few clock ticks for all 4 instructions. This is not enough of a delay, so the routine thinks the timer is not working. This fix uses up to a max number of cycle counter ticks for the delay and uses back_to_back_c0_hazard() instead of irq_disable_hazard() to handle the hazard condition between cp0 writes and cp0 reads. Signed-off-by: Al Cooper Cc: linux-mips@linux-mips.org Cc: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/2911/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/cevt-r4k.c | 38 ++++++++++++++++++------------------- 1 file changed, 19 insertions(+), 19 deletions(-) diff --git a/arch/mips/kernel/cevt-r4k.c b/arch/mips/kernel/cevt-r4k.c index 98c5a9737c14d..e2d8e199be323 100644 --- a/arch/mips/kernel/cevt-r4k.c +++ b/arch/mips/kernel/cevt-r4k.c @@ -103,19 +103,10 @@ static int c0_compare_int_pending(void) /* * Compare interrupt can be routed and latched outside the core, - * so a single execution hazard barrier may not be enough to give - * it time to clear as seen in the Cause register. 4 time the - * pipeline depth seems reasonably conservative, and empirically - * works better in configurations with high CPU/bus clock ratios. + * so wait up to worst case number of cycle counter ticks for timer interrupt + * changes to propagate to the cause register. */ - -#define compare_change_hazard() \ - do { \ - irq_disable_hazard(); \ - irq_disable_hazard(); \ - irq_disable_hazard(); \ - irq_disable_hazard(); \ - } while (0) +#define COMPARE_INT_SEEN_TICKS 50 int c0_compare_int_usable(void) { @@ -126,8 +117,12 @@ int c0_compare_int_usable(void) * IP7 already pending? Try to clear it by acking the timer. */ if (c0_compare_int_pending()) { - write_c0_compare(read_c0_count()); - compare_change_hazard(); + cnt = read_c0_count(); + write_c0_compare(cnt); + back_to_back_c0_hazard(); + while (read_c0_count() < (cnt + COMPARE_INT_SEEN_TICKS)) + if (!c0_compare_int_pending()) + break; if (c0_compare_int_pending()) return 0; } @@ -136,7 +131,7 @@ int c0_compare_int_usable(void) cnt = read_c0_count(); cnt += delta; write_c0_compare(cnt); - compare_change_hazard(); + back_to_back_c0_hazard(); if ((int)(read_c0_count() - cnt) < 0) break; /* increase delta if the timer was already expired */ @@ -145,12 +140,17 @@ int c0_compare_int_usable(void) while ((int)(read_c0_count() - cnt) <= 0) ; /* Wait for expiry */ - compare_change_hazard(); + while (read_c0_count() < (cnt + COMPARE_INT_SEEN_TICKS)) + if (c0_compare_int_pending()) + break; if (!c0_compare_int_pending()) return 0; - - write_c0_compare(read_c0_count()); - compare_change_hazard(); + cnt = read_c0_count(); + write_c0_compare(cnt); + back_to_back_c0_hazard(); + while (read_c0_count() < (cnt + COMPARE_INT_SEEN_TICKS)) + if (!c0_compare_int_pending()) + break; if (c0_compare_int_pending()) return 0; -- GitLab From 71e63e748ee6f4bad482b8021386eeb74f6e47f1 Mon Sep 17 00:00:00 2001 From: Ulf Hansson Date: Fri, 4 Nov 2011 08:10:09 +0100 Subject: [PATCH 0126/3089] ARM: 7149/1: spi/pl022: Enable clock in probe Make sure we enable the clock before leaving probe. Signed-off-by: Ulf Hansson Acked-by: Linus Walleij Signed-off-by: Russell King --- drivers/spi/spi-pl022.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/spi/spi-pl022.c b/drivers/spi/spi-pl022.c index f103e470cb636..5559b22991987 100644 --- a/drivers/spi/spi-pl022.c +++ b/drivers/spi/spi-pl022.c @@ -2184,6 +2184,12 @@ pl022_probe(struct amba_device *adev, const struct amba_id *id) goto err_clk_prep; } + status = clk_enable(pl022->clk); + if (status) { + dev_err(&adev->dev, "could not enable SSP/SPI bus clock\n"); + goto err_no_clk_en; + } + /* Disable SSP */ writew((readw(SSP_CR1(pl022->virtbase)) & (~SSP_CR1_MASK_SSE)), SSP_CR1(pl022->virtbase)); @@ -2237,6 +2243,8 @@ pl022_probe(struct amba_device *adev, const struct amba_id *id) free_irq(adev->irq[0], pl022); err_no_irq: + clk_disable(pl022->clk); + err_no_clk_en: clk_unprepare(pl022->clk); err_clk_prep: clk_put(pl022->clk); -- GitLab From 8428e84d42179c2a00f5f6450866e70d802d1d05 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Mon, 7 Nov 2011 18:05:53 +0100 Subject: [PATCH 0127/3089] ARM: 7150/1: Allow kernel unaligned accesses on ARMv6+ processors Recent gcc versions generate unaligned accesses by default on ARMv6 and later processors. This patch ensures that the SCTLR.A bit is always cleared on such processors to avoid kernel traping before alignment_init() is called. Signed-off-by: Catalin Marinas Tested-by: John Linn Acked-by: Nicolas Pitre Cc: stable@vger.kernel.org Signed-off-by: Russell King --- arch/arm/kernel/head.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/kernel/head.S b/arch/arm/kernel/head.S index 566c54c2a1fef..08c82fd844a86 100644 --- a/arch/arm/kernel/head.S +++ b/arch/arm/kernel/head.S @@ -360,7 +360,7 @@ __secondary_data: * r13 = *virtual* address to jump to upon completion */ __enable_mmu: -#ifdef CONFIG_ALIGNMENT_TRAP +#if defined(CONFIG_ALIGNMENT_TRAP) && __LINUX_ARM_ARCH__ < 6 orr r0, r0, #CR_A #else bic r0, r0, #CR_A -- GitLab From 5f300acd8ae9f3d4585154370012ffc5c665330f Mon Sep 17 00:00:00 2001 From: Dirk B Date: Tue, 8 Nov 2011 18:16:39 +0100 Subject: [PATCH 0128/3089] ARM: 7152/1: distclean: Remove generated .dtb files The patch 'arm/dt: Add dtb make rule' adds support to create a .dtb file. But this is never removed afterwards. Remove the generated .dtb file if 'distclean' is called. Signed-off-by: Dirk Behme Acked-by: Rob Herring Acked-by: Grant Likely Cc: Shawn Guo Cc: Jason Liu Signed-off-by: Russell King --- arch/arm/boot/Makefile | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/boot/Makefile b/arch/arm/boot/Makefile index 176062ac7f073..5df26a9976a26 100644 --- a/arch/arm/boot/Makefile +++ b/arch/arm/boot/Makefile @@ -65,6 +65,8 @@ $(obj)/%.dtb: $(src)/dts/%.dts $(obj)/dtbs: $(addprefix $(obj)/, $(dtb-y)) +clean-files := *.dtb + quiet_cmd_uimage = UIMAGE $@ cmd_uimage = $(CONFIG_SHELL) $(MKIMAGE) -A arm -O linux -T kernel \ -C none -a $(LOADADDR) -e $(STARTADDR) \ -- GitLab From 042f36e1560cedfe524607791fa44607a3121f63 Mon Sep 17 00:00:00 2001 From: Mike Marciniszyn Date: Tue, 8 Nov 2011 13:27:31 -0500 Subject: [PATCH 0129/3089] IB/qib: Don't use schedule_work() It was mistakenly introduced by dde05cbdf8b1 ("IB/qib: Hold links until tuning data is available"). Signed-off-by: Mike Marciniszyn Signed-off-by: Roland Dreier --- drivers/infiniband/hw/qib/qib_iba7322.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/qib/qib_iba7322.c b/drivers/infiniband/hw/qib/qib_iba7322.c index 5bd2162b95dcb..8b46b608c02f3 100644 --- a/drivers/infiniband/hw/qib/qib_iba7322.c +++ b/drivers/infiniband/hw/qib/qib_iba7322.c @@ -5241,7 +5241,7 @@ static int qib_7322_ib_updown(struct qib_pportdata *ppd, int ibup, u64 ibcs) off */ if (ppd->dd->flags & QIB_HAS_QSFP) { qd->t_insert = get_jiffies_64(); - schedule_work(&qd->work); + queue_work(ib_wq, &qd->work); } spin_lock_irqsave(&ppd->sdma_lock, flags); if (__qib_sdma_running(ppd)) -- GitLab From e50e705ca71db82a78a3e13fb7abf6b2171ac9c7 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 8 Nov 2011 13:59:44 -0500 Subject: [PATCH 0130/3089] l2tp: fix l2tp_udp_recv_core() pskb_may_pull() can change skb->data, so we have to load ptr/optr at the right place. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/l2tp/l2tp_core.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index bf8d50c67931e..cf0f308abf5e7 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -756,9 +756,6 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb, goto error; } - /* Point to L2TP header */ - optr = ptr = skb->data; - /* Trace packet contents, if enabled */ if (tunnel->debug & L2TP_MSG_DATA) { length = min(32u, skb->len); @@ -769,12 +766,15 @@ static int l2tp_udp_recv_core(struct l2tp_tunnel *tunnel, struct sk_buff *skb, offset = 0; do { - printk(" %02X", ptr[offset]); + printk(" %02X", skb->data[offset]); } while (++offset < length); printk("\n"); } + /* Point to L2TP header */ + optr = ptr = skb->data; + /* Get L2TP header flags */ hdrflags = ntohs(*(__be16 *) ptr); -- GitLab From 59caa5612ce916c4902c753aa61429eb76ac2b21 Mon Sep 17 00:00:00 2001 From: oftedal Date: Mon, 7 Nov 2011 11:47:53 +0000 Subject: [PATCH 0131/3089] sunhme: Allow usage on SBI based SBus systems To prevent the SBus driver for Sun Happy Meal cards from being loaded for PCI cards utilizing the same chipset, a filter was added to the probe function in commit 0b492fce3d72d982a7981905f85484a1e1ba7fde. The filter was implemented by checking the name of the parent node in the OF tree. This patch extends this filter, so that the driver will load on SBus systems that are based upon SBI SBus Bridges. Signed-off-by: Kjetil Oftedal Signed-off-by: David S. Miller --- drivers/net/ethernet/sun/sunhme.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c index c517dac02ae18..cf14ab9db5768 100644 --- a/drivers/net/ethernet/sun/sunhme.c +++ b/drivers/net/ethernet/sun/sunhme.c @@ -2637,7 +2637,7 @@ static int __devinit happy_meal_sbus_probe_one(struct platform_device *op, int i sbus_dp = op->dev.parent->of_node; /* We can match PCI devices too, do not accept those here. */ - if (strcmp(sbus_dp->name, "sbus")) + if (strcmp(sbus_dp->name, "sbus") && strcmp(sbus_dp->name, "sbi")) return err; if (is_qfe) { -- GitLab From 6d67e9beb68bc6712b042c9723cecc87dc8e3f4c Mon Sep 17 00:00:00 2001 From: Feng King Date: Sat, 5 Nov 2011 04:23:23 +0000 Subject: [PATCH 0132/3089] tcp: Fix comments for Nagle algorithm TCP_NODELAY is weaker than TCP_CORK, when TCP_CORK was set, small segments will always pass Nagle test regardless of TCP_NODELAY option. Signed-off-by: Feng King Signed-off-by: David S. Miller --- net/ipv4/tcp_output.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 980b98f6288c3..63170e2975400 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1382,7 +1382,7 @@ static inline int tcp_minshall_check(const struct tcp_sock *tp) /* Return 0, if packet can be sent now without violation Nagle's rules: * 1. It is full sized. * 2. Or it contains FIN. (already checked by caller) - * 3. Or TCP_NODELAY was set. + * 3. Or TCP_CORK is not set, and TCP_NODELAY is set. * 4. Or TCP_CORK is not set, and all sent packets are ACKed. * With Minshall's modification: all sent small packets are ACKed. */ -- GitLab From 20db93c34095553a01a9c31136658917bf1fa5d5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 8 Nov 2011 14:21:44 -0500 Subject: [PATCH 0133/3089] net: min_pmtu default is 552 Small fix in Documentation, since min_pmtu is 512 + 20 + 20 = 552 Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index cb7f3148035db..f049a1ca186fb 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -20,7 +20,7 @@ ip_no_pmtu_disc - BOOLEAN default FALSE min_pmtu - INTEGER - default 562 - minimum discovered Path MTU + default 552 - minimum discovered Path MTU route/max_size - INTEGER Maximum number of routes allowed in the kernel. Increase -- GitLab From 2bc8ca40f951163b3bb75949479e2755c12c1b96 Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Tue, 11 Oct 2011 01:12:02 +0000 Subject: [PATCH 0134/3089] ipv4: Fix inetpeer expire time information As we update the learned pmtu informations on demand, we might report a nagative expiration time value to userspace if the pmtu informations are already expired and we have not send a packet to that inetpeer after expiration. With this patch we send a expire time of null to userspace after expiration until the next packet is send to that inetpeer. Signed-off-by: Steffen Klassert Signed-off-by: David S. Miller --- net/ipv4/route.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 155138d8ec8bb..511f4a75149c8 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2845,7 +2845,7 @@ static int rt_fill_info(struct net *net, struct rtable *rt = skb_rtable(skb); struct rtmsg *r; struct nlmsghdr *nlh; - long expires = 0; + unsigned long expires = 0; const struct inet_peer *peer = rt->peer; u32 id = 0, ts = 0, tsage = 0, error; @@ -2902,8 +2902,12 @@ static int rt_fill_info(struct net *net, tsage = get_seconds() - peer->tcp_ts_stamp; } expires = ACCESS_ONCE(peer->pmtu_expires); - if (expires) - expires -= jiffies; + if (expires) { + if (time_before(jiffies, expires)) + expires -= jiffies; + else + expires = 0; + } } if (rt_is_input_route(rt)) { -- GitLab From 8ce120f11898c921329a5f618d01dcc1e8e69cac Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 4 Nov 2011 23:19:28 +0000 Subject: [PATCH 0135/3089] net: better pcpu data alignment Tunnels can force an alignment of their percpu data to reduce number of cache lines used in fast path, or read in .ndo_get_stats() percpu_alloc() is a very fine grained allocator, so any small hole will be used anyway. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- drivers/net/veth.c | 2 +- net/ipv4/ip_gre.c | 2 +- net/ipv4/ipip.c | 2 +- net/ipv6/ip6_tunnel.c | 2 +- net/ipv6/sit.c | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/net/veth.c b/drivers/net/veth.c index ef883e97cee08..726c790ec74c7 100644 --- a/drivers/net/veth.c +++ b/drivers/net/veth.c @@ -27,8 +27,8 @@ struct veth_net_stats { u64 rx_packets; - u64 tx_packets; u64 rx_bytes; + u64 tx_packets; u64 tx_bytes; u64 rx_dropped; struct u64_stats_sync syncp; diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c index d55110e931202..38f7c07d12ab6 100644 --- a/net/ipv4/ip_gre.c +++ b/net/ipv4/ip_gre.c @@ -171,7 +171,7 @@ struct pcpu_tstats { unsigned long rx_bytes; unsigned long tx_packets; unsigned long tx_bytes; -}; +} __attribute__((aligned(4*sizeof(unsigned long)))); static struct net_device_stats *ipgre_get_stats(struct net_device *dev) { diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c index 065effd8349a8..94906908a416d 100644 --- a/net/ipv4/ipip.c +++ b/net/ipv4/ipip.c @@ -148,7 +148,7 @@ struct pcpu_tstats { unsigned long rx_bytes; unsigned long tx_packets; unsigned long tx_bytes; -}; +} __attribute__((aligned(4*sizeof(unsigned long)))); static struct net_device_stats *ipip_get_stats(struct net_device *dev) { diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index bdc15c9003d78..f36ca13975135 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -93,7 +93,7 @@ struct pcpu_tstats { unsigned long rx_bytes; unsigned long tx_packets; unsigned long tx_bytes; -}; +} __attribute__((aligned(4*sizeof(unsigned long)))); static struct net_device_stats *ip6_get_stats(struct net_device *dev) { diff --git a/net/ipv6/sit.c b/net/ipv6/sit.c index a7a18602a046e..cec09382282d3 100644 --- a/net/ipv6/sit.c +++ b/net/ipv6/sit.c @@ -91,7 +91,7 @@ struct pcpu_tstats { unsigned long rx_bytes; unsigned long tx_packets; unsigned long tx_bytes; -}; +} __attribute__((aligned(4*sizeof(unsigned long)))); static struct net_device_stats *ipip6_get_stats(struct net_device *dev) { -- GitLab From ddc4bbee6ef1ed20314be3888dd39ceefe233e79 Mon Sep 17 00:00:00 2001 From: Michio Honda Date: Fri, 17 Jun 2011 11:03:23 +0900 Subject: [PATCH 0136/3089] sctp: fasthandoff with ASCONF at mobile-node Fast retransmission after changing the last address with ASCONF negotiation Signed-off-by: Michio Honda Signed-off-by: David S. Miller --- include/net/sctp/structs.h | 1 + net/sctp/sm_make_chunk.c | 4 +++- net/sctp/transport.c | 16 ++++++++++++++++ 3 files changed, 20 insertions(+), 1 deletion(-) diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h index e90e7a9935ddc..3382615bd7104 100644 --- a/include/net/sctp/structs.h +++ b/include/net/sctp/structs.h @@ -1085,6 +1085,7 @@ void sctp_transport_burst_reset(struct sctp_transport *); unsigned long sctp_transport_timeout(struct sctp_transport *); void sctp_transport_reset(struct sctp_transport *); void sctp_transport_update_pmtu(struct sctp_transport *, u32); +void sctp_transport_immediate_rtx(struct sctp_transport *); /* This is the structure we use to queue packets as they come into diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 0121e0ab03516..a85eeeb55dd00 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -3400,8 +3400,10 @@ int sctp_process_asconf_ack(struct sctp_association *asoc, asconf_len -= length; } - if (no_err && asoc->src_out_of_asoc_ok) + if (no_err && asoc->src_out_of_asoc_ok) { asoc->src_out_of_asoc_ok = 0; + sctp_transport_immediate_rtx(asoc->peer.primary_path); + } /* Free the cached last sent asconf chunk. */ list_del_init(&asconf->transmitted_list); diff --git a/net/sctp/transport.c b/net/sctp/transport.c index 394c57ca2f542..3889330b7b04c 100644 --- a/net/sctp/transport.c +++ b/net/sctp/transport.c @@ -641,3 +641,19 @@ void sctp_transport_reset(struct sctp_transport *t) t->cacc.next_tsn_at_change = 0; t->cacc.cacc_saw_newack = 0; } + +/* Schedule retransmission on the given transport */ +void sctp_transport_immediate_rtx(struct sctp_transport *t) +{ + /* Stop pending T3_rtx_timer */ + if (timer_pending(&t->T3_rtx_timer)) { + (void)del_timer(&t->T3_rtx_timer); + sctp_transport_put(t); + } + sctp_retransmit(&t->asoc->outqueue, t, SCTP_RTXR_T3_RTX); + if (!timer_pending(&t->T3_rtx_timer)) { + if (!mod_timer(&t->T3_rtx_timer, jiffies + t->rto)) + sctp_transport_hold(t); + } + return; +} -- GitLab From 34d2d89f2d7da3b72b3157e778bbf709047ded97 Mon Sep 17 00:00:00 2001 From: Michio Honda Date: Fri, 17 Jun 2011 11:22:35 +0900 Subject: [PATCH 0137/3089] sctp: fasthandoff with ASCONF at server-node Retransmit chunks to newly confirmed destination when ASCONF and HEARTBEAT negotiation has success with a single-homed peer. Signed-off-by: Michio Honda Signed-off-by: David S. Miller --- net/sctp/sm_sideeffect.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 76388b083f283..1ff51c9d18d5d 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -666,6 +666,7 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds, struct sctp_chunk *chunk) { sctp_sender_hb_info_t *hbinfo; + int was_unconfirmed = 0; /* 8.3 Upon the receipt of the HEARTBEAT ACK, the sender of the * HEARTBEAT should clear the error counter of the destination @@ -692,9 +693,11 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds, /* Mark the destination transport address as active if it is not so * marked. */ - if ((t->state == SCTP_INACTIVE) || (t->state == SCTP_UNCONFIRMED)) + if ((t->state == SCTP_INACTIVE) || (t->state == SCTP_UNCONFIRMED)) { + was_unconfirmed = 1; sctp_assoc_control_transport(asoc, t, SCTP_TRANSPORT_UP, SCTP_HEARTBEAT_SUCCESS); + } /* The receiver of the HEARTBEAT ACK should also perform an * RTT measurement for that destination transport address @@ -712,6 +715,9 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds, /* Update the heartbeat timer. */ if (!mod_timer(&t->hb_timer, sctp_transport_timeout(t))) sctp_transport_hold(t); + + if (was_unconfirmed && asoc->peer.transport_count == 1) + sctp_transport_immediate_rtx(t); } -- GitLab From 68aad78c5023b8aa82da99b47f9d8cf40e8ca453 Mon Sep 17 00:00:00 2001 From: Rick Jones Date: Mon, 7 Nov 2011 13:29:27 +0000 Subject: [PATCH 0138/3089] sweep the floors and convert some .get_drvinfo routines to strlcpy Per the mention made by Ben Hutchings that strlcpy is now the preferred string copy routine for a .get_drvinfo routine, do a bit of floor sweeping and convert some of the as-yet unconverted ethernet drivers to it. Signed-off-by: Rick Jones Signed-off-by: David S. Miller --- drivers/net/ethernet/3com/3c589_cs.c | 7 ++++--- drivers/net/ethernet/3com/3c59x.c | 12 +++++++----- drivers/net/ethernet/3com/typhoon.c | 16 +++++++++------- drivers/net/ethernet/8390/ne2k-pci.c | 6 +++--- drivers/net/ethernet/adaptec/starfire.c | 6 +++--- .../net/ethernet/atheros/atl1e/atl1e_ethtool.c | 10 ++++++---- drivers/net/ethernet/atheros/atlx/atl2.c | 10 ++++++---- drivers/net/ethernet/broadcom/bnx2.c | 8 ++++---- .../net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c | 8 ++++---- drivers/net/ethernet/broadcom/tg3.c | 8 ++++---- drivers/net/ethernet/brocade/bna/bnad_ethtool.c | 11 ++++++----- drivers/net/ethernet/dec/tulip/de2104x.c | 6 +++--- drivers/net/ethernet/dec/tulip/dmfe.c | 7 ++++--- drivers/net/ethernet/dec/tulip/tulip_core.c | 6 +++--- drivers/net/ethernet/dec/tulip/uli526x.c | 7 ++++--- drivers/net/ethernet/dec/tulip/winbond-840.c | 6 +++--- drivers/net/ethernet/dlink/sundance.c | 6 +++--- drivers/net/ethernet/dnet.c | 6 +++--- drivers/net/ethernet/emulex/benet/be_ethtool.c | 7 ++++--- drivers/net/ethernet/fealnx.c | 6 +++--- drivers/net/ethernet/i825xx/eepro.c | 7 ++++--- drivers/net/ethernet/marvell/mv643xx_eth.c | 9 +++++---- drivers/net/ethernet/marvell/skge.c | 9 +++++---- drivers/net/ethernet/marvell/sky2.c | 9 +++++---- drivers/net/ethernet/natsemi/natsemi.c | 6 +++--- drivers/net/ethernet/natsemi/ns83820.c | 6 +++--- drivers/net/ethernet/neterion/s2io.c | 8 ++++---- drivers/net/ethernet/nvidia/forcedeth.c | 6 +++--- .../ethernet/qlogic/netxen/netxen_nic_ethtool.c | 12 ++++++++---- drivers/net/ethernet/qlogic/qla3xxx.c | 10 ++++++---- .../net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c | 13 ++++++++----- drivers/net/ethernet/qlogic/qlge/qlge_ethtool.c | 11 +++++++---- drivers/net/ethernet/realtek/8139cp.c | 6 +++--- drivers/net/ethernet/realtek/8139too.c | 6 +++--- drivers/net/ethernet/realtek/r8169.c | 10 +++++----- drivers/net/ethernet/smsc/epic100.c | 6 +++--- drivers/net/ethernet/smsc/smsc9420.c | 7 ++++--- drivers/net/ethernet/xircom/xirc2ps_cs.c | 2 +- 38 files changed, 166 insertions(+), 136 deletions(-) diff --git a/drivers/net/ethernet/3com/3c589_cs.c b/drivers/net/ethernet/3com/3c589_cs.c index 972f80ecc510a..da410f036869c 100644 --- a/drivers/net/ethernet/3com/3c589_cs.c +++ b/drivers/net/ethernet/3com/3c589_cs.c @@ -468,9 +468,10 @@ static void tc589_reset(struct net_device *dev) static void netdev_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - sprintf(info->bus_info, "PCMCIA 0x%lx", dev->base_addr); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + snprintf(info->bus_info, sizeof(info->bus_info), + "PCMCIA 0x%lx", dev->base_addr); } static const struct ethtool_ops netdev_ethtool_ops = { diff --git a/drivers/net/ethernet/3com/3c59x.c b/drivers/net/ethernet/3com/3c59x.c index b42c06baba898..8153a3e0a1a44 100644 --- a/drivers/net/ethernet/3com/3c59x.c +++ b/drivers/net/ethernet/3com/3c59x.c @@ -2929,15 +2929,17 @@ static void vortex_get_drvinfo(struct net_device *dev, { struct vortex_private *vp = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); if (VORTEX_PCI(vp)) { - strcpy(info->bus_info, pci_name(VORTEX_PCI(vp))); + strlcpy(info->bus_info, pci_name(VORTEX_PCI(vp)), + sizeof(info->bus_info)); } else { if (VORTEX_EISA(vp)) - strcpy(info->bus_info, dev_name(vp->gendev)); + strlcpy(info->bus_info, dev_name(vp->gendev), + sizeof(info->bus_info)); else - sprintf(info->bus_info, "EISA 0x%lx %d", - dev->base_addr, dev->irq); + snprintf(info->bus_info, sizeof(info->bus_info), + "EISA 0x%lx %d", dev->base_addr, dev->irq); } } diff --git a/drivers/net/ethernet/3com/typhoon.c b/drivers/net/ethernet/3com/typhoon.c index 20ea07508ac74..6d6bc754b1a8e 100644 --- a/drivers/net/ethernet/3com/typhoon.c +++ b/drivers/net/ethernet/3com/typhoon.c @@ -988,21 +988,23 @@ typhoon_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) smp_rmb(); if(tp->card_state == Sleeping) { - strcpy(info->fw_version, "Sleep image"); + strlcpy(info->fw_version, "Sleep image", + sizeof(info->fw_version)); } else { INIT_COMMAND_WITH_RESPONSE(&xp_cmd, TYPHOON_CMD_READ_VERSIONS); if(typhoon_issue_command(tp, 1, &xp_cmd, 3, xp_resp) < 0) { - strcpy(info->fw_version, "Unknown runtime"); + strlcpy(info->fw_version, "Unknown runtime", + sizeof(info->fw_version)); } else { u32 sleep_ver = le32_to_cpu(xp_resp[0].parm2); - snprintf(info->fw_version, 32, "%02x.%03x.%03x", - sleep_ver >> 24, (sleep_ver >> 12) & 0xfff, - sleep_ver & 0xfff); + snprintf(info->fw_version, sizeof(info->fw_version), + "%02x.%03x.%03x", sleep_ver >> 24, + (sleep_ver >> 12) & 0xfff, sleep_ver & 0xfff); } } - strcpy(info->driver, KBUILD_MODNAME); - strcpy(info->bus_info, pci_name(pci_dev)); + strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver)); + strlcpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info)); } static int diff --git a/drivers/net/ethernet/8390/ne2k-pci.c b/drivers/net/ethernet/8390/ne2k-pci.c index 39923425ba254..3fab04a0034a3 100644 --- a/drivers/net/ethernet/8390/ne2k-pci.c +++ b/drivers/net/ethernet/8390/ne2k-pci.c @@ -639,9 +639,9 @@ static void ne2k_pci_get_drvinfo(struct net_device *dev, struct ei_device *ei = netdev_priv(dev); struct pci_dev *pci_dev = (struct pci_dev *) ei->priv; - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(pci_dev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info)); } static const struct ethtool_ops ne2k_pci_ethtool_ops = { diff --git a/drivers/net/ethernet/adaptec/starfire.c b/drivers/net/ethernet/adaptec/starfire.c index 6d9f6911000ff..a446e251908bf 100644 --- a/drivers/net/ethernet/adaptec/starfire.c +++ b/drivers/net/ethernet/adaptec/starfire.c @@ -1842,9 +1842,9 @@ static int check_if_running(struct net_device *dev) static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct netdev_private *np = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(np->pci_dev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info)); } static int get_settings(struct net_device *dev, struct ethtool_cmd *ecmd) diff --git a/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c b/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c index 6269438d365f9..6e61f9f9ebb53 100644 --- a/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c +++ b/drivers/net/ethernet/atheros/atl1e/atl1e_ethtool.c @@ -310,10 +310,12 @@ static void atl1e_get_drvinfo(struct net_device *netdev, { struct atl1e_adapter *adapter = netdev_priv(netdev); - strncpy(drvinfo->driver, atl1e_driver_name, 32); - strncpy(drvinfo->version, atl1e_driver_version, 32); - strncpy(drvinfo->fw_version, "L1e", 32); - strncpy(drvinfo->bus_info, pci_name(adapter->pdev), 32); + strlcpy(drvinfo->driver, atl1e_driver_name, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, atl1e_driver_version, + sizeof(drvinfo->version)); + strlcpy(drvinfo->fw_version, "L1e", sizeof(drvinfo->fw_version)); + strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); drvinfo->n_stats = 0; drvinfo->testinfo_len = 0; drvinfo->regdump_len = atl1e_get_regs_len(netdev); diff --git a/drivers/net/ethernet/atheros/atlx/atl2.c b/drivers/net/ethernet/atheros/atlx/atl2.c index 1feae5928a4b0..db3f43046d321 100644 --- a/drivers/net/ethernet/atheros/atlx/atl2.c +++ b/drivers/net/ethernet/atheros/atlx/atl2.c @@ -2049,10 +2049,12 @@ static void atl2_get_drvinfo(struct net_device *netdev, { struct atl2_adapter *adapter = netdev_priv(netdev); - strncpy(drvinfo->driver, atl2_driver_name, 32); - strncpy(drvinfo->version, atl2_driver_version, 32); - strncpy(drvinfo->fw_version, "L2", 32); - strncpy(drvinfo->bus_info, pci_name(adapter->pdev), 32); + strlcpy(drvinfo->driver, atl2_driver_name, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, atl2_driver_version, + sizeof(drvinfo->version)); + strlcpy(drvinfo->fw_version, "L2", sizeof(drvinfo->fw_version)); + strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); drvinfo->n_stats = 0; drvinfo->testinfo_len = 0; drvinfo->regdump_len = atl2_get_regs_len(netdev); diff --git a/drivers/net/ethernet/broadcom/bnx2.c b/drivers/net/ethernet/broadcom/bnx2.c index 965c7235804d6..32d1f92a2479c 100644 --- a/drivers/net/ethernet/broadcom/bnx2.c +++ b/drivers/net/ethernet/broadcom/bnx2.c @@ -6873,10 +6873,10 @@ bnx2_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct bnx2 *bp = netdev_priv(dev); - strcpy(info->driver, DRV_MODULE_NAME); - strcpy(info->version, DRV_MODULE_VERSION); - strcpy(info->bus_info, pci_name(bp->pdev)); - strcpy(info->fw_version, bp->fw_version); + strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(bp->pdev), sizeof(info->bus_info)); + strlcpy(info->fw_version, bp->fw_version, sizeof(info->fw_version)); } #define BNX2_REGDUMP_LEN (32 * 1024) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c index f0ca8b27a55ee..f6402fac26fd0 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c @@ -761,8 +761,8 @@ static void bnx2x_get_drvinfo(struct net_device *dev, struct bnx2x *bp = netdev_priv(dev); u8 phy_fw_ver[PHY_FW_VER_LEN]; - strcpy(info->driver, DRV_MODULE_NAME); - strcpy(info->version, DRV_MODULE_VERSION); + strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); phy_fw_ver[0] = '\0'; if (bp->port.pmf) { @@ -773,14 +773,14 @@ static void bnx2x_get_drvinfo(struct net_device *dev, bnx2x_release_phy_lock(bp); } - strncpy(info->fw_version, bp->fw_ver, 32); + strlcpy(info->fw_version, bp->fw_ver, sizeof(info->fw_version)); snprintf(info->fw_version + strlen(bp->fw_ver), 32 - strlen(bp->fw_ver), "bc %d.%d.%d%s%s", (bp->common.bc_ver & 0xff0000) >> 16, (bp->common.bc_ver & 0xff00) >> 8, (bp->common.bc_ver & 0xff), ((phy_fw_ver[0] != '\0') ? " phy " : ""), phy_fw_ver); - strcpy(info->bus_info, pci_name(bp->pdev)); + strlcpy(info->bus_info, pci_name(bp->pdev), sizeof(info->bus_info)); info->n_stats = BNX2X_NUM_STATS; info->testinfo_len = BNX2X_NUM_TESTS; info->eedump_len = bp->common.flash_size; diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c index bf4074167d6a3..cd3623416a4e9 100644 --- a/drivers/net/ethernet/broadcom/tg3.c +++ b/drivers/net/ethernet/broadcom/tg3.c @@ -10428,10 +10428,10 @@ static void tg3_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info { struct tg3 *tp = netdev_priv(dev); - strcpy(info->driver, DRV_MODULE_NAME); - strcpy(info->version, DRV_MODULE_VERSION); - strcpy(info->fw_version, tp->fw_ver); - strcpy(info->bus_info, pci_name(tp->pdev)); + strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); + strlcpy(info->fw_version, tp->fw_ver, sizeof(info->fw_version)); + strlcpy(info->bus_info, pci_name(tp->pdev), sizeof(info->bus_info)); } static void tg3_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) diff --git a/drivers/net/ethernet/brocade/bna/bnad_ethtool.c b/drivers/net/ethernet/brocade/bna/bnad_ethtool.c index fd3dcc1e91453..38d5c66075f90 100644 --- a/drivers/net/ethernet/brocade/bna/bnad_ethtool.c +++ b/drivers/net/ethernet/brocade/bna/bnad_ethtool.c @@ -296,8 +296,8 @@ bnad_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) struct bfa_ioc_attr *ioc_attr; unsigned long flags; - strcpy(drvinfo->driver, BNAD_NAME); - strcpy(drvinfo->version, BNAD_VERSION); + strlcpy(drvinfo->driver, BNAD_NAME, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, BNAD_VERSION, sizeof(drvinfo->version)); ioc_attr = kzalloc(sizeof(*ioc_attr), GFP_KERNEL); if (ioc_attr) { @@ -305,12 +305,13 @@ bnad_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) bfa_nw_ioc_get_attr(&bnad->bna.ioceth.ioc, ioc_attr); spin_unlock_irqrestore(&bnad->bna_lock, flags); - strncpy(drvinfo->fw_version, ioc_attr->adapter_attr.fw_ver, - sizeof(drvinfo->fw_version) - 1); + strlcpy(drvinfo->fw_version, ioc_attr->adapter_attr.fw_ver, + sizeof(drvinfo->fw_version)); kfree(ioc_attr); } - strncpy(drvinfo->bus_info, pci_name(bnad->pcidev), ETHTOOL_BUSINFO_LEN); + strlcpy(drvinfo->bus_info, pci_name(bnad->pcidev), + sizeof(drvinfo->bus_info)); } static void diff --git a/drivers/net/ethernet/dec/tulip/de2104x.c b/drivers/net/ethernet/dec/tulip/de2104x.c index 1427739d9a514..1eb46a0bb488f 100644 --- a/drivers/net/ethernet/dec/tulip/de2104x.c +++ b/drivers/net/ethernet/dec/tulip/de2104x.c @@ -1598,9 +1598,9 @@ static void de_get_drvinfo (struct net_device *dev,struct ethtool_drvinfo *info) { struct de_private *de = netdev_priv(dev); - strcpy (info->driver, DRV_NAME); - strcpy (info->version, DRV_VERSION); - strcpy (info->bus_info, pci_name(de->pdev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(de->pdev), sizeof(info->bus_info)); info->eedump_len = DE_EEPROM_SIZE; } diff --git a/drivers/net/ethernet/dec/tulip/dmfe.c b/drivers/net/ethernet/dec/tulip/dmfe.c index 17b11ee1745a3..51f7542eb451a 100644 --- a/drivers/net/ethernet/dec/tulip/dmfe.c +++ b/drivers/net/ethernet/dec/tulip/dmfe.c @@ -1085,10 +1085,11 @@ static void dmfe_ethtool_get_drvinfo(struct net_device *dev, { struct dmfe_board_info *np = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); if (np->pdev) - strcpy(info->bus_info, pci_name(np->pdev)); + strlcpy(info->bus_info, pci_name(np->pdev), + sizeof(info->bus_info)); else sprintf(info->bus_info, "EISA 0x%lx %d", dev->base_addr, dev->irq); diff --git a/drivers/net/ethernet/dec/tulip/tulip_core.c b/drivers/net/ethernet/dec/tulip/tulip_core.c index 9656dd0647d98..4eb0d76145c23 100644 --- a/drivers/net/ethernet/dec/tulip/tulip_core.c +++ b/drivers/net/ethernet/dec/tulip/tulip_core.c @@ -871,9 +871,9 @@ static struct net_device_stats *tulip_get_stats(struct net_device *dev) static void tulip_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct tulip_private *np = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(np->pdev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(np->pdev), sizeof(info->bus_info)); } diff --git a/drivers/net/ethernet/dec/tulip/uli526x.c b/drivers/net/ethernet/dec/tulip/uli526x.c index 7a44a7a6adc8a..48b0b6566eef2 100644 --- a/drivers/net/ethernet/dec/tulip/uli526x.c +++ b/drivers/net/ethernet/dec/tulip/uli526x.c @@ -960,10 +960,11 @@ static void netdev_get_drvinfo(struct net_device *dev, { struct uli526x_board_info *np = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); if (np->pdev) - strcpy(info->bus_info, pci_name(np->pdev)); + strlcpy(info->bus_info, pci_name(np->pdev), + sizeof(info->bus_info)); else sprintf(info->bus_info, "EISA 0x%lx %d", dev->base_addr, dev->irq); diff --git a/drivers/net/ethernet/dec/tulip/winbond-840.c b/drivers/net/ethernet/dec/tulip/winbond-840.c index 4d01219ba22ff..52da7b2fe3b61 100644 --- a/drivers/net/ethernet/dec/tulip/winbond-840.c +++ b/drivers/net/ethernet/dec/tulip/winbond-840.c @@ -1390,9 +1390,9 @@ static void netdev_get_drvinfo (struct net_device *dev, struct ethtool_drvinfo * { struct netdev_private *np = netdev_priv(dev); - strcpy (info->driver, DRV_NAME); - strcpy (info->version, DRV_VERSION); - strcpy (info->bus_info, pci_name(np->pci_dev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info)); } static int netdev_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) diff --git a/drivers/net/ethernet/dlink/sundance.c b/drivers/net/ethernet/dlink/sundance.c index dcd7f7a71ad4a..28a3a9b50b8b2 100644 --- a/drivers/net/ethernet/dlink/sundance.c +++ b/drivers/net/ethernet/dlink/sundance.c @@ -1634,9 +1634,9 @@ static int check_if_running(struct net_device *dev) static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct netdev_private *np = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(np->pci_dev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info)); } static int get_settings(struct net_device *dev, struct ethtool_cmd *ecmd) diff --git a/drivers/net/ethernet/dnet.c b/drivers/net/ethernet/dnet.c index c1063d1540c2a..d94b9686b80cc 100644 --- a/drivers/net/ethernet/dnet.c +++ b/drivers/net/ethernet/dnet.c @@ -804,9 +804,9 @@ static int dnet_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) static void dnet_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, "0"); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, "0", sizeof(info->bus_info)); } static const struct ethtool_ops dnet_ethtool_ops = { diff --git a/drivers/net/ethernet/emulex/benet/be_ethtool.c b/drivers/net/ethernet/emulex/benet/be_ethtool.c index bf8153ea4ed81..1ad7a28fae6ad 100644 --- a/drivers/net/ethernet/emulex/benet/be_ethtool.c +++ b/drivers/net/ethernet/emulex/benet/be_ethtool.c @@ -127,8 +127,8 @@ static void be_get_drvinfo(struct net_device *netdev, memset(fw_on_flash, 0 , sizeof(fw_on_flash)); be_cmd_get_fw_ver(adapter, adapter->fw_ver, fw_on_flash); - strcpy(drvinfo->driver, DRV_NAME); - strcpy(drvinfo->version, DRV_VER); + strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, DRV_VER, sizeof(drvinfo->version)); strncpy(drvinfo->fw_version, adapter->fw_ver, FW_VER_LEN); if (memcmp(adapter->fw_ver, fw_on_flash, FW_VER_LEN) != 0) { strcat(drvinfo->fw_version, " ["); @@ -136,7 +136,8 @@ static void be_get_drvinfo(struct net_device *netdev, strcat(drvinfo->fw_version, "]"); } - strcpy(drvinfo->bus_info, pci_name(adapter->pdev)); + strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); drvinfo->testinfo_len = 0; drvinfo->regdump_len = 0; drvinfo->eedump_len = 0; diff --git a/drivers/net/ethernet/fealnx.c b/drivers/net/ethernet/fealnx.c index 61d2bddec1fa9..c82d444b582d5 100644 --- a/drivers/net/ethernet/fealnx.c +++ b/drivers/net/ethernet/fealnx.c @@ -1818,9 +1818,9 @@ static void netdev_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *i { struct netdev_private *np = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(np->pci_dev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info)); } static int netdev_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) diff --git a/drivers/net/ethernet/i825xx/eepro.c b/drivers/net/ethernet/i825xx/eepro.c index 067c46069a113..114cda7721fe5 100644 --- a/drivers/net/ethernet/i825xx/eepro.c +++ b/drivers/net/ethernet/i825xx/eepro.c @@ -1726,9 +1726,10 @@ static int eepro_ethtool_get_settings(struct net_device *dev, static void eepro_ethtool_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo) { - strcpy(drvinfo->driver, DRV_NAME); - strcpy(drvinfo->version, DRV_VERSION); - sprintf(drvinfo->bus_info, "ISA 0x%lx", dev->base_addr); + strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version)); + snprintf(drvinfo->bus_info, sizeof(drvinfo->bus_info), + "ISA 0x%lx", dev->base_addr); } static const struct ethtool_ops eepro_ethtool_ops = { diff --git a/drivers/net/ethernet/marvell/mv643xx_eth.c b/drivers/net/ethernet/marvell/mv643xx_eth.c index 194a03113802f..f6b4304ca4593 100644 --- a/drivers/net/ethernet/marvell/mv643xx_eth.c +++ b/drivers/net/ethernet/marvell/mv643xx_eth.c @@ -1502,10 +1502,11 @@ mv643xx_eth_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) static void mv643xx_eth_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo) { - strncpy(drvinfo->driver, mv643xx_eth_driver_name, 32); - strncpy(drvinfo->version, mv643xx_eth_driver_version, 32); - strncpy(drvinfo->fw_version, "N/A", 32); - strncpy(drvinfo->bus_info, "platform", 32); + strlcpy(drvinfo->driver, mv643xx_eth_driver_name, sizeof(info->driver)); + strlcpy(drvinfo->version, mv643xx_eth_driver_version, + sizeof(info->version)); + strlcpy(drvinfo->fw_version, "N/A", sizeof(info->fw_version)); + strlcpy(drvinfo->bus_info, "platform", sizeof(info->bus_info)); drvinfo->n_stats = ARRAY_SIZE(mv643xx_eth_stats); } diff --git a/drivers/net/ethernet/marvell/skge.c b/drivers/net/ethernet/marvell/skge.c index c7b60839ac995..3943f5f7bb4a4 100644 --- a/drivers/net/ethernet/marvell/skge.c +++ b/drivers/net/ethernet/marvell/skge.c @@ -394,10 +394,11 @@ static void skge_get_drvinfo(struct net_device *dev, { struct skge_port *skge = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->fw_version, "N/A"); - strcpy(info->bus_info, pci_name(skge->hw->pdev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->fw_version, "N/A", sizeof(info->fw_version)); + strlcpy(info->bus_info, pci_name(skge->hw->pdev), + sizeof(info->bus_info)); } static const struct skge_stat { diff --git a/drivers/net/ethernet/marvell/sky2.c b/drivers/net/ethernet/marvell/sky2.c index fdc6c394c683e..553d1a315b3af 100644 --- a/drivers/net/ethernet/marvell/sky2.c +++ b/drivers/net/ethernet/marvell/sky2.c @@ -3623,10 +3623,11 @@ static void sky2_get_drvinfo(struct net_device *dev, { struct sky2_port *sky2 = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->fw_version, "N/A"); - strcpy(info->bus_info, pci_name(sky2->hw->pdev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->fw_version, "N/A", sizeof(info->fw_version)); + strlcpy(info->bus_info, pci_name(sky2->hw->pdev), + sizeof(info->bus_info)); } static const struct sky2_stat { diff --git a/drivers/net/ethernet/natsemi/natsemi.c b/drivers/net/ethernet/natsemi/natsemi.c index 6ca047aab7938..ac7b16b6e7af3 100644 --- a/drivers/net/ethernet/natsemi/natsemi.c +++ b/drivers/net/ethernet/natsemi/natsemi.c @@ -2555,9 +2555,9 @@ static void set_rx_mode(struct net_device *dev) static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct netdev_private *np = netdev_priv(dev); - strncpy(info->driver, DRV_NAME, ETHTOOL_BUSINFO_LEN); - strncpy(info->version, DRV_VERSION, ETHTOOL_BUSINFO_LEN); - strncpy(info->bus_info, pci_name(np->pci_dev), ETHTOOL_BUSINFO_LEN); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info)); } static int get_regs_len(struct net_device *dev) diff --git a/drivers/net/ethernet/natsemi/ns83820.c b/drivers/net/ethernet/natsemi/ns83820.c index 2b8f64ddfb553..c24b46cbfe27a 100644 --- a/drivers/net/ethernet/natsemi/ns83820.c +++ b/drivers/net/ethernet/natsemi/ns83820.c @@ -1364,9 +1364,9 @@ static int ns83820_set_settings(struct net_device *ndev, static void ns83820_get_drvinfo(struct net_device *ndev, struct ethtool_drvinfo *info) { struct ns83820 *dev = PRIV(ndev); - strcpy(info->driver, "ns83820"); - strcpy(info->version, VERSION); - strcpy(info->bus_info, pci_name(dev->pci_dev)); + strlcpy(info->driver, "ns83820", sizeof(info->driver)); + strlcpy(info->version, VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(dev->pci_dev), sizeof(info->bus_info)); } static u32 ns83820_get_link(struct net_device *ndev) diff --git a/drivers/net/ethernet/neterion/s2io.c b/drivers/net/ethernet/neterion/s2io.c index c27fb3dda9f49..e6c90a5ac5d47 100644 --- a/drivers/net/ethernet/neterion/s2io.c +++ b/drivers/net/ethernet/neterion/s2io.c @@ -5391,10 +5391,10 @@ static void s2io_ethtool_gdrvinfo(struct net_device *dev, { struct s2io_nic *sp = netdev_priv(dev); - strncpy(info->driver, s2io_driver_name, sizeof(info->driver)); - strncpy(info->version, s2io_driver_version, sizeof(info->version)); - strncpy(info->fw_version, "", sizeof(info->fw_version)); - strncpy(info->bus_info, pci_name(sp->pdev), sizeof(info->bus_info)); + strlcpy(info->driver, s2io_driver_name, sizeof(info->driver)); + strlcpy(info->version, s2io_driver_version, sizeof(info->version)); + strlcpy(info->fw_version, "", sizeof(info->fw_version)); + strlcpy(info->bus_info, pci_name(sp->pdev), sizeof(info->bus_info)); info->regdump_len = XENA_REG_SPACE; info->eedump_len = XENA_EEPROM_SPACE; } diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c index 1dca57013cb2d..d24c45bec508b 100644 --- a/drivers/net/ethernet/nvidia/forcedeth.c +++ b/drivers/net/ethernet/nvidia/forcedeth.c @@ -3900,9 +3900,9 @@ static void nv_do_stats_poll(unsigned long data) static void nv_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct fe_priv *np = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, FORCEDETH_VERSION); - strcpy(info->bus_info, pci_name(np->pci_dev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, FORCEDETH_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info)); } static void nv_get_wol(struct net_device *dev, struct ethtool_wolinfo *wolinfo) diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c index e09ea83b8c47e..8a371985319f6 100644 --- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c +++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_ethtool.c @@ -83,14 +83,18 @@ netxen_nic_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo) u32 fw_minor = 0; u32 fw_build = 0; - strncpy(drvinfo->driver, netxen_nic_driver_name, 32); - strncpy(drvinfo->version, NETXEN_NIC_LINUX_VERSIONID, 32); + strlcpy(drvinfo->driver, netxen_nic_driver_name, + sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, NETXEN_NIC_LINUX_VERSIONID, + sizeof(drvinfo->version)); fw_major = NXRD32(adapter, NETXEN_FW_VERSION_MAJOR); fw_minor = NXRD32(adapter, NETXEN_FW_VERSION_MINOR); fw_build = NXRD32(adapter, NETXEN_FW_VERSION_SUB); - sprintf(drvinfo->fw_version, "%d.%d.%d", fw_major, fw_minor, fw_build); + snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), + "%d.%d.%d", fw_major, fw_minor, fw_build); - strncpy(drvinfo->bus_info, pci_name(adapter->pdev), 32); + strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); drvinfo->regdump_len = NETXEN_NIC_REGS_LEN; drvinfo->eedump_len = netxen_nic_get_eeprom_len(dev); } diff --git a/drivers/net/ethernet/qlogic/qla3xxx.c b/drivers/net/ethernet/qlogic/qla3xxx.c index a4bdff438a5e5..9416f297100e1 100644 --- a/drivers/net/ethernet/qlogic/qla3xxx.c +++ b/drivers/net/ethernet/qlogic/qla3xxx.c @@ -1735,10 +1735,12 @@ static void ql_get_drvinfo(struct net_device *ndev, struct ethtool_drvinfo *drvinfo) { struct ql3_adapter *qdev = netdev_priv(ndev); - strncpy(drvinfo->driver, ql3xxx_driver_name, 32); - strncpy(drvinfo->version, ql3xxx_driver_version, 32); - strncpy(drvinfo->fw_version, "N/A", 32); - strncpy(drvinfo->bus_info, pci_name(qdev->pdev), 32); + strlcpy(drvinfo->driver, ql3xxx_driver_name, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, ql3xxx_driver_version, + sizeof(drvinfo->version)); + strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version)); + strlcpy(drvinfo->bus_info, pci_name(qdev->pdev), + sizeof(drvinfo->bus_info)); drvinfo->regdump_len = 0; drvinfo->eedump_len = 0; } diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c index 8aa1c6e8667be..cc228cf3d84bc 100644 --- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c +++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_ethtool.c @@ -140,11 +140,14 @@ qlcnic_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo) fw_major = QLCRD32(adapter, QLCNIC_FW_VERSION_MAJOR); fw_minor = QLCRD32(adapter, QLCNIC_FW_VERSION_MINOR); fw_build = QLCRD32(adapter, QLCNIC_FW_VERSION_SUB); - sprintf(drvinfo->fw_version, "%d.%d.%d", fw_major, fw_minor, fw_build); - - strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), 32); - strlcpy(drvinfo->driver, qlcnic_driver_name, 32); - strlcpy(drvinfo->version, QLCNIC_LINUX_VERSIONID, 32); + snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), + "%d.%d.%d", fw_major, fw_minor, fw_build); + + strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); + strlcpy(drvinfo->driver, qlcnic_driver_name, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, QLCNIC_LINUX_VERSIONID, + sizeof(drvinfo->version)); } static int diff --git a/drivers/net/ethernet/qlogic/qlge/qlge_ethtool.c b/drivers/net/ethernet/qlogic/qlge/qlge_ethtool.c index 9b67bfea035ff..8e2c2a74f3a5b 100644 --- a/drivers/net/ethernet/qlogic/qlge/qlge_ethtool.c +++ b/drivers/net/ethernet/qlogic/qlge/qlge_ethtool.c @@ -366,13 +366,16 @@ static void ql_get_drvinfo(struct net_device *ndev, struct ethtool_drvinfo *drvinfo) { struct ql_adapter *qdev = netdev_priv(ndev); - strncpy(drvinfo->driver, qlge_driver_name, 32); - strncpy(drvinfo->version, qlge_driver_version, 32); - snprintf(drvinfo->fw_version, 32, "v%d.%d.%d", + strlcpy(drvinfo->driver, qlge_driver_name, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, qlge_driver_version, + sizeof(drvinfo->version)); + snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), + "v%d.%d.%d", (qdev->fw_rev_id & 0x00ff0000) >> 16, (qdev->fw_rev_id & 0x0000ff00) >> 8, (qdev->fw_rev_id & 0x000000ff)); - strncpy(drvinfo->bus_info, pci_name(qdev->pdev), 32); + strlcpy(drvinfo->bus_info, pci_name(qdev->pdev), + sizeof(drvinfo->bus_info)); drvinfo->n_stats = 0; drvinfo->testinfo_len = 0; if (!test_bit(QL_FRC_COREDUMP, &qdev->flags)) diff --git a/drivers/net/ethernet/realtek/8139cp.c b/drivers/net/ethernet/realtek/8139cp.c index ee5da9293ce00..6cfc5dc0f76e4 100644 --- a/drivers/net/ethernet/realtek/8139cp.c +++ b/drivers/net/ethernet/realtek/8139cp.c @@ -1319,9 +1319,9 @@ static void cp_get_drvinfo (struct net_device *dev, struct ethtool_drvinfo *info { struct cp_private *cp = netdev_priv(dev); - strcpy (info->driver, DRV_NAME); - strcpy (info->version, DRV_VERSION); - strcpy (info->bus_info, pci_name(cp->pdev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(cp->pdev), sizeof(info->bus_info)); } static void cp_get_ringparam(struct net_device *dev, diff --git a/drivers/net/ethernet/realtek/8139too.c b/drivers/net/ethernet/realtek/8139too.c index 4d6b254fc6c16..d9c72273e4288 100644 --- a/drivers/net/ethernet/realtek/8139too.c +++ b/drivers/net/ethernet/realtek/8139too.c @@ -2330,9 +2330,9 @@ static int rtl8139_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) static void rtl8139_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct rtl8139_private *tp = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(tp->pci_dev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(tp->pci_dev), sizeof(info->bus_info)); info->regdump_len = tp->regs_len; } diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c index 6f06aa10f0d72..cdf66d68d8495 100644 --- a/drivers/net/ethernet/realtek/r8169.c +++ b/drivers/net/ethernet/realtek/r8169.c @@ -1404,12 +1404,12 @@ static void rtl8169_get_drvinfo(struct net_device *dev, struct rtl8169_private *tp = netdev_priv(dev); struct rtl_fw *rtl_fw = tp->rtl_fw; - strcpy(info->driver, MODULENAME); - strcpy(info->version, RTL8169_VERSION); - strcpy(info->bus_info, pci_name(tp->pci_dev)); + strlcpy(info->driver, MODULENAME, sizeof(info->driver)); + strlcpy(info->version, RTL8169_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(tp->pci_dev), sizeof(info->bus_info)); BUILD_BUG_ON(sizeof(info->fw_version) < sizeof(rtl_fw->version)); - strcpy(info->fw_version, IS_ERR_OR_NULL(rtl_fw) ? "N/A" : - rtl_fw->version); + strlcpy(info->fw_version, IS_ERR_OR_NULL(rtl_fw) ? "N/A" : + rtl_fw->version, sizeof(info->fw_version)); } static int rtl8169_get_regs_len(struct net_device *dev) diff --git a/drivers/net/ethernet/smsc/epic100.c b/drivers/net/ethernet/smsc/epic100.c index 0a5dfb814157f..2c077ce0b6d64 100644 --- a/drivers/net/ethernet/smsc/epic100.c +++ b/drivers/net/ethernet/smsc/epic100.c @@ -1414,9 +1414,9 @@ static void netdev_get_drvinfo (struct net_device *dev, struct ethtool_drvinfo * { struct epic_private *np = netdev_priv(dev); - strcpy (info->driver, DRV_NAME); - strcpy (info->version, DRV_VERSION); - strcpy (info->bus_info, pci_name(np->pci_dev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(np->pci_dev), sizeof(info->bus_info)); } static int netdev_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) diff --git a/drivers/net/ethernet/smsc/smsc9420.c b/drivers/net/ethernet/smsc/smsc9420.c index edb24b0e337be..a9efbdfe5302d 100644 --- a/drivers/net/ethernet/smsc/smsc9420.c +++ b/drivers/net/ethernet/smsc/smsc9420.c @@ -279,9 +279,10 @@ static void smsc9420_ethtool_get_drvinfo(struct net_device *netdev, { struct smsc9420_pdata *pd = netdev_priv(netdev); - strcpy(drvinfo->driver, DRV_NAME); - strcpy(drvinfo->bus_info, pci_name(pd->pdev)); - strcpy(drvinfo->version, DRV_VERSION); + strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver)); + strlcpy(drvinfo->bus_info, pci_name(pd->pdev), + sizeof(drvinfo->bus_info)); + strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version)); } static u32 smsc9420_ethtool_get_msglevel(struct net_device *netdev) diff --git a/drivers/net/ethernet/xircom/xirc2ps_cs.c b/drivers/net/ethernet/xircom/xirc2ps_cs.c index bbe8b7dbf3f38..33979c3ac943e 100644 --- a/drivers/net/ethernet/xircom/xirc2ps_cs.c +++ b/drivers/net/ethernet/xircom/xirc2ps_cs.c @@ -1411,7 +1411,7 @@ do_open(struct net_device *dev) static void netdev_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { - strcpy(info->driver, "xirc2ps_cs"); + strlcpy(info->driver, "xirc2ps_cs", sizeof(info->driver)); sprintf(info->bus_info, "PCMCIA 0x%lx", dev->base_addr); } -- GitLab From f74024d9f05caa570dcf7582b498bbf011943491 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Maciej=20=C5=BBenczykowski?= Date: Mon, 7 Nov 2011 14:57:21 +0000 Subject: [PATCH 0139/3089] net: make ipv6 bind honour freebind MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This makes native ipv6 bind follow the precedent set by: - native ipv4 bind behaviour - dual stack ipv4-mapped ipv6 bind behaviour. This does allow an unpriviledged process to spoof its source IPv6 address, just like it currently can spoof its source IPv4 address (for example when using UDP). Signed-off-by: Maciej Å»enczykowski Signed-off-by: David S. Miller --- net/ipv6/af_inet6.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index d27c797f9f05e..1040424c788fe 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -347,7 +347,7 @@ int inet6_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) */ v4addr = LOOPBACK4_IPV6; if (!(addr_type & IPV6_ADDR_MULTICAST)) { - if (!inet->transparent && + if (!(inet->freebind || inet->transparent) && !ipv6_chk_addr(net, &addr->sin6_addr, dev, 0)) { err = -EADDRNOTAVAIL; -- GitLab From 2563fa595447bba6a73e6c58c4bbf11ac0f28931 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Maciej=20=C5=BBenczykowski?= Date: Mon, 7 Nov 2011 14:57:22 +0000 Subject: [PATCH 0140/3089] net: make ipv6 PKTINFO honour freebind MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This just makes it possible to spoof source IPv6 address on a socket without having to create and bind a new socket for every source IP we wish to spoof. Signed-off-by: Maciej Å»enczykowski Signed-off-by: David S. Miller --- net/ipv6/datagram.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index e2480691c220a..83037af4fa7b5 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -654,7 +654,7 @@ int datagram_send_ctl(struct net *net, struct sock *sk, if (addr_type != IPV6_ADDR_ANY) { int strict = __ipv6_addr_src_scope(addr_type) <= IPV6_ADDR_SCOPE_LINKLOCAL; - if (!inet_sk(sk)->transparent && + if (!(inet_sk(sk)->freebind || inet_sk(sk)->transparent) && !ipv6_chk_addr(net, &src_info->ipi6_addr, strict ? dev : NULL, 0)) err = -EINVAL; -- GitLab From 8d8bdfe8034399357df58b5f3e4da638a9e9a257 Mon Sep 17 00:00:00 2001 From: Ricardo Ribalda Date: Mon, 7 Nov 2011 23:47:45 +0000 Subject: [PATCH 0141/3089] ll_temac: Add support for phy_mii_ioctl This patch enables the ioctl support for the driver. So userspace programs like mii-tool can work. Resend in merge window Signed-off-by: Ricardo Ribalda Delgado Signed-off-by: David S. Miller --- drivers/net/ethernet/xilinx/ll_temac_main.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index caf3659e173cd..bbfcb59f69ab9 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -915,12 +915,26 @@ temac_poll_controller(struct net_device *ndev) } #endif +static int temac_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd) +{ + struct temac_local *lp = netdev_priv(ndev); + + if (!netif_running(ndev)) + return -EINVAL; + + if (!lp->phy_dev) + return -EINVAL; + + return phy_mii_ioctl(lp->phy_dev, rq, cmd); +} + static const struct net_device_ops temac_netdev_ops = { .ndo_open = temac_open, .ndo_stop = temac_stop, .ndo_start_xmit = temac_start_xmit, .ndo_set_mac_address = netdev_set_mac_address, .ndo_validate_addr = eth_validate_addr, + .ndo_do_ioctl = temac_ioctl, #ifdef CONFIG_NET_POLL_CONTROLLER .ndo_poll_controller = temac_poll_controller, #endif -- GitLab From 9ecd04bc04af7df98b3a93c571e31b6ef6a90681 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 8 Nov 2011 10:45:04 +0000 Subject: [PATCH 0142/3089] sch_choke: use skb_header_pointer() Remove the assumption that skb_get_rxhash() makes IP header and ports linear, and use skb_header_pointer() instead in choke_match_flow() This permits __skb_get_rxhash() to use skb_header_pointer() eventually. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- net/sched/sch_choke.c | 27 +++++++++++++++++---------- 1 file changed, 17 insertions(+), 10 deletions(-) diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c index 3422b25df9e46..061bcb744bbd3 100644 --- a/net/sched/sch_choke.c +++ b/net/sched/sch_choke.c @@ -152,15 +152,14 @@ static bool choke_match_flow(struct sk_buff *skb1, { int off1, off2, poff; const u32 *ports1, *ports2; + u32 _ports1, _ports2; u8 ip_proto; __u32 hash1; if (skb1->protocol != skb2->protocol) return false; - /* Use hash value as quick check - * Assumes that __skb_get_rxhash makes IP header and ports linear - */ + /* Use rxhash value as quick check */ hash1 = skb_get_rxhash(skb1); if (!hash1 || hash1 != skb_get_rxhash(skb2)) return false; @@ -172,10 +171,12 @@ static bool choke_match_flow(struct sk_buff *skb1, switch (skb1->protocol) { case __constant_htons(ETH_P_IP): { const struct iphdr *ip1, *ip2; + struct iphdr _ip1, _ip2; - ip1 = (const struct iphdr *) (skb1->data + off1); - ip2 = (const struct iphdr *) (skb2->data + off2); - + ip1 = skb_header_pointer(skb1, off1, sizeof(_ip1), &_ip1); + ip2 = skb_header_pointer(skb2, off2, sizeof(_ip2), &_ip2); + if (!ip1 || !ip2) + return false; ip_proto = ip1->protocol; if (ip_proto != ip2->protocol || ip1->saddr != ip2->saddr || ip1->daddr != ip2->daddr) @@ -190,9 +191,12 @@ static bool choke_match_flow(struct sk_buff *skb1, case __constant_htons(ETH_P_IPV6): { const struct ipv6hdr *ip1, *ip2; + struct ipv6hdr _ip1, _ip2; - ip1 = (const struct ipv6hdr *) (skb1->data + off1); - ip2 = (const struct ipv6hdr *) (skb2->data + off2); + ip1 = skb_header_pointer(skb1, off1, sizeof(_ip1), &_ip1); + ip2 = skb_header_pointer(skb2, off2, sizeof(_ip2), &_ip2); + if (!ip1 || !ip2) + return false; ip_proto = ip1->nexthdr; if (ip_proto != ip2->nexthdr || @@ -214,8 +218,11 @@ static bool choke_match_flow(struct sk_buff *skb1, off1 += poff; off2 += poff; - ports1 = (__force u32 *)(skb1->data + off1); - ports2 = (__force u32 *)(skb2->data + off2); + ports1 = skb_header_pointer(skb1, off1, sizeof(_ports1), &_ports1); + ports2 = skb_header_pointer(skb2, off2, sizeof(_ports2), &_ports2); + if (!ports1 || !ports2) + return false; + return *ports1 == *ports2; } -- GitLab From e56c57d0d3fdbbdf583d3af96bfb803b8dfa713e Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 8 Nov 2011 17:07:07 -0500 Subject: [PATCH 0143/3089] net: rename sk_clone to sk_clone_lock Make clear that sk_clone() and inet_csk_clone() return a locked socket. Add _lock() prefix and kerneldoc. Suggested-by: Linus Torvalds Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/inet_connection_sock.h | 6 +++--- include/net/sock.h | 4 ++-- net/core/sock.c | 11 +++++++++-- net/dccp/minisocks.c | 2 +- net/ipv4/inet_connection_sock.c | 17 +++++++++++++---- net/ipv4/tcp_minisocks.c | 2 +- 6 files changed, 29 insertions(+), 13 deletions(-) diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h index e6db62e756dc5..dbf9aab34c827 100644 --- a/include/net/inet_connection_sock.h +++ b/include/net/inet_connection_sock.h @@ -143,9 +143,9 @@ static inline void *inet_csk_ca(const struct sock *sk) return (void *)inet_csk(sk)->icsk_ca_priv; } -extern struct sock *inet_csk_clone(struct sock *sk, - const struct request_sock *req, - const gfp_t priority); +extern struct sock *inet_csk_clone_lock(const struct sock *sk, + const struct request_sock *req, + const gfp_t priority); enum inet_csk_ack_state_t { ICSK_ACK_SCHED = 1, diff --git a/include/net/sock.h b/include/net/sock.h index abb6e0f0c3c3e..67cd4581b6dad 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1089,8 +1089,8 @@ extern struct sock *sk_alloc(struct net *net, int family, struct proto *prot); extern void sk_free(struct sock *sk); extern void sk_release_kernel(struct sock *sk); -extern struct sock *sk_clone(const struct sock *sk, - const gfp_t priority); +extern struct sock *sk_clone_lock(const struct sock *sk, + const gfp_t priority); extern struct sk_buff *sock_wmalloc(struct sock *sk, unsigned long size, int force, diff --git a/net/core/sock.c b/net/core/sock.c index 4ed7b1d12f5ec..2de9dc2959561 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -1204,7 +1204,14 @@ void sk_release_kernel(struct sock *sk) } EXPORT_SYMBOL(sk_release_kernel); -struct sock *sk_clone(const struct sock *sk, const gfp_t priority) +/** + * sk_clone_lock - clone a socket, and lock its clone + * @sk: the socket to clone + * @priority: for allocation (%GFP_KERNEL, %GFP_ATOMIC, etc) + * + * Caller must unlock socket even in error path (bh_unlock_sock(newsk)) + */ +struct sock *sk_clone_lock(const struct sock *sk, const gfp_t priority) { struct sock *newsk; @@ -1297,7 +1304,7 @@ struct sock *sk_clone(const struct sock *sk, const gfp_t priority) out: return newsk; } -EXPORT_SYMBOL_GPL(sk_clone); +EXPORT_SYMBOL_GPL(sk_clone_lock); void sk_setup_caps(struct sock *sk, struct dst_entry *dst) { diff --git a/net/dccp/minisocks.c b/net/dccp/minisocks.c index d7041a0963af9..563b7c74e49de 100644 --- a/net/dccp/minisocks.c +++ b/net/dccp/minisocks.c @@ -100,7 +100,7 @@ struct sock *dccp_create_openreq_child(struct sock *sk, * (* Generate a new socket and switch to that socket *) * Set S := new socket for this port pair */ - struct sock *newsk = inet_csk_clone(sk, req, GFP_ATOMIC); + struct sock *newsk = inet_csk_clone_lock(sk, req, GFP_ATOMIC); if (newsk != NULL) { struct dccp_request_sock *dreq = dccp_rsk(req); diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index c14d88ad348d3..a598768c616c1 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -588,10 +588,19 @@ void inet_csk_reqsk_queue_prune(struct sock *parent, } EXPORT_SYMBOL_GPL(inet_csk_reqsk_queue_prune); -struct sock *inet_csk_clone(struct sock *sk, const struct request_sock *req, - const gfp_t priority) +/** + * inet_csk_clone_lock - clone an inet socket, and lock its clone + * @sk: the socket to clone + * @req: request_sock + * @priority: for allocation (%GFP_KERNEL, %GFP_ATOMIC, etc) + * + * Caller must unlock socket even in error path (bh_unlock_sock(newsk)) + */ +struct sock *inet_csk_clone_lock(const struct sock *sk, + const struct request_sock *req, + const gfp_t priority) { - struct sock *newsk = sk_clone(sk, priority); + struct sock *newsk = sk_clone_lock(sk, priority); if (newsk != NULL) { struct inet_connection_sock *newicsk = inet_csk(newsk); @@ -615,7 +624,7 @@ struct sock *inet_csk_clone(struct sock *sk, const struct request_sock *req, } return newsk; } -EXPORT_SYMBOL_GPL(inet_csk_clone); +EXPORT_SYMBOL_GPL(inet_csk_clone_lock); /* * At this point, there should be no process reference to this diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c index 66363b689ad65..0a7e3398c4616 100644 --- a/net/ipv4/tcp_minisocks.c +++ b/net/ipv4/tcp_minisocks.c @@ -425,7 +425,7 @@ static inline void TCP_ECN_openreq_child(struct tcp_sock *tp, */ struct sock *tcp_create_openreq_child(struct sock *sk, struct request_sock *req, struct sk_buff *skb) { - struct sock *newsk = inet_csk_clone(sk, req, GFP_ATOMIC); + struct sock *newsk = inet_csk_clone_lock(sk, req, GFP_ATOMIC); if (newsk != NULL) { const struct inet_request_sock *ireq = inet_rsk(req); -- GitLab From 8ff8584e51d4d3fbe08ede413c4a221223766323 Mon Sep 17 00:00:00 2001 From: David Daney Date: Tue, 8 Nov 2011 14:54:55 -0800 Subject: [PATCH 0144/3089] MIPS: Hook up process_vm_readv and process_vm_writev system calls. Signed-off-by: David Daney To: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2918/ Signed-off-by: Ralf Baechle --- arch/mips/include/asm/unistd.h | 18 ++++++++++++------ arch/mips/kernel/scall32-o32.S | 2 ++ arch/mips/kernel/scall64-64.S | 2 ++ arch/mips/kernel/scall64-n32.S | 2 ++ arch/mips/kernel/scall64-o32.S | 2 ++ 5 files changed, 20 insertions(+), 6 deletions(-) diff --git a/arch/mips/include/asm/unistd.h b/arch/mips/include/asm/unistd.h index ecea7871dec28..d8dad5340ea30 100644 --- a/arch/mips/include/asm/unistd.h +++ b/arch/mips/include/asm/unistd.h @@ -365,16 +365,18 @@ #define __NR_syncfs (__NR_Linux + 342) #define __NR_sendmmsg (__NR_Linux + 343) #define __NR_setns (__NR_Linux + 344) +#define __NR_process_vm_readv (__NR_Linux + 345) +#define __NR_process_vm_writev (__NR_Linux + 346) /* * Offset of the last Linux o32 flavoured syscall */ -#define __NR_Linux_syscalls 344 +#define __NR_Linux_syscalls 346 #endif /* _MIPS_SIM == _MIPS_SIM_ABI32 */ #define __NR_O32_Linux 4000 -#define __NR_O32_Linux_syscalls 344 +#define __NR_O32_Linux_syscalls 346 #if _MIPS_SIM == _MIPS_SIM_ABI64 @@ -686,16 +688,18 @@ #define __NR_syncfs (__NR_Linux + 301) #define __NR_sendmmsg (__NR_Linux + 302) #define __NR_setns (__NR_Linux + 303) +#define __NR_process_vm_readv (__NR_Linux + 304) +#define __NR_process_vm_writev (__NR_Linux + 305) /* * Offset of the last Linux 64-bit flavoured syscall */ -#define __NR_Linux_syscalls 303 +#define __NR_Linux_syscalls 305 #endif /* _MIPS_SIM == _MIPS_SIM_ABI64 */ #define __NR_64_Linux 5000 -#define __NR_64_Linux_syscalls 303 +#define __NR_64_Linux_syscalls 305 #if _MIPS_SIM == _MIPS_SIM_NABI32 @@ -1012,16 +1016,18 @@ #define __NR_syncfs (__NR_Linux + 306) #define __NR_sendmmsg (__NR_Linux + 307) #define __NR_setns (__NR_Linux + 308) +#define __NR_process_vm_readv (__NR_Linux + 309) +#define __NR_process_vm_writev (__NR_Linux + 310) /* * Offset of the last N32 flavoured syscall */ -#define __NR_Linux_syscalls 308 +#define __NR_Linux_syscalls 310 #endif /* _MIPS_SIM == _MIPS_SIM_NABI32 */ #define __NR_N32_Linux 6000 -#define __NR_N32_Linux_syscalls 308 +#define __NR_N32_Linux_syscalls 310 #ifdef __KERNEL__ diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S index 47920657968d2..a632bc144efa1 100644 --- a/arch/mips/kernel/scall32-o32.S +++ b/arch/mips/kernel/scall32-o32.S @@ -591,6 +591,8 @@ einval: li v0, -ENOSYS sys sys_syncfs 1 sys sys_sendmmsg 4 sys sys_setns 2 + sys sys_process_vm_readv 6 /* 4345 */ + sys sys_process_vm_writev 6 .endm /* We pre-compute the number of _instruction_ bytes needed to diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S index fb7334bea7316..3b5a5e9ae49c1 100644 --- a/arch/mips/kernel/scall64-64.S +++ b/arch/mips/kernel/scall64-64.S @@ -430,4 +430,6 @@ sys_call_table: PTR sys_syncfs PTR sys_sendmmsg PTR sys_setns + PTR sys_process_vm_readv + PTR sys_process_vm_writev /* 5305 */ .size sys_call_table,.-sys_call_table diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index 6de1f598346e3..6be6f7020923f 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -430,4 +430,6 @@ EXPORT(sysn32_call_table) PTR sys_syncfs PTR compat_sys_sendmmsg PTR sys_setns + PTR compat_sys_process_vm_readv + PTR compat_sys_process_vm_writev /* 6310 */ .size sysn32_call_table,.-sysn32_call_table diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index 1d813169e453e..54228553691d6 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -548,4 +548,6 @@ sys_call_table: PTR sys_syncfs PTR compat_sys_sendmmsg PTR sys_setns + PTR compat_sys_process_vm_readv /* 4345 */ + PTR compat_sys_process_vm_writev .size sys_call_table,.-sys_call_table -- GitLab From 76ee4557137ce12a60a1e1a070ed341182163a64 Mon Sep 17 00:00:00 2001 From: Kyungmin Park Date: Tue, 8 Nov 2011 19:57:59 +0900 Subject: [PATCH 0145/3089] ARM: EXYNOS: Fix compiler error with THIS_MODULE CC arch/arm/mach-exynos/cpuidle.o arch/arm/mach-exynos/cpuidle.c:36: error: 'THIS_MODULE' undeclared here (not in a function) arch/arm/mach-exynos/cpuidle.c: In function 'exynos4_enter_idle': arch/arm/mach-exynos/cpuidle.c:42: error: storage size of 'before' isn't known arch/arm/mach-exynos/cpuidle.c:42: error: storage size of 'after' isn't known arch/arm/mach-exynos/cpuidle.c:46: error: implicit declaration of function 'do_gettimeofday' arch/arm/mach-exynos/cpuidle.c:52: error: 'USEC_PER_SEC' undeclared (first use in this function) arch/arm/mach-exynos/cpuidle.c:52: error: (Each undeclared identifier is reported only once arch/arm/mach-exynos/cpuidle.c:52: error: for each function it appears in.) arch/arm/mach-exynos/cpuidle.c:42: warning: unused variable 'after' arch/arm/mach-exynos/cpuidle.c:42: warning: unused variable 'before' make[1]: *** [arch/arm/mach-exynos/cpuidle.o] Error 1 Signed-off-by: Kyungmin Park [kgene.kim@samsung.com: Fixed as per Stephen's suggestion] Signed-off-by: Kukjin Kim --- arch/arm/mach-exynos/cpuidle.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/mach-exynos/cpuidle.c b/arch/arm/mach-exynos/cpuidle.c index 35f6502144ae1..4ebb382c59791 100644 --- a/arch/arm/mach-exynos/cpuidle.c +++ b/arch/arm/mach-exynos/cpuidle.c @@ -12,6 +12,8 @@ #include #include #include +#include +#include #include -- GitLab From 8d804d4fdf90b3af62a41a20bf8b21b75529a003 Mon Sep 17 00:00:00 2001 From: David Daney Date: Mon, 7 Nov 2011 12:49:30 -0800 Subject: [PATCH 0146/3089] STAGING: octeon-ethernet: Fix compile error caused by skb_frag_struct change Evidently the definition of struct skb_frag_struct has changed, so we need to change to match it. Signed-off-by: David Daney To: netdev@vger.kernel.org To: gregkh@suse.de To: devel@driverdev.osuosl.org Patchwork: https://patchwork.linux-mips.org/patch/2909/ Signed-off-by: Ralf Baechle --- drivers/staging/octeon/ethernet-tx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/staging/octeon/ethernet-tx.c b/drivers/staging/octeon/ethernet-tx.c index b445cd63f901d..2542c37439049 100644 --- a/drivers/staging/octeon/ethernet-tx.c +++ b/drivers/staging/octeon/ethernet-tx.c @@ -275,7 +275,7 @@ int cvm_oct_xmit(struct sk_buff *skb, struct net_device *dev) CVM_OCT_SKB_CB(skb)[0] = hw_buffer.u64; for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) { struct skb_frag_struct *fs = skb_shinfo(skb)->frags + i; - hw_buffer.s.addr = XKPHYS_TO_PHYS((u64)(page_address(fs->page) + fs->page_offset)); + hw_buffer.s.addr = XKPHYS_TO_PHYS((u64)(page_address(fs->page.p) + fs->page_offset)); hw_buffer.s.size = fs->size; CVM_OCT_SKB_CB(skb)[i + 1] = hw_buffer.u64; } -- GitLab From 4e6c82b3614a18740ef63109d58743a359266daf Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Mon, 7 Nov 2011 08:51:24 -0600 Subject: [PATCH 0147/3089] [SCSI] fix WARNING: at drivers/scsi/scsi_lib.c:1704 On Mon, 2011-11-07 at 17:24 +1100, Stephen Rothwell wrote: > Hi all, > > Starting some time last week I am getting the following during boot on > our PPC970 blade: > > calling .ipr_init+0x0/0x68 @ 1 > ipr: IBM Power RAID SCSI Device Driver version: 2.5.2 (April 27, 2011) > ipr 0000:01:01.0: Found IOA with IRQ: 26 > ipr 0000:01:01.0: Starting IOA initialization sequence. > ipr 0000:01:01.0: Adapter firmware version: 06160039 > ipr 0000:01:01.0: IOA initialized. > scsi0 : IBM 572E Storage Adapter > ------------[ cut here ]------------ > WARNING: at drivers/scsi/scsi_lib.c:1704 > Modules linked in: > NIP: c00000000053b3d4 LR: c00000000053e5b0 CTR: c000000000541d70 > REGS: c0000000783c2f60 TRAP: 0700 Not tainted (3.1.0-autokern1) > MSR: 8000000000029032 CR: 24002024 XER: 20000002 > TASK = c0000000783b8000[1] 'swapper' THREAD: c0000000783c0000 CPU: 0 > GPR00: 0000000000000001 c0000000783c31e0 c000000000cf38b0 c00000000239a9d0 > GPR04: c000000000cbe8f8 0000000000000000 c0000000783c3040 0000000000000000 > GPR08: c000000075daf488 c000000078a3b7ff c000000000bcacc8 0000000000000000 > GPR12: 0000000044002028 c000000007ffb000 0000000002e40000 000000000099b800 > GPR16: 0000000000000000 c000000000bba5fc c000000000a61db8 0000000000000000 > GPR20: 0000000001b77200 0000000000000000 c000000078990000 0000000000000001 > GPR24: c000000002396828 0000000000000000 0000000000000000 c000000078a3b938 > GPR28: fffffffffffffffa c0000000008ad2c0 c000000000c7faa8 c00000000239a9d0 > NIP [c00000000053b3d4] .scsi_free_queue+0x24/0x90 > LR [c00000000053e5b0] .scsi_alloc_sdev+0x280/0x2e0 > Call Trace: > [c0000000783c31e0] [c000000000c7faa8] wireless_seq_fops+0x278d0/0x2eb88 (unreliable) > [c0000000783c3270] [c00000000053e5b0] .scsi_alloc_sdev+0x280/0x2e0 > [c0000000783c3330] [c00000000053eba0] .scsi_probe_and_add_lun+0x390/0xb40 > [c0000000783c34a0] [c00000000053f7ec] .__scsi_scan_target+0x16c/0x650 > [c0000000783c35f0] [c00000000053fd90] .scsi_scan_channel+0xc0/0x100 > [c0000000783c36a0] [c00000000053fefc] .scsi_scan_host_selected+0x12c/0x1c0 > [c0000000783c3750] [c00000000083dcb4] .ipr_probe+0x2c0/0x390 > [c0000000783c3830] [c0000000003f50b4] .local_pci_probe+0x34/0x50 > [c0000000783c38a0] [c0000000003f5f78] .pci_device_probe+0x148/0x150 > [c0000000783c3950] [c0000000004e1e8c] .driver_probe_device+0xdc/0x210 > [c0000000783c39f0] [c0000000004e20cc] .__driver_attach+0x10c/0x110 > [c0000000783c3a80] [c0000000004e1228] .bus_for_each_dev+0x98/0xf0 > [c0000000783c3b30] [c0000000004e1bf8] .driver_attach+0x28/0x40 > [c0000000783c3bb0] [c0000000004e07d8] .bus_add_driver+0x218/0x340 > [c0000000783c3c60] [c0000000004e2a2c] .driver_register+0x9c/0x1b0 > [c0000000783c3d00] [c0000000003f62d4] .__pci_register_driver+0x64/0x140 > [c0000000783c3da0] [c000000000b99f88] .ipr_init+0x4c/0x68 > [c0000000783c3e20] [c00000000000ad24] .do_one_initcall+0x1a4/0x1e0 > [c0000000783c3ee0] [c000000000b512d0] .kernel_init+0x14c/0x1fc > [c0000000783c3f90] [c000000000022468] .kernel_thread+0x54/0x70 > Instruction dump: > ebe1fff8 7c0803a6 4e800020 7c0802a6 fba1ffe8 fbe1fff8 7c7f1b78 f8010010 > f821ff71 e8030398 3120ffff 7c090110 <0b000000> e86303b0 482de065 60000000 > ---[ end trace 759bed76a85e8dec ]--- > scsi 0:0:1:0: Direct-Access IBM-ESXS MAY2036RC T106 PQ: 0 ANSI: 5 > ------------[ cut here ]------------ > > I get lots more of these. The obvious commit to point the finger at > is 3308511c93e6 ("[SCSI] Make scsi_free_queue() kill pending SCSI > commands") but the root cause may be something different. Caused by commit f7c9c6bb14f3104608a3a83cadea10a6943d2804 Author: Anton Blanchard Date: Thu Nov 3 08:56:22 2011 +1100 [SCSI] Fix block queue and elevator memory leak in scsi_alloc_sdev Doesn't completely do the teardown. The true fix is to do a proper teardown instead of hand rolling it Reported-by: Stephen Rothwell Tested-by: Stephen Rothwell Cc: stable@kernel.org #2.6.38+ Signed-off-by: James Bottomley --- drivers/scsi/scsi_scan.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index 72273a0e56662..b3c6d957fbd8a 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -319,11 +319,7 @@ static struct scsi_device *scsi_alloc_sdev(struct scsi_target *starget, return sdev; out_device_destroy: - scsi_device_set_state(sdev, SDEV_DEL); - transport_destroy_device(&sdev->sdev_gendev); - put_device(&sdev->sdev_dev); - scsi_free_queue(sdev->request_queue); - put_device(&sdev->sdev_gendev); + __scsi_remove_device(sdev); out: if (display_failure_msg) printk(ALLOC_FAILURE_MSG, __func__); -- GitLab From 745718132c3c7cac98a622b610e239dcd5217f71 Mon Sep 17 00:00:00 2001 From: Hannes Reinecke Date: Wed, 9 Nov 2011 08:39:24 +0100 Subject: [PATCH 0148/3089] [SCSI] Silencing 'killing requests for dead queue' When we tear down a device we try to flush all outstanding commands in scsi_free_queue(). However the check in scsi_request_fn() is imperfect as it only signals that we _might start_ aborting commands, not that we've actually aborted some. So move the printk inside the scsi_kill_request function, this will also give us a hint about which commands are aborted. Signed-off-by: Hannes Reinecke Signed-off-by: James Bottomley --- drivers/scsi/scsi_lib.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 06bc26554a673..f85cfa6c47b52 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -1409,6 +1409,8 @@ static void scsi_kill_request(struct request *req, struct request_queue *q) blk_start_request(req); + scmd_printk(KERN_INFO, cmd, "killing request\n"); + sdev = cmd->device; starget = scsi_target(sdev); shost = sdev->host; @@ -1490,7 +1492,6 @@ static void scsi_request_fn(struct request_queue *q) struct request *req; if (!sdev) { - printk("scsi: killing requests for dead queue\n"); while ((req = blk_peek_request(q)) != NULL) scsi_kill_request(req, q); return; -- GitLab From 50ec1538fac0e39078d45ca5f8a5186621830058 Mon Sep 17 00:00:00 2001 From: Ricardo Ribalda Date: Mon, 7 Nov 2011 23:31:58 +0000 Subject: [PATCH 0149/3089] net/temac: FIX segfault when process old irqs Do not enable the irq until the scatter gather registers are ready to handle the data. Otherwise an irq from a packet send/received before last close can lead to an access to an invalid memory region on the irq handler. Also, stop the dma engine on close. Signed-off-by: Ricardo Ribalda Delgado Signed-off-by: David S. Miller --- drivers/net/ethernet/xilinx/ll_temac_main.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index caf3659e173cd..05a1ffad20d2f 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -203,6 +203,9 @@ static void temac_dma_bd_release(struct net_device *ndev) struct temac_local *lp = netdev_priv(ndev); int i; + /* Reset Local Link (DMA) */ + lp->dma_out(lp, DMA_CONTROL_REG, DMA_CONTROL_RST); + for (i = 0; i < RX_BD_NUM; i++) { if (!lp->rx_skb[i]) break; @@ -860,6 +863,8 @@ static int temac_open(struct net_device *ndev) phy_start(lp->phy_dev); } + temac_device_reset(ndev); + rc = request_irq(lp->tx_irq, ll_temac_tx_irq, 0, ndev->name, ndev); if (rc) goto err_tx_irq; @@ -867,7 +872,6 @@ static int temac_open(struct net_device *ndev) if (rc) goto err_rx_irq; - temac_device_reset(ndev); return 0; err_rx_irq: -- GitLab From f79d7e6f6ae397caf219cef1392ca39b3ff10833 Mon Sep 17 00:00:00 2001 From: Ricardo Ribalda Date: Mon, 7 Nov 2011 23:39:57 +0000 Subject: [PATCH 0150/3089] net/ll_temac: FIX : Wait for indirect wait to end While tracing down a connectivity problem on the temac I connected a probe to the Cross bar irq, and it was triggered when doing ifdown->ifup. This is fixed once waiting for the indirect write to end. Since it is not on the hot path there is no performance loss. Signed-off-by: Ricardo Ribalda Delgado Signed-off-by: David S. Miller --- drivers/net/ethernet/xilinx/ll_temac_main.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/xilinx/ll_temac_main.c b/drivers/net/ethernet/xilinx/ll_temac_main.c index 05a1ffad20d2f..2681b53820eef 100644 --- a/drivers/net/ethernet/xilinx/ll_temac_main.c +++ b/drivers/net/ethernet/xilinx/ll_temac_main.c @@ -114,6 +114,7 @@ void temac_indirect_out32(struct temac_local *lp, int reg, u32 value) return; temac_iow(lp, XTE_LSW0_OFFSET, value); temac_iow(lp, XTE_CTL0_OFFSET, CNTLREG_WRITE_ENABLE_MASK | reg); + temac_indirect_busywait(lp); } /** -- GitLab From fecc73519439361eab21fb3cebec504672ef0e03 Mon Sep 17 00:00:00 2001 From: Paul Bolle Date: Tue, 8 Nov 2011 10:31:10 +0000 Subject: [PATCH 0151/3089] net: drivers/net/hippi/Kconfig should be sourced Commit ff5a3b509e ("hippi: Move the HIPPI driver") moved the HIPPI driver into drivers/net/hippi. It didn't source drivers/net/hippi/Kconfig though, so it didn't make all necessary Kconfig changes. So let drivers/net/kconfig source HIPPI's Kconfig file. [ Fix syntax error at the end of HIPP's Kconfig file. -DaveM ] Signed-off-by: Paul Bolle Acked-by: Jeff Kirsher Signed-off-by: David S. Miller --- drivers/net/Kconfig | 2 ++ drivers/net/hippi/Kconfig | 2 +- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 583f66cd5bbd6..654a5e94e0e7c 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -245,6 +245,8 @@ source "drivers/net/ethernet/Kconfig" source "drivers/net/fddi/Kconfig" +source "drivers/net/hippi/Kconfig" + config NET_SB1000 tristate "General Instruments Surfboard 1000" depends on PNP diff --git a/drivers/net/hippi/Kconfig b/drivers/net/hippi/Kconfig index 7393eb732ee67..95eb34fdbba7b 100644 --- a/drivers/net/hippi/Kconfig +++ b/drivers/net/hippi/Kconfig @@ -36,4 +36,4 @@ config ROADRUNNER_LARGE_RINGS kernel code or by user space programs. Say Y here only if you have the memory. -endif /* HIPPI */ +endif # HIPPI -- GitLab From 069294e813ed5f27f82613b027609bcda5f1b914 Mon Sep 17 00:00:00 2001 From: Nick Bowler Date: Tue, 8 Nov 2011 12:12:44 +0000 Subject: [PATCH 0152/3089] ah: Correctly pass error codes in ahash output callback. The AH4/6 ahash output callbacks pass nexthdr to xfrm_output_resume instead of the error code. This appears to be a copy+paste error from the input case, where nexthdr is expected. This causes the driver to continuously add AH headers to the datagram until either an allocation fails and the packet is dropped or the ahash driver hits a synchronous fallback and the resulting monstrosity is transmitted. Correct this issue by simply passing the error code unadulterated. Signed-off-by: Nick Bowler Signed-off-by: David S. Miller --- net/ipv4/ah4.c | 2 -- net/ipv6/ah6.c | 2 -- 2 files changed, 4 deletions(-) diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c index c1f4154552fc5..33ca18603e3bd 100644 --- a/net/ipv4/ah4.c +++ b/net/ipv4/ah4.c @@ -136,8 +136,6 @@ static void ah_output_done(struct crypto_async_request *base, int err) memcpy(top_iph+1, iph+1, top_iph->ihl*4 - sizeof(struct iphdr)); } - err = ah->nexthdr; - kfree(AH_SKB_CB(skb)->tmp); xfrm_output_resume(skb, err); } diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c index 2195ae651923e..ede4d9d6cc2b6 100644 --- a/net/ipv6/ah6.c +++ b/net/ipv6/ah6.c @@ -324,8 +324,6 @@ static void ah6_output_done(struct crypto_async_request *base, int err) #endif } - err = ah->nexthdr; - kfree(AH_SKB_CB(skb)->tmp); xfrm_output_resume(skb, err); } -- GitLab From b7ea81a58adc123a4e980cb0eff9eb5c144b5dc7 Mon Sep 17 00:00:00 2001 From: Nick Bowler Date: Tue, 8 Nov 2011 12:12:45 +0000 Subject: [PATCH 0153/3089] ah: Read nexthdr value before overwriting it in ahash input callback. The AH4/6 ahash input callbacks read out the nexthdr field from the AH header *after* they overwrite that header. This is obviously not going to end well. Fix it up. Signed-off-by: Nick Bowler Signed-off-by: David S. Miller --- net/ipv4/ah4.c | 4 ++-- net/ipv6/ah6.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c index 33ca18603e3bd..c7056b2e831f6 100644 --- a/net/ipv4/ah4.c +++ b/net/ipv4/ah4.c @@ -262,12 +262,12 @@ static void ah_input_done(struct crypto_async_request *base, int err) if (err) goto out; + err = ah->nexthdr; + skb->network_header += ah_hlen; memcpy(skb_network_header(skb), work_iph, ihl); __skb_pull(skb, ah_hlen + ihl); skb_set_transport_header(skb, -ihl); - - err = ah->nexthdr; out: kfree(AH_SKB_CB(skb)->tmp); xfrm_input_resume(skb, err); diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c index ede4d9d6cc2b6..7a33aaa002278 100644 --- a/net/ipv6/ah6.c +++ b/net/ipv6/ah6.c @@ -464,12 +464,12 @@ static void ah6_input_done(struct crypto_async_request *base, int err) if (err) goto out; + err = ah->nexthdr; + skb->network_header += ah_hlen; memcpy(skb_network_header(skb), work_iph, hdr_len); __skb_pull(skb, ah_hlen + hdr_len); skb_set_transport_header(skb, -hdr_len); - - err = ah->nexthdr; out: kfree(AH_SKB_CB(skb)->tmp); xfrm_input_resume(skb, err); -- GitLab From b12f62efb8ec0b9523bdb6c2d412c07193086de9 Mon Sep 17 00:00:00 2001 From: Li Wei Date: Tue, 8 Nov 2011 21:39:28 +0000 Subject: [PATCH 0154/3089] ipv4: fix for ip_options_rcv_srr() daddr update. When opt->srr_is_hit is set skb_rtable(skb) has been updated for 'nexthop' and iph->daddr should always equals to skb_rtable->rt_dst holds, We need update iph->daddr either. Signed-off-by: Li Wei Signed-off-by: David S. Miller --- net/ipv4/ip_options.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index ec93335901ddc..05d20cca9d66e 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -640,6 +640,7 @@ int ip_options_rcv_srr(struct sk_buff *skb) } if (srrptr <= srrspace) { opt->srr_is_hit = 1; + iph->daddr = nexthop; opt->is_changed = 1; } return 0; -- GitLab From acb32ba3dee66d58704caeeb8c6ff95f60efdc66 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 8 Nov 2011 13:04:43 +0000 Subject: [PATCH 0155/3089] ipv4: reduce percpu needs for icmpmsg mibs Reading /proc/net/snmp on a machine with a lot of cpus is very expensive (can be ~88000 us). This is because ICMPMSG MIB uses 4096 bytes per cpu, and folding values for all possible cpus can read 16 Mbytes of memory. ICMP messages are not considered as fast path on a typical server, and eventually few cpus handle them anyway. We can afford an atomic operation instead of using percpu data. This saves 4096 bytes per cpu and per network namespace. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/icmp.h | 4 ++-- include/net/netns/mib.h | 2 +- include/net/snmp.h | 2 +- net/ipv4/af_inet.c | 8 ++++---- net/ipv4/proc.c | 9 ++++----- 5 files changed, 12 insertions(+), 13 deletions(-) diff --git a/include/net/icmp.h b/include/net/icmp.h index f0698b955b73c..75d615649071e 100644 --- a/include/net/icmp.h +++ b/include/net/icmp.h @@ -31,8 +31,8 @@ struct icmp_err { extern const struct icmp_err icmp_err_convert[]; #define ICMP_INC_STATS(net, field) SNMP_INC_STATS((net)->mib.icmp_statistics, field) #define ICMP_INC_STATS_BH(net, field) SNMP_INC_STATS_BH((net)->mib.icmp_statistics, field) -#define ICMPMSGOUT_INC_STATS(net, field) SNMP_INC_STATS((net)->mib.icmpmsg_statistics, field+256) -#define ICMPMSGIN_INC_STATS_BH(net, field) SNMP_INC_STATS_BH((net)->mib.icmpmsg_statistics, field) +#define ICMPMSGOUT_INC_STATS(net, field) SNMP_INC_STATS_ATOMIC_LONG((net)->mib.icmpmsg_statistics, field+256) +#define ICMPMSGIN_INC_STATS_BH(net, field) SNMP_INC_STATS_ATOMIC_LONG((net)->mib.icmpmsg_statistics, field) struct dst_entry; struct net_proto_family; diff --git a/include/net/netns/mib.h b/include/net/netns/mib.h index 0b44112e2366e..f360135cb69ff 100644 --- a/include/net/netns/mib.h +++ b/include/net/netns/mib.h @@ -10,7 +10,7 @@ struct netns_mib { DEFINE_SNMP_STAT(struct udp_mib, udp_statistics); DEFINE_SNMP_STAT(struct udp_mib, udplite_statistics); DEFINE_SNMP_STAT(struct icmp_mib, icmp_statistics); - DEFINE_SNMP_STAT(struct icmpmsg_mib, icmpmsg_statistics); + DEFINE_SNMP_STAT_ATOMIC(struct icmpmsg_mib, icmpmsg_statistics); #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) struct proc_dir_entry *proc_net_devsnmp6; diff --git a/include/net/snmp.h b/include/net/snmp.h index 8f0f9ac0307ff..0feafa68da01e 100644 --- a/include/net/snmp.h +++ b/include/net/snmp.h @@ -67,7 +67,7 @@ struct icmp_mib { #define ICMPMSG_MIB_MAX __ICMPMSG_MIB_MAX struct icmpmsg_mib { - unsigned long mibs[ICMPMSG_MIB_MAX]; + atomic_long_t mibs[ICMPMSG_MIB_MAX]; }; /* ICMP6 (IPv6-ICMP) */ diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 1b5096a9875aa..b2bbcd0ebd19c 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1572,9 +1572,9 @@ static __net_init int ipv4_mib_init_net(struct net *net) sizeof(struct icmp_mib), __alignof__(struct icmp_mib)) < 0) goto err_icmp_mib; - if (snmp_mib_init((void __percpu **)net->mib.icmpmsg_statistics, - sizeof(struct icmpmsg_mib), - __alignof__(struct icmpmsg_mib)) < 0) + net->mib.icmpmsg_statistics = kzalloc(sizeof(struct icmpmsg_mib), + GFP_KERNEL); + if (!net->mib.icmpmsg_statistics) goto err_icmpmsg_mib; tcp_mib_init(net); @@ -1598,7 +1598,7 @@ err_tcp_mib: static __net_exit void ipv4_mib_exit_net(struct net *net) { - snmp_mib_free((void __percpu **)net->mib.icmpmsg_statistics); + kfree(net->mib.icmpmsg_statistics); snmp_mib_free((void __percpu **)net->mib.icmp_statistics); snmp_mib_free((void __percpu **)net->mib.udplite_statistics); snmp_mib_free((void __percpu **)net->mib.udp_statistics); diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c index 466ea8bb7a4d9..961eed4f510a2 100644 --- a/net/ipv4/proc.c +++ b/net/ipv4/proc.c @@ -288,7 +288,7 @@ static void icmpmsg_put(struct seq_file *seq) count = 0; for (i = 0; i < ICMPMSG_MIB_MAX; i++) { - val = snmp_fold_field((void __percpu **) net->mib.icmpmsg_statistics, i); + val = atomic_long_read(&net->mib.icmpmsg_statistics->mibs[i]); if (val) { type[count] = i; vals[count++] = val; @@ -307,6 +307,7 @@ static void icmp_put(struct seq_file *seq) { int i; struct net *net = seq->private; + atomic_long_t *ptr = net->mib.icmpmsg_statistics->mibs; seq_puts(seq, "\nIcmp: InMsgs InErrors"); for (i=0; icmpmibmap[i].name != NULL; i++) @@ -319,15 +320,13 @@ static void icmp_put(struct seq_file *seq) snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_INERRORS)); for (i=0; icmpmibmap[i].name != NULL; i++) seq_printf(seq, " %lu", - snmp_fold_field((void __percpu **) net->mib.icmpmsg_statistics, - icmpmibmap[i].index)); + atomic_long_read(ptr + icmpmibmap[i].index)); seq_printf(seq, " %lu %lu", snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_OUTMSGS), snmp_fold_field((void __percpu **) net->mib.icmp_statistics, ICMP_MIB_OUTERRORS)); for (i=0; icmpmibmap[i].name != NULL; i++) seq_printf(seq, " %lu", - snmp_fold_field((void __percpu **) net->mib.icmpmsg_statistics, - icmpmibmap[i].index | 0x100)); + atomic_long_read(ptr + (icmpmibmap[i].index | 0x100))); } /* -- GitLab From d826eb14ecef3574b6b3be55e5f4329f4a76fbf3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 9 Nov 2011 07:24:35 +0000 Subject: [PATCH 0156/3089] ipv4: PKTINFO doesnt need dst reference MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Le lundi 07 novembre 2011 à 15:33 +0100, Eric Dumazet a écrit : > At least, in recent kernels we dont change dst->refcnt in forwarding > patch (usinf NOREF skb->dst) > > One particular point is the atomic_inc(dst->refcnt) we have to perform > when queuing an UDP packet if socket asked PKTINFO stuff (for example a > typical DNS server has to setup this option) > > I have one patch somewhere that stores the information in skb->cb[] and > avoid the atomic_{inc|dec}(dst->refcnt). > OK I found it, I did some extra tests and believe its ready. [PATCH net-next] ipv4: IP_PKTINFO doesnt need dst reference When a socket uses IP_PKTINFO notifications, we currently force a dst reference for each received skb. Reader has to access dst to get needed information (rt_iif & rt_spec_dst) and must release dst reference. We also forced a dst reference if skb was put in socket backlog, even without IP_PKTINFO handling. This happens under stress/load. We can instead store the needed information in skb->cb[], so that only softirq handler really access dst, improving cache hit ratios. This removes two atomic operations per packet, and false sharing as well. On a benchmark using a mono threaded receiver (doing only recvmsg() calls), I can reach 720.000 pps instead of 570.000 pps. IP_PKTINFO is typically used by DNS servers, and any multihomed aware UDP application. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ip.h | 2 +- net/ipv4/ip_sockglue.c | 35 ++++++++++++++++++----------------- net/ipv4/raw.c | 3 ++- net/ipv4/udp.c | 3 ++- net/ipv6/raw.c | 3 ++- net/ipv6/udp.c | 4 +++- 6 files changed, 28 insertions(+), 22 deletions(-) diff --git a/include/net/ip.h b/include/net/ip.h index eca0ef7a495e9..fd1561e88a1a5 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -450,7 +450,7 @@ extern int ip_options_rcv_srr(struct sk_buff *skb); * Functions provided by ip_sockglue.c */ -extern int ip_queue_rcv_skb(struct sock *sk, struct sk_buff *skb); +extern void ipv4_pktinfo_prepare(struct sk_buff *skb); extern void ip_cmsg_recv(struct msghdr *msg, struct sk_buff *skb); extern int ip_cmsg_send(struct net *net, struct msghdr *msg, struct ipcm_cookie *ipc); diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c index 09ff51bf16a4c..80d5fa4502104 100644 --- a/net/ipv4/ip_sockglue.c +++ b/net/ipv4/ip_sockglue.c @@ -55,20 +55,13 @@ /* * SOL_IP control messages. */ +#define PKTINFO_SKB_CB(__skb) ((struct in_pktinfo *)((__skb)->cb)) static void ip_cmsg_recv_pktinfo(struct msghdr *msg, struct sk_buff *skb) { - struct in_pktinfo info; - struct rtable *rt = skb_rtable(skb); + struct in_pktinfo info = *PKTINFO_SKB_CB(skb); info.ipi_addr.s_addr = ip_hdr(skb)->daddr; - if (rt) { - info.ipi_ifindex = rt->rt_iif; - info.ipi_spec_dst.s_addr = rt->rt_spec_dst; - } else { - info.ipi_ifindex = 0; - info.ipi_spec_dst.s_addr = 0; - } put_cmsg(msg, SOL_IP, IP_PKTINFO, sizeof(info), &info); } @@ -992,20 +985,28 @@ e_inval: } /** - * ip_queue_rcv_skb - Queue an skb into sock receive queue + * ipv4_pktinfo_prepare - transfert some info from rtable to skb * @sk: socket * @skb: buffer * - * Queues an skb into socket receive queue. If IP_CMSG_PKTINFO option - * is not set, we drop skb dst entry now, while dst cache line is hot. + * To support IP_CMSG_PKTINFO option, we store rt_iif and rt_spec_dst + * in skb->cb[] before dst drop. + * This way, receiver doesnt make cache line misses to read rtable. */ -int ip_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) +void ipv4_pktinfo_prepare(struct sk_buff *skb) { - if (!(inet_sk(sk)->cmsg_flags & IP_CMSG_PKTINFO)) - skb_dst_drop(skb); - return sock_queue_rcv_skb(sk, skb); + struct in_pktinfo *pktinfo = PKTINFO_SKB_CB(skb); + const struct rtable *rt = skb_rtable(skb); + + if (rt) { + pktinfo->ipi_ifindex = rt->rt_iif; + pktinfo->ipi_spec_dst.s_addr = rt->rt_spec_dst; + } else { + pktinfo->ipi_ifindex = 0; + pktinfo->ipi_spec_dst.s_addr = 0; + } + skb_dst_drop(skb); } -EXPORT_SYMBOL(ip_queue_rcv_skb); int ip_setsockopt(struct sock *sk, int level, int optname, char __user *optval, unsigned int optlen) diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index 007e2eb769d33..7a8410d1b4b17 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -292,7 +292,8 @@ static int raw_rcv_skb(struct sock * sk, struct sk_buff * skb) { /* Charge it to the socket. */ - if (ip_queue_rcv_skb(sk, skb) < 0) { + ipv4_pktinfo_prepare(skb); + if (sock_queue_rcv_skb(sk, skb) < 0) { kfree_skb(skb); return NET_RX_DROP; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index ab0966df1e2a8..6854f581313f6 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1357,7 +1357,7 @@ static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) if (inet_sk(sk)->inet_daddr) sock_rps_save_rxhash(sk, skb); - rc = ip_queue_rcv_skb(sk, skb); + rc = sock_queue_rcv_skb(sk, skb); if (rc < 0) { int is_udplite = IS_UDPLITE(sk); @@ -1473,6 +1473,7 @@ int udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) rc = 0; + ipv4_pktinfo_prepare(skb); bh_lock_sock(sk); if (!sock_owned_by_user(sk)) rc = __udp_queue_rcv_skb(sk, skb); diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 331af3b882ac2..204f2e833f04a 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -383,7 +383,8 @@ static inline int rawv6_rcv_skb(struct sock *sk, struct sk_buff *skb) } /* Charge it to the socket. */ - if (ip_queue_rcv_skb(sk, skb) < 0) { + skb_dst_drop(skb); + if (sock_queue_rcv_skb(sk, skb) < 0) { kfree_skb(skb); return NET_RX_DROP; } diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 846f4757eb8d4..b4a4a15fa96fe 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -538,7 +538,9 @@ int udpv6_queue_rcv_skb(struct sock * sk, struct sk_buff *skb) goto drop; } - if ((rc = ip_queue_rcv_skb(sk, skb)) < 0) { + skb_dst_drop(skb); + rc = sock_queue_rcv_skb(sk, skb); + if (rc < 0) { /* Note that an ENOMEM error is charged twice */ if (rc == -ENOMEM) UDP6_INC_STATS_BH(sock_net(sk), -- GitLab From 8658ca730fa66fd6b8c559315915c774330f35bb Mon Sep 17 00:00:00 2001 From: "Maxin B. John" Date: Wed, 9 Nov 2011 11:12:58 +0100 Subject: [PATCH 0157/3089] ARM: 7153/1: mach-bcmring: fix build error in core.c Fixes this build error: CC arch/arm/mach-bcmring/core.o arch/arm/mach-bcmring/core.c: In function 'bcmring_init_timer': arch/arm/mach-bcmring/core.c:238: error: implicit declaration of function 'sp804_clockevents_register' make[1]: *** [arch/arm/mach-bcmring/core.o] Error 1 make: *** [arch/arm/mach-bcmring] Error 2 Signed-off-by: Maxin B. John Reviewed-by: Jiandong Zheng Signed-off-by: Russell King --- arch/arm/mach-bcmring/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-bcmring/core.c b/arch/arm/mach-bcmring/core.c index 43eadbcc29ede..430da120a297f 100644 --- a/arch/arm/mach-bcmring/core.c +++ b/arch/arm/mach-bcmring/core.c @@ -235,7 +235,7 @@ void __init bcmring_init_timer(void) */ bcmring_clocksource_init(); - sp804_clockevents_register(TIMER0_VA_BASE, IRQ_TIMER0, "timer0"); + sp804_clockevents_init(TIMER0_VA_BASE, IRQ_TIMER0, "timer0"); } struct sys_timer bcmring_timer = { -- GitLab From 5fc86281b9eba5eb9cfab999bcb00deb335da762 Mon Sep 17 00:00:00 2001 From: "Maxin B. John" Date: Wed, 9 Nov 2011 11:27:36 +0100 Subject: [PATCH 0158/3089] ARM: 7154/1: mach-bcmring: fix build error in dma.c Fixes this build error: CC arch/arm/mach-bcmring/dma.o arch/arm/mach-bcmring/dma.c: In function 'dma_request_channel_dbg': arch/arm/mach-bcmring/dma.c:1022: error: 'TASK_INTERRUPTIBLE' undeclared (first use in this function) arch/arm/mach-bcmring/dma.c:1022: error: (Each undeclared identifier is reported only once arch/arm/mach-bcmring/dma.c:1022: error: for each function it appears in.) arch/arm/mach-bcmring/dma.c:1024: error: implicit declaration of function 'schedule' arch/arm/mach-bcmring/dma.c:1027: error: implicit declaration of function 'signal_pending' arch/arm/mach-bcmring/dma.c: In function 'dma_free_channel': arch/arm/mach-bcmring/dma.c:1092: error: 'TASK_INTERRUPTIBLE' undeclared (first use in this function) arch/arm/mach-bcmring/dma.c: In function 'dma_map_add_region': arch/arm/mach-bcmring/dma.c:1946: error: dereferencing pointer to incomplete type arch/arm/mach-bcmring/dma.c:1948: error: dereferencing pointer to incomplete type arch/arm/mach-bcmring/dma.c:1955: error: dereferencing pointer to incomplete type make[1]: *** [arch/arm/mach-bcmring/dma.o] Error 1 make: *** [arch/arm/mach-bcmring] Error 2 Signed-off-by: Maxin B. John Reviewed-by: Jiandong Zheng Signed-off-by: Russell King --- arch/arm/mach-bcmring/dma.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/mach-bcmring/dma.c b/arch/arm/mach-bcmring/dma.c index b52b8de91bde7..f4d4d6d174d06 100644 --- a/arch/arm/mach-bcmring/dma.c +++ b/arch/arm/mach-bcmring/dma.c @@ -36,6 +36,7 @@ #include #include #include +#include #include /* I don't quite understand why dc4 fails when this is set to 1 and DMA is enabled */ -- GitLab From 1a1f2be2a4d7f6f538169dca534c94b221b21923 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Wed, 9 Nov 2011 19:13:43 +0100 Subject: [PATCH 0159/3089] ARM: 7155/1: arch.h: Declare 'pt_regs' locally Fix the following warning when building imx_v4_v5_defconfig target: CC arch/arm/mach-imx/mach-imx27ipcam.o In file included from arch/arm/mach-imx/mach-imx27ipcam.c:18: /home/fabio/linus/linux-2.6/arch/arm/include/asm/mach/arch.h:47: warning: 'struct pt_regs' declared inside parameter list /home/fabio/linus/linux-2.6/arch/arm/include/asm/mach/arch.h:47: warning: its scope is only this definition or declaration, which is probably not what you want arch/arm/mach-imx/mach-imx27ipcam.c:78: warning: initialization from incompatible pointer type Declare 'struct pt_regs' locally. Cc: Eric Miao Signed-off-by: Fabio Estevam Signed-off-by: Russell King --- arch/arm/include/asm/mach/arch.h | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/include/asm/mach/arch.h b/arch/arm/include/asm/mach/arch.h index 7d19425dd496a..2b0efc3104ac6 100644 --- a/arch/arm/include/asm/mach/arch.h +++ b/arch/arm/include/asm/mach/arch.h @@ -13,6 +13,7 @@ struct tag; struct meminfo; struct sys_timer; +struct pt_regs; struct machine_desc { unsigned int nr; /* architecture number */ -- GitLab From 5bb01a991897d74295ff8de548d350400d7370a9 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Wed, 9 Nov 2011 15:29:31 -0800 Subject: [PATCH 0160/3089] ARM: OMAP: Fix map_io for Amstrad E3 Commit 7b88e62f5d219a86d81bdf4388012c97dc42e8f8 (ARM: OMAP1: Use generic map_io, init_early and init_irq) changed omap1 to use generic map_io. Looks like I missed one board though. Fix this by adding a custom map_io for Amstrad E3. Reported-by: Aaro Koskinen Signed-off-by: Tony Lindgren --- arch/arm/mach-omap1/board-ams-delta.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/arch/arm/mach-omap1/board-ams-delta.c b/arch/arm/mach-omap1/board-ams-delta.c index 51bae31cf3612..b0f15d234a12b 100644 --- a/arch/arm/mach-omap1/board-ams-delta.c +++ b/arch/arm/mach-omap1/board-ams-delta.c @@ -302,8 +302,6 @@ static void __init ams_delta_init(void) omap_cfg_reg(J19_1610_CAM_D6); omap_cfg_reg(J18_1610_CAM_D7); - iotable_init(ams_delta_io_desc, ARRAY_SIZE(ams_delta_io_desc)); - omap_board_config = ams_delta_config; omap_board_config_size = ARRAY_SIZE(ams_delta_config); omap_serial_init(); @@ -373,10 +371,16 @@ static int __init ams_delta_modem_init(void) } arch_initcall(ams_delta_modem_init); +static void __init ams_delta_map_io(void) +{ + omap15xx_map_io(); + iotable_init(ams_delta_io_desc, ARRAY_SIZE(ams_delta_io_desc)); +} + MACHINE_START(AMS_DELTA, "Amstrad E3 (Delta)") /* Maintainer: Jonathan McDowell */ .atag_offset = 0x100, - .map_io = omap15xx_map_io, + .map_io = ams_delta_map_io, .init_early = omap1_init_early, .reserve = omap_reserve, .init_irq = omap1_init_irq, -- GitLab From be2658ffe1d2a4ab4e045b22a317459595abc2c4 Mon Sep 17 00:00:00 2001 From: Guennadi Liakhovetski Date: Thu, 10 Nov 2011 00:43:52 +0100 Subject: [PATCH 0161/3089] PM / shmobile: Don't skip debugging output in pd_power_up() Don't skip debugging output upon earlier loop termination in pd_power_up(). Signed-off-by: Guennadi Liakhovetski Acked-by: Magnus Damm Signed-off-by: Rafael J. Wysocki --- arch/arm/mach-shmobile/pm-sh7372.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-shmobile/pm-sh7372.c b/arch/arm/mach-shmobile/pm-sh7372.c index 79612737c5b23..e26138352f07a 100644 --- a/arch/arm/mach-shmobile/pm-sh7372.c +++ b/arch/arm/mach-shmobile/pm-sh7372.c @@ -123,13 +123,13 @@ static int pd_power_up(struct generic_pm_domain *genpd) for (retry_count = 2 * PSTR_RETRIES; retry_count; retry_count--) { if (!(__raw_readl(SWUCR) & mask)) - goto out; + break; if (retry_count > PSTR_RETRIES) udelay(PSTR_DELAY_US); else cpu_relax(); } - if (__raw_readl(SWUCR) & mask) + if (!retry_count) ret = -EIO; if (!sh7372_pd->no_debug) -- GitLab From 056879d2f244001b2888cdc8cf868c33ca8b23a0 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Thu, 10 Nov 2011 00:44:01 +0100 Subject: [PATCH 0162/3089] ARM: mach-shmobile: sh7372 A3SP no_suspend_console fix If "no_suspend_irq" is used on the sh7372 Mackerel board with v3.2-rc1 then Suspend-to-RAM fails because the serial console tries to write to the SCIF driver even though the power domain is turned off. This patch checks the state of "console_suspend_enabled" to see if A3SP should be forced enabled. Signed-off-by: Magnus Damm Signed-off-by: Rafael J. Wysocki --- arch/arm/mach-shmobile/pm-sh7372.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/arch/arm/mach-shmobile/pm-sh7372.c b/arch/arm/mach-shmobile/pm-sh7372.c index e26138352f07a..78c98a8474335 100644 --- a/arch/arm/mach-shmobile/pm-sh7372.c +++ b/arch/arm/mach-shmobile/pm-sh7372.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include #include @@ -469,6 +470,11 @@ void __init sh7372_pm_init(void) /* do not convert A3SM, A3SP, A3SG, A4R power down into A4S */ __raw_writel(0, PDNSEL); + /* serial consoles make use of SCIF hardware located in A3SP, + * keep such power domain on if "no_console_suspend" is set. + */ + sh7372_a3sp.stay_on = !console_suspend_enabled; + sh7372_suspend_init(); sh7372_cpuidle_init(); } -- GitLab From 24050956e029a9ecff096e1992869ada4492963c Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Thu, 10 Nov 2011 00:44:10 +0100 Subject: [PATCH 0163/3089] PM / Clocks: Only disable enabled clocks in pm_clk_suspend() Refrain from running clk_disable() on clocks that have not been enabled. A typical case when this can happen is during Suspend-to-RAM for devices that have no driver associated with them. In such case the clock may be in default ACQUIRED state. Without this patch the sh7372 Mackerel board crashes in __clk_disable() during Suspend-to-RAM with: "Trying to disable clock 0xdeadbeef with 0 usecount" This happens for the CEU device which is added during boot. The test case has no CEU driver included in the kernel configuration. Needed for v3.2-rc1. Signed-off-by: Magnus Damm Signed-off-by: Rafael J. Wysocki --- drivers/base/power/clock_ops.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/base/power/clock_ops.c b/drivers/base/power/clock_ops.c index 5f0f85d5c5765..428e55e012dcd 100644 --- a/drivers/base/power/clock_ops.c +++ b/drivers/base/power/clock_ops.c @@ -229,7 +229,8 @@ int pm_clk_suspend(struct device *dev) list_for_each_entry_reverse(ce, &psd->clock_list, node) { if (ce->status < PCE_STATUS_ERROR) { - clk_disable(ce->clk); + if (ce->status == PCE_STATUS_ENABLED) + clk_disable(ce->clk); ce->status = PCE_STATUS_ACQUIRED; } } -- GitLab From af4c720efc0507e01b89774fed936087baac4107 Mon Sep 17 00:00:00 2001 From: Guennadi Liakhovetski Date: Thu, 10 Nov 2011 00:44:18 +0100 Subject: [PATCH 0164/3089] PM / QoS: Properly use the WARN() macro in dev_pm_qos_add_request() Make dev_pm_qos_add_request() use WARN() in a better way and do not hardcode the function's name into the message (use __func__ instead). Signed-off-by: Guennadi Liakhovetski Signed-off-by: Rafael J. Wysocki --- drivers/base/power/qos.c | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) diff --git a/drivers/base/power/qos.c b/drivers/base/power/qos.c index 30a94eadc200c..86de6c50fc418 100644 --- a/drivers/base/power/qos.c +++ b/drivers/base/power/qos.c @@ -212,11 +212,9 @@ int dev_pm_qos_add_request(struct device *dev, struct dev_pm_qos_request *req, if (!dev || !req) /*guard against callers passing in null */ return -EINVAL; - if (dev_pm_qos_request_active(req)) { - WARN(1, KERN_ERR "dev_pm_qos_add_request() called for already " - "added request\n"); + if (WARN(dev_pm_qos_request_active(req), + "%s() called for already added request\n", __func__)) return -EINVAL; - } req->dev = dev; @@ -271,11 +269,9 @@ int dev_pm_qos_update_request(struct dev_pm_qos_request *req, if (!req) /*guard against callers passing in null */ return -EINVAL; - if (!dev_pm_qos_request_active(req)) { - WARN(1, KERN_ERR "dev_pm_qos_update_request() called for " - "unknown object\n"); + if (WARN(!dev_pm_qos_request_active(req), + "%s() called for unknown object\n", __func__)) return -EINVAL; - } mutex_lock(&dev_pm_qos_mtx); @@ -312,11 +308,9 @@ int dev_pm_qos_remove_request(struct dev_pm_qos_request *req) if (!req) /*guard against callers passing in null */ return -EINVAL; - if (!dev_pm_qos_request_active(req)) { - WARN(1, KERN_ERR "dev_pm_qos_remove_request() called for " - "unknown object\n"); + if (WARN(!dev_pm_qos_request_active(req), + "%s() called for unknown object\n", __func__)) return -EINVAL; - } mutex_lock(&dev_pm_qos_mtx); -- GitLab From 1e49570171117e547e6324c58371db4a0dc2f1db Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 9 Nov 2011 00:14:11 -0500 Subject: [PATCH 0165/3089] net: Fix references to deleted NET_ETHERNET Kconfig setting. Change them over to plain "ETHERNET" Reported-by: Paul Bolle Signed-off-by: David S. Miller Acked-by: Jeff Kirsher --- arch/cris/arch-v10/drivers/Kconfig | 2 +- arch/cris/arch-v32/drivers/Kconfig | 2 +- drivers/s390/net/Kconfig | 2 +- drivers/s390/net/lcs.c | 6 +++--- 4 files changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/cris/arch-v10/drivers/Kconfig b/arch/cris/arch-v10/drivers/Kconfig index 32d90867a9841..5f2cdb3e428cd 100644 --- a/arch/cris/arch-v10/drivers/Kconfig +++ b/arch/cris/arch-v10/drivers/Kconfig @@ -3,7 +3,7 @@ if ETRAX_ARCH_V10 config ETRAX_ETHERNET bool "Ethernet support" depends on ETRAX_ARCH_V10 - select NET_ETHERNET + select ETHERNET select NET_CORE select MII help diff --git a/arch/cris/arch-v32/drivers/Kconfig b/arch/cris/arch-v32/drivers/Kconfig index e47e9c3401b08..de43aadcdbc47 100644 --- a/arch/cris/arch-v32/drivers/Kconfig +++ b/arch/cris/arch-v32/drivers/Kconfig @@ -3,7 +3,7 @@ if ETRAX_ARCH_V32 config ETRAX_ETHERNET bool "Ethernet support" depends on ETRAX_ARCH_V32 - select NET_ETHERNET + select ETHERNET select NET_CORE select MII help diff --git a/drivers/s390/net/Kconfig b/drivers/s390/net/Kconfig index fa80ba1f03443..9b66d2d1809b3 100644 --- a/drivers/s390/net/Kconfig +++ b/drivers/s390/net/Kconfig @@ -4,7 +4,7 @@ menu "S/390 network device drivers" config LCS def_tristate m prompt "Lan Channel Station Interface" - depends on CCW && NETDEVICES && (NET_ETHERNET || TR || FDDI) + depends on CCW && NETDEVICES && (ETHERNET || TR || FDDI) help Select this option if you want to use LCS networking on IBM System z. This device driver supports Token Ring (IEEE 802.5), diff --git a/drivers/s390/net/lcs.c b/drivers/s390/net/lcs.c index c28713da1ec5d..863fc2197155c 100644 --- a/drivers/s390/net/lcs.c +++ b/drivers/s390/net/lcs.c @@ -50,7 +50,7 @@ #include "lcs.h" -#if !defined(CONFIG_NET_ETHERNET) && \ +#if !defined(CONFIG_ETHERNET) && \ !defined(CONFIG_TR) && !defined(CONFIG_FDDI) #error Cannot compile lcs.c without some net devices switched on. #endif @@ -1634,7 +1634,7 @@ lcs_startlan_auto(struct lcs_card *card) int rc; LCS_DBF_TEXT(2, trace, "strtauto"); -#ifdef CONFIG_NET_ETHERNET +#ifdef CONFIG_ETHERNET card->lan_type = LCS_FRAME_TYPE_ENET; rc = lcs_send_startlan(card, LCS_INITIATOR_TCPIP); if (rc == 0) @@ -2166,7 +2166,7 @@ lcs_new_device(struct ccwgroup_device *ccwgdev) goto netdev_out; } switch (card->lan_type) { -#ifdef CONFIG_NET_ETHERNET +#ifdef CONFIG_ETHERNET case LCS_FRAME_TYPE_ENET: card->lan_type_trans = eth_type_trans; dev = alloc_etherdev(0); -- GitLab From 3940d6185a982a970ff562e085caccbdd62f40bb Mon Sep 17 00:00:00 2001 From: JJ Ding Date: Tue, 8 Nov 2011 22:13:14 -0800 Subject: [PATCH 0166/3089] Input: elantech - adjust hw_version detection logic MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch fixes some v3 hardware (fw_version: 0x150500) wrongly detected as v2 hardware. Reported-by: Marc Dietrich Signed-off-by: JJ Ding Tested-By: Marc Dietrich Acked-by: Éric Piel Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/elantech.c | 26 ++++++++++++++++++-------- 1 file changed, 18 insertions(+), 8 deletions(-) diff --git a/drivers/input/mouse/elantech.c b/drivers/input/mouse/elantech.c index 09b93b11a2742..e2a9867c19d52 100644 --- a/drivers/input/mouse/elantech.c +++ b/drivers/input/mouse/elantech.c @@ -1210,18 +1210,28 @@ static int elantech_reconnect(struct psmouse *psmouse) */ static int elantech_set_properties(struct elantech_data *etd) { + /* This represents the version of IC body. */ int ver = (etd->fw_version & 0x0f0000) >> 16; + /* Early version of Elan touchpads doesn't obey the rule. */ if (etd->fw_version < 0x020030 || etd->fw_version == 0x020600) etd->hw_version = 1; - else if (etd->fw_version < 0x150600) - etd->hw_version = 2; - else if (ver == 5) - etd->hw_version = 3; - else if (ver == 6) - etd->hw_version = 4; - else - return -1; + else { + switch (ver) { + case 2: + case 4: + etd->hw_version = 2; + break; + case 5: + etd->hw_version = 3; + break; + case 6: + etd->hw_version = 4; + break; + default: + return -1; + } + } /* * Turn on packet checking by default. -- GitLab From 77f6ca5a6283a5d34c4bd9dc809fec9e37fdd80d Mon Sep 17 00:00:00 2001 From: Aaro Koskinen Date: Wed, 9 Nov 2011 10:13:33 -0800 Subject: [PATCH 0167/3089] Input: ams_delta_serio - include linux/module.h Fix the following compilation failure with v3.2-rc1 by including module.h: CC drivers/input/serio/ams_delta_serio.o drivers/input/serio/ams_delta_serio.c:33:15: error: expected declaration specifiers or '...' before string constant drivers/input/serio/ams_delta_serio.c:34:20: error: expected declaration specifiers or '...' before string constant drivers/input/serio/ams_delta_serio.c:35:16: error: expected declaration specifiers or '...' before string constant drivers/input/serio/ams_delta_serio.c: In function 'ams_delta_serio_init': drivers/input/serio/ams_delta_serio.c:155:2: error: 'THIS_MODULE' undeclared (first use in this function) drivers/input/serio/ams_delta_serio.c:155:2: note: each undeclared identifier is reported only once for each function it appears in Signed-off-by: Aaro Koskinen Signed-off-by: Dmitry Torokhov --- drivers/input/serio/ams_delta_serio.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/serio/ams_delta_serio.c b/drivers/input/serio/ams_delta_serio.c index 4b2a42f9f0bb4..d4d08bd9205b8 100644 --- a/drivers/input/serio/ams_delta_serio.c +++ b/drivers/input/serio/ams_delta_serio.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include -- GitLab From c0a39151a4055332897cba615623d3de2f3896df Mon Sep 17 00:00:00 2001 From: Haojian Zhuang Date: Thu, 10 Nov 2011 07:13:07 +0800 Subject: [PATCH 0168/3089] ARM: pxa: fix inconsistent CONFIG_USB_PXA27X Since CONFIG_USB_GADGET_PXA27X and other macros are renamed to CONFIG_USB_PXA27X. Update them in arch/arm/mach-pxa and arch/arm/configs to keep consistent. Cc: stable@kernel.org Signed-off-by: Haojian Zhuang Acked-by: Felipe Balbi Signed-off-by: Eric Miao --- arch/arm/configs/ezx_defconfig | 2 +- arch/arm/configs/imote2_defconfig | 2 +- arch/arm/configs/magician_defconfig | 2 +- arch/arm/configs/zeus_defconfig | 2 +- arch/arm/mach-pxa/balloon3.c | 2 +- arch/arm/mach-pxa/colibri-pxa320.c | 2 +- arch/arm/mach-pxa/gumstix.c | 2 +- arch/arm/mach-pxa/include/mach/palm27x.h | 4 ++-- arch/arm/mach-pxa/palm27x.c | 4 ++-- arch/arm/mach-pxa/palmtc.c | 2 +- arch/arm/mach-pxa/vpac270.c | 2 +- 11 files changed, 13 insertions(+), 13 deletions(-) diff --git a/arch/arm/configs/ezx_defconfig b/arch/arm/configs/ezx_defconfig index 227a477346edb..d95763d5f0d83 100644 --- a/arch/arm/configs/ezx_defconfig +++ b/arch/arm/configs/ezx_defconfig @@ -287,7 +287,7 @@ CONFIG_USB=y # CONFIG_USB_DEVICE_CLASS is not set CONFIG_USB_OHCI_HCD=y CONFIG_USB_GADGET=y -CONFIG_USB_GADGET_PXA27X=y +CONFIG_USB_PXA27X=y CONFIG_USB_ETH=m # CONFIG_USB_ETH_RNDIS is not set CONFIG_MMC=y diff --git a/arch/arm/configs/imote2_defconfig b/arch/arm/configs/imote2_defconfig index 176ec22af0342..fd996bb130228 100644 --- a/arch/arm/configs/imote2_defconfig +++ b/arch/arm/configs/imote2_defconfig @@ -263,7 +263,7 @@ CONFIG_USB=y # CONFIG_USB_DEVICE_CLASS is not set CONFIG_USB_OHCI_HCD=y CONFIG_USB_GADGET=y -CONFIG_USB_GADGET_PXA27X=y +CONFIG_USB_PXA27X=y CONFIG_USB_ETH=m # CONFIG_USB_ETH_RNDIS is not set CONFIG_MMC=y diff --git a/arch/arm/configs/magician_defconfig b/arch/arm/configs/magician_defconfig index a88e64d4e9a58..443675d317e6d 100644 --- a/arch/arm/configs/magician_defconfig +++ b/arch/arm/configs/magician_defconfig @@ -132,7 +132,7 @@ CONFIG_USB_MON=m CONFIG_USB_OHCI_HCD=y CONFIG_USB_GADGET=y CONFIG_USB_GADGET_VBUS_DRAW=500 -CONFIG_USB_GADGET_PXA27X=y +CONFIG_USB_PXA27X=y CONFIG_USB_ETH=m # CONFIG_USB_ETH_RNDIS is not set CONFIG_USB_GADGETFS=m diff --git a/arch/arm/configs/zeus_defconfig b/arch/arm/configs/zeus_defconfig index 59577ad3f4efd..547a3c1e59dbc 100644 --- a/arch/arm/configs/zeus_defconfig +++ b/arch/arm/configs/zeus_defconfig @@ -140,7 +140,7 @@ CONFIG_USB_SERIAL=m CONFIG_USB_SERIAL_GENERIC=y CONFIG_USB_SERIAL_MCT_U232=m CONFIG_USB_GADGET=m -CONFIG_USB_GADGET_PXA27X=y +CONFIG_USB_PXA27X=y CONFIG_USB_ETH=m CONFIG_USB_GADGETFS=m CONFIG_USB_FILE_STORAGE=m diff --git a/arch/arm/mach-pxa/balloon3.c b/arch/arm/mach-pxa/balloon3.c index fc0b8544e1745..4b81f59a4cbaf 100644 --- a/arch/arm/mach-pxa/balloon3.c +++ b/arch/arm/mach-pxa/balloon3.c @@ -307,7 +307,7 @@ static inline void balloon3_mmc_init(void) {} /****************************************************************************** * USB Gadget ******************************************************************************/ -#if defined(CONFIG_USB_GADGET_PXA27X)||defined(CONFIG_USB_GADGET_PXA27X_MODULE) +#if defined(CONFIG_USB_PXA27X)||defined(CONFIG_USB_PXA27X_MODULE) static void balloon3_udc_command(int cmd) { if (cmd == PXA2XX_UDC_CMD_CONNECT) diff --git a/arch/arm/mach-pxa/colibri-pxa320.c b/arch/arm/mach-pxa/colibri-pxa320.c index 692e1ffc55862..d23b92b804882 100644 --- a/arch/arm/mach-pxa/colibri-pxa320.c +++ b/arch/arm/mach-pxa/colibri-pxa320.c @@ -146,7 +146,7 @@ static void __init colibri_pxa320_init_eth(void) static inline void __init colibri_pxa320_init_eth(void) {} #endif /* CONFIG_AX88796 */ -#if defined(CONFIG_USB_GADGET_PXA27X)||defined(CONFIG_USB_GADGET_PXA27X_MODULE) +#if defined(CONFIG_USB_PXA27X)||defined(CONFIG_USB_PXA27X_MODULE) static struct gpio_vbus_mach_info colibri_pxa320_gpio_vbus_info = { .gpio_vbus = mfp_to_gpio(MFP_PIN_GPIO96), .gpio_pullup = -1, diff --git a/arch/arm/mach-pxa/gumstix.c b/arch/arm/mach-pxa/gumstix.c index 9c8208ca04150..ffdd70dad327d 100644 --- a/arch/arm/mach-pxa/gumstix.c +++ b/arch/arm/mach-pxa/gumstix.c @@ -106,7 +106,7 @@ static void __init gumstix_mmc_init(void) } #endif -#ifdef CONFIG_USB_GADGET_PXA25X +#ifdef CONFIG_USB_PXA25X static struct gpio_vbus_mach_info gumstix_udc_info = { .gpio_vbus = GPIO_GUMSTIX_USB_GPIOn, .gpio_pullup = GPIO_GUMSTIX_USB_GPIOx, diff --git a/arch/arm/mach-pxa/include/mach/palm27x.h b/arch/arm/mach-pxa/include/mach/palm27x.h index f80bbe246afe5..d4eac3d6ffb5e 100644 --- a/arch/arm/mach-pxa/include/mach/palm27x.h +++ b/arch/arm/mach-pxa/include/mach/palm27x.h @@ -37,8 +37,8 @@ extern void __init palm27x_lcd_init(int power, #define palm27x_lcd_init(power, mode) do {} while (0) #endif -#if defined(CONFIG_USB_GADGET_PXA27X) || \ - defined(CONFIG_USB_GADGET_PXA27X_MODULE) +#if defined(CONFIG_USB_PXA27X) || \ + defined(CONFIG_USB_PXA27X_MODULE) extern void __init palm27x_udc_init(int vbus, int pullup, int vbus_inverted); #else diff --git a/arch/arm/mach-pxa/palm27x.c b/arch/arm/mach-pxa/palm27x.c index 325c245c0a0dd..fbc10d7b95d1e 100644 --- a/arch/arm/mach-pxa/palm27x.c +++ b/arch/arm/mach-pxa/palm27x.c @@ -164,8 +164,8 @@ void __init palm27x_lcd_init(int power, struct pxafb_mode_info *mode) /****************************************************************************** * USB Gadget ******************************************************************************/ -#if defined(CONFIG_USB_GADGET_PXA27X) || \ - defined(CONFIG_USB_GADGET_PXA27X_MODULE) +#if defined(CONFIG_USB_PXA27X) || \ + defined(CONFIG_USB_PXA27X_MODULE) static struct gpio_vbus_mach_info palm27x_udc_info = { .gpio_vbus_inverted = 1, }; diff --git a/arch/arm/mach-pxa/palmtc.c b/arch/arm/mach-pxa/palmtc.c index 6ec7caefb37c8..2c24c67fd92b6 100644 --- a/arch/arm/mach-pxa/palmtc.c +++ b/arch/arm/mach-pxa/palmtc.c @@ -338,7 +338,7 @@ static inline void palmtc_mkp_init(void) {} /****************************************************************************** * UDC ******************************************************************************/ -#if defined(CONFIG_USB_GADGET_PXA25X)||defined(CONFIG_USB_GADGET_PXA25X_MODULE) +#if defined(CONFIG_USB_PXA25X)||defined(CONFIG_USB_PXA25X_MODULE) static struct gpio_vbus_mach_info palmtc_udc_info = { .gpio_vbus = GPIO_NR_PALMTC_USB_DETECT_N, .gpio_vbus_inverted = 1, diff --git a/arch/arm/mach-pxa/vpac270.c b/arch/arm/mach-pxa/vpac270.c index a7539a6ed1ff2..ca0c6615028c4 100644 --- a/arch/arm/mach-pxa/vpac270.c +++ b/arch/arm/mach-pxa/vpac270.c @@ -343,7 +343,7 @@ static inline void vpac270_uhc_init(void) {} /****************************************************************************** * USB Gadget ******************************************************************************/ -#if defined(CONFIG_USB_GADGET_PXA27X)||defined(CONFIG_USB_GADGET_PXA27X_MODULE) +#if defined(CONFIG_USB_PXA27X)||defined(CONFIG_USB_PXA27X_MODULE) static struct gpio_vbus_mach_info vpac270_gpio_vbus_info = { .gpio_vbus = GPIO41_VPAC270_UDC_DETECT, .gpio_pullup = -1, -- GitLab From d0985394e7fee6b25a7cc8335d45bc1c1a8ab2d3 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 10 Nov 2011 09:03:55 +0100 Subject: [PATCH 0169/3089] block: Revert "[SCSI] genhd: add a new attribute "alias" in gendisk" This reverts commit a72c5e5eb738033938ab30d6a634b74d1d060f10. The commit introduced alias for block devices which is intended to be used during logging although actual usage hasn't been committed yet. This approach adds very limited benefit (raw log might be easier to follow) which can be trivially implemented in userland but has a lot of problems. It is much worse than netif renames because it doesn't rename the actual device but just adds conveninence name which isn't used universally or enforced. Everything internal including device lookup and sysfs still uses the internal name and nothing prevents two devices from using conflicting alias - ie. sda can have sdb as its alias. This has been nacked by people working on device driver core, block layer and kernel-userland interface and shouldn't have been upstreamed. Revert it. http://thread.gmane.org/gmane.linux.kernel/1155104 http://thread.gmane.org/gmane.linux.scsi/68632 http://thread.gmane.org/gmane.linux.scsi/69776 Signed-off-by: Tejun Heo Acked-by: Greg Kroah-Hartman Acked-by: Kay Sievers Cc: "James E.J. Bottomley" Cc: Nao Nishijima Cc: Alan Cox Cc: Al Viro Signed-off-by: Jens Axboe --- Documentation/ABI/testing/sysfs-block | 13 ----- block/genhd.c | 71 --------------------------- include/linux/genhd.h | 4 -- 3 files changed, 88 deletions(-) diff --git a/Documentation/ABI/testing/sysfs-block b/Documentation/ABI/testing/sysfs-block index 2b5d56127fce4..c1eb41cb98760 100644 --- a/Documentation/ABI/testing/sysfs-block +++ b/Documentation/ABI/testing/sysfs-block @@ -206,16 +206,3 @@ Description: when a discarded area is read the discard_zeroes_data parameter will be set to one. Otherwise it will be 0 and the result of reading a discarded area is undefined. -What: /sys/block//alias -Date: Aug 2011 -Contact: Nao Nishijima -Description: - A raw device name of a disk does not always point a same disk - each boot-up time. Therefore, users have to use persistent - device names, which udev creates when the kernel finds a disk, - instead of raw device name. However, kernel doesn't show those - persistent names on its messages (e.g. dmesg). - This file can store an alias of the disk and it would be - appeared in kernel messages if it is set. A disk can have an - alias which length is up to 255bytes. Users can use alphabets, - numbers, "-" and "_" in alias name. This file is writeonce. diff --git a/block/genhd.c b/block/genhd.c index 9253839714ff9..02e9fca808256 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -19,7 +19,6 @@ #include #include #include -#include #include "blk.h" @@ -916,74 +915,6 @@ static int __init genhd_device_init(void) subsys_initcall(genhd_device_init); -static ssize_t alias_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct gendisk *disk = dev_to_disk(dev); - ssize_t ret = 0; - - if (disk->alias) - ret = snprintf(buf, ALIAS_LEN, "%s\n", disk->alias); - return ret; -} - -static ssize_t alias_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct gendisk *disk = dev_to_disk(dev); - char *alias; - char *envp[] = { NULL, NULL }; - unsigned char c; - int i; - ssize_t ret = count; - - if (!count) - return -EINVAL; - - if (count >= ALIAS_LEN) { - printk(KERN_ERR "alias: alias is too long\n"); - return -EINVAL; - } - - /* Validation check */ - for (i = 0; i < count; i++) { - c = buf[i]; - if (i == count - 1 && c == '\n') - break; - if (!isalnum(c) && c != '_' && c != '-') { - printk(KERN_ERR "alias: invalid alias\n"); - return -EINVAL; - } - } - - if (disk->alias) { - printk(KERN_INFO "alias: %s is already assigned (%s)\n", - disk->disk_name, disk->alias); - return -EINVAL; - } - - alias = kasprintf(GFP_KERNEL, "%s", buf); - if (!alias) - return -ENOMEM; - - if (alias[count - 1] == '\n') - alias[count - 1] = '\0'; - - envp[0] = kasprintf(GFP_KERNEL, "ALIAS=%s", alias); - if (!envp[0]) { - kfree(alias); - return -ENOMEM; - } - - disk->alias = alias; - printk(KERN_INFO "alias: assigned %s to %s\n", alias, disk->disk_name); - - kobject_uevent_env(&dev->kobj, KOBJ_ADD, envp); - - kfree(envp[0]); - return ret; -} - static ssize_t disk_range_show(struct device *dev, struct device_attribute *attr, char *buf) { @@ -1043,7 +974,6 @@ static ssize_t disk_discard_alignment_show(struct device *dev, return sprintf(buf, "%d\n", queue_discard_alignment(disk->queue)); } -static DEVICE_ATTR(alias, S_IRUGO|S_IWUSR, alias_show, alias_store); static DEVICE_ATTR(range, S_IRUGO, disk_range_show, NULL); static DEVICE_ATTR(ext_range, S_IRUGO, disk_ext_range_show, NULL); static DEVICE_ATTR(removable, S_IRUGO, disk_removable_show, NULL); @@ -1066,7 +996,6 @@ static struct device_attribute dev_attr_fail_timeout = #endif static struct attribute *disk_attrs[] = { - &dev_attr_alias.attr, &dev_attr_range.attr, &dev_attr_ext_range.attr, &dev_attr_removable.attr, diff --git a/include/linux/genhd.h b/include/linux/genhd.h index 9de31bc98c880..6d18f3531f180 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -21,8 +21,6 @@ #define dev_to_part(device) container_of((device), struct hd_struct, __dev) #define disk_to_dev(disk) (&(disk)->part0.__dev) #define part_to_dev(part) (&((part)->__dev)) -#define alias_name(disk) ((disk)->alias ? (disk)->alias : \ - (disk)->disk_name) extern struct device_type part_type; extern struct kobject *block_depr; @@ -60,7 +58,6 @@ enum { #define DISK_MAX_PARTS 256 #define DISK_NAME_LEN 32 -#define ALIAS_LEN 256 #include #include @@ -166,7 +163,6 @@ struct gendisk { * disks that can't be partitioned. */ char disk_name[DISK_NAME_LEN]; /* name of major driver */ - char *alias; /* alias name of disk */ char *(*devnode)(struct gendisk *gd, mode_t *mode); unsigned int events; /* supported events */ -- GitLab From 952f6d1386b21c5e8db346b805380bf2432e5e9b Mon Sep 17 00:00:00 2001 From: MyungJoo Ham Date: Thu, 10 Nov 2011 10:16:23 +0100 Subject: [PATCH 0170/3089] PM / devfreq: Remove compiler error after irq.h update Added and to avoid a compiler error because linux/irq.h no longer includes linux/module.h after Linux 3.2. Signed-off-by: MyungJoo Ham Signed-off-by: Kyungmin Park Signed-off-by: Rafael J. Wysocki --- drivers/devfreq/devfreq.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c index 5d15b812377bc..d0659253387a3 100644 --- a/drivers/devfreq/devfreq.c +++ b/drivers/devfreq/devfreq.c @@ -15,7 +15,9 @@ #include #include #include +#include #include +#include #include #include #include -- GitLab From 1ec454baf1245df4fdb5dae728da3363630ce6de Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Thu, 20 Oct 2011 23:01:09 +0900 Subject: [PATCH 0171/3089] x86, perf: Add a build-time sanity test to the x86 decoder Add a sanity test of x86 insn decoder against a stream of randomly generated input, at build time. This test is also able to reproduce any bug that might trigger by allowing the passing of random-seed and iteration-number to the test, or by passing input which has invalid byte code. Changes in V2: - Code cleanup. - Show how to reproduce the error by insn_sanity test. Signed-off-by: Masami Hiramatsu Cc: acme@redhat.com Cc: ming.m.lin@intel.com Cc: robert.richter@amd.com Cc: ravitillo@lbl.gov Cc: yrl.pp-manager.tt@hitachi.com Cc: Andi Kleen Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Andi Kleen Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/20111020140109.20938.92572.stgit@localhost.localdomain Signed-off-by: Ingo Molnar --- arch/x86/include/asm/insn.h | 7 + arch/x86/tools/Makefile | 10 +- arch/x86/tools/insn_sanity.c | 275 +++++++++++++++++++++++++++++++++++ 3 files changed, 291 insertions(+), 1 deletion(-) create mode 100644 arch/x86/tools/insn_sanity.c diff --git a/arch/x86/include/asm/insn.h b/arch/x86/include/asm/insn.h index 88c765e16410b..74df3f1eddfd5 100644 --- a/arch/x86/include/asm/insn.h +++ b/arch/x86/include/asm/insn.h @@ -137,6 +137,13 @@ static inline int insn_is_avx(struct insn *insn) return (insn->vex_prefix.value != 0); } +/* Ensure this instruction is decoded completely */ +static inline int insn_complete(struct insn *insn) +{ + return insn->opcode.got && insn->modrm.got && insn->sib.got && + insn->displacement.got && insn->immediate.got; +} + static inline insn_byte_t insn_vex_m_bits(struct insn *insn) { if (insn->vex_prefix.nbytes == 2) /* 2 bytes VEX */ diff --git a/arch/x86/tools/Makefile b/arch/x86/tools/Makefile index f82082677337e..3255c3df67f45 100644 --- a/arch/x86/tools/Makefile +++ b/arch/x86/tools/Makefile @@ -18,14 +18,22 @@ chkobjdump = $(srctree)/arch/x86/tools/chkobjdump.awk quiet_cmd_posttest = TEST $@ cmd_posttest = ($(OBJDUMP) -v | $(AWK) -f $(chkobjdump)) || $(OBJDUMP) -d -j .text $(objtree)/vmlinux | $(AWK) -f $(distill_awk) | $(obj)/test_get_len $(posttest_64bit) $(posttest_verbose) -posttest: $(obj)/test_get_len vmlinux +quiet_cmd_sanitytest = TEST $@ + cmd_sanitytest = $(obj)/insn_sanity $(posttest_64bit) -m 1000000 + +posttest: $(obj)/test_get_len vmlinux $(obj)/insn_sanity $(call cmd,posttest) + $(call cmd,sanitytest) hostprogs-y := test_get_len +hostprogs-y := insn_sanity # -I needed for generated C source and C source which in the kernel tree. HOSTCFLAGS_test_get_len.o := -Wall -I$(objtree)/arch/x86/lib/ -I$(srctree)/arch/x86/include/ -I$(srctree)/arch/x86/lib/ -I$(srctree)/include/ +HOSTCFLAGS_insn_sanity.o := -Wall -I$(objtree)/arch/x86/lib/ -I$(srctree)/arch/x86/include/ -I$(srctree)/arch/x86/lib/ -I$(srctree)/include/ + # Dependencies are also needed. $(obj)/test_get_len.o: $(srctree)/arch/x86/lib/insn.c $(srctree)/arch/x86/lib/inat.c $(srctree)/arch/x86/include/asm/inat_types.h $(srctree)/arch/x86/include/asm/inat.h $(srctree)/arch/x86/include/asm/insn.h $(objtree)/arch/x86/lib/inat-tables.c +$(obj)/insn_sanity.o: $(srctree)/arch/x86/lib/insn.c $(srctree)/arch/x86/lib/inat.c $(srctree)/arch/x86/include/asm/inat_types.h $(srctree)/arch/x86/include/asm/inat.h $(srctree)/arch/x86/include/asm/insn.h $(objtree)/arch/x86/lib/inat-tables.c diff --git a/arch/x86/tools/insn_sanity.c b/arch/x86/tools/insn_sanity.c new file mode 100644 index 0000000000000..334d9de7d0cac --- /dev/null +++ b/arch/x86/tools/insn_sanity.c @@ -0,0 +1,275 @@ +/* + * x86 decoder sanity test - based on test_get_insn.c + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. + * + * Copyright (C) IBM Corporation, 2009 + * Copyright (C) Hitachi, Ltd., 2011 + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#define unlikely(cond) (cond) +#define ARRAY_SIZE(a) (sizeof(a)/sizeof(a[0])) + +#include +#include +#include + +/* + * Test of instruction analysis against tampering. + * Feed random binary to instruction decoder and ensure not to + * access out-of-instruction-buffer. + */ + +#define DEFAULT_MAX_ITER 10000 +#define INSN_NOP 0x90 + +static const char *prog; /* Program name */ +static int verbose; /* Verbosity */ +static int x86_64; /* x86-64 bit mode flag */ +static unsigned int seed; /* Random seed */ +static unsigned long iter_start; /* Start of iteration number */ +static unsigned long iter_end = DEFAULT_MAX_ITER; /* End of iteration number */ +static FILE *input_file; /* Input file name */ + +static void usage(const char *err) +{ + if (err) + fprintf(stderr, "Error: %s\n\n", err); + fprintf(stderr, "Usage: %s [-y|-n|-v] [-s seed[,no]] [-m max] [-i input]\n", prog); + fprintf(stderr, "\t-y 64bit mode\n"); + fprintf(stderr, "\t-n 32bit mode\n"); + fprintf(stderr, "\t-v Verbose mode\n"); + fprintf(stderr, "\t-s Give a random seed (and iteration number)\n"); + fprintf(stderr, "\t-m Give a maximum iteration number\n"); + fprintf(stderr, "\t-i Give an input file with decoded binary\n"); + exit(1); +} + +static void dump_field(FILE *fp, const char *name, const char *indent, + struct insn_field *field) +{ + fprintf(fp, "%s.%s = {\n", indent, name); + fprintf(fp, "%s\t.value = %d, bytes[] = {%x, %x, %x, %x},\n", + indent, field->value, field->bytes[0], field->bytes[1], + field->bytes[2], field->bytes[3]); + fprintf(fp, "%s\t.got = %d, .nbytes = %d},\n", indent, + field->got, field->nbytes); +} + +static void dump_insn(FILE *fp, struct insn *insn) +{ + fprintf(fp, "Instruction = {\n"); + dump_field(fp, "prefixes", "\t", &insn->prefixes); + dump_field(fp, "rex_prefix", "\t", &insn->rex_prefix); + dump_field(fp, "vex_prefix", "\t", &insn->vex_prefix); + dump_field(fp, "opcode", "\t", &insn->opcode); + dump_field(fp, "modrm", "\t", &insn->modrm); + dump_field(fp, "sib", "\t", &insn->sib); + dump_field(fp, "displacement", "\t", &insn->displacement); + dump_field(fp, "immediate1", "\t", &insn->immediate1); + dump_field(fp, "immediate2", "\t", &insn->immediate2); + fprintf(fp, "\t.attr = %x, .opnd_bytes = %d, .addr_bytes = %d,\n", + insn->attr, insn->opnd_bytes, insn->addr_bytes); + fprintf(fp, "\t.length = %d, .x86_64 = %d, .kaddr = %p}\n", + insn->length, insn->x86_64, insn->kaddr); +} + +static void dump_stream(FILE *fp, const char *msg, unsigned long nr_iter, + unsigned char *insn_buf, struct insn *insn) +{ + int i; + + fprintf(fp, "%s:\n", msg); + + dump_insn(stderr, insn); + + fprintf(fp, "You can reproduce this with below command(s);\n"); + + /* Input a decoded instruction sequence directly */ + fprintf(fp, " $ echo "); + for (i = 0; i < MAX_INSN_SIZE; i++) + fprintf(fp, " %02x", insn_buf[i]); + fprintf(fp, " | %s -i -\n", prog); + + if (!input_file) { + fprintf(fp, "Or \n"); + /* Give a seed and iteration number */ + fprintf(fp, " $ %s -s 0x%x,%lu\n", prog, seed, nr_iter); + } +} + +static void init_random_seed(void) +{ + int fd; + + fd = open("/dev/urandom", O_RDONLY); + if (fd < 0) + goto fail; + + if (read(fd, &seed, sizeof(seed)) != sizeof(seed)) + goto fail; + + close(fd); + return; +fail: + usage("Failed to open /dev/urandom"); +} + +/* Read given instruction sequence from the input file */ +static int read_next_insn(unsigned char *insn_buf) +{ + char buf[256] = "", *tmp; + int i; + + tmp = fgets(buf, ARRAY_SIZE(buf), input_file); + if (tmp == NULL || feof(input_file)) + return 0; + + for (i = 0; i < MAX_INSN_SIZE; i++) { + insn_buf[i] = (unsigned char)strtoul(tmp, &tmp, 16); + if (*tmp != ' ') + break; + } + + return i; +} + +static int generate_insn(unsigned char *insn_buf) +{ + int i; + + if (input_file) + return read_next_insn(insn_buf); + + /* Fills buffer with random binary up to MAX_INSN_SIZE */ + for (i = 0; i < MAX_INSN_SIZE - 1; i += 2) + *(unsigned short *)(&insn_buf[i]) = random() & 0xffff; + + while (i < MAX_INSN_SIZE) + insn_buf[i++] = random() & 0xff; + + return i; +} + +static void parse_args(int argc, char **argv) +{ + int c; + char *tmp = NULL; + int set_seed = 0; + + prog = argv[0]; + while ((c = getopt(argc, argv, "ynvs:m:i:")) != -1) { + switch (c) { + case 'y': + x86_64 = 1; + break; + case 'n': + x86_64 = 0; + break; + case 'v': + verbose = 1; + break; + case 'i': + if (strcmp("-", optarg) == 0) + input_file = stdin; + else + input_file = fopen(optarg, "r"); + if (!input_file) + usage("Failed to open input file"); + break; + case 's': + seed = (unsigned int)strtoul(optarg, &tmp, 0); + if (*tmp == ',') { + optarg = tmp + 1; + iter_start = strtoul(optarg, &tmp, 0); + } + if (*tmp != '\0' || tmp == optarg) + usage("Failed to parse seed"); + set_seed = 1; + break; + case 'm': + iter_end = strtoul(optarg, &tmp, 0); + if (*tmp != '\0' || tmp == optarg) + usage("Failed to parse max_iter"); + break; + default: + usage(NULL); + } + } + + /* Check errors */ + if (iter_end < iter_start) + usage("Max iteration number must be bigger than iter-num"); + + if (set_seed && input_file) + usage("Don't use input file (-i) with random seed (-s)"); + + /* Initialize random seed */ + if (!input_file) { + if (!set_seed) /* No seed is given */ + init_random_seed(); + srand(seed); + } +} + +int main(int argc, char **argv) +{ + struct insn insn; + int insns = 0; + int errors = 0; + unsigned long i; + unsigned char insn_buf[MAX_INSN_SIZE * 2]; + + parse_args(argc, argv); + + /* Prepare stop bytes with NOPs */ + memset(insn_buf + MAX_INSN_SIZE, INSN_NOP, MAX_INSN_SIZE); + + for (i = 0; i < iter_end; i++) { + if (generate_insn(insn_buf) <= 0) + break; + + if (i < iter_start) /* Skip to given iteration number */ + continue; + + /* Decode an instruction */ + insn_init(&insn, insn_buf, x86_64); + insn_get_length(&insn); + + if (verbose && !insn_complete(&insn)) + dump_stream(stdout, "Info: Found an undecodable input", i, insn_buf, &insn); + + if (insn.next_byte <= insn.kaddr || + insn.kaddr + MAX_INSN_SIZE < insn.next_byte) { + /* Access out-of-range memory */ + dump_stream(stdout, "Error: Found an access violation", i, insn_buf, &insn); + errors++; + } + insns++; + } + + fprintf(stdout, "%s: decoded and checked %d %s instructions with %d errors (seed:0x%x)\n", (errors) ? "Failure" : "Success", insns, (input_file) ? "given" : "random", errors, seed); + + return errors ? 1 : 0; +} -- GitLab From 98d97019c88bd832da1457729739cf739ece493f Mon Sep 17 00:00:00 2001 From: Jassi Brar Date: Thu, 10 Nov 2011 17:19:07 +0530 Subject: [PATCH 0172/3089] MAINTAINERS: Drop inactive Samsung ASoC maintainer Signed-off-by: Jassi Brar Signed-off-by: Mark Brown --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index c802e5fa2d11b..fd7e441b5ea7f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -5648,7 +5648,6 @@ F: drivers/media/video/*7146* F: include/media/*7146* SAMSUNG AUDIO (ASoC) DRIVERS -M: Jassi Brar M: Sangbeom Kim L: alsa-devel@alsa-project.org (moderated for non-subscribers) S: Supported -- GitLab From cf8ff6b6ab0e99dd3058852f4ec76a6140abadec Mon Sep 17 00:00:00 2001 From: Feng Tang Date: Thu, 10 Nov 2011 13:42:02 +0000 Subject: [PATCH 0173/3089] x86/platform: Add a wallclock_init func to x86_platforms ops Some wall clock devices use MMIO based HW register, this new function will give them a chance to do some initialization work before their get/set_time service get called. Signed-off-by: Feng Tang Signed-off-by: Jacob Pan Signed-off-by: Alan Cox Signed-off-by: Dirk Brandewie Signed-off-by: Ingo Molnar --- arch/x86/include/asm/x86_init.h | 2 ++ arch/x86/kernel/setup.c | 2 ++ arch/x86/kernel/x86_init.c | 2 ++ 3 files changed, 6 insertions(+) diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h index d3d859035af9e..f864fbe474c6b 100644 --- a/arch/x86/include/asm/x86_init.h +++ b/arch/x86/include/asm/x86_init.h @@ -152,6 +152,7 @@ struct x86_cpuinit_ops { /** * struct x86_platform_ops - platform specific runtime functions * @calibrate_tsc: calibrate TSC + * @wallclock_init: init the wallclock device * @get_wallclock: get time from HW clock like RTC etc. * @set_wallclock: set time back to HW clock * @is_untracked_pat_range exclude from PAT logic @@ -160,6 +161,7 @@ struct x86_cpuinit_ops { */ struct x86_platform_ops { unsigned long (*calibrate_tsc)(void); + void (*wallclock_init)(void); unsigned long (*get_wallclock)(void); int (*set_wallclock)(unsigned long nowtime); void (*iommu_shutdown)(void); diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index afaf38447ef5f..cf0ef986cb6df 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -1045,6 +1045,8 @@ void __init setup_arch(char **cmdline_p) x86_init.timers.wallclock_init(); + x86_platform.wallclock_init(); + mcheck_init(); arch_init_ideal_nops(); diff --git a/arch/x86/kernel/x86_init.c b/arch/x86/kernel/x86_init.c index 6f164bd5e14d1..701c7be442f1c 100644 --- a/arch/x86/kernel/x86_init.c +++ b/arch/x86/kernel/x86_init.c @@ -27,6 +27,7 @@ void __init x86_init_uint_noop(unsigned int unused) { } void __init x86_init_pgd_noop(pgd_t *unused) { } int __init iommu_init_noop(void) { return 0; } void iommu_shutdown_noop(void) { } +void wallclock_init_noop(void) { } /* * The platform setup functions are preset with the default functions @@ -97,6 +98,7 @@ static int default_i8042_detect(void) { return 1; }; struct x86_platform_ops x86_platform = { .calibrate_tsc = native_calibrate_tsc, + .wallclock_init = wallclock_init_noop, .get_wallclock = mach_get_cmos_time, .set_wallclock = mach_set_rtc_mmss, .iommu_shutdown = iommu_shutdown_noop, -- GitLab From bb84ac2d3a603f8f6c7cc553a260e8ceaf871df2 Mon Sep 17 00:00:00 2001 From: Jacob Pan Date: Thu, 10 Nov 2011 13:42:21 +0000 Subject: [PATCH 0174/3089] x86/apic: Do not clear nr_irqs_gsi if no legacy irqs nr_legacy_irqs is set in probe_nr_irqs_gsi, we should not clear it after that. Otherwise, the result is that MSI irqs will be allocated from the wrong range for the systems without legacy PIC. Signed-off-by: Jacob Pan Signed-off-by: Dirk Brandewie Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 3c31fa98af6dc..841b8da40525a 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -193,10 +193,8 @@ int __init arch_early_irq_init(void) struct irq_cfg *cfg; int count, node, i; - if (!legacy_pic->nr_legacy_irqs) { - nr_irqs_gsi = 0; + if (!legacy_pic->nr_legacy_irqs) io_apic_irqs = ~0UL; - } for (i = 0; i < nr_ioapics; i++) { ioapics[i].saved_registers = -- GitLab From 1ade93efd0a3dda5b0c0afda8ab8f4bd12938c1b Mon Sep 17 00:00:00 2001 From: Jacob Pan Date: Thu, 10 Nov 2011 13:42:40 +0000 Subject: [PATCH 0175/3089] x86/apic: Allow use of lapic timer early calibration result lapic timer calibration can be combined with tsc in platform specific calibration functions. if such calibration result is obtained early, we can skip the redundant calibration loops. Signed-off-by: Jacob Pan Signed-off-by: Jacob Pan Signed-off-by: Alan Cox Signed-off-by: Dirk Brandewie Signed-off-by: Ingo Molnar --- arch/x86/include/asm/apic.h | 1 + arch/x86/kernel/apic/apic.c | 33 ++++++++++++++++++++++++++------- 2 files changed, 27 insertions(+), 7 deletions(-) diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h index 9b7273cb21937..1a6c09af048fb 100644 --- a/arch/x86/include/asm/apic.h +++ b/arch/x86/include/asm/apic.h @@ -49,6 +49,7 @@ extern unsigned int apic_verbosity; extern int local_apic_timer_c2_ok; extern int disable_apic; +extern unsigned int lapic_timer_frequency; #ifdef CONFIG_SMP extern void __inquire_remote_apic(int apicid); diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index a2fd72e0ab35b..f98d84caf94cf 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -186,7 +186,7 @@ static struct resource lapic_resource = { .flags = IORESOURCE_MEM | IORESOURCE_BUSY, }; -static unsigned int calibration_result; +unsigned int lapic_timer_frequency = 0; static void apic_pm_activate(void); @@ -454,7 +454,7 @@ static void lapic_timer_setup(enum clock_event_mode mode, switch (mode) { case CLOCK_EVT_MODE_PERIODIC: case CLOCK_EVT_MODE_ONESHOT: - __setup_APIC_LVTT(calibration_result, + __setup_APIC_LVTT(lapic_timer_frequency, mode != CLOCK_EVT_MODE_PERIODIC, 1); break; case CLOCK_EVT_MODE_UNUSED: @@ -638,6 +638,25 @@ static int __init calibrate_APIC_clock(void) long delta, deltatsc; int pm_referenced = 0; + /** + * check if lapic timer has already been calibrated by platform + * specific routine, such as tsc calibration code. if so, we just fill + * in the clockevent structure and return. + */ + + if (lapic_timer_frequency) { + apic_printk(APIC_VERBOSE, "lapic timer already calibrated %d\n", + lapic_timer_frequency); + lapic_clockevent.mult = div_sc(lapic_timer_frequency/APIC_DIVISOR, + TICK_NSEC, lapic_clockevent.shift); + lapic_clockevent.max_delta_ns = + clockevent_delta2ns(0x7FFFFF, &lapic_clockevent); + lapic_clockevent.min_delta_ns = + clockevent_delta2ns(0xF, &lapic_clockevent); + lapic_clockevent.features &= ~CLOCK_EVT_FEAT_DUMMY; + return 0; + } + local_irq_disable(); /* Replace the global interrupt handler */ @@ -679,12 +698,12 @@ static int __init calibrate_APIC_clock(void) lapic_clockevent.min_delta_ns = clockevent_delta2ns(0xF, &lapic_clockevent); - calibration_result = (delta * APIC_DIVISOR) / LAPIC_CAL_LOOPS; + lapic_timer_frequency = (delta * APIC_DIVISOR) / LAPIC_CAL_LOOPS; apic_printk(APIC_VERBOSE, "..... delta %ld\n", delta); apic_printk(APIC_VERBOSE, "..... mult: %u\n", lapic_clockevent.mult); apic_printk(APIC_VERBOSE, "..... calibration result: %u\n", - calibration_result); + lapic_timer_frequency); if (cpu_has_tsc) { apic_printk(APIC_VERBOSE, "..... CPU clock speed is " @@ -695,13 +714,13 @@ static int __init calibrate_APIC_clock(void) apic_printk(APIC_VERBOSE, "..... host bus clock speed is " "%u.%04u MHz.\n", - calibration_result / (1000000 / HZ), - calibration_result % (1000000 / HZ)); + lapic_timer_frequency / (1000000 / HZ), + lapic_timer_frequency % (1000000 / HZ)); /* * Do a sanity check on the APIC calibration result */ - if (calibration_result < (1000000 / HZ)) { + if (lapic_timer_frequency < (1000000 / HZ)) { local_irq_enable(); pr_warning("APIC frequency too slow, disabling apic timer\n"); return -1; -- GitLab From 0a9153261d54c432bc0bdc88607f24c835ac729c Mon Sep 17 00:00:00 2001 From: Dirk Brandewie Date: Thu, 10 Nov 2011 13:42:53 +0000 Subject: [PATCH 0176/3089] x86/mrst: Add support for Penwell clock calibration Signed-off-by: Dirk Brandewie Signed-off-by: Alan Cox Signed-off-by: Ingo Molnar --- arch/x86/include/asm/mrst.h | 7 +++++++ arch/x86/platform/mrst/mrst.c | 33 ++++++++++++++++++++++++++++----- 2 files changed, 35 insertions(+), 5 deletions(-) diff --git a/arch/x86/include/asm/mrst.h b/arch/x86/include/asm/mrst.h index 719f00b28ff53..e6283129c8210 100644 --- a/arch/x86/include/asm/mrst.h +++ b/arch/x86/include/asm/mrst.h @@ -44,6 +44,13 @@ enum mrst_timer_options { extern enum mrst_timer_options mrst_timer_options; +/* + * Penwell uses spread spectrum clock, so the freq number is not exactly + * the same as reported by MSR based on SDM. + */ +#define PENWELL_FSB_FREQ_83SKU 83200 +#define PENWELL_FSB_FREQ_100SKU 99840 + #define SFI_MTMR_MAX_NUM 8 #define SFI_MRTC_MAX 8 diff --git a/arch/x86/platform/mrst/mrst.c b/arch/x86/platform/mrst/mrst.c index 6ed7afdaf4afa..b7f14e5b2c661 100644 --- a/arch/x86/platform/mrst/mrst.c +++ b/arch/x86/platform/mrst/mrst.c @@ -187,11 +187,34 @@ int __init sfi_parse_mrtc(struct sfi_table_header *table) static unsigned long __init mrst_calibrate_tsc(void) { unsigned long flags, fast_calibrate; - - local_irq_save(flags); - fast_calibrate = apbt_quick_calibrate(); - local_irq_restore(flags); - + if (__mrst_cpu_chip == MRST_CPU_CHIP_PENWELL) { + u32 lo, hi, ratio, fsb; + + rdmsr(MSR_IA32_PERF_STATUS, lo, hi); + pr_debug("IA32 perf status is 0x%x, 0x%0x\n", lo, hi); + ratio = (hi >> 8) & 0x1f; + pr_debug("ratio is %d\n", ratio); + if (!ratio) { + pr_err("read a zero ratio, should be incorrect!\n"); + pr_err("force tsc ratio to 16 ...\n"); + ratio = 16; + } + rdmsr(MSR_FSB_FREQ, lo, hi); + if ((lo & 0x7) == 0x7) + fsb = PENWELL_FSB_FREQ_83SKU; + else + fsb = PENWELL_FSB_FREQ_100SKU; + fast_calibrate = ratio * fsb; + pr_debug("read penwell tsc %lu khz\n", fast_calibrate); + lapic_timer_frequency = fsb * 1000 / HZ; + /* mark tsc clocksource as reliable */ + set_cpu_cap(&boot_cpu_data, X86_FEATURE_TSC_RELIABLE); + } else { + local_irq_save(flags); + fast_calibrate = apbt_quick_calibrate(); + local_irq_restore(flags); + } + if (fast_calibrate) return fast_calibrate; -- GitLab From 064a59b6dd1f341cc478c212bb436e3da9cb8d04 Mon Sep 17 00:00:00 2001 From: Jacob Pan Date: Thu, 10 Nov 2011 13:43:05 +0000 Subject: [PATCH 0177/3089] x86/mrst: Avoid reporting wrong nmi status Moorestown/Medfield platform does not have port 0x61 to report NMI status, nor does it have external NMI sources. The only NMI sources are from lapic, as results of perf counter overflow or IPI, e.g. NMI watchdog or spin lock debug. Reading port 0x61 on Moorestown will return 0xff which misled NMI handlers to false critical errors such memory parity error. The subsequent ioport access for NMI handling can also cause undefined behavior on Moorestown. This patch allows kernel process NMI due to watchdog or backrace dump without unnecessary hangs. Signed-off-by: Jacob Pan Signed-off-by: Ingo Molnar [hand applied] Signed-off-by: Alan Cox --- arch/x86/include/asm/mach_traps.h | 2 +- arch/x86/include/asm/x86_init.h | 1 + arch/x86/kernel/nmi.c | 2 +- arch/x86/kernel/x86_init.c | 2 ++ arch/x86/platform/mrst/mrst.c | 13 +++++++++++++ 5 files changed, 18 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/mach_traps.h b/arch/x86/include/asm/mach_traps.h index 72a8b52e7dfd0..a01e7ec7d2377 100644 --- a/arch/x86/include/asm/mach_traps.h +++ b/arch/x86/include/asm/mach_traps.h @@ -17,7 +17,7 @@ #define NMI_REASON_CLEAR_IOCHK 0x08 #define NMI_REASON_CLEAR_MASK 0x0f -static inline unsigned char get_nmi_reason(void) +static inline unsigned char default_get_nmi_reason(void) { return inb(NMI_REASON_PORT); } diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h index f864fbe474c6b..1971e652d24be 100644 --- a/arch/x86/include/asm/x86_init.h +++ b/arch/x86/include/asm/x86_init.h @@ -167,6 +167,7 @@ struct x86_platform_ops { void (*iommu_shutdown)(void); bool (*is_untracked_pat_range)(u64 start, u64 end); void (*nmi_init)(void); + unsigned char (*get_nmi_reason)(void); int (*i8042_detect)(void); }; diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c index b9c8628974af3..27d1e7cbdb6c0 100644 --- a/arch/x86/kernel/nmi.c +++ b/arch/x86/kernel/nmi.c @@ -348,7 +348,7 @@ static notrace __kprobes void default_do_nmi(struct pt_regs *regs) /* Non-CPU-specific NMI: NMI sources can be processed on any CPU */ raw_spin_lock(&nmi_reason_lock); - reason = get_nmi_reason(); + reason = x86_platform.get_nmi_reason(); if (reason & NMI_REASON_MASK) { if (reason & NMI_REASON_SERR) diff --git a/arch/x86/kernel/x86_init.c b/arch/x86/kernel/x86_init.c index 701c7be442f1c..c1d6cd549397a 100644 --- a/arch/x86/kernel/x86_init.c +++ b/arch/x86/kernel/x86_init.c @@ -21,6 +21,7 @@ #include #include #include +#include void __cpuinit x86_init_noop(void) { } void __init x86_init_uint_noop(unsigned int unused) { } @@ -104,6 +105,7 @@ struct x86_platform_ops x86_platform = { .iommu_shutdown = iommu_shutdown_noop, .is_untracked_pat_range = is_ISA_range, .nmi_init = default_nmi_init, + .get_nmi_reason = default_get_nmi_reason, .i8042_detect = default_i8042_detect }; diff --git a/arch/x86/platform/mrst/mrst.c b/arch/x86/platform/mrst/mrst.c index b7f14e5b2c661..9b9ee292c107f 100644 --- a/arch/x86/platform/mrst/mrst.c +++ b/arch/x86/platform/mrst/mrst.c @@ -276,6 +276,17 @@ static void mrst_reboot(void) intel_scu_ipc_simple_command(0xf1, 0); } +/* + * Moorestown does not have external NMI source nor port 0x61 to report + * NMI status. The possible NMI sources are from pmu as a result of NMI + * watchdog or lock debug. Reading io port 0x61 results in 0xff which + * misled NMI handler. + */ +static unsigned char mrst_get_nmi_reason(void) +{ + return 0; +} + /* * Moorestown specific x86_init function overrides and early setup * calls. @@ -297,6 +308,8 @@ void __init x86_mrst_early_setup(void) x86_platform.calibrate_tsc = mrst_calibrate_tsc; x86_platform.i8042_detect = mrst_i8042_detect; x86_init.timers.wallclock_init = mrst_rtc_init; + x86_platform.get_nmi_reason = mrst_get_nmi_reason; + x86_init.pci.init = pci_mrst_init; x86_init.pci.fixup_irqs = x86_init_noop; -- GitLab From 181a9d795ab763b03b15f73402691abde98d1803 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 4 Nov 2011 21:25:01 +0300 Subject: [PATCH 0178/3089] [SCSI] mpt2sas: add missing allocation. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit There was supposed to be a kzalloc() here and the compiler complained about it. mpt2sas_scsih.c: In function ‘mpt2sas_scsih_reset_handler’: mpt2sas_scsih.c:2807:21: warning: ‘fw_event’ may be used uninitialized in this function [-Wuninitialized] Signed-off-by: Dan Carpenter Acked-by: "Nandigama, Nagalakshmi" Signed-off-by: James Bottomley --- drivers/scsi/mpt2sas/mpt2sas_scsih.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/scsi/mpt2sas/mpt2sas_scsih.c b/drivers/scsi/mpt2sas/mpt2sas_scsih.c index 8889b1babcaca..4e041f6d808cd 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_scsih.c +++ b/drivers/scsi/mpt2sas/mpt2sas_scsih.c @@ -2802,6 +2802,11 @@ _scsih_error_recovery_delete_devices(struct MPT2SAS_ADAPTER *ioc) if (ioc->is_driver_loading) return; + + fw_event = kzalloc(sizeof(struct fw_event_work), GFP_ATOMIC); + if (!fw_event) + return; + fw_event->event = MPT2SAS_REMOVE_UNRESPONDING_DEVICES; fw_event->ioc = ioc; _scsih_fw_event_add(ioc, fw_event); -- GitLab From 4415e63b13c68c2f56d16d400a1ae345f68cf655 Mon Sep 17 00:00:00 2001 From: Keith Packard Date: Wed, 9 Nov 2011 09:57:50 -0800 Subject: [PATCH 0179/3089] drm/i915: Module parameters using '-1' as default must be signed type Testing i915_panel_use_ssc for the default value was broken, so the driver would never autodetect the correct value. Signed-off-by: Keith Packard Reviewed-by: Michel Alexandre Salim Tested-by: Michel Alexandre Salim Cc: stable@kernel.org --- drivers/gpu/drm/i915/i915_drv.c | 4 ++-- drivers/gpu/drm/i915/i915_drv.h | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 548e04bade3c3..13488be4ae4b0 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -67,7 +67,7 @@ module_param_named(i915_enable_rc6, i915_enable_rc6, int, 0600); MODULE_PARM_DESC(i915_enable_rc6, "Enable power-saving render C-state 6 (default: true)"); -unsigned int i915_enable_fbc __read_mostly = -1; +int i915_enable_fbc __read_mostly = -1; module_param_named(i915_enable_fbc, i915_enable_fbc, int, 0600); MODULE_PARM_DESC(i915_enable_fbc, "Enable frame buffer compression for power savings " @@ -79,7 +79,7 @@ MODULE_PARM_DESC(lvds_downclock, "Use panel (LVDS/eDP) downclocking for power savings " "(default: false)"); -unsigned int i915_panel_use_ssc __read_mostly = -1; +int i915_panel_use_ssc __read_mostly = -1; module_param_named(lvds_use_ssc, i915_panel_use_ssc, int, 0600); MODULE_PARM_DESC(lvds_use_ssc, "Use Spread Spectrum Clock with panels [LVDS/eDP] " diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index d2da91f902528..4a9c1b9798049 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -1000,10 +1000,10 @@ extern int i915_panel_ignore_lid __read_mostly; extern unsigned int i915_powersave __read_mostly; extern unsigned int i915_semaphores __read_mostly; extern unsigned int i915_lvds_downclock __read_mostly; -extern unsigned int i915_panel_use_ssc __read_mostly; +extern int i915_panel_use_ssc __read_mostly; extern int i915_vbt_sdvo_panel_type __read_mostly; extern unsigned int i915_enable_rc6 __read_mostly; -extern unsigned int i915_enable_fbc __read_mostly; +extern int i915_enable_fbc __read_mostly; extern bool i915_enable_hangcheck __read_mostly; extern int i915_suspend(struct drm_device *dev, pm_message_t state); -- GitLab From 6fd36ba02132c61f67ebefff77fe710bd38ba95a Mon Sep 17 00:00:00 2001 From: Mathias Nyman Date: Thu, 10 Nov 2011 13:45:24 +0000 Subject: [PATCH 0180/3089] x86, ioapic: Only print ioapic debug information for IRQs belonging to an ioapic chip with "apic=verbose" the print_IO_APIC() function tries to print IRQ to pin mappings for every active irq. It assumes chip_data is of type irq_cfg and may cause an oops if not. As the print_IO_APIC() is called from a late_initcall other chained irq chips may already be registered with custom chip_data information, causing an oops. This is the case with intel MID SoC devices with gpio demuxers registered as irq_chips. Signed-off-by: Mathias Nyman Signed-off-by: Alan Cox [ -v2: fixed build failure ] Signed-off-by: Ingo Molnar --- arch/x86/kernel/apic/io_apic.c | 5 +++++ arch/x86/kernel/nmi.c | 1 + 2 files changed, 6 insertions(+) diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 841b8da40525a..6d939d7847e29 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -1694,6 +1694,7 @@ __apicdebuginit(void) print_IO_APICs(void) int ioapic_idx; struct irq_cfg *cfg; unsigned int irq; + struct irq_chip *chip; printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries); for (ioapic_idx = 0; ioapic_idx < nr_ioapics; ioapic_idx++) @@ -1714,6 +1715,10 @@ __apicdebuginit(void) print_IO_APICs(void) for_each_active_irq(irq) { struct irq_pin_list *entry; + chip = irq_get_chip(irq); + if (chip != &ioapic_chip) + continue; + cfg = irq_get_chip_data(irq); if (!cfg) continue; diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c index 27d1e7cbdb6c0..e88f37b58ddde 100644 --- a/arch/x86/kernel/nmi.c +++ b/arch/x86/kernel/nmi.c @@ -29,6 +29,7 @@ #include #include #include +#include #define NMI_MAX_NAMELEN 16 struct nmiaction { -- GitLab From 54b2edf487d497cc522bce91b3bd79538f001d13 Mon Sep 17 00:00:00 2001 From: Aaro Koskinen Date: Thu, 10 Nov 2011 17:59:45 +0000 Subject: [PATCH 0181/3089] MIPS: errloongson2_clock: Fix build error by including linux/module.h Fix the following compilation failure with v3.2-rc1 by including module.h: CC [M] arch/mips/kernel/cpufreq/loongson2_clock.o arch/mips/kernel/cpufreq/loongson2_clock.c:39:1: error: data definition has no type or storage class [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:39:1: error: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL' [-Werror=implicit-int] arch/mips/kernel/cpufreq/loongson2_clock.c:39:1: error: parameter names (without types) in function declaration [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:51:1: error: data definition has no type or storage class [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:51:1: error: type defaults to 'int' in declaration of 'EXPORT_SYMBOL' [-Werror=implicit-int] arch/mips/kernel/cpufreq/loongson2_clock.c:51:1: error: parameter names (without types) in function declaration [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:71:1: error: data definition has no type or storage class [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:71:1: error: type defaults to 'int' in declaration of 'EXPORT_SYMBOL' [-Werror=implicit-int] arch/mips/kernel/cpufreq/loongson2_clock.c:71:1: error: parameter names (without types) in function declaration [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:76:1: error: data definition has no type or storage class [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:76:1: error: type defaults to 'int' in declaration of 'EXPORT_SYMBOL' [-Werror=implicit-int] arch/mips/kernel/cpufreq/loongson2_clock.c:76:1: error: parameter names (without types) in function declaration [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:82:1: error: data definition has no type or storage class [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:82:1: error: type defaults to 'int' in declaration of 'EXPORT_SYMBOL' [-Werror=implicit-int] arch/mips/kernel/cpufreq/loongson2_clock.c:82:1: error: parameter names (without types) in function declaration [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:87:1: error: data definition has no type or storage class [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:87:1: error: type defaults to 'int' in declaration of 'EXPORT_SYMBOL' [-Werror=implicit-int] arch/mips/kernel/cpufreq/loongson2_clock.c:87:1: error: parameter names (without types) in function declaration [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:93:1: error: data definition has no type or storage class [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:93:1: error: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL' [-Werror=implicit-int] arch/mips/kernel/cpufreq/loongson2_clock.c:93:1: error: parameter names (without types) in function declaration [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:131:1: error: data definition has no type or storage class [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:131:1: error: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL' [-Werror=implicit-int] arch/mips/kernel/cpufreq/loongson2_clock.c:131:1: error: parameter names (without types) in function declaration [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:147:1: error: data definition has no type or storage class [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:147:1: error: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL' [-Werror=implicit-int] arch/mips/kernel/cpufreq/loongson2_clock.c:147:1: error: parameter names (without types) in function declaration [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:166:1: error: data definition has no type or storage class [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:166:1: error: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL' [-Werror=implicit-int] arch/mips/kernel/cpufreq/loongson2_clock.c:166:1: error: parameter names (without types) in function declaration [-Werror] arch/mips/kernel/cpufreq/loongson2_clock.c:168:15: error: expected declaration specifiers or '...' before string constant arch/mips/kernel/cpufreq/loongson2_clock.c:169:20: error: expected declaration specifiers or '...' before string constant arch/mips/kernel/cpufreq/loongson2_clock.c:170:16: error: expected declaration specifiers or '...' before string constant Signed-off-by: Aaro Koskinen To: linux-mips@linux-mips.org To: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/2922/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/cpufreq/loongson2_clock.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/mips/kernel/cpufreq/loongson2_clock.c b/arch/mips/kernel/cpufreq/loongson2_clock.c index cefc6e259bafd..5426779d9fdb7 100644 --- a/arch/mips/kernel/cpufreq/loongson2_clock.c +++ b/arch/mips/kernel/cpufreq/loongson2_clock.c @@ -7,6 +7,7 @@ * for more details. */ +#include #include #include -- GitLab From b2f909419b72cf043814bf17b96c93d4695378a1 Mon Sep 17 00:00:00 2001 From: David Daney Date: Thu, 10 Nov 2011 17:59:45 +0000 Subject: [PATCH 0182/3089] MIPS: Octeon: Fix compile error in arch/mips/cavium-octeon/flash_setup.c The parse_mtd_partitions() and mtd_device_register() functions were combined into mtd_device_parse_register(). So call that instead. Signed-off-by: David Daney To: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2923/ Signed-off-by: Ralf Baechle --- arch/mips/cavium-octeon/flash_setup.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/arch/mips/cavium-octeon/flash_setup.c b/arch/mips/cavium-octeon/flash_setup.c index 975c20327bb15..0a430e06f5e5e 100644 --- a/arch/mips/cavium-octeon/flash_setup.c +++ b/arch/mips/cavium-octeon/flash_setup.c @@ -17,8 +17,6 @@ static struct map_info flash_map; static struct mtd_info *mymtd; -static int nr_parts; -static struct mtd_partition *parts; static const char *part_probe_types[] = { "cmdlinepart", #ifdef CONFIG_MTD_REDBOOT_PARTS @@ -61,11 +59,8 @@ static int __init flash_init(void) mymtd = do_map_probe("cfi_probe", &flash_map); if (mymtd) { mymtd->owner = THIS_MODULE; - - nr_parts = parse_mtd_partitions(mymtd, - part_probe_types, - &parts, 0); - mtd_device_register(mymtd, parts, nr_parts); + mtd_device_parse_register(mymtd, part_probe_types, + 0, NULL, 0); } else { pr_err("Failed to register MTD device for flash\n"); } -- GitLab From 5c200197130e307de6eba72fc335c83c9dd6a5bc Mon Sep 17 00:00:00 2001 From: Maksim Rayskiy Date: Thu, 10 Nov 2011 17:59:45 +0000 Subject: [PATCH 0183/3089] MIPS: ASID conflict after CPU hotplug I am running SMP Linux 2.6.37-rc1 on BMIPS5000 (single core dual thread) and observe some abnormalities when doing system suspend/resume which I narrowed down to cpu hotplugging. The suspend brings the second thread processor down and then restarts it, after which I see memory corruption in userspace. I started digging and found out that problem occurs because while doing execve() the child process is getting the same ASID as the parent, which obviously corrupts parent's address space. Further digging showed that activate_mm() calls get_new_mmu_context() to get a new ASID, but at this time ASID field in entryHi is 1, and asid_cache(cpu) is 0x100 (it was just reset to ASID_FIRST_VERSION when the secondary TP was booting). So, get_new_mmu_context() increments the asid_cache(cpu) value to 0x101, and thus puts 0x01 into entryHi. The result - ASID field does not get changed as it was supposed to. My solution is very simple - do not reset asid_cache(cpu) on TP warm restart. Patchwork: https://patchwork.linux-mips.org/patch/1797/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/traps.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c index 261ccbc077401..5c8a49d55054d 100644 --- a/arch/mips/kernel/traps.c +++ b/arch/mips/kernel/traps.c @@ -1596,7 +1596,8 @@ void __cpuinit per_cpu_trap_init(void) } #endif /* CONFIG_MIPS_MT_SMTC */ - cpu_data[cpu].asid_cache = ASID_FIRST_VERSION; + if (!cpu_data[cpu].asid_cache) + cpu_data[cpu].asid_cache = ASID_FIRST_VERSION; atomic_inc(&init_mm.mm_count); current->active_mm = &init_mm; -- GitLab From d65670a78cdbfae94f20a9e05ec705871d7cdf2b Mon Sep 17 00:00:00 2001 From: John Stultz Date: Mon, 31 Oct 2011 17:06:35 -0400 Subject: [PATCH 0184/3089] clocksource: Avoid selecting mult values that might overflow when adjusted For some frequencies, the clocks_calc_mult_shift() function will unfortunately select mult values very close to 0xffffffff. This has the potential to overflow when NTP adjusts the clock, adding to the mult value. This patch adds a clocksource.maxadj value, which provides an approximation of an 11% adjustment(NTP limits adjustments to 500ppm and the tick adjustment is limited to 10%), which could be made to the clocksource.mult value. This is then used to both check that the current mult value won't overflow/underflow, as well as warning us if the timekeeping_adjust() code pushes over that 11% boundary. v2: Fix max_adjustment calculation, and improve WARN_ONCE messages. v3: Don't warn before maxadj has actually been set CC: Yong Zhang CC: David Daney CC: Thomas Gleixner CC: Chen Jie CC: zhangfx CC: stable@kernel.org Reported-by: Chen Jie Reported-by: zhangfx Tested-by: Yong Zhang Signed-off-by: John Stultz --- include/linux/clocksource.h | 3 +- kernel/time/clocksource.c | 58 ++++++++++++++++++++++++++++++------- kernel/time/timekeeping.c | 7 +++++ 3 files changed, 57 insertions(+), 11 deletions(-) diff --git a/include/linux/clocksource.h b/include/linux/clocksource.h index 139c4db55f173..c86c940d1de3a 100644 --- a/include/linux/clocksource.h +++ b/include/linux/clocksource.h @@ -156,6 +156,7 @@ extern u64 timecounter_cyc2time(struct timecounter *tc, * @mult: cycle to nanosecond multiplier * @shift: cycle to nanosecond divisor (power of two) * @max_idle_ns: max idle time permitted by the clocksource (nsecs) + * @maxadj maximum adjustment value to mult (~11%) * @flags: flags describing special properties * @archdata: arch-specific data * @suspend: suspend function for the clocksource, if necessary @@ -172,7 +173,7 @@ struct clocksource { u32 mult; u32 shift; u64 max_idle_ns; - + u32 maxadj; #ifdef CONFIG_ARCH_CLOCKSOURCE_DATA struct arch_clocksource_data archdata; #endif diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index cf52fda2e0966..cfc65e1eb9fbb 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -491,6 +491,22 @@ void clocksource_touch_watchdog(void) clocksource_resume_watchdog(); } +/** + * clocksource_max_adjustment- Returns max adjustment amount + * @cs: Pointer to clocksource + * + */ +static u32 clocksource_max_adjustment(struct clocksource *cs) +{ + u64 ret; + /* + * We won't try to correct for more then 11% adjustments (110,000 ppm), + */ + ret = (u64)cs->mult * 11; + do_div(ret,100); + return (u32)ret; +} + /** * clocksource_max_deferment - Returns max time the clocksource can be deferred * @cs: Pointer to clocksource @@ -503,25 +519,28 @@ static u64 clocksource_max_deferment(struct clocksource *cs) /* * Calculate the maximum number of cycles that we can pass to the * cyc2ns function without overflowing a 64-bit signed result. The - * maximum number of cycles is equal to ULLONG_MAX/cs->mult which - * is equivalent to the below. - * max_cycles < (2^63)/cs->mult - * max_cycles < 2^(log2((2^63)/cs->mult)) - * max_cycles < 2^(log2(2^63) - log2(cs->mult)) - * max_cycles < 2^(63 - log2(cs->mult)) - * max_cycles < 1 << (63 - log2(cs->mult)) + * maximum number of cycles is equal to ULLONG_MAX/(cs->mult+cs->maxadj) + * which is equivalent to the below. + * max_cycles < (2^63)/(cs->mult + cs->maxadj) + * max_cycles < 2^(log2((2^63)/(cs->mult + cs->maxadj))) + * max_cycles < 2^(log2(2^63) - log2(cs->mult + cs->maxadj)) + * max_cycles < 2^(63 - log2(cs->mult + cs->maxadj)) + * max_cycles < 1 << (63 - log2(cs->mult + cs->maxadj)) * Please note that we add 1 to the result of the log2 to account for * any rounding errors, ensure the above inequality is satisfied and * no overflow will occur. */ - max_cycles = 1ULL << (63 - (ilog2(cs->mult) + 1)); + max_cycles = 1ULL << (63 - (ilog2(cs->mult + cs->maxadj) + 1)); /* * The actual maximum number of cycles we can defer the clocksource is * determined by the minimum of max_cycles and cs->mask. + * Note: Here we subtract the maxadj to make sure we don't sleep for + * too long if there's a large negative adjustment. */ max_cycles = min_t(u64, max_cycles, (u64) cs->mask); - max_nsecs = clocksource_cyc2ns(max_cycles, cs->mult, cs->shift); + max_nsecs = clocksource_cyc2ns(max_cycles, cs->mult - cs->maxadj, + cs->shift); /* * To ensure that the clocksource does not wrap whilst we are idle, @@ -640,7 +659,6 @@ static void clocksource_enqueue(struct clocksource *cs) void __clocksource_updatefreq_scale(struct clocksource *cs, u32 scale, u32 freq) { u64 sec; - /* * Calc the maximum number of seconds which we can run before * wrapping around. For clocksources which have a mask > 32bit @@ -661,6 +679,20 @@ void __clocksource_updatefreq_scale(struct clocksource *cs, u32 scale, u32 freq) clocks_calc_mult_shift(&cs->mult, &cs->shift, freq, NSEC_PER_SEC / scale, sec * scale); + + /* + * for clocksources that have large mults, to avoid overflow. + * Since mult may be adjusted by ntp, add an safety extra margin + * + */ + cs->maxadj = clocksource_max_adjustment(cs); + while ((cs->mult + cs->maxadj < cs->mult) + || (cs->mult - cs->maxadj > cs->mult)) { + cs->mult >>= 1; + cs->shift--; + cs->maxadj = clocksource_max_adjustment(cs); + } + cs->max_idle_ns = clocksource_max_deferment(cs); } EXPORT_SYMBOL_GPL(__clocksource_updatefreq_scale); @@ -701,6 +733,12 @@ EXPORT_SYMBOL_GPL(__clocksource_register_scale); */ int clocksource_register(struct clocksource *cs) { + /* calculate max adjustment for given mult/shift */ + cs->maxadj = clocksource_max_adjustment(cs); + WARN_ONCE(cs->mult + cs->maxadj < cs->mult, + "Clocksource %s might overflow on 11%% adjustment\n", + cs->name); + /* calculate max idle time permitted for this clocksource */ cs->max_idle_ns = clocksource_max_deferment(cs); diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 2b021b0e8507e..e65ff31711028 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -820,6 +820,13 @@ static void timekeeping_adjust(s64 offset) } else return; + WARN_ONCE(timekeeper.clock->maxadj && + (timekeeper.mult + adj > timekeeper.clock->mult + + timekeeper.clock->maxadj), + "Adjusting %s more then 11%% (%ld vs %ld)\n", + timekeeper.clock->name, (long)timekeeper.mult + adj, + (long)timekeeper.clock->mult + + timekeeper.clock->maxadj); timekeeper.mult += adj; timekeeper.xtime_interval += interval; timekeeper.xtime_nsec -= offset; -- GitLab From 2aa13531bbbc6582874bedfcd853e1058b0fb4f9 Mon Sep 17 00:00:00 2001 From: Stanislav Kinsbursky Date: Thu, 10 Nov 2011 14:33:23 +0300 Subject: [PATCH 0185/3089] SUNRPC: destroy freshly allocated transport in case of sockaddr init error Otherwise we will leak xprt structure and struct net reference. Signed-off-by: Stanislav Kinsbursky Signed-off-by: Trond Myklebust --- net/sunrpc/xprtsock.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c index d7f97ef265904..2d78d95955ab5 100644 --- a/net/sunrpc/xprtsock.c +++ b/net/sunrpc/xprtsock.c @@ -2530,8 +2530,10 @@ static struct rpc_xprt *xs_setup_xprt(struct xprt_create *args, int err; err = xs_init_anyaddr(args->dstaddr->sa_family, (struct sockaddr *)&new->srcaddr); - if (err != 0) + if (err != 0) { + xprt_free(xprt); return ERR_PTR(err); + } } return xprt; -- GitLab From 62e4a76987eab2b7fa952546614bc83e5bfc9d3e Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 10 Nov 2011 14:30:37 -0500 Subject: [PATCH 0186/3089] NFS: Revert pnfs ugliness from the generic NFS read code path pNFS-specific code belongs in the pnfs layer. It should not be hijacking generic NFS read or write code paths. Signed-off-by: Trond Myklebust --- fs/nfs/internal.h | 2 ++ fs/nfs/pnfs.c | 26 +++++++++++++++++++++----- fs/nfs/read.c | 14 ++------------ 3 files changed, 25 insertions(+), 17 deletions(-) diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index c1a1bd8ddf1cb..3f4d95751d52f 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -299,6 +299,8 @@ extern void nfs_read_prepare(struct rpc_task *task, void *calldata); extern int nfs_generic_pagein(struct nfs_pageio_descriptor *desc, struct list_head *head); +extern void nfs_pageio_init_read_mds(struct nfs_pageio_descriptor *pgio, + struct inode *inode); extern void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio); extern void nfs_readdata_release(struct nfs_read_data *rdata); diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index a2478bc74442a..e40a3ca22eaf5 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -1259,6 +1259,25 @@ pnfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc) } EXPORT_SYMBOL_GPL(pnfs_generic_pg_writepages); +static void pnfs_ld_handle_read_error(struct nfs_read_data *data) +{ + struct nfs_pageio_descriptor pgio; + + put_lseg(data->lseg); + data->lseg = NULL; + dprintk("pnfs write error = %d\n", data->pnfs_error); + + nfs_pageio_init_read_mds(&pgio, data->inode); + + while (!list_empty(&data->pages)) { + struct nfs_page *req = nfs_list_entry(data->pages.next); + + nfs_list_remove_request(req); + nfs_pageio_add_request(&pgio, req); + } + nfs_pageio_complete(&pgio); +} + /* * Called by non rpc-based layout drivers */ @@ -1267,11 +1286,8 @@ void pnfs_ld_read_done(struct nfs_read_data *data) if (likely(!data->pnfs_error)) { __nfs4_read_done_cb(data); data->mds_ops->rpc_call_done(&data->task, data); - } else { - put_lseg(data->lseg); - data->lseg = NULL; - dprintk("pnfs write error = %d\n", data->pnfs_error); - } + } else + pnfs_ld_handle_read_error(data); data->mds_ops->rpc_release(data); } EXPORT_SYMBOL_GPL(pnfs_ld_read_done); diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 8b48ec63f7229..cfa175c223dcf 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -109,7 +109,7 @@ static void nfs_readpage_truncate_uninitialised_page(struct nfs_read_data *data) } } -static void nfs_pageio_init_read_mds(struct nfs_pageio_descriptor *pgio, +void nfs_pageio_init_read_mds(struct nfs_pageio_descriptor *pgio, struct inode *inode) { nfs_pageio_init(pgio, inode, &nfs_pageio_read_ops, @@ -534,23 +534,13 @@ static void nfs_readpage_result_full(struct rpc_task *task, void *calldata) static void nfs_readpage_release_full(void *calldata) { struct nfs_read_data *data = calldata; - struct nfs_pageio_descriptor pgio; - if (data->pnfs_error) { - nfs_pageio_init_read_mds(&pgio, data->inode); - pgio.pg_recoalesce = 1; - } while (!list_empty(&data->pages)) { struct nfs_page *req = nfs_list_entry(data->pages.next); nfs_list_remove_request(req); - if (!data->pnfs_error) - nfs_readpage_release(req); - else - nfs_pageio_add_request(&pgio, req); + nfs_readpage_release(req); } - if (data->pnfs_error) - nfs_pageio_complete(&pgio); nfs_readdata_release(calldata); } -- GitLab From 7be620f79714b6598275bd5c71d89bb24dfe332e Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Thu, 10 Nov 2011 12:12:55 -0800 Subject: [PATCH 0187/3089] ARM: OMAP: Fix dpll_data compile error when omap2 only is selected Without this patch we get the following error: arch/arm/mach-omap2/clkt_dpll.c: In function '_dpll_test_fint': arch/arm/mach-omap2/clkt_dpll.c:98: error: 'struct dpll_data' has no member named 'flags' Signed-off-by: Tony Lindgren --- arch/arm/plat-omap/include/plat/clock.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/plat-omap/include/plat/clock.h b/arch/arm/plat-omap/include/plat/clock.h index 197ca03c3f7d8..eb73ab40e9556 100644 --- a/arch/arm/plat-omap/include/plat/clock.h +++ b/arch/arm/plat-omap/include/plat/clock.h @@ -165,8 +165,8 @@ struct dpll_data { u8 auto_recal_bit; u8 recal_en_bit; u8 recal_st_bit; - u8 flags; # endif + u8 flags; }; #endif -- GitLab From a528721da08a793e0cec6ef6281d404eddd6e27b Mon Sep 17 00:00:00 2001 From: Russell King Date: Fri, 4 Nov 2011 15:05:24 +0000 Subject: [PATCH 0188/3089] ARM: restart: add restart hook to machine_desc record Add a restart hook to the machine_desc record so we don't have to populate all platforms with init_early methods to initialize the arm_pm_restart function pointer. Acked-by: Nicolas Pitre Acked-by: Will Deacon Acked-by: H Hartley Sweeten Acked-by: Tony Lindgren Signed-off-by: Russell King --- arch/arm/include/asm/mach/arch.h | 1 + arch/arm/kernel/setup.c | 3 +++ 2 files changed, 4 insertions(+) diff --git a/arch/arm/include/asm/mach/arch.h b/arch/arm/include/asm/mach/arch.h index 7d19425dd496a..8aa397a5ea727 100644 --- a/arch/arm/include/asm/mach/arch.h +++ b/arch/arm/include/asm/mach/arch.h @@ -45,6 +45,7 @@ struct machine_desc { #ifdef CONFIG_MULTI_IRQ_HANDLER void (*handle_irq)(struct pt_regs *); #endif + void (*restart)(char, const char *); }; /* diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 7e7977ab994ff..4755e2827dc3f 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -922,6 +922,9 @@ void __init setup_arch(char **cmdline_p) paging_init(mdesc); request_standard_resources(mdesc); + if (mdesc->restart) + arm_pm_restart = mdesc->restart; + unflatten_device_tree(); #ifdef CONFIG_SMP -- GitLab From b44c350d4104265cf3a6b4355dc1ee05c16aa5de Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 1 Nov 2011 14:27:33 +0000 Subject: [PATCH 0189/3089] ARM: restart: allow platforms more flexibility specifying restart mode Change 'soft_reboot' into a more generic 'restart_mode' variable, allowing the default restart mode to be specified. Acked-by: Nicolas Pitre Acked-by: Will Deacon Acked-by: H Hartley Sweeten Signed-off-by: Russell King --- arch/arm/include/asm/mach/arch.h | 8 ++++---- arch/arm/kernel/setup.c | 4 ++-- arch/arm/mach-ebsa110/core.c | 2 +- arch/arm/mach-footbridge/cats-hw.c | 2 +- 4 files changed, 8 insertions(+), 8 deletions(-) diff --git a/arch/arm/include/asm/mach/arch.h b/arch/arm/include/asm/mach/arch.h index 8aa397a5ea727..6fe6cf0895c94 100644 --- a/arch/arm/include/asm/mach/arch.h +++ b/arch/arm/include/asm/mach/arch.h @@ -30,10 +30,10 @@ struct machine_desc { unsigned int video_start; /* start of video RAM */ unsigned int video_end; /* end of video RAM */ - unsigned int reserve_lp0 :1; /* never has lp0 */ - unsigned int reserve_lp1 :1; /* never has lp1 */ - unsigned int reserve_lp2 :1; /* never has lp2 */ - unsigned int soft_reboot :1; /* soft reboot */ + unsigned char reserve_lp0 :1; /* never has lp0 */ + unsigned char reserve_lp1 :1; /* never has lp1 */ + unsigned char reserve_lp2 :1; /* never has lp2 */ + char restart_mode; /* default restart mode */ void (*fixup)(struct tag *, char **, struct meminfo *); void (*reserve)(void);/* reserve mem blocks */ diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 4755e2827dc3f..a753880e984b9 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -902,8 +902,8 @@ void __init setup_arch(char **cmdline_p) machine_desc = mdesc; machine_name = mdesc->name; - if (mdesc->soft_reboot) - reboot_setup("s"); + if (mdesc->restart_mode) + reboot_setup(&mdesc->restart_mode); init_mm.start_code = (unsigned long) _text; init_mm.end_code = (unsigned long) _etext; diff --git a/arch/arm/mach-ebsa110/core.c b/arch/arm/mach-ebsa110/core.c index d0ce8abdd4b67..ce3ed244c4b03 100644 --- a/arch/arm/mach-ebsa110/core.c +++ b/arch/arm/mach-ebsa110/core.c @@ -283,7 +283,7 @@ MACHINE_START(EBSA110, "EBSA110") .atag_offset = 0x400, .reserve_lp0 = 1, .reserve_lp2 = 1, - .soft_reboot = 1, + .restart_mode = 's', .map_io = ebsa110_map_io, .init_irq = ebsa110_init_irq, .timer = &ebsa110_timer, diff --git a/arch/arm/mach-footbridge/cats-hw.c b/arch/arm/mach-footbridge/cats-hw.c index d5f1785409286..60b6774e1eaa0 100644 --- a/arch/arm/mach-footbridge/cats-hw.c +++ b/arch/arm/mach-footbridge/cats-hw.c @@ -86,7 +86,7 @@ fixup_cats(struct tag *tags, char **cmdline, struct meminfo *mi) MACHINE_START(CATS, "Chalice-CATS") /* Maintainer: Philip Blundell */ .atag_offset = 0x100, - .soft_reboot = 1, + .restart_mode = 's', .fixup = fixup_cats, .map_io = footbridge_map_io, .init_irq = footbridge_init_irq, -- GitLab From 34adeda943e1fb10e8de022b834996e04241fa7d Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 1 Nov 2011 16:24:18 +0000 Subject: [PATCH 0190/3089] ARM: restart: add default restart modes for PXA mioa701, spitz and tosa Add a default restart mode to reflect the hard-coded restart mode found in these files. Acked-by: Nicolas Pitre Acked-by: Will Deacon Acked-by: H Hartley Sweeten Acked-by: Robert Jarzmik Signed-off-by: Russell King --- arch/arm/mach-pxa/mioa701.c | 1 + arch/arm/mach-pxa/spitz.c | 3 +++ arch/arm/mach-pxa/tosa.c | 1 + 3 files changed, 5 insertions(+) diff --git a/arch/arm/mach-pxa/mioa701.c b/arch/arm/mach-pxa/mioa701.c index b938fc2c316a7..4f47a760398ff 100644 --- a/arch/arm/mach-pxa/mioa701.c +++ b/arch/arm/mach-pxa/mioa701.c @@ -752,6 +752,7 @@ static void mioa701_machine_exit(void) MACHINE_START(MIOA701, "MIO A701") .atag_offset = 0x100, + .restart_mode = 's', .map_io = &pxa27x_map_io, .init_irq = &pxa27x_init_irq, .handle_irq = &pxa27x_handle_irq, diff --git a/arch/arm/mach-pxa/spitz.c b/arch/arm/mach-pxa/spitz.c index 953a9195f9e5b..2f57d94de727a 100644 --- a/arch/arm/mach-pxa/spitz.c +++ b/arch/arm/mach-pxa/spitz.c @@ -982,6 +982,7 @@ static void __init spitz_fixup(struct tag *tags, char **cmdline, #ifdef CONFIG_MACH_SPITZ MACHINE_START(SPITZ, "SHARP Spitz") + .restart_mode = 'g', .fixup = spitz_fixup, .map_io = pxa27x_map_io, .init_irq = pxa27x_init_irq, @@ -993,6 +994,7 @@ MACHINE_END #ifdef CONFIG_MACH_BORZOI MACHINE_START(BORZOI, "SHARP Borzoi") + .restart_mode = 'g', .fixup = spitz_fixup, .map_io = pxa27x_map_io, .init_irq = pxa27x_init_irq, @@ -1004,6 +1006,7 @@ MACHINE_END #ifdef CONFIG_MACH_AKITA MACHINE_START(AKITA, "SHARP Akita") + .restart_mode = 'g', .fixup = spitz_fixup, .map_io = pxa27x_map_io, .init_irq = pxa27x_init_irq, diff --git a/arch/arm/mach-pxa/tosa.c b/arch/arm/mach-pxa/tosa.c index 402b0c96613ba..ef6453041cf13 100644 --- a/arch/arm/mach-pxa/tosa.c +++ b/arch/arm/mach-pxa/tosa.c @@ -970,6 +970,7 @@ static void __init fixup_tosa(struct tag *tags, char **cmdline, } MACHINE_START(TOSA, "SHARP Tosa") + .restart_mode = 'g', .fixup = fixup_tosa, .map_io = pxa25x_map_io, .nr_irqs = TOSA_NR_IRQS, -- GitLab From ac15e00b1efe705b66a36d1a6a9db7f6ed524c43 Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 31 Oct 2011 09:22:22 +0000 Subject: [PATCH 0191/3089] ARM: restart: move reboot failure handing into machine_restart() Move the failure to reboot into machine_restart() to always catch this condition, even if a platform decides to hook the restarting via arm_pm_restart(). Acked-by: Nicolas Pitre Acked-by: Will Deacon Acked-by: H Hartley Sweeten Acked-by: Tony Lindgren Signed-off-by: Russell King --- arch/arm/kernel/process.c | 20 +++++++++----------- 1 file changed, 9 insertions(+), 11 deletions(-) diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 75316f0dd02ae..3bda1c3797762 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -114,18 +114,8 @@ void arm_machine_restart(char mode, const char *cmd) /* Push out any further dirty data, and ensure cache is empty */ flush_cache_all(); - /* - * Now call the architecture specific reboot code. - */ + /* Now call the architecture specific reboot code. */ arch_reset(mode, cmd); - - /* - * Whoops - the architecture was unable to reboot. - * Tell the user! - */ - mdelay(1000); - printk("Reboot failed -- System halted\n"); - while (1); } /* @@ -250,7 +240,15 @@ void machine_power_off(void) void machine_restart(char *cmd) { machine_shutdown(); + arm_pm_restart(reboot_mode, cmd); + + /* Give a grace period for failure to restart of 1s */ + mdelay(1000); + + /* Whoops - the platform was unable to reboot. Tell the user! */ + printk("Reboot failed -- System halted\n"); + while (1); } void __show_regs(struct pt_regs *regs) -- GitLab From 5aafec15bdc54cf0722696c95091d7bd674bfcad Mon Sep 17 00:00:00 2001 From: Russell King Date: Tue, 1 Nov 2011 10:15:27 +0000 Subject: [PATCH 0192/3089] ARM: restart: remove argument to setup_mm_for_reboot() setup_mm_for_reboot() doesn't make use of its argument, so remove it. Acked-by: Nicolas Pitre Acked-by: Will Deacon Acked-by: H Hartley Sweeten Acked-by: Tony Lindgren Signed-off-by: Russell King --- arch/arm/kernel/machine_kexec.c | 4 ++-- arch/arm/kernel/process.c | 4 ++-- arch/arm/mm/idmap.c | 2 +- arch/arm/mm/nommu.c | 2 +- 4 files changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/arm/kernel/machine_kexec.c b/arch/arm/kernel/machine_kexec.c index c1b4463dcc839..cc40b965d42a5 100644 --- a/arch/arm/kernel/machine_kexec.c +++ b/arch/arm/kernel/machine_kexec.c @@ -16,7 +16,7 @@ extern const unsigned char relocate_new_kernel[]; extern const unsigned int relocate_new_kernel_size; -extern void setup_mm_for_reboot(char mode); +extern void setup_mm_for_reboot(void); extern unsigned long kexec_start_address; extern unsigned long kexec_indirection_page; @@ -114,7 +114,7 @@ void machine_kexec(struct kimage *image) kexec_reinit(); local_irq_disable(); local_fiq_disable(); - setup_mm_for_reboot(0); /* mode is not used, so just pass 0*/ + setup_mm_for_reboot(); flush_cache_all(); outer_flush_all(); outer_disable(); diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c index 3bda1c3797762..4181738452fc8 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -57,7 +57,7 @@ static const char *isa_modes[] = { "ARM" , "Thumb" , "Jazelle", "ThumbEE" }; -extern void setup_mm_for_reboot(char mode); +extern void setup_mm_for_reboot(void); static volatile int hlt_counter; @@ -103,7 +103,7 @@ void arm_machine_restart(char mode, const char *cmd) * we may need it to insert some 1:1 mappings so that * soft boot works. */ - setup_mm_for_reboot(mode); + setup_mm_for_reboot(); /* Clean and invalidate caches */ flush_cache_all(); diff --git a/arch/arm/mm/idmap.c b/arch/arm/mm/idmap.c index 2be9139a4ef3c..296ad2eaddb05 100644 --- a/arch/arm/mm/idmap.c +++ b/arch/arm/mm/idmap.c @@ -78,7 +78,7 @@ void identity_mapping_del(pgd_t *pgd, unsigned long addr, unsigned long end) * the user-mode pages. This will then ensure that we have predictable * results when turning the mmu off */ -void setup_mm_for_reboot(char mode) +void setup_mm_for_reboot(void) { /* * We need to access to user-mode page tables here. For kernel threads diff --git a/arch/arm/mm/nommu.c b/arch/arm/mm/nommu.c index 941a98c9e8aaf..88417514b2c66 100644 --- a/arch/arm/mm/nommu.c +++ b/arch/arm/mm/nommu.c @@ -43,7 +43,7 @@ void __init paging_init(struct machine_desc *mdesc) /* * We don't need to do anything here for nommu machines. */ -void setup_mm_for_reboot(char mode) +void setup_mm_for_reboot(void) { } -- GitLab From afef2cfa0ecf45dec7e88db9fa312ee82d347111 Mon Sep 17 00:00:00 2001 From: Charles Chin Date: Fri, 11 Nov 2011 08:05:28 +0100 Subject: [PATCH 0193/3089] ALSA: hda - pwr_nids cleanup for IDT codecs Clean up and fix pwr_nids for 92HD71 / 73 / 83 family codecs; remove pwr_mapping which was incorrect. The original pwr_nids support of 92HD83xxx was incorrect and never actually worked before. Now we should have things working correctly without having to hack by DID anymore. It is also not necessary to explicitly turn on all the pins near the beginning of patch_stac92hd83xxx() now, the pins will go though initialization properly. Tested on 92HD66 / 71 / 73 / 75 / 83 / 89 / 91 demo boards. Signed-off-by: Charles Chin Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_sigmatel.c | 33 +++++++++------------------------ 1 file changed, 9 insertions(+), 24 deletions(-) diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index edc2b7bc177c6..470f6f286e814 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -227,7 +227,6 @@ struct sigmatel_spec { /* power management */ unsigned int num_pwrs; - const unsigned int *pwr_mapping; const hda_nid_t *pwr_nids; const hda_nid_t *dac_list; @@ -374,18 +373,15 @@ static const unsigned long stac92hd73xx_capvols[] = { #define STAC92HD83_DAC_COUNT 3 -static const hda_nid_t stac92hd83xxx_pwr_nids[4] = { - 0xa, 0xb, 0xd, 0xe, +static const hda_nid_t stac92hd83xxx_pwr_nids[7] = { + 0x0a, 0x0b, 0x0c, 0xd, 0x0e, + 0x0f, 0x10 }; static const hda_nid_t stac92hd83xxx_slave_dig_outs[2] = { 0x1e, 0, }; -static const unsigned int stac92hd83xxx_pwr_mapping[4] = { - 0x03, 0x0c, 0x20, 0x40, -}; - static const hda_nid_t stac92hd83xxx_dmic_nids[] = { 0x11, 0x20, }; @@ -4470,8 +4466,12 @@ static int stac92xx_init(struct hda_codec *codec) stac_toggle_power_map(codec, nid, 1); continue; } - if (enable_pin_detect(codec, nid, STAC_PWR_EVENT)) + if (enable_pin_detect(codec, nid, STAC_PWR_EVENT)) { stac_issue_unsol_event(codec, nid); + continue; + } + /* none of the above, turn the port OFF */ + stac_toggle_power_map(codec, nid, 0); } /* sync mute LED */ @@ -4727,11 +4727,7 @@ static void stac_toggle_power_map(struct hda_codec *codec, hda_nid_t nid, if (idx >= spec->num_pwrs) return; - /* several codecs have two power down bits */ - if (spec->pwr_mapping) - idx = spec->pwr_mapping[idx]; - else - idx = 1 << idx; + idx = 1 << idx; val = snd_hda_codec_read(codec, codec->afg, 0, 0x0fec, 0x0) & 0xff; if (enable) @@ -5629,9 +5625,6 @@ static int patch_stac92hd83xxx(struct hda_codec *codec) snd_hda_codec_set_pincfg(codec, 0xf, 0x2181205e); } - /* reset pin power-down; Windows may leave these bits after reboot */ - snd_hda_codec_write_cache(codec, codec->afg, 0, 0x7EC, 0); - snd_hda_codec_write_cache(codec, codec->afg, 0, 0x7ED, 0); codec->no_trigger_sense = 1; codec->spec = spec; @@ -5641,7 +5634,6 @@ static int patch_stac92hd83xxx(struct hda_codec *codec) codec->slave_dig_outs = stac92hd83xxx_slave_dig_outs; spec->digbeep_nid = 0x21; spec->pwr_nids = stac92hd83xxx_pwr_nids; - spec->pwr_mapping = stac92hd83xxx_pwr_mapping; spec->num_pwrs = ARRAY_SIZE(stac92hd83xxx_pwr_nids); spec->multiout.dac_nids = spec->dac_nids; spec->init = stac92hd83xxx_core_init; @@ -5658,9 +5650,6 @@ again: stac92xx_set_config_regs(codec, stac92hd83xxx_brd_tbl[spec->board_config]); - if (spec->board_config != STAC_92HD83XXX_PWR_REF) - spec->num_pwrs = 0; - codec->patch_ops = stac92xx_patch_ops; if (find_mute_led_gpio(codec, 0)) @@ -5869,8 +5858,6 @@ again: (codec->revision_id & 0xf) == 1) spec->stream_delay = 40; /* 40 milliseconds */ - /* no output amps */ - spec->num_pwrs = 0; /* disable VSW */ spec->init = stac92hd71bxx_core_init; unmute_init++; @@ -5885,8 +5872,6 @@ again: if ((codec->revision_id & 0xf) == 1) spec->stream_delay = 40; /* 40 milliseconds */ - /* no output amps */ - spec->num_pwrs = 0; /* fallthru */ default: spec->init = stac92hd71bxx_core_init; -- GitLab From 7a401a972df8e184b3d1a3fc958c0a4ddee8d312 Mon Sep 17 00:00:00 2001 From: Rabin Vincent Date: Fri, 11 Nov 2011 13:29:04 +0100 Subject: [PATCH 0194/3089] backing-dev: ensure wakeup_timer is deleted bdi_prune_sb() in bdi_unregister() attempts to removes the bdi links from all super_blocks and then del_timer_sync() the writeback timer. However, this can race with __mark_inode_dirty(), leading to bdi_wakeup_thread_delayed() rearming the writeback timer on the bdi we're unregistering, after we've called del_timer_sync(). This can end up with the bdi being freed with an active timer inside it, as in the case of the following dump after the removal of an SD card. Fix this by redoing the del_timer_sync() in bdi_destory(). ------------[ cut here ]------------ WARNING: at /home/rabin/kernel/arm/lib/debugobjects.c:262 debug_print_object+0x9c/0xc8() ODEBUG: free active (active state 0) object type: timer_list hint: wakeup_timer_fn+0x0/0x180 Modules linked in: Backtrace: [] (dump_backtrace+0x0/0x110) from [] (dump_stack+0x18/0x1c) r6:c02bc638 r5:00000106 r4:c79f5d18 r3:00000000 [] (dump_stack+0x0/0x1c) from [] (warn_slowpath_common+0x54/0x6c) [] (warn_slowpath_common+0x0/0x6c) from [] (warn_slowpath_fmt+0x38/0x40) r8:20000013 r7:c780c6f0 r6:c031613c r5:c780c6f0 r4:c02b1b29 r3:00000009 [] (warn_slowpath_fmt+0x0/0x40) from [] (debug_print_object+0x9c/0xc8) r3:c02b1b29 r2:c02bc662 [] (debug_print_object+0x0/0xc8) from [] (debug_check_no_obj_freed+0xac/0x1dc) r6:c7964000 r5:00000001 r4:c7964000 [] (debug_check_no_obj_freed+0x0/0x1dc) from [] (kmem_cache_free+0x88/0x1f8) [] (kmem_cache_free+0x0/0x1f8) from [] (blk_release_queue+0x70/0x78) [] (blk_release_queue+0x0/0x78) from [] (kobject_release+0x70/0x84) r5:c79641f0 r4:c796420c [] (kobject_release+0x0/0x84) from [] (kref_put+0x68/0x80) r7:00000083 r6:c74083d0 r5:c015289c r4:c796420c [] (kref_put+0x0/0x80) from [] (kobject_put+0x48/0x5c) r5:c79643b4 r4:c79641f0 [] (kobject_put+0x0/0x5c) from [] (blk_cleanup_queue+0x68/0x74) r4:c7964000 [] (blk_cleanup_queue+0x0/0x74) from [] (mmc_blk_put+0x78/0xe8) r5:00000000 r4:c794c400 [] (mmc_blk_put+0x0/0xe8) from [] (mmc_blk_release+0x24/0x38) r5:c794c400 r4:c0322824 [] (mmc_blk_release+0x0/0x38) from [] (__blkdev_put+0xe8/0x170) r5:c78d5e00 r4:c74083c0 [] (__blkdev_put+0x0/0x170) from [] (blkdev_put+0x11c/0x12c) r8:c79f5f70 r7:00000001 r6:c74083d0 r5:00000083 r4:c74083c0 r3:00000000 [] (blkdev_put+0x0/0x12c) from [] (kill_block_super+0x60/0x6c) r7:c7942300 r6:c79f4000 r5:00000083 r4:c74083c0 [] (kill_block_super+0x0/0x6c) from [] (deactivate_locked_super+0x44/0x70) r6:c79f4000 r5:c031af64 r4:c794dc00 r3:c00b06c4 [] (deactivate_locked_super+0x0/0x70) from [] (deactivate_super+0x6c/0x70) r5:c794dc00 r4:c794dc00 [] (deactivate_super+0x0/0x70) from [] (mntput_no_expire+0x188/0x194) r5:c794dc00 r4:c7942300 [] (mntput_no_expire+0x0/0x194) from [] (sys_umount+0x2e4/0x310) r6:c7942300 r5:00000000 r4:00000000 r3:00000000 [] (sys_umount+0x0/0x310) from [] (ret_fast_syscall+0x0/0x30) ---[ end trace e5c83c92ada51c76 ]--- Cc: stable@kernel.org Signed-off-by: Rabin Vincent Signed-off-by: Linus Walleij Signed-off-by: Jens Axboe --- mm/backing-dev.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/mm/backing-dev.c b/mm/backing-dev.c index a0860640378d8..71034f41a2ba9 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -724,6 +724,14 @@ void bdi_destroy(struct backing_dev_info *bdi) bdi_unregister(bdi); + /* + * If bdi_unregister() had already been called earlier, the + * wakeup_timer could still be armed because bdi_prune_sb() + * can race with the bdi_wakeup_thread_delayed() calls from + * __mark_inode_dirty(). + */ + del_timer_sync(&bdi->wb.wakeup_timer); + for (i = 0; i < NR_BDI_STAT_ITEMS; i++) percpu_counter_destroy(&bdi->bdi_stat[i]); -- GitLab From cf16123c9c8e346ed1dd171295a678d77648d7f8 Mon Sep 17 00:00:00 2001 From: Vasily Averin Date: Fri, 11 Nov 2011 13:42:16 +0400 Subject: [PATCH 0195/3089] [SCSI] aacraid: controller hangs if kernel uses non-default ASPM policy Aacraid controller can hang on some nodes if kernel uses non-default (powersave) ASPM policy. Controller hangs shortly after successful load and hardware detection. Scsi error handler detects this hang and tries to restart hardware but it does not help. Initially it was noticed on RHEL6-based openVZ kernel after backporting aacraid driver from mainline (RHEL6 kernel with original driver works well) http://bugzilla.openvz.org/show_bug.cgi?id=2043 This issue happens because default ASPM policy was changed in Red Hat kernels. Therefore guys from Red Hat have noticed this problem long time ago: on Fedora 12 https://bugzilla.redhat.com/show_bug.cgi?id=540478 on Fedora 14 https://bugzilla.redhat.com/show_bug.cgi?id=679385 In RHEL6 kernel this issue was fixed, ASPM was disabled in aacraid driver. In kernel changelog I've found that seems it was done by Matthew Garrett: - [scsi] aacraid: Disable ASPM by default (Matthew Garrett) [599735] However seems this patch was not submitted to mainline. I've reproduced this issue on vanilla 3.1.0 kernel booted with "pcie_aspm.policy=powersave" option, So I believe it makes sense to do it now. Signed-off-by: Vasily Averin [mjg: Checking the Windows drivers indicates that they disable ASPM under all circumstances, so:] Acked-by: Matthew Garrett Acked-by: Achim Leubner Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/scsi/aacraid/linit.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c index 4aa76d6f11dfe..705e13e470af1 100644 --- a/drivers/scsi/aacraid/linit.c +++ b/drivers/scsi/aacraid/linit.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #include #include @@ -1109,6 +1110,9 @@ static int __devinit aac_probe_one(struct pci_dev *pdev, unique_id++; } + pci_disable_link_state(pdev, PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1 | + PCIE_LINK_STATE_CLKPM); + error = pci_enable_device(pdev); if (error) goto out; -- GitLab From 6a53fc531dcf4719e80d0152ca044e35b0f92591 Mon Sep 17 00:00:00 2001 From: Fabio Estevam Date: Thu, 25 Aug 2011 08:52:36 -0300 Subject: [PATCH 0196/3089] ARM: imx: Remove unused chip revision strings Since commit (167a19d2: ARM: imx: Introduce generic function for displaying silicon revision) we no longer need the chip revision strings, so remove them. Signed-off-by: Fabio Estevam Signed-off-by: Sascha Hauer --- arch/arm/plat-mxc/include/mach/mxc.h | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/arch/arm/plat-mxc/include/mach/mxc.h b/arch/arm/plat-mxc/include/mach/mxc.h index 00a78193c681a..a4d36d601d55a 100644 --- a/arch/arm/plat-mxc/include/mach/mxc.h +++ b/arch/arm/plat-mxc/include/mach/mxc.h @@ -50,20 +50,6 @@ #define IMX_CHIP_REVISION_3_3 0x33 #define IMX_CHIP_REVISION_UNKNOWN 0xff -#define IMX_CHIP_REVISION_1_0_STRING "1.0" -#define IMX_CHIP_REVISION_1_1_STRING "1.1" -#define IMX_CHIP_REVISION_1_2_STRING "1.2" -#define IMX_CHIP_REVISION_1_3_STRING "1.3" -#define IMX_CHIP_REVISION_2_0_STRING "2.0" -#define IMX_CHIP_REVISION_2_1_STRING "2.1" -#define IMX_CHIP_REVISION_2_2_STRING "2.2" -#define IMX_CHIP_REVISION_2_3_STRING "2.3" -#define IMX_CHIP_REVISION_3_0_STRING "3.0" -#define IMX_CHIP_REVISION_3_1_STRING "3.1" -#define IMX_CHIP_REVISION_3_2_STRING "3.2" -#define IMX_CHIP_REVISION_3_3_STRING "3.3" -#define IMX_CHIP_REVISION_UNKNOWN_STRING "unknown" - #ifndef __ASSEMBLY__ extern unsigned int __mxc_cpu_type; #endif -- GitLab From 8c6d8319eca370f804fb8e1fcf7070b587d1b81d Mon Sep 17 00:00:00 2001 From: Shawn Guo Date: Fri, 11 Nov 2011 13:09:18 +0800 Subject: [PATCH 0197/3089] arm/imx: remove imx_idle hook and use pm_idle instead The patch removes imx_idle hook and use pm_idle instead to get imx arch_idle prepared for the cleanup. It's suggested by Russel King as below. > The final removal of mach/system.h depends on getting rid of the arch_idle > thing. While going through these headers, I was dismayed to find these: > > arch/arm/mach-s3c2410/include/mach/system.h:void (*s3c24xx_idle)(void); > arch/arm/plat-mxc/include/mach/system.h:extern void (*imx_idle)(void); > > when we have a perfectly good pm_idle hook already in place - so there's > no excuse for these especially when other platforms are already using > pm_idle to hook their platform specific idle function into. This is > something that better be gone at the next merge window! Suggested-by: Russell King Signed-off-by: Shawn Guo Acked-by: Russell King Signed-off-by: Sascha Hauer --- arch/arm/mach-imx/mm-imx3.c | 53 +++++++++++++------------ arch/arm/mach-mx5/mm.c | 6 ++- arch/arm/plat-mxc/include/mach/common.h | 1 - arch/arm/plat-mxc/include/mach/system.h | 7 +--- arch/arm/plat-mxc/system.c | 1 - 5 files changed, 33 insertions(+), 35 deletions(-) diff --git a/arch/arm/mach-imx/mm-imx3.c b/arch/arm/mach-imx/mm-imx3.c index 9f0e82ec3398d..6d01de326f906 100644 --- a/arch/arm/mach-imx/mm-imx3.c +++ b/arch/arm/mach-imx/mm-imx3.c @@ -33,29 +33,32 @@ static void imx3_idle(void) { unsigned long reg = 0; - __asm__ __volatile__( - /* disable I and D cache */ - "mrc p15, 0, %0, c1, c0, 0\n" - "bic %0, %0, #0x00001000\n" - "bic %0, %0, #0x00000004\n" - "mcr p15, 0, %0, c1, c0, 0\n" - /* invalidate I cache */ - "mov %0, #0\n" - "mcr p15, 0, %0, c7, c5, 0\n" - /* clear and invalidate D cache */ - "mov %0, #0\n" - "mcr p15, 0, %0, c7, c14, 0\n" - /* WFI */ - "mov %0, #0\n" - "mcr p15, 0, %0, c7, c0, 4\n" - "nop\n" "nop\n" "nop\n" "nop\n" - "nop\n" "nop\n" "nop\n" - /* enable I and D cache */ - "mrc p15, 0, %0, c1, c0, 0\n" - "orr %0, %0, #0x00001000\n" - "orr %0, %0, #0x00000004\n" - "mcr p15, 0, %0, c1, c0, 0\n" - : "=r" (reg)); + + if (!need_resched()) + __asm__ __volatile__( + /* disable I and D cache */ + "mrc p15, 0, %0, c1, c0, 0\n" + "bic %0, %0, #0x00001000\n" + "bic %0, %0, #0x00000004\n" + "mcr p15, 0, %0, c1, c0, 0\n" + /* invalidate I cache */ + "mov %0, #0\n" + "mcr p15, 0, %0, c7, c5, 0\n" + /* clear and invalidate D cache */ + "mov %0, #0\n" + "mcr p15, 0, %0, c7, c14, 0\n" + /* WFI */ + "mov %0, #0\n" + "mcr p15, 0, %0, c7, c0, 4\n" + "nop\n" "nop\n" "nop\n" "nop\n" + "nop\n" "nop\n" "nop\n" + /* enable I and D cache */ + "mrc p15, 0, %0, c1, c0, 0\n" + "orr %0, %0, #0x00001000\n" + "orr %0, %0, #0x00000004\n" + "mcr p15, 0, %0, c1, c0, 0\n" + : "=r" (reg)); + local_irq_enable(); } static void __iomem *imx3_ioremap(unsigned long phys_addr, size_t size, @@ -143,7 +146,7 @@ void __init imx31_init_early(void) { mxc_set_cpu_type(MXC_CPU_MX31); mxc_arch_reset_init(MX31_IO_ADDRESS(MX31_WDOG_BASE_ADDR)); - imx_idle = imx3_idle; + pm_idle = imx3_idle; imx_ioremap = imx3_ioremap; } @@ -152,7 +155,7 @@ void __init imx35_init_early(void) mxc_set_cpu_type(MXC_CPU_MX35); mxc_iomux_v3_init(MX35_IO_ADDRESS(MX35_IOMUXC_BASE_ADDR)); mxc_arch_reset_init(MX35_IO_ADDRESS(MX35_WDOG_BASE_ADDR)); - imx_idle = imx3_idle; + pm_idle = imx3_idle; imx_ioremap = imx3_ioremap; } diff --git a/arch/arm/mach-mx5/mm.c b/arch/arm/mach-mx5/mm.c index 26eacc9d0d90f..df4a508f240a0 100644 --- a/arch/arm/mach-mx5/mm.c +++ b/arch/arm/mach-mx5/mm.c @@ -23,7 +23,9 @@ static void imx5_idle(void) { - mx5_cpu_lp_set(WAIT_UNCLOCKED_POWER_OFF); + if (!need_resched()) + mx5_cpu_lp_set(WAIT_UNCLOCKED_POWER_OFF); + local_irq_enable(); } /* @@ -89,7 +91,7 @@ void __init imx51_init_early(void) mxc_set_cpu_type(MXC_CPU_MX51); mxc_iomux_v3_init(MX51_IO_ADDRESS(MX51_IOMUXC_BASE_ADDR)); mxc_arch_reset_init(MX51_IO_ADDRESS(MX51_WDOG1_BASE_ADDR)); - imx_idle = imx5_idle; + pm_idle = imx5_idle; } void __init imx53_init_early(void) diff --git a/arch/arm/plat-mxc/include/mach/common.h b/arch/arm/plat-mxc/include/mach/common.h index 83b745a5e1b72..7143d52c25e18 100644 --- a/arch/arm/plat-mxc/include/mach/common.h +++ b/arch/arm/plat-mxc/include/mach/common.h @@ -85,7 +85,6 @@ enum mxc_cpu_pwr_mode { }; extern void mx5_cpu_lp_set(enum mxc_cpu_pwr_mode mode); -extern void (*imx_idle)(void); extern void imx_print_silicon_rev(const char *cpu, int srev); void avic_handle_irq(struct pt_regs *); diff --git a/arch/arm/plat-mxc/include/mach/system.h b/arch/arm/plat-mxc/include/mach/system.h index cf88b3593fba7..b9895d250167c 100644 --- a/arch/arm/plat-mxc/include/mach/system.h +++ b/arch/arm/plat-mxc/include/mach/system.h @@ -17,14 +17,9 @@ #ifndef __ASM_ARCH_MXC_SYSTEM_H__ #define __ASM_ARCH_MXC_SYSTEM_H__ -extern void (*imx_idle)(void); - static inline void arch_idle(void) { - if (imx_idle != NULL) - (imx_idle)(); - else - cpu_do_idle(); + cpu_do_idle(); } void arch_reset(char mode, const char *cmd); diff --git a/arch/arm/plat-mxc/system.c b/arch/arm/plat-mxc/system.c index 9dad8dcc2ea9d..a0bcccf9b640d 100644 --- a/arch/arm/plat-mxc/system.c +++ b/arch/arm/plat-mxc/system.c @@ -28,7 +28,6 @@ #include #include -void (*imx_idle)(void) = NULL; void __iomem *(*imx_ioremap)(unsigned long, size_t, unsigned int) = NULL; static void __iomem *wdog_base; -- GitLab From 71ca86937513f44151454cb6aa8a66e9ed86275a Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 10 Nov 2011 14:15:57 +0000 Subject: [PATCH 0198/3089] MIPS: Fix build error due to missing inclusion of . Signed-off-by: Ralf Baechle --- arch/mips/pci/pci-alchemy.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/mips/pci/pci-alchemy.c b/arch/mips/pci/pci-alchemy.c index 4ee57104e47bb..b5ce041cdafb5 100644 --- a/arch/mips/pci/pci-alchemy.c +++ b/arch/mips/pci/pci-alchemy.c @@ -7,6 +7,7 @@ * Support for all devices (greater than 16) added by David Gathright. */ +#include #include #include #include -- GitLab From 0027cb3e1947d0f453fece40ed16764fb362bac6 Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Thu, 10 Nov 2011 16:40:37 +0900 Subject: [PATCH 0199/3089] PCI: pciehp: wait 1000 ms before Link Training check We need to wait for 1000 ms after Data Link Layer Link Active (DLLLA) bit reads 1b before sending configuration request. Currently pciehp does this wait after checking Link Training (LT) bit. But we need it before checking LT bit because LT is still set even after DLLLA bit is set on some platforms. Acked-by: Yinghai Lu Tested-by: Yinghai Lu Signed-off-by: Kenji Kaneshige Signed-off-by: Jesse Barnes --- drivers/pci/hotplug/pciehp_ctrl.c | 3 --- drivers/pci/hotplug/pciehp_hpc.c | 8 ++++++++ 2 files changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/pci/hotplug/pciehp_ctrl.c b/drivers/pci/hotplug/pciehp_ctrl.c index 1e9c9aacc3a6c..085dbb5fc168b 100644 --- a/drivers/pci/hotplug/pciehp_ctrl.c +++ b/drivers/pci/hotplug/pciehp_ctrl.c @@ -213,9 +213,6 @@ static int board_added(struct slot *p_slot) goto err_exit; } - /* Wait for 1 second after checking link training status */ - msleep(1000); - /* Check for a power fault */ if (ctrl->power_fault_detected || pciehp_query_power_fault(p_slot)) { ctrl_err(ctrl, "Power fault on slot %s\n", slot_name(p_slot)); diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c index 6692832c9c81d..81a177a5f0322 100644 --- a/drivers/pci/hotplug/pciehp_hpc.c +++ b/drivers/pci/hotplug/pciehp_hpc.c @@ -280,6 +280,14 @@ int pciehp_check_link_status(struct controller *ctrl) else msleep(1000); + /* + * Need to wait for 1000 ms after Data Link Layer Link Active + * (DLLLA) bit reads 1b before sending configuration request. + * We need it before checking Link Training (LT) bit becuase + * LT is still set even after DLLLA bit is set on some platform. + */ + msleep(1000); + retval = pciehp_readw(ctrl, PCI_EXP_LNKSTA, &lnk_status); if (retval) { ctrl_err(ctrl, "Cannot read LNKSTATUS register\n"); -- GitLab From b3c004542229099e18198061c737e13eafc8d4d6 Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Thu, 10 Nov 2011 16:42:16 +0900 Subject: [PATCH 0200/3089] PCI: pciehp: wait 100 ms after Link Training check If the port supports Link speeds greater than 5.0 GT/s, we must wait for 100 ms after Link training completes before sending configuration request. Acked-by: Yinghai Lu Tested-by: Yinghai Lu Signed-off-by: Kenji Kaneshige Signed-off-by: Jesse Barnes --- drivers/pci/hotplug/pciehp_hpc.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/pci/hotplug/pciehp_hpc.c b/drivers/pci/hotplug/pciehp_hpc.c index 81a177a5f0322..7b1414810ae3e 100644 --- a/drivers/pci/hotplug/pciehp_hpc.c +++ b/drivers/pci/hotplug/pciehp_hpc.c @@ -302,6 +302,14 @@ int pciehp_check_link_status(struct controller *ctrl) return retval; } + /* + * If the port supports Link speeds greater than 5.0 GT/s, we + * must wait for 100 ms after Link training completes before + * sending configuration request. + */ + if (ctrl->pcie->port->subordinate->max_bus_speed > PCIE_SPEED_5_0GT) + msleep(100); + pcie_update_link_speed(ctrl->pcie->port->subordinate, lnk_status); return retval; -- GitLab From 774ac21da76f5c3018428725074e27a3fd40b128 Mon Sep 17 00:00:00 2001 From: Sage Weil Date: Fri, 11 Nov 2011 09:48:08 -0800 Subject: [PATCH 0201/3089] ceph: initialize root dentry Set up d_fsdata on the root dentry. This fixes a NULL pointer dereference in ceph_d_prune on umount. It also means we can eventually strip out all of the conditional checks on d_fsdata because it is now set unconditionally (prior to setting up the d_ops). Fix the ceph_d_prune debug print while we're here. Signed-off-by: Sage Weil --- fs/ceph/dir.c | 2 +- fs/ceph/super.c | 6 ++++-- 2 files changed, 5 insertions(+), 3 deletions(-) diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 2abd0dfad7f80..bca3948e9dbf6 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -1143,7 +1143,7 @@ static void ceph_d_prune(struct dentry *dentry) { struct ceph_dentry_info *di; - dout("d_release %p\n", dentry); + dout("ceph_d_prune %p\n", dentry); /* do we have a valid parent? */ if (!dentry->d_parent || IS_ROOT(dentry)) diff --git a/fs/ceph/super.c b/fs/ceph/super.c index a90846fac759b..8dc73a594a90c 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -638,10 +638,12 @@ static struct dentry *open_root_dentry(struct ceph_fs_client *fsc, if (err == 0) { dout("open_root_inode success\n"); if (ceph_ino(req->r_target_inode) == CEPH_INO_ROOT && - fsc->sb->s_root == NULL) + fsc->sb->s_root == NULL) { root = d_alloc_root(req->r_target_inode); - else + ceph_init_dentry(root); + } else { root = d_obtain_alias(req->r_target_inode); + } req->r_target_inode = NULL; dout("open_root_inode success, root dentry is %p\n", root); } else { -- GitLab From 224736d9113ab4a7cf3f05c05377492bd99b4b02 Mon Sep 17 00:00:00 2001 From: Stratos Psomadakis Date: Thu, 10 Nov 2011 15:45:37 +0200 Subject: [PATCH 0202/3089] libceph: Allocate larger oid buffer in request msgs ceph_osd_request struct allocates a 40-byte buffer for object names. RBD image names can be up to 96 chars long (100 with the .rbd suffix), which results in the object name for the image being truncated, and a subsequent map failure. Increase the oid buffer in request messages, in order to avoid the truncation. Signed-off-by: Stratos Psomadakis Signed-off-by: Sage Weil --- include/linux/ceph/osd_client.h | 8 +++++++- net/ceph/osd_client.c | 2 +- 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/include/linux/ceph/osd_client.h b/include/linux/ceph/osd_client.h index f88eacb111d41..7c05ac202d906 100644 --- a/include/linux/ceph/osd_client.h +++ b/include/linux/ceph/osd_client.h @@ -10,6 +10,12 @@ #include "osdmap.h" #include "messenger.h" +/* + * Maximum object name size + * (must be at least as big as RBD_MAX_MD_NAME_LEN -- currently 100) + */ +#define MAX_OBJ_NAME_SIZE 100 + struct ceph_msg; struct ceph_snap_context; struct ceph_osd_request; @@ -75,7 +81,7 @@ struct ceph_osd_request { struct inode *r_inode; /* for use by callbacks */ void *r_priv; /* ditto */ - char r_oid[40]; /* object name */ + char r_oid[MAX_OBJ_NAME_SIZE]; /* object name */ int r_oid_len; unsigned long r_stamp; /* send OR check time */ diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c index 733e46008b89d..f4f3f58f5234c 100644 --- a/net/ceph/osd_client.c +++ b/net/ceph/osd_client.c @@ -244,7 +244,7 @@ struct ceph_osd_request *ceph_osdc_alloc_request(struct ceph_osd_client *osdc, ceph_pagelist_init(req->r_trail); } /* create request message; allow space for oid */ - msg_size += 40; + msg_size += MAX_OBJ_NAME_SIZE; if (snapc) msg_size += sizeof(u64) * snapc->num_snaps; if (use_mempool) -- GitLab From e9b7086b80c4d9e354f4edc9e280ae85a60df408 Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Fri, 11 Nov 2011 10:15:11 -0800 Subject: [PATCH 0203/3089] ARM: OMAP: Fix reprogramming of dpll1 rate Commit a66cb3454f220f49f900646ebdc76cb943319eb7 (ARM: OMAP: Map SRAM later on with ioremap_exec()) moved the SRAM init to happen later to remove a dependency to early SoC detection for map_io. This broke booting on some boards not using Kconfig option for OMAP_CLOCKS_SET_BY_BOOTLOADER as the dpll1 reprogramming would cause the following error: kernel BUG at arch/arm/plat-omap/sram.c:226! Internal error: Oops - undefined instruction: 0 [#1] PREEMPT Modules linked in: CPU: 0 Not tainted (3.2.0-rc1-e3 #9) PC is at omap_sram_reprogram_clock+0x28/0x30 LR is at omap1_select_table_rate+0x88/0xb4 pc : [] lr : [] psr: 600000d3 sp : c035bf10 ip : c035bf20 fp : c035bf1c r10: c035bfd4 r9 : 54029252 r8 : c03f8120 r7 : c0362b50 r6 : 00b71b00 r5 : c03873cc r4 : c0362b40 r3 : 00000000 r2 : c0362b40 r1 : 0000010a r0 : 00002cb0 Flags: nZCv IRQs off FIQs off Mode SVC_32 ISA ARM Segment kernel Control: 0000317f Table: 10004000 DAC: 00000017 Process swapper (pid: 0, stack limit = 0xc035a270) Stack: (0xc035bf10 to 0xc035c000) bf00: c035bf3c c035bf20 c0019f54 c001b0ac bf20: 00001000 00002cb3 00000004 c035ed4c c035bf74 c035bf40 c033ea24 c0019edc bf40: c02f526c 00000002 00000015 bc058c9b 93111a16 c035335c 02000000 c035ed4c bf60: c035ed4c c03f8120 c035bf84 c035bf78 c00194c4 c033e8ec c035bfc4 c035bf88 bf80: c033bc24 c00194a0 c035bf90 c035bf98 00000000 00000000 00000000 00000000 bfa0: 00000001 00000000 c0354678 c035ece4 10004000 103532f4 c035bff4 c035bfc8 bfc0: c0338574 c033b598 00000000 00000000 00000000 c035467c 0000317d c035c03c bfe0: c0354678 c035ece4 00000000 c035bff8 10008040 c0338508 00000000 00000000 Backtrace: [] (omap_sram_reprogram_clock+0x0/0x30) from [] (omap1_select_table_rate+0x88/0xb4) [] (omap1_select_table_rate+0x0/0xb4) from [] (omap1_clk_init+0x148/0x334) r7:c035ed4c r6:00000004 r5:00002cb3 r4:00001000 [] (omap1_clk_init+0x0/0x334) from [] (omap1_init_early+0x34/0x48) r8:c03f8120 r7:c035ed4c r6:c035ed4c r5:02000000 r4:c035335c [] (omap1_init_early+0x0/0x48) from [] (setup_arch+0x69c/0x79c) [] (setup_arch+0x0/0x79c) from [] (start_kernel+0x7c/0x2f4) [] (start_kernel+0x0/0x2f4) from [<10008040>] (0x10008040) r7:c035ece4 r6:c0354678 r5:c035c03c r4:0000317d Code: 0a000002 e1a0e00f e12fff13 e89da800 (e7f001f2) Fix this by adding omap1_clk_late_init() that only reprograms dpll1 if the bootloader rate is less than 60MHz. This also allows removing of the OMAP_CLOCKS_SET_BY_BOOTLOADER option. Reported-by: Aaro Koskinen Tested-by: Aaro Koskinen Signed-off-by: Tony Lindgren --- arch/arm/configs/omap1_defconfig | 1 - arch/arm/mach-omap1/Kconfig | 8 ----- arch/arm/mach-omap1/clock.h | 3 +- arch/arm/mach-omap1/clock_data.c | 53 ++++++++++++++++++++------------ arch/arm/mach-omap1/devices.c | 3 ++ 5 files changed, 39 insertions(+), 29 deletions(-) diff --git a/arch/arm/configs/omap1_defconfig b/arch/arm/configs/omap1_defconfig index 7b63462b349d7..a7e7775813788 100644 --- a/arch/arm/configs/omap1_defconfig +++ b/arch/arm/configs/omap1_defconfig @@ -48,7 +48,6 @@ CONFIG_MACH_SX1=y CONFIG_MACH_NOKIA770=y CONFIG_MACH_AMS_DELTA=y CONFIG_MACH_OMAP_GENERIC=y -CONFIG_OMAP_CLOCKS_SET_BY_BOOTLOADER=y CONFIG_OMAP_ARM_216MHZ=y CONFIG_OMAP_ARM_195MHZ=y CONFIG_OMAP_ARM_192MHZ=y diff --git a/arch/arm/mach-omap1/Kconfig b/arch/arm/mach-omap1/Kconfig index e0a028161ddee..73f287d6429b6 100644 --- a/arch/arm/mach-omap1/Kconfig +++ b/arch/arm/mach-omap1/Kconfig @@ -171,14 +171,6 @@ config MACH_OMAP_GENERIC comment "OMAP CPU Speed" depends on ARCH_OMAP1 -config OMAP_CLOCKS_SET_BY_BOOTLOADER - bool "OMAP clocks set by bootloader" - depends on ARCH_OMAP1 - help - Enable this option to prevent the kernel from overriding the clock - frequencies programmed by bootloader for MPU, DSP, MMUs, TC, - internal LCD controller and MPU peripherals. - config OMAP_ARM_216MHZ bool "OMAP ARM 216 MHz CPU (1710 only)" depends on ARCH_OMAP1 && ARCH_OMAP16XX diff --git a/arch/arm/mach-omap1/clock.h b/arch/arm/mach-omap1/clock.h index eaf09efb91cae..16b1423b454a3 100644 --- a/arch/arm/mach-omap1/clock.h +++ b/arch/arm/mach-omap1/clock.h @@ -17,7 +17,8 @@ #include -extern int __init omap1_clk_init(void); +int omap1_clk_init(void); +void omap1_clk_late_init(void); extern int omap1_clk_enable(struct clk *clk); extern void omap1_clk_disable(struct clk *clk); extern long omap1_clk_round_rate(struct clk *clk, unsigned long rate); diff --git a/arch/arm/mach-omap1/clock_data.c b/arch/arm/mach-omap1/clock_data.c index 92400b9eb69f1..1297bb58869cf 100644 --- a/arch/arm/mach-omap1/clock_data.c +++ b/arch/arm/mach-omap1/clock_data.c @@ -767,6 +767,15 @@ static struct clk_functions omap1_clk_functions = { .clk_disable_unused = omap1_clk_disable_unused, }; +static void __init omap1_show_rates(void) +{ + pr_notice("Clocking rate (xtal/DPLL1/MPU): " + "%ld.%01ld/%ld.%01ld/%ld.%01ld MHz\n", + ck_ref.rate / 1000000, (ck_ref.rate / 100000) % 10, + ck_dpll1.rate / 1000000, (ck_dpll1.rate / 100000) % 10, + arm_ck.rate / 1000000, (arm_ck.rate / 100000) % 10); +} + int __init omap1_clk_init(void) { struct omap_clk *c; @@ -835,9 +844,12 @@ int __init omap1_clk_init(void) /* We want to be in syncronous scalable mode */ omap_writew(0x1000, ARM_SYSST); -#ifdef CONFIG_OMAP_CLOCKS_SET_BY_BOOTLOADER - /* Use values set by bootloader. Determine PLL rate and recalculate - * dependent clocks as if kernel had changed PLL or divisors. + + /* + * Initially use the values set by bootloader. Determine PLL rate and + * recalculate dependent clocks as if kernel had changed PLL or + * divisors. See also omap1_clk_late_init() that can reprogram dpll1 + * after the SRAM is initialized. */ { unsigned pll_ctl_val = omap_readw(DPLL_CTL); @@ -862,25 +874,10 @@ int __init omap1_clk_init(void) } } } -#else - /* Find the highest supported frequency and enable it */ - if (omap1_select_table_rate(&virtual_ck_mpu, ~0)) { - printk(KERN_ERR "System frequencies not set. Check your config.\n"); - /* Guess sane values (60MHz) */ - omap_writew(0x2290, DPLL_CTL); - omap_writew(cpu_is_omap7xx() ? 0x3005 : 0x1005, ARM_CKCTL); - ck_dpll1.rate = 60000000; - } -#endif propagate_rate(&ck_dpll1); /* Cache rates for clocks connected to ck_ref (not dpll1) */ propagate_rate(&ck_ref); - printk(KERN_INFO "Clocking rate (xtal/DPLL1/MPU): " - "%ld.%01ld/%ld.%01ld/%ld.%01ld MHz\n", - ck_ref.rate / 1000000, (ck_ref.rate / 100000) % 10, - ck_dpll1.rate / 1000000, (ck_dpll1.rate / 100000) % 10, - arm_ck.rate / 1000000, (arm_ck.rate / 100000) % 10); - + omap1_show_rates(); if (machine_is_omap_perseus2() || machine_is_omap_fsample()) { /* Select slicer output as OMAP input clock */ omap_writew(omap_readw(OMAP7XX_PCC_UPLD_CTRL) & ~0x1, @@ -925,3 +922,21 @@ int __init omap1_clk_init(void) return 0; } + +#define OMAP1_DPLL1_SANE_VALUE 60000000 + +void __init omap1_clk_late_init(void) +{ + if (ck_dpll1.rate >= OMAP1_DPLL1_SANE_VALUE) + return; + + /* Find the highest supported frequency and enable it */ + if (omap1_select_table_rate(&virtual_ck_mpu, ~0)) { + pr_err("System frequencies not set, using default. Check your config.\n"); + omap_writew(0x2290, DPLL_CTL); + omap_writew(cpu_is_omap7xx() ? 0x3005 : 0x1005, ARM_CKCTL); + ck_dpll1.rate = OMAP1_DPLL1_SANE_VALUE; + } + propagate_rate(&ck_dpll1); + omap1_show_rates(); +} diff --git a/arch/arm/mach-omap1/devices.c b/arch/arm/mach-omap1/devices.c index 48ef9888e820e..475cb2f50d872 100644 --- a/arch/arm/mach-omap1/devices.c +++ b/arch/arm/mach-omap1/devices.c @@ -30,6 +30,8 @@ #include #include +#include "clock.h" + /*-------------------------------------------------------------------------*/ #if defined(CONFIG_RTC_DRV_OMAP) || defined(CONFIG_RTC_DRV_OMAP_MODULE) @@ -293,6 +295,7 @@ static int __init omap1_init_devices(void) return -ENODEV; omap_sram_init(); + omap1_clk_late_init(); /* please keep these calls, and their implementations above, * in alphabetical order so they're easier to sort through. -- GitLab From 40809565ca57a8e94bae20b22da014c44ec233f6 Mon Sep 17 00:00:00 2001 From: Casey Schaufler Date: Thu, 10 Nov 2011 15:02:22 -0800 Subject: [PATCH 0204/3089] Smack: smackfs cipso seq read repair Commit 272cd7a8c67dd40a31ecff76a503bbb84707f757 introduced a change to the way rule lists are handled and reported in the smackfs filesystem. One of the issues addressed had to do with the termination of read requests on /smack/load. This change introduced a error in /smack/cipso, which shares some of the same list processing code. This patch updates all the file access list handling in smackfs to use the code introduced for /smack/load. Signed-off-by: Casey Schaufler --- security/smack/smackfs.c | 115 +++++++++++++-------------------------- 1 file changed, 39 insertions(+), 76 deletions(-) diff --git a/security/smack/smackfs.c b/security/smack/smackfs.c index 6aceef518a41f..5c32f36ff7061 100644 --- a/security/smack/smackfs.c +++ b/security/smack/smackfs.c @@ -102,9 +102,6 @@ static int smk_cipso_doi_value = SMACK_CIPSO_DOI_DEFAULT; const char *smack_cipso_option = SMACK_CIPSO_OPTION; - -#define SEQ_READ_FINISHED ((loff_t)-1) - /* * Values for parsing cipso rules * SMK_DIGITLEN: Length of a digit field in a rule. @@ -357,10 +354,12 @@ static ssize_t smk_write_load_list(struct file *file, const char __user *buf, rc = count; /* + * If this is "load" as opposed to "load-self" and a new rule + * it needs to get added for reporting. * smk_set_access returns true if there was already a rule * for the subject/object pair, and false if it was new. */ - if (!smk_set_access(rule, rule_list, rule_lock)) { + if (load && !smk_set_access(rule, rule_list, rule_lock)) { smlp = kzalloc(sizeof(*smlp), GFP_KERNEL); if (smlp != NULL) { smlp->smk_rule = rule; @@ -377,12 +376,12 @@ out: return rc; } - /* - * Seq_file read operations for /smack/load + * Core logic for smackfs seq list operations. */ -static void *load_seq_start(struct seq_file *s, loff_t *pos) +static void *smk_seq_start(struct seq_file *s, loff_t *pos, + struct list_head *head) { struct list_head *list; @@ -390,7 +389,7 @@ static void *load_seq_start(struct seq_file *s, loff_t *pos) * This is 0 the first time through. */ if (s->index == 0) - s->private = &smack_rule_list; + s->private = head; if (s->private == NULL) return NULL; @@ -404,11 +403,12 @@ static void *load_seq_start(struct seq_file *s, loff_t *pos) return list; } -static void *load_seq_next(struct seq_file *s, void *v, loff_t *pos) +static void *smk_seq_next(struct seq_file *s, void *v, loff_t *pos, + struct list_head *head) { struct list_head *list = v; - if (list_is_last(list, &smack_rule_list)) { + if (list_is_last(list, head)) { s->private = NULL; return NULL; } @@ -416,6 +416,25 @@ static void *load_seq_next(struct seq_file *s, void *v, loff_t *pos) return list->next; } +static void smk_seq_stop(struct seq_file *s, void *v) +{ + /* No-op */ +} + +/* + * Seq_file read operations for /smack/load + */ + +static void *load_seq_start(struct seq_file *s, loff_t *pos) +{ + return smk_seq_start(s, pos, &smack_rule_list); +} + +static void *load_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + return smk_seq_next(s, v, pos, &smack_rule_list); +} + static int load_seq_show(struct seq_file *s, void *v) { struct list_head *list = v; @@ -446,16 +465,11 @@ static int load_seq_show(struct seq_file *s, void *v) return 0; } -static void load_seq_stop(struct seq_file *s, void *v) -{ - /* No-op */ -} - static const struct seq_operations load_seq_ops = { .start = load_seq_start, .next = load_seq_next, .show = load_seq_show, - .stop = load_seq_stop, + .stop = smk_seq_stop, }; /** @@ -574,28 +588,12 @@ static void smk_unlbl_ambient(char *oldambient) static void *cipso_seq_start(struct seq_file *s, loff_t *pos) { - if (*pos == SEQ_READ_FINISHED) - return NULL; - if (list_empty(&smack_known_list)) - return NULL; - - return smack_known_list.next; + return smk_seq_start(s, pos, &smack_known_list); } static void *cipso_seq_next(struct seq_file *s, void *v, loff_t *pos) { - struct list_head *list = v; - - /* - * labels with no associated cipso value wont be printed - * in cipso_seq_show - */ - if (list_is_last(list, &smack_known_list)) { - *pos = SEQ_READ_FINISHED; - return NULL; - } - - return list->next; + return smk_seq_next(s, v, pos, &smack_known_list); } /* @@ -634,16 +632,11 @@ static int cipso_seq_show(struct seq_file *s, void *v) return 0; } -static void cipso_seq_stop(struct seq_file *s, void *v) -{ - /* No-op */ -} - static const struct seq_operations cipso_seq_ops = { .start = cipso_seq_start, - .stop = cipso_seq_stop, .next = cipso_seq_next, .show = cipso_seq_show, + .stop = smk_seq_stop, }; /** @@ -788,23 +781,12 @@ static const struct file_operations smk_cipso_ops = { static void *netlbladdr_seq_start(struct seq_file *s, loff_t *pos) { - if (*pos == SEQ_READ_FINISHED) - return NULL; - if (list_empty(&smk_netlbladdr_list)) - return NULL; - return smk_netlbladdr_list.next; + return smk_seq_start(s, pos, &smk_netlbladdr_list); } static void *netlbladdr_seq_next(struct seq_file *s, void *v, loff_t *pos) { - struct list_head *list = v; - - if (list_is_last(list, &smk_netlbladdr_list)) { - *pos = SEQ_READ_FINISHED; - return NULL; - } - - return list->next; + return smk_seq_next(s, v, pos, &smk_netlbladdr_list); } #define BEBITS (sizeof(__be32) * 8) @@ -828,16 +810,11 @@ static int netlbladdr_seq_show(struct seq_file *s, void *v) return 0; } -static void netlbladdr_seq_stop(struct seq_file *s, void *v) -{ - /* No-op */ -} - static const struct seq_operations netlbladdr_seq_ops = { .start = netlbladdr_seq_start, - .stop = netlbladdr_seq_stop, .next = netlbladdr_seq_next, .show = netlbladdr_seq_show, + .stop = smk_seq_stop, }; /** @@ -1405,23 +1382,14 @@ static void *load_self_seq_start(struct seq_file *s, loff_t *pos) { struct task_smack *tsp = current_security(); - if (*pos == SEQ_READ_FINISHED) - return NULL; - if (list_empty(&tsp->smk_rules)) - return NULL; - return tsp->smk_rules.next; + return smk_seq_start(s, pos, &tsp->smk_rules); } static void *load_self_seq_next(struct seq_file *s, void *v, loff_t *pos) { struct task_smack *tsp = current_security(); - struct list_head *list = v; - if (list_is_last(list, &tsp->smk_rules)) { - *pos = SEQ_READ_FINISHED; - return NULL; - } - return list->next; + return smk_seq_next(s, v, pos, &tsp->smk_rules); } static int load_self_seq_show(struct seq_file *s, void *v) @@ -1453,16 +1421,11 @@ static int load_self_seq_show(struct seq_file *s, void *v) return 0; } -static void load_self_seq_stop(struct seq_file *s, void *v) -{ - /* No-op */ -} - static const struct seq_operations load_self_seq_ops = { .start = load_self_seq_start, .next = load_self_seq_next, .show = load_self_seq_show, - .stop = load_self_seq_stop, + .stop = smk_seq_stop, }; -- GitLab From 193733585692301f38d489b8ad8724c2f88349c0 Mon Sep 17 00:00:00 2001 From: Matthew Garrett Date: Fri, 11 Nov 2011 22:05:54 +0100 Subject: [PATCH 0205/3089] The Windows driver .inf disables ASPM on all cciss devices. Do the same. Signed-off-by: Matthew Garrett Cc: iss_storagedev@hp.com Acked-by: Mike Miller Signed-off-by: Jens Axboe --- drivers/block/cciss.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index 486f94ef24d49..5b690194dd991 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -24,6 +24,7 @@ #include #include #include +#include #include #include #include @@ -4319,6 +4320,10 @@ static int __devinit cciss_pci_init(ctlr_info_t *h) dev_warn(&h->pdev->dev, "controller appears to be disabled\n"); return -ENODEV; } + + pci_disable_link_state(h->pdev, PCIE_LINK_STATE_L0S | + PCIE_LINK_STATE_L1 | PCIE_LINK_STATE_CLKPM); + err = pci_enable_device(h->pdev); if (err) { dev_warn(&h->pdev->dev, "Unable to Enable PCI device\n"); -- GitLab From e7c86c7b264c57b4d6d121c48ce3be98a2e6f8c5 Mon Sep 17 00:00:00 2001 From: Olof Johansson Date: Wed, 9 Nov 2011 22:10:00 +0100 Subject: [PATCH 0206/3089] ARM: 7156/1: l2x0: fix compile error on !CONFIG_USE_OF fae2b89ab1 (ARM: l2x0: add empty l2x0_of_init) adds a static inline function that returns -ENODEV, but at least on tegra cache-l2x0.h is included without errno.h being pulled in first, resulting in compile errors if OF isn't enabled: In file included from arch/arm/mach-tegra/common.c:26: arch/arm/include/asm/hardware/cache-l2x0.h: In function 'l2x0_of_init': arch/arm/include/asm/hardware/cache-l2x0.h:110: error: 'ENODEV' undeclared (first use in this function) arch/arm/include/asm/hardware/cache-l2x0.h:110: error: (Each undeclared identifier is reported only once arch/arm/include/asm/hardware/cache-l2x0.h:110: error: for each function it appears in.) Add errno.h to the include file to make it self-contained. Signed-off-by: Olof Johansson Acked-by: Rob Herring Signed-off-by: Russell King --- arch/arm/include/asm/hardware/cache-l2x0.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/include/asm/hardware/cache-l2x0.h b/arch/arm/include/asm/hardware/cache-l2x0.h index 1db1143a94838..7df239bcdf274 100644 --- a/arch/arm/include/asm/hardware/cache-l2x0.h +++ b/arch/arm/include/asm/hardware/cache-l2x0.h @@ -20,6 +20,8 @@ #ifndef __ASM_ARM_HARDWARE_L2X0_H #define __ASM_ARM_HARDWARE_L2X0_H +#include + #define L2X0_CACHE_ID 0x000 #define L2X0_CACHE_TYPE 0x004 #define L2X0_CTRL 0x100 -- GitLab From a6dc54460e5405248773abef704c7980ab31af77 Mon Sep 17 00:00:00 2001 From: wanzongshun Date: Thu, 10 Nov 2011 07:05:41 +0100 Subject: [PATCH 0207/3089] ARM: 7157/1: fix a building WARNING for nuc900 This patch is to fix a WARNING which is the "modpost: Found 1 section mismatch(es)" for nuc900 platform. Signed-off-by: Wan ZongShun Signed-off-by: Russell King --- arch/arm/mach-w90x900/dev.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/mach-w90x900/dev.c b/arch/arm/mach-w90x900/dev.c index 7a1fa6adb7c32..5b0c38abacc10 100644 --- a/arch/arm/mach-w90x900/dev.c +++ b/arch/arm/mach-w90x900/dev.c @@ -422,7 +422,7 @@ struct platform_device nuc900_device_kpi = { /* LCD controller*/ -static struct nuc900fb_display __initdata nuc900_lcd_info[] = { +static struct nuc900fb_display nuc900_lcd_info[] = { /* Giantplus Technology GPM1040A0 320x240 Color TFT LCD */ [0] = { .type = LCM_DCCS_VA_SRC_RGB565, @@ -445,7 +445,7 @@ static struct nuc900fb_display __initdata nuc900_lcd_info[] = { }, }; -static struct nuc900fb_mach_info nuc900_fb_info __initdata = { +static struct nuc900fb_mach_info nuc900_fb_info = { #if defined(CONFIG_GPM1040A0_320X240) .displays = &nuc900_lcd_info[0], #else -- GitLab From 798681bf2127e60f16e9decc5e3a0b901f3ff3fe Mon Sep 17 00:00:00 2001 From: wanzongshun Date: Thu, 10 Nov 2011 07:06:29 +0100 Subject: [PATCH 0208/3089] ARM: 7158/1: add new MFP implement for NUC900 This patch is to add new MFP implement in mfp.c,since nuc900 mmc driver needs this function support. Signed-off-by: Wan ZongShun Signed-off-by: Russell King --- arch/arm/mach-w90x900/include/mach/mfp.h | 3 +- .../mach-w90x900/include/mach/nuc900_spi.h | 2 +- arch/arm/mach-w90x900/mfp.c | 48 ++++++++++++++++--- 3 files changed, 45 insertions(+), 8 deletions(-) diff --git a/arch/arm/mach-w90x900/include/mach/mfp.h b/arch/arm/mach-w90x900/include/mach/mfp.h index 94c0e71617c6c..23ef1f573abd3 100644 --- a/arch/arm/mach-w90x900/include/mach/mfp.h +++ b/arch/arm/mach-w90x900/include/mach/mfp.h @@ -19,6 +19,7 @@ extern void mfp_set_groupf(struct device *dev); extern void mfp_set_groupc(struct device *dev); extern void mfp_set_groupi(struct device *dev); -extern void mfp_set_groupg(struct device *dev); +extern void mfp_set_groupg(struct device *dev, const char *subname); +extern void mfp_set_groupd(struct device *dev, const char *subname); #endif /* __ASM_ARCH_MFP_H */ diff --git a/arch/arm/mach-w90x900/include/mach/nuc900_spi.h b/arch/arm/mach-w90x900/include/mach/nuc900_spi.h index bd94819e314fe..2c4e0c1285010 100644 --- a/arch/arm/mach-w90x900/include/mach/nuc900_spi.h +++ b/arch/arm/mach-w90x900/include/mach/nuc900_spi.h @@ -14,7 +14,7 @@ #ifndef __ASM_ARCH_SPI_H #define __ASM_ARCH_SPI_H -extern void mfp_set_groupg(struct device *dev); +extern void mfp_set_groupg(struct device *dev, const char *subname); struct nuc900_spi_info { unsigned int num_cs; diff --git a/arch/arm/mach-w90x900/mfp.c b/arch/arm/mach-w90x900/mfp.c index fb7fb627b1a56..9dd74612bb870 100644 --- a/arch/arm/mach-w90x900/mfp.c +++ b/arch/arm/mach-w90x900/mfp.c @@ -26,10 +26,8 @@ #define REG_MFSEL (W90X900_VA_GCR + 0xC) #define GPSELF (0x01 << 1) - #define GPSELC (0x03 << 2) -#define ENKPI (0x02 << 2) -#define ENNAND (0x01 << 2) +#define GPSELD (0x0f << 4) #define GPSELEI0 (0x01 << 26) #define GPSELEI1 (0x01 << 27) @@ -37,11 +35,16 @@ #define GPIOG0TO1 (0x03 << 14) #define GPIOG2TO3 (0x03 << 16) #define GPIOG22TO23 (0x03 << 22) +#define GPIOG18TO20 (0x07 << 18) #define ENSPI (0x0a << 14) #define ENI2C0 (0x01 << 14) #define ENI2C1 (0x01 << 16) #define ENAC97 (0x02 << 22) +#define ENSD1 (0x02 << 18) +#define ENSD0 (0x0a << 4) +#define ENKPI (0x02 << 2) +#define ENNAND (0x01 << 2) static DEFINE_MUTEX(mfp_mutex); @@ -127,16 +130,19 @@ void mfp_set_groupi(struct device *dev) } EXPORT_SYMBOL(mfp_set_groupi); -void mfp_set_groupg(struct device *dev) +void mfp_set_groupg(struct device *dev, const char *subname) { unsigned long mfpen; const char *dev_id; - BUG_ON(!dev); + BUG_ON((!dev) && (!subname)); mutex_lock(&mfp_mutex); - dev_id = dev_name(dev); + if (subname != NULL) + dev_id = subname; + else + dev_id = dev_name(dev); mfpen = __raw_readl(REG_MFSEL); @@ -152,6 +158,9 @@ void mfp_set_groupg(struct device *dev) } else if (strcmp(dev_id, "nuc900-audio") == 0) { mfpen &= ~(GPIOG22TO23); mfpen |= ENAC97;/*enable AC97*/ + } else if (strcmp(dev_id, "nuc900-mmc-port1") == 0) { + mfpen &= ~(GPIOG18TO20); + mfpen |= (ENSD1 | 0x01);/*enable sd1*/ } else { mfpen &= ~(GPIOG0TO1 | GPIOG2TO3);/*GPIOG[3:0]*/ } @@ -162,3 +171,30 @@ void mfp_set_groupg(struct device *dev) } EXPORT_SYMBOL(mfp_set_groupg); +void mfp_set_groupd(struct device *dev, const char *subname) +{ + unsigned long mfpen; + const char *dev_id; + + BUG_ON((!dev) && (!subname)); + + mutex_lock(&mfp_mutex); + + if (subname != NULL) + dev_id = subname; + else + dev_id = dev_name(dev); + + mfpen = __raw_readl(REG_MFSEL); + + if (strcmp(dev_id, "nuc900-mmc-port0") == 0) { + mfpen &= ~GPSELD;/*enable sd0*/ + mfpen |= ENSD0; + } else + mfpen &= (~GPSELD); + + __raw_writel(mfpen, REG_MFSEL); + + mutex_unlock(&mfp_mutex); +} +EXPORT_SYMBOL(mfp_set_groupd); -- GitLab From a34dbfb03f472ad3e063d3b1d1e88f7ef8f88bf1 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Fri, 11 Nov 2011 11:35:58 +0100 Subject: [PATCH 0209/3089] ARM: 7160/1: setup: avoid overflowing {elf,arch}_name from proc_info_list setup_processor copies the arch_name and elf_name fields out of the selected proc_info_list into two fixed size buffers. Since the proc_info_list structure is defined in a proc_*.S assembly file, this can lead to subtle errors if the strings defined there are too long (for example, corrupting the machine ID). This patch uses snprintf instead of sprintf to ensure that these buffers are not overrun. Signed-off-by: Will Deacon Signed-off-by: Russell King --- arch/arm/kernel/setup.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index 7e7977ab994ff..3448a3f9cc8c9 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -461,8 +461,10 @@ static void __init setup_processor(void) cpu_name, read_cpuid_id(), read_cpuid_id() & 15, proc_arch[cpu_architecture()], cr_alignment); - sprintf(init_utsname()->machine, "%s%c", list->arch_name, ENDIANNESS); - sprintf(elf_platform, "%s%c", list->elf_name, ENDIANNESS); + snprintf(init_utsname()->machine, __NEW_UTS_LEN + 1, "%s%c", + list->arch_name, ENDIANNESS); + snprintf(elf_platform, ELF_PLATFORM_SIZE, "%s%c", + list->elf_name, ENDIANNESS); elf_hwcap = list->elf_hwcap; #ifndef CONFIG_ARM_THUMB elf_hwcap &= ~HWCAP_THUMB; -- GitLab From 30128031d71741ef7d0e32c345e3bf02aa8a0704 Mon Sep 17 00:00:00 2001 From: Sathya Perla Date: Thu, 10 Nov 2011 19:17:57 +0000 Subject: [PATCH 0210/3089] be2net: init (vf)_if_handle/vf_pmac_id to handle failure scenarios Initialize if_handle, vf_if_handle and vf_pmac_id with "-1" so that in failure cases when be_clear() is called, we can skip over if_destroy/pmac_del cmds if they have not been created. Signed-off-by: Sathya Perla Signed-off-by: David S. Miller --- drivers/net/ethernet/emulex/benet/be.h | 8 ++- drivers/net/ethernet/emulex/benet/be_cmds.c | 9 ++-- drivers/net/ethernet/emulex/benet/be_cmds.h | 4 +- drivers/net/ethernet/emulex/benet/be_main.c | 55 ++++++++++++--------- 4 files changed, 42 insertions(+), 34 deletions(-) diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index 644e8fed83649..4163980d6bcdc 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -289,14 +289,12 @@ struct be_drv_stats { struct be_vf_cfg { unsigned char vf_mac_addr[ETH_ALEN]; - u32 vf_if_handle; - u32 vf_pmac_id; + int vf_if_handle; + int vf_pmac_id; u16 vf_vlan_tag; u32 vf_tx_rate; }; -#define BE_INVALID_PMAC_ID 0xffffffff - struct be_adapter { struct pci_dev *pdev; struct net_device *netdev; @@ -347,7 +345,7 @@ struct be_adapter { /* Ethtool knobs and info */ char fw_ver[FW_VER_LEN]; - u32 if_handle; /* Used to configure filtering */ + int if_handle; /* Used to configure filtering */ u32 pmac_id; /* MAC addr handle used by BE card */ u32 beacon_state; /* for set_phys_id */ diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c index 2c7b36673dfc2..c5912c4ed24f5 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.c +++ b/drivers/net/ethernet/emulex/benet/be_cmds.c @@ -695,12 +695,15 @@ err: } /* Uses synchronous MCCQ */ -int be_cmd_pmac_del(struct be_adapter *adapter, u32 if_id, u32 pmac_id, u32 dom) +int be_cmd_pmac_del(struct be_adapter *adapter, u32 if_id, int pmac_id, u32 dom) { struct be_mcc_wrb *wrb; struct be_cmd_req_pmac_del *req; int status; + if (pmac_id == -1) + return 0; + spin_lock_bh(&adapter->mcc_lock); wrb = wrb_from_mccq(adapter); @@ -1136,7 +1139,7 @@ err: } /* Uses MCCQ */ -int be_cmd_if_destroy(struct be_adapter *adapter, u32 interface_id, u32 domain) +int be_cmd_if_destroy(struct be_adapter *adapter, int interface_id, u32 domain) { struct be_mcc_wrb *wrb; struct be_cmd_req_if_destroy *req; @@ -1145,7 +1148,7 @@ int be_cmd_if_destroy(struct be_adapter *adapter, u32 interface_id, u32 domain) if (adapter->eeh_err) return -EIO; - if (!interface_id) + if (interface_id == -1) return 0; spin_lock_bh(&adapter->mcc_lock); diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h index a35cd03fac4e8..08180396891ca 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.h +++ b/drivers/net/ethernet/emulex/benet/be_cmds.h @@ -1417,11 +1417,11 @@ extern int be_cmd_mac_addr_query(struct be_adapter *adapter, u8 *mac_addr, extern int be_cmd_pmac_add(struct be_adapter *adapter, u8 *mac_addr, u32 if_id, u32 *pmac_id, u32 domain); extern int be_cmd_pmac_del(struct be_adapter *adapter, u32 if_id, - u32 pmac_id, u32 domain); + int pmac_id, u32 domain); extern int be_cmd_if_create(struct be_adapter *adapter, u32 cap_flags, u32 en_flags, u8 *mac, u32 *if_handle, u32 *pmac_id, u32 domain); -extern int be_cmd_if_destroy(struct be_adapter *adapter, u32 if_handle, +extern int be_cmd_if_destroy(struct be_adapter *adapter, int if_handle, u32 domain); extern int be_cmd_eq_create(struct be_adapter *adapter, struct be_queue_info *eq, int eq_delay); diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index bf266a00c7742..83d971d8a5a33 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -848,15 +848,11 @@ static int be_set_vf_mac(struct net_device *netdev, int vf, u8 *mac) if (!is_valid_ether_addr(mac) || (vf >= num_vfs)) return -EINVAL; - if (adapter->vf_cfg[vf].vf_pmac_id != BE_INVALID_PMAC_ID) - status = be_cmd_pmac_del(adapter, - adapter->vf_cfg[vf].vf_if_handle, - adapter->vf_cfg[vf].vf_pmac_id, vf + 1); + status = be_cmd_pmac_del(adapter, adapter->vf_cfg[vf].vf_if_handle, + adapter->vf_cfg[vf].vf_pmac_id, vf + 1); - status = be_cmd_pmac_add(adapter, mac, - adapter->vf_cfg[vf].vf_if_handle, + status = be_cmd_pmac_add(adapter, mac, adapter->vf_cfg[vf].vf_if_handle, &adapter->vf_cfg[vf].vf_pmac_id, vf + 1); - if (status) dev_err(&adapter->pdev->dev, "MAC %pM set on VF %d Failed\n", mac, vf); @@ -2488,17 +2484,13 @@ static void be_vf_clear(struct be_adapter *adapter) { u32 vf; - for (vf = 0; vf < num_vfs; vf++) { - if (adapter->vf_cfg[vf].vf_pmac_id != BE_INVALID_PMAC_ID) - be_cmd_pmac_del(adapter, - adapter->vf_cfg[vf].vf_if_handle, - adapter->vf_cfg[vf].vf_pmac_id, vf + 1); - } + for (vf = 0; vf < num_vfs; vf++) + be_cmd_pmac_del(adapter, adapter->vf_cfg[vf].vf_if_handle, + adapter->vf_cfg[vf].vf_pmac_id, vf + 1); for (vf = 0; vf < num_vfs; vf++) - if (adapter->vf_cfg[vf].vf_if_handle) - be_cmd_if_destroy(adapter, - adapter->vf_cfg[vf].vf_if_handle, vf + 1); + be_cmd_if_destroy(adapter, adapter->vf_cfg[vf].vf_if_handle, + vf + 1); } static int be_clear(struct be_adapter *adapter) @@ -2511,22 +2503,30 @@ static int be_clear(struct be_adapter *adapter) be_mcc_queues_destroy(adapter); be_rx_queues_destroy(adapter); be_tx_queues_destroy(adapter); - adapter->eq_next_idx = 0; - - adapter->be3_native = false; - adapter->promiscuous = false; /* tell fw we're done with firing cmds */ be_cmd_fw_clean(adapter); return 0; } +static void be_vf_setup_init(struct be_adapter *adapter) +{ + int vf; + + for (vf = 0; vf < num_vfs; vf++) { + adapter->vf_cfg[vf].vf_if_handle = -1; + adapter->vf_cfg[vf].vf_pmac_id = -1; + } +} + static int be_vf_setup(struct be_adapter *adapter) { u32 cap_flags, en_flags, vf; u16 lnk_speed; int status; + be_vf_setup_init(adapter); + cap_flags = en_flags = BE_IF_FLAGS_UNTAGGED | BE_IF_FLAGS_BROADCAST; for (vf = 0; vf < num_vfs; vf++) { status = be_cmd_if_create(adapter, cap_flags, en_flags, NULL, @@ -2534,7 +2534,6 @@ static int be_vf_setup(struct be_adapter *adapter) NULL, vf+1); if (status) goto err; - adapter->vf_cfg[vf].vf_pmac_id = BE_INVALID_PMAC_ID; } if (!lancer_chip(adapter)) { @@ -2555,6 +2554,16 @@ err: return status; } +static void be_setup_init(struct be_adapter *adapter) +{ + adapter->vlan_prio_bmap = 0xff; + adapter->link_speed = -1; + adapter->if_handle = -1; + adapter->be3_native = false; + adapter->promiscuous = false; + adapter->eq_next_idx = 0; +} + static int be_setup(struct be_adapter *adapter) { struct net_device *netdev = adapter->netdev; @@ -2563,9 +2572,7 @@ static int be_setup(struct be_adapter *adapter) int status; u8 mac[ETH_ALEN]; - /* Allow all priorities by default. A GRP5 evt may modify this */ - adapter->vlan_prio_bmap = 0xff; - adapter->link_speed = -1; + be_setup_init(adapter); be_cmd_req_native_mode(adapter); -- GitLab From 72f02485626b3e71955e54d227ede1b54021d571 Mon Sep 17 00:00:00 2001 From: Sathya Perla Date: Thu, 10 Nov 2011 19:17:58 +0000 Subject: [PATCH 0211/3089] be2net: stop checking the UE registers after an EEH error Signed-off-by: Sathya Perla Signed-off-by: David S. Miller --- drivers/net/ethernet/emulex/benet/be_main.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 83d971d8a5a33..99da07ff45102 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -1978,6 +1978,9 @@ void be_detect_dump_ue(struct be_adapter *adapter) u32 sliport_status = 0, sliport_err1 = 0, sliport_err2 = 0; u32 i; + if (adapter->eeh_err || adapter->ue_detected) + return; + if (lancer_chip(adapter)) { sliport_status = ioread32(adapter->db + SLIPORT_STATUS_OFFSET); if (sliport_status & SLIPORT_STATUS_ERR_MASK) { @@ -2039,8 +2042,7 @@ static void be_worker(struct work_struct *work) struct be_rx_obj *rxo; int i; - if (!adapter->ue_detected) - be_detect_dump_ue(adapter); + be_detect_dump_ue(adapter); /* when interrupts are not yet enabled, just reap any pending * mcc completions */ -- GitLab From 434b3648e9a58600cea5f3a1a0a7a89048e4df61 Mon Sep 17 00:00:00 2001 From: Sathya Perla Date: Thu, 10 Nov 2011 19:17:59 +0000 Subject: [PATCH 0212/3089] be2net: don't log more than one error on detecting EEH/UE errors Currently we're spamming error messages each time a FW cmd call is made while in EEH/UE error state. One log msg on error detection is enough. Signed-off-by: Sathya Perla Signed-off-by: David S. Miller --- drivers/net/ethernet/emulex/benet/be_cmds.c | 15 +++------------ drivers/net/ethernet/emulex/benet/be_main.c | 3 ++- 2 files changed, 5 insertions(+), 13 deletions(-) diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c index c5912c4ed24f5..94cd77ca367ae 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.c +++ b/drivers/net/ethernet/emulex/benet/be_cmds.c @@ -31,11 +31,8 @@ static void be_mcc_notify(struct be_adapter *adapter) struct be_queue_info *mccq = &adapter->mcc_obj.q; u32 val = 0; - if (adapter->eeh_err) { - dev_info(&adapter->pdev->dev, - "Error in Card Detected! Cannot issue commands\n"); + if (adapter->eeh_err) return; - } val |= mccq->id & DB_MCCQ_RING_ID_MASK; val |= 1 << DB_MCCQ_NUM_POSTED_SHIFT; @@ -298,19 +295,13 @@ static int be_mbox_db_ready_wait(struct be_adapter *adapter, void __iomem *db) int msecs = 0; u32 ready; - if (adapter->eeh_err) { - dev_err(&adapter->pdev->dev, - "Error detected in card.Cannot issue commands\n"); + if (adapter->eeh_err) return -EIO; - } do { ready = ioread32(db); - if (ready == 0xffffffff) { - dev_err(&adapter->pdev->dev, - "pci slot disconnected\n"); + if (ready == 0xffffffff) return -1; - } ready &= MPU_MAILBOX_DB_RDY_MASK; if (ready) diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 99da07ff45102..0e97b6d01c79b 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -2007,7 +2007,8 @@ void be_detect_dump_ue(struct be_adapter *adapter) sliport_status & SLIPORT_STATUS_ERR_MASK) { adapter->ue_detected = true; adapter->eeh_err = true; - dev_err(&adapter->pdev->dev, "UE Detected!!\n"); + dev_err(&adapter->pdev->dev, + "Unrecoverable error in the card\n"); } if (ue_lo) { -- GitLab From 6589ade019dcab245d3bb847370f855b56cdf6ad Mon Sep 17 00:00:00 2001 From: Sathya Perla Date: Thu, 10 Nov 2011 19:18:00 +0000 Subject: [PATCH 0213/3089] be2net: stop issuing FW cmds if any cmd times out A FW cmd timeout (with a sufficiently large timeout value in the order of tens of seconds) indicates an unresponsive FW. In this state issuing further cmds and waiting for a completion will only stall the process. Signed-off-by: Sathya Perla Signed-off-by: David S. Miller --- drivers/net/ethernet/emulex/benet/be.h | 8 +++++- drivers/net/ethernet/emulex/benet/be_cmds.c | 29 ++++++++------------- drivers/net/ethernet/emulex/benet/be_main.c | 2 ++ 3 files changed, 20 insertions(+), 19 deletions(-) diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h index 4163980d6bcdc..34f162db9f2e3 100644 --- a/drivers/net/ethernet/emulex/benet/be.h +++ b/drivers/net/ethernet/emulex/benet/be.h @@ -350,6 +350,8 @@ struct be_adapter { u32 beacon_state; /* for set_phys_id */ bool eeh_err; + bool ue_detected; + bool fw_timeout; u32 port_num; bool promiscuous; bool wol; @@ -357,7 +359,6 @@ struct be_adapter { u32 function_caps; u32 rx_fc; /* Rx flow control */ u32 tx_fc; /* Tx flow control */ - bool ue_detected; bool stats_cmd_sent; int link_speed; u8 port_type; @@ -522,6 +523,11 @@ static inline bool be_multi_rxq(const struct be_adapter *adapter) return adapter->num_rx_qs > 1; } +static inline bool be_error(struct be_adapter *adapter) +{ + return adapter->eeh_err || adapter->ue_detected || adapter->fw_timeout; +} + extern void be_cq_notify(struct be_adapter *adapter, u16 qid, bool arm, u16 num_popped); extern void be_link_status_update(struct be_adapter *adapter, u32 link_status); diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.c b/drivers/net/ethernet/emulex/benet/be_cmds.c index 94cd77ca367ae..ad3eef0beea02 100644 --- a/drivers/net/ethernet/emulex/benet/be_cmds.c +++ b/drivers/net/ethernet/emulex/benet/be_cmds.c @@ -31,7 +31,7 @@ static void be_mcc_notify(struct be_adapter *adapter) struct be_queue_info *mccq = &adapter->mcc_obj.q; u32 val = 0; - if (adapter->eeh_err) + if (be_error(adapter)) return; val |= mccq->id & DB_MCCQ_RING_ID_MASK; @@ -263,10 +263,10 @@ static int be_mcc_wait_compl(struct be_adapter *adapter) int i, num, status = 0; struct be_mcc_obj *mcc_obj = &adapter->mcc_obj; - if (adapter->eeh_err) - return -EIO; - for (i = 0; i < mcc_timeout; i++) { + if (be_error(adapter)) + return -EIO; + num = be_process_mcc(adapter, &status); if (num) be_cq_notify(adapter, mcc_obj->cq.id, @@ -277,7 +277,8 @@ static int be_mcc_wait_compl(struct be_adapter *adapter) udelay(100); } if (i == mcc_timeout) { - dev_err(&adapter->pdev->dev, "mccq poll timed out\n"); + dev_err(&adapter->pdev->dev, "FW not responding\n"); + adapter->fw_timeout = true; return -1; } return status; @@ -295,10 +296,10 @@ static int be_mbox_db_ready_wait(struct be_adapter *adapter, void __iomem *db) int msecs = 0; u32 ready; - if (adapter->eeh_err) - return -EIO; - do { + if (be_error(adapter)) + return -EIO; + ready = ioread32(db); if (ready == 0xffffffff) return -1; @@ -308,7 +309,8 @@ static int be_mbox_db_ready_wait(struct be_adapter *adapter, void __iomem *db) break; if (msecs > 4000) { - dev_err(&adapter->pdev->dev, "mbox poll timed out\n"); + dev_err(&adapter->pdev->dev, "FW not responding\n"); + adapter->fw_timeout = true; be_detect_dump_ue(adapter); return -1; } @@ -546,9 +548,6 @@ int be_cmd_fw_clean(struct be_adapter *adapter) u8 *wrb; int status; - if (adapter->eeh_err) - return -EIO; - if (mutex_lock_interruptible(&adapter->mbox_lock)) return -1; @@ -1012,9 +1011,6 @@ int be_cmd_q_destroy(struct be_adapter *adapter, struct be_queue_info *q, u8 subsys = 0, opcode = 0; int status; - if (adapter->eeh_err) - return -EIO; - if (mutex_lock_interruptible(&adapter->mbox_lock)) return -1; @@ -1136,9 +1132,6 @@ int be_cmd_if_destroy(struct be_adapter *adapter, int interface_id, u32 domain) struct be_cmd_req_if_destroy *req; int status; - if (adapter->eeh_err) - return -EIO; - if (interface_id == -1) return 0; diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c index 0e97b6d01c79b..ce20d64d1f95e 100644 --- a/drivers/net/ethernet/emulex/benet/be_main.c +++ b/drivers/net/ethernet/emulex/benet/be_main.c @@ -3569,6 +3569,8 @@ static pci_ers_result_t be_eeh_reset(struct pci_dev *pdev) dev_info(&adapter->pdev->dev, "EEH reset\n"); adapter->eeh_err = false; + adapter->ue_detected = false; + adapter->fw_timeout = false; status = pci_enable_device(pdev); if (status) -- GitLab From 4b90a603a1b21d63cf743cc833680cb195a729f6 Mon Sep 17 00:00:00 2001 From: Nick Bowler Date: Thu, 10 Nov 2011 09:01:27 +0000 Subject: [PATCH 0214/3089] ah: Don't return NET_XMIT_DROP on input. When the ahash driver returns -EBUSY, AH4/6 input functions return NET_XMIT_DROP, presumably copied from the output code path. But returning transmit codes on input doesn't make a lot of sense. Since NET_XMIT_DROP is a positive int, this gets interpreted as the next header type (i.e., success). As that can only end badly, remove the check. Signed-off-by: Nick Bowler Signed-off-by: David S. Miller --- net/ipv4/ah4.c | 2 -- net/ipv6/ah6.c | 2 -- 2 files changed, 4 deletions(-) diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c index c7056b2e831f6..36d14406261e8 100644 --- a/net/ipv4/ah4.c +++ b/net/ipv4/ah4.c @@ -369,8 +369,6 @@ static int ah_input(struct xfrm_state *x, struct sk_buff *skb) if (err == -EINPROGRESS) goto out; - if (err == -EBUSY) - err = NET_XMIT_DROP; goto out_free; } diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c index 7a33aaa002278..4c0f894d08432 100644 --- a/net/ipv6/ah6.c +++ b/net/ipv6/ah6.c @@ -581,8 +581,6 @@ static int ah6_input(struct xfrm_state *x, struct sk_buff *skb) if (err == -EINPROGRESS) goto out; - if (err == -EBUSY) - err = NET_XMIT_DROP; goto out_free; } -- GitLab From 6b76106d8ef31111d6fc469564b83b5f5542794f Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Sun, 13 Nov 2011 19:58:09 +0100 Subject: [PATCH 0215/3089] block: Always check length of all iov entries in blk_rq_map_user_iov() Even after commit 5478755616ae2ef1ce144dded589b62b2a50d575 ("block: check for proper length of iov entries earlier ...") we still won't check for zero-length entries after an unaligned entry. Remove the break-statement, so all entries are checked. Signed-off-by: Ben Hutchings Cc: stable@vger.kernel.org Signed-off-by: Jens Axboe --- block/blk-map.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/block/blk-map.c b/block/blk-map.c index e663ac2d8e68f..164cd00597062 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -204,10 +204,11 @@ int blk_rq_map_user_iov(struct request_queue *q, struct request *rq, if (!iov[i].iov_len) return -EINVAL; - if (uaddr & queue_dma_alignment(q)) { + /* + * Keep going so we check length of all segments + */ + if (uaddr & queue_dma_alignment(q)) unaligned = 1; - break; - } } if (unaligned || (q->dma_pad_mask & len) || map_data) -- GitLab From 62ac0dc9ec0b90b83103ebb659e0696c344e4be4 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Sun, 13 Nov 2011 04:34:21 +0000 Subject: [PATCH 0216/3089] bnx2x: allow FCoE and DCB for 578xx Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c | 2 +- drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c index 51bd7485ab189..5cba9d76b3c00 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c @@ -923,7 +923,7 @@ static void bnx2x_dcbx_admin_mib_updated_params(struct bnx2x *bp, void bnx2x_dcbx_set_state(struct bnx2x *bp, bool dcb_on, u32 dcbx_enabled) { - if (!CHIP_IS_E1x(bp) && !CHIP_IS_E3(bp)) { + if (!CHIP_IS_E1x(bp)) { bp->dcb_state = dcb_on; bp->dcbx_enabled = dcbx_enabled; } else { diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 6486ab8c8fc87..6f3a78468bad3 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -10817,8 +10817,8 @@ static int __devinit bnx2x_init_one(struct pci_dev *pdev, bp->qm_cid_count = bnx2x_set_qm_cid_count(bp); #ifdef BCM_CNIC - /* disable FCOE L2 queue for E1x and E3*/ - if (CHIP_IS_E1x(bp) || CHIP_IS_E3(bp)) + /* disable FCOE L2 queue for E1x */ + if (CHIP_IS_E1x(bp)) bp->flags |= NO_FCOE_FLAG; #endif -- GitLab From f233cafe1a9df8de75f446bc6f5dc715cc564325 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Sun, 13 Nov 2011 04:34:22 +0000 Subject: [PATCH 0217/3089] bnx2x: use rx_queue index for skb_record_rx_queue() Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 1 + drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 2 +- drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h | 1 + drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 2 +- 4 files changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index aec7212ac9835..e17a739c6753e 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -507,6 +507,7 @@ struct bnx2x_fastpath { __le16 fp_hc_idx; u8 index; /* number in fp array */ + u8 rx_queue; /* index for skb_record */ u8 cl_id; /* eth client id */ u8 cl_qzone_id; u8 fw_sb_id; /* status block number in FW */ diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 580b44edb066e..1ace946e1c337 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -755,7 +755,7 @@ reuse_rx: } } - skb_record_rx_queue(skb, fp->index); + skb_record_rx_queue(skb, fp->rx_queue); if (le16_to_cpu(cqe_fp->pars_flags.flags) & PARSING_FLAGS_VLAN) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h index 283d663da1802..4a16757793f42 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h @@ -1318,6 +1318,7 @@ static inline void bnx2x_init_fcoe_fp(struct bnx2x *bp) struct bnx2x_fastpath *fp = bnx2x_fcoe_fp(bp); unsigned long q_type = 0; + bnx2x_fcoe(bp, rx_queue) = BNX2X_NUM_ETH_QUEUES(bp); bnx2x_fcoe(bp, cl_id) = bnx2x_cnic_eth_cl_id(bp, BNX2X_FCOE_ETH_CL_ID_IDX); /** Current BNX2X_FCOE_ETH_CID deffinition implies not more than diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 6f3a78468bad3..1d185f2fc1e8e 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -5247,7 +5247,7 @@ static void bnx2x_init_eth_fp(struct bnx2x *bp, int fp_idx) u8 cos; unsigned long q_type = 0; u32 cids[BNX2X_MULTI_TX_COS] = { 0 }; - + fp->rx_queue = fp_idx; fp->cid = fp_idx; fp->cl_id = bnx2x_fp_cl_id(fp); fp->fw_sb_id = bnx2x_fp_fw_sb_id(fp); -- GitLab From ad756594a8d88ffc048d14b8d5c02971e08856ce Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Sun, 13 Nov 2011 04:34:23 +0000 Subject: [PATCH 0218/3089] bnx2x: remove unused variable Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 1ace946e1c337..f946a6ea89097 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -1094,13 +1094,11 @@ static void bnx2x_free_tx_skbs(struct bnx2x *bp) for_each_cos_in_tx_queue(fp, cos) { struct bnx2x_fp_txdata *txdata = &fp->txdata[cos]; - u16 bd_cons = txdata->tx_bd_cons; u16 sw_prod = txdata->tx_pkt_prod; u16 sw_cons = txdata->tx_pkt_cons; while (sw_cons != sw_prod) { - bd_cons = bnx2x_free_tx_pkt(bp, txdata, - TX_BD(sw_cons)); + bnx2x_free_tx_pkt(bp, txdata, TX_BD(sw_cons)); sw_cons++; } } -- GitLab From b306f5edf6615d3abeba16914872c24c9be29051 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Sun, 13 Nov 2011 04:34:24 +0000 Subject: [PATCH 0219/3089] bnx2x: separate FCoE and iSCSI license initialization. FCoE license info must be initialized at probe(), but iSCSI at open(). Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- .../net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 2 + .../net/ethernet/broadcom/bnx2x/bnx2x_cmn.h | 10 +++++ .../net/ethernet/broadcom/bnx2x/bnx2x_main.c | 45 ++++++++++++++----- 3 files changed, 46 insertions(+), 11 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index f946a6ea89097..3f80c11a29a7f 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -1934,6 +1934,8 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode) mod_timer(&bp->timer, jiffies + bp->current_interval); #ifdef BCM_CNIC + /* re-read iscsi info */ + bnx2x_get_iscsi_info(bp); bnx2x_setup_cnic_irq_info(bp); if (bp->state == BNX2X_STATE_OPEN) bnx2x_cnic_notify(bp, CNIC_CTL_START_CMD); diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h index 4a16757793f42..c1d7833c42e6f 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h @@ -1489,4 +1489,14 @@ static inline u16 bnx2x_extract_max_cfg(struct bnx2x *bp, u32 mf_cfg) return max_cfg; } +#ifdef BCM_CNIC +/** + * bnx2x_get_iscsi_info - update iSCSI params according to licensing info. + * + * @bp: driver handle + * + */ +void bnx2x_get_iscsi_info(struct bnx2x *bp); +#endif + #endif /* BNX2X_CMN_H */ diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 1d185f2fc1e8e..26dc539ef2900 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -9268,21 +9268,38 @@ static void __devinit bnx2x_get_port_hwinfo(struct bnx2x *bp) } #ifdef BCM_CNIC -static void __devinit bnx2x_get_cnic_info(struct bnx2x *bp) +void bnx2x_get_iscsi_info(struct bnx2x *bp) { int port = BP_PORT(bp); - int func = BP_ABS_FUNC(bp); u32 max_iscsi_conn = FW_ENCODE_32BIT_PATTERN ^ SHMEM_RD(bp, drv_lic_key[port].max_iscsi_conn); - u32 max_fcoe_conn = FW_ENCODE_32BIT_PATTERN ^ SHMEM_RD(bp, - drv_lic_key[port].max_fcoe_conn); - /* Get the number of maximum allowed iSCSI and FCoE connections */ + /* Get the number of maximum allowed iSCSI connections */ bp->cnic_eth_dev.max_iscsi_conn = (max_iscsi_conn & BNX2X_MAX_ISCSI_INIT_CONN_MASK) >> BNX2X_MAX_ISCSI_INIT_CONN_SHIFT; + BNX2X_DEV_INFO("max_iscsi_conn 0x%x\n", + bp->cnic_eth_dev.max_iscsi_conn); + + /* + * If maximum allowed number of connections is zero - + * disable the feature. + */ + if (!bp->cnic_eth_dev.max_iscsi_conn) + bp->flags |= NO_ISCSI_FLAG; +} + +static void __devinit bnx2x_get_fcoe_info(struct bnx2x *bp) +{ + int port = BP_PORT(bp); + int func = BP_ABS_FUNC(bp); + + u32 max_fcoe_conn = FW_ENCODE_32BIT_PATTERN ^ SHMEM_RD(bp, + drv_lic_key[port].max_fcoe_conn); + + /* Get the number of maximum allowed FCoE connections */ bp->cnic_eth_dev.max_fcoe_conn = (max_fcoe_conn & BNX2X_MAX_FCOE_INIT_CONN_MASK) >> BNX2X_MAX_FCOE_INIT_CONN_SHIFT; @@ -9334,20 +9351,26 @@ static void __devinit bnx2x_get_cnic_info(struct bnx2x *bp) } } - BNX2X_DEV_INFO("max_iscsi_conn 0x%x max_fcoe_conn 0x%x\n", - bp->cnic_eth_dev.max_iscsi_conn, - bp->cnic_eth_dev.max_fcoe_conn); + BNX2X_DEV_INFO("max_fcoe_conn 0x%x\n", bp->cnic_eth_dev.max_fcoe_conn); /* * If maximum allowed number of connections is zero - * disable the feature. */ - if (!bp->cnic_eth_dev.max_iscsi_conn) - bp->flags |= NO_ISCSI_OOO_FLAG | NO_ISCSI_FLAG; - if (!bp->cnic_eth_dev.max_fcoe_conn) bp->flags |= NO_FCOE_FLAG; } + +static void __devinit bnx2x_get_cnic_info(struct bnx2x *bp) +{ + /* + * iSCSI may be dynamically disabled but reading + * info here we will decrease memory usage by driver + * if the feature is disabled for good + */ + bnx2x_get_iscsi_info(bp); + bnx2x_get_fcoe_info(bp); +} #endif static void __devinit bnx2x_get_mac_hwinfo(struct bnx2x *bp) -- GitLab From 00253a8cf3119af6cb07c9de2c08a50d39fc7201 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Sun, 13 Nov 2011 04:34:25 +0000 Subject: [PATCH 0220/3089] bnx2x: propagate DCBX negotiation We need propagate the DCBX results from PMF to other functions on the same port, in order to properly update netdev structure and allow following new ETS and PFC configurations. Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- .../net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 4 +- .../net/ethernet/broadcom/bnx2x/bnx2x_cmn.h | 54 +++++++++++++++++++ .../net/ethernet/broadcom/bnx2x/bnx2x_dcb.c | 53 ++++++++++-------- .../net/ethernet/broadcom/bnx2x/bnx2x_main.c | 25 ++------- 4 files changed, 90 insertions(+), 46 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 3f80c11a29a7f..e9a91a347c759 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -1927,7 +1927,9 @@ int bnx2x_nic_load(struct bnx2x *bp, int load_mode) break; } - if (!bp->port.pmf) + if (bp->port.pmf) + bnx2x_update_drv_flags(bp, DRV_FLAGS_DCB_CONFIGURED, 0); + else bnx2x__link_status_update(bp); /* start the timer */ diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h index c1d7833c42e6f..59f1291b2f782 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h @@ -1499,4 +1499,58 @@ static inline u16 bnx2x_extract_max_cfg(struct bnx2x *bp, u32 mf_cfg) void bnx2x_get_iscsi_info(struct bnx2x *bp); #endif +/* returns func by VN for current port */ +static inline int func_by_vn(struct bnx2x *bp, int vn) +{ + return 2 * vn + BP_PORT(bp); +} + +/** + * bnx2x_link_sync_notify - send notification to other functions. + * + * @bp: driver handle + * + */ +static inline void bnx2x_link_sync_notify(struct bnx2x *bp) +{ + int func; + int vn; + + /* Set the attention towards other drivers on the same port */ + for (vn = VN_0; vn < BP_MAX_VN_NUM(bp); vn++) { + if (vn == BP_VN(bp)) + continue; + + func = func_by_vn(bp, vn); + REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_0 + + (LINK_SYNC_ATTENTION_BIT_FUNC_0 + func)*4, 1); + } +} + +/** + * bnx2x_update_drv_flags - update flags in shmem + * + * @bp: driver handle + * @flags: flags to update + * @set: set or clear + * + */ +static inline void bnx2x_update_drv_flags(struct bnx2x *bp, u32 flags, u32 set) +{ + if (SHMEM2_HAS(bp, drv_flags)) { + u32 drv_flags; + bnx2x_acquire_hw_lock(bp, HW_LOCK_DRV_FLAGS); + drv_flags = SHMEM2_RD(bp, drv_flags); + + if (set) + SET_FLAGS(drv_flags, flags); + else + RESET_FLAGS(drv_flags, flags); + + SHMEM2_WR(bp, drv_flags, drv_flags); + DP(NETIF_MSG_HW, "drv_flags 0x%08x\n", drv_flags); + bnx2x_release_hw_lock(bp, HW_LOCK_DRV_FLAGS); + } +} + #endif /* BNX2X_CMN_H */ diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c index 5cba9d76b3c00..a0598fd9be428 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c @@ -685,24 +685,6 @@ int bnx2x_dcbnl_update_applist(struct bnx2x *bp, bool delall) } #endif -static inline void bnx2x_update_drv_flags(struct bnx2x *bp, u32 flags, u32 set) -{ - if (SHMEM2_HAS(bp, drv_flags)) { - u32 drv_flags; - bnx2x_acquire_hw_lock(bp, HW_LOCK_DRV_FLAGS); - drv_flags = SHMEM2_RD(bp, drv_flags); - - if (set) - SET_FLAGS(drv_flags, flags); - else - RESET_FLAGS(drv_flags, flags); - - SHMEM2_WR(bp, drv_flags, drv_flags); - DP(NETIF_MSG_HW, "drv_flags 0x%08x\n", drv_flags); - bnx2x_release_hw_lock(bp, HW_LOCK_DRV_FLAGS); - } -} - static inline void bnx2x_dcbx_update_tc_mapping(struct bnx2x *bp) { u8 prio, cos; @@ -755,18 +737,26 @@ void bnx2x_dcbx_set_params(struct bnx2x *bp, u32 state) /* mark DCBX result for PMF migration */ bnx2x_update_drv_flags(bp, DRV_FLAGS_DCB_CONFIGURED, 1); #ifdef BCM_DCBNL - /** + /* * Add new app tlvs to dcbnl */ bnx2x_dcbnl_update_applist(bp, false); #endif - bnx2x_dcbx_stop_hw_tx(bp); - - /* reconfigure the netdevice with the results of the new + /* + * reconfigure the netdevice with the results of the new * dcbx negotiation. */ bnx2x_dcbx_update_tc_mapping(bp); + /* + * allow other funtions to update their netdevices + * accordingly + */ + if (IS_MF(bp)) + bnx2x_link_sync_notify(bp); + + bnx2x_dcbx_stop_hw_tx(bp); + return; } case BNX2X_DCBX_STATE_TX_PAUSED: @@ -775,6 +765,7 @@ void bnx2x_dcbx_set_params(struct bnx2x *bp, u32 state) bnx2x_dcbx_update_ets_params(bp); bnx2x_dcbx_resume_hw_tx(bp); + return; case BNX2X_DCBX_STATE_TX_RELEASED: DP(NETIF_MSG_LINK, "BNX2X_DCBX_STATE_TX_RELEASED\n"); @@ -1863,7 +1854,7 @@ static void bnx2x_dcbx_fw_struct(struct bnx2x *bp, void bnx2x_dcbx_pmf_update(struct bnx2x *bp) { /* if we need to syncronize DCBX result from prev PMF - * read it from shmem and update bp accordingly + * read it from shmem and update bp and netdev accordingly */ if (SHMEM2_HAS(bp, drv_flags) && GET_FLAGS(SHMEM2_RD(bp, drv_flags), DRV_FLAGS_DCB_CONFIGURED)) { @@ -1875,6 +1866,22 @@ void bnx2x_dcbx_pmf_update(struct bnx2x *bp) bp->dcbx_error); bnx2x_get_dcbx_drv_param(bp, &bp->dcbx_local_feat, bp->dcbx_error); +#ifdef BCM_DCBNL + /* + * Add new app tlvs to dcbnl + */ + bnx2x_dcbnl_update_applist(bp, false); + /* + * Send a notification for the new negotiated parameters + */ + dcbnl_cee_notify(bp->dev, RTM_GETDCB, DCB_CMD_CEE_GET, 0, 0); +#endif + /* + * reconfigure the netdevice with the results of the new + * dcbx negotiation. + */ + bnx2x_dcbx_update_tc_mapping(bp); + } } diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 26dc539ef2900..967c41b8b1728 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -2318,12 +2318,6 @@ static void bnx2x_calc_vn_weight_sum(struct bnx2x *bp) CMNG_FLAGS_PER_PORT_FAIRNESS_VN; } -/* returns func by VN for current port */ -static inline int func_by_vn(struct bnx2x *bp, int vn) -{ - return 2 * vn + BP_PORT(bp); -} - static void bnx2x_init_vn_minmax(struct bnx2x *bp, int vn) { struct rate_shaping_vars_per_vn m_rs_vn; @@ -2475,22 +2469,6 @@ static void bnx2x_cmng_fns_init(struct bnx2x *bp, u8 read_cfg, u8 cmng_type) "rate shaping and fairness are disabled\n"); } -static inline void bnx2x_link_sync_notify(struct bnx2x *bp) -{ - int func; - int vn; - - /* Set the attention towards other drivers on the same port */ - for (vn = VN_0; vn < BP_MAX_VN_NUM(bp); vn++) { - if (vn == BP_VN(bp)) - continue; - - func = func_by_vn(bp, vn); - REG_WR(bp, MISC_REG_AEU_GENERAL_ATTN_0 + - (LINK_SYNC_ATTENTION_BIT_FUNC_0 + func)*4, 1); - } -} - /* This function is called upon link interrupt */ static void bnx2x_link_attn(struct bnx2x *bp) { @@ -2549,6 +2527,9 @@ void bnx2x__link_status_update(struct bnx2x *bp) if (bp->state != BNX2X_STATE_OPEN) return; + /* read updated dcb configuration */ + bnx2x_dcbx_pmf_update(bp); + bnx2x_link_status_update(&bp->link_params, &bp->link_vars); if (bp->link_vars.link_up) -- GitLab From f9c058b633000e64fba05fc14ba94dd09a20e674 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Sun, 13 Nov 2011 04:34:26 +0000 Subject: [PATCH 0221/3089] bnx2x: DCBX: use #define instead of magic Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c | 6 +++--- drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.h | 4 +++- 2 files changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c index a0598fd9be428..5051cf3deb209 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.c @@ -874,7 +874,7 @@ static void bnx2x_dcbx_admin_mib_updated_params(struct bnx2x *bp, /*For IEEE admin_recommendation_bw_precentage *For IEEE admin_recommendation_ets_pg */ af->pfc.pri_en_bitmap = (u8)dp->admin_pfc_bitmap; - for (i = 0; i < 4; i++) { + for (i = 0; i < DCBX_CONFIG_MAX_APP_PROTOCOL; i++) { if (dp->admin_priority_app_table[i].valid) { struct bnx2x_admin_priority_app_table *table = dp->admin_priority_app_table; @@ -2249,7 +2249,7 @@ static int bnx2x_set_admin_app_up(struct bnx2x *bp, u8 idtype, u16 idval, u8 up) int i, ff; /* iterate over the app entries looking for idtype and idval */ - for (i = 0, ff = -1; i < 4; i++) { + for (i = 0, ff = -1; i < DCBX_CONFIG_MAX_APP_PROTOCOL; i++) { struct bnx2x_admin_priority_app_table *app_ent = &bp->dcbx_config_params.admin_priority_app_table[i]; if (bnx2x_admin_app_is_equal(app_ent, idtype, idval)) @@ -2258,7 +2258,7 @@ static int bnx2x_set_admin_app_up(struct bnx2x *bp, u8 idtype, u16 idval, u8 up) if (ff < 0 && !app_ent->valid) ff = i; } - if (i < 4) + if (i < DCBX_CONFIG_MAX_APP_PROTOCOL) /* if found overwrite up */ bp->dcbx_config_params. admin_priority_app_table[i].priority = up; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.h index 2c6a3bca6f284..2ab9254e2d5ef 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_dcb.h @@ -90,6 +90,7 @@ struct bnx2x_admin_priority_app_table { u32 app_id; }; +#define DCBX_CONFIG_MAX_APP_PROTOCOL 4 struct bnx2x_config_dcbx_params { u32 overwrite_settings; u32 admin_dcbx_version; @@ -109,7 +110,8 @@ struct bnx2x_config_dcbx_params { u32 admin_recommendation_bw_precentage[8]; u32 admin_recommendation_ets_pg[8]; u32 admin_pfc_bitmap; - struct bnx2x_admin_priority_app_table admin_priority_app_table[4]; + struct bnx2x_admin_priority_app_table + admin_priority_app_table[DCBX_CONFIG_MAX_APP_PROTOCOL]; u32 admin_default_priority; }; -- GitLab From b363782761eea7076619fe44063b915dae0b4cc7 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Sun, 13 Nov 2011 04:34:27 +0000 Subject: [PATCH 0222/3089] bnx2x: simplify definition of RX_SGE_MASK_LEN and use it. Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 3 +-- drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index e17a739c6753e..b78c3844e8385 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -411,8 +411,7 @@ union db_prod { /* Number of u64 elements in SGE mask array */ -#define RX_SGE_MASK_LEN ((NUM_RX_SGE_PAGES * RX_SGE_CNT) / \ - BIT_VEC64_ELEM_SZ) +#define RX_SGE_MASK_LEN (NUM_RX_SGE / BIT_VEC64_ELEM_SZ) #define RX_SGE_MASK_LEN_MASK (RX_SGE_MASK_LEN - 1) #define NEXT_SGE_MASK_ELEM(el) (((el) + 1) & RX_SGE_MASK_LEN_MASK) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h index 59f1291b2f782..260226d099168 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h @@ -874,8 +874,7 @@ static inline void bnx2x_clear_sge_mask_next_elems(struct bnx2x_fastpath *fp) static inline void bnx2x_init_sge_ring_bit_mask(struct bnx2x_fastpath *fp) { /* Set the mask to all 1-s: it's faster to compare to 0 than to 0xf-s */ - memset(fp->sge_mask, 0xff, - (NUM_RX_SGE >> BIT_VEC64_ELEM_SHIFT)*sizeof(u64)); + memset(fp->sge_mask, 0xff, sizeof(fp->sge_mask)); /* Clear the two last indices in the page to 1: these are the indices that correspond to the "next" element, -- GitLab From 46fa1309fe5b05249447df903ec256c3509eb985 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Sun, 13 Nov 2011 04:34:28 +0000 Subject: [PATCH 0223/3089] bnx2x: remove unused #define Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 7 ------- 1 file changed, 7 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index b78c3844e8385..23e91f4c1485b 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -1984,13 +1984,6 @@ static inline u32 reg_poll(struct bnx2x *bp, u32 reg, u32 expected, int ms, #define HW_PRTY_ASSERT_SET_4 (AEU_INPUTS_ATTN_BITS_PGLUE_PARITY_ERROR | \ AEU_INPUTS_ATTN_BITS_ATC_PARITY_ERROR) -#define RSS_FLAGS(bp) \ - (TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_CAPABILITY | \ - TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV4_TCP_CAPABILITY | \ - TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_CAPABILITY | \ - TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_IPV6_TCP_CAPABILITY | \ - (bp->multi_mode << \ - TSTORM_ETH_FUNCTION_COMMON_CONFIG_RSS_MODE_SHIFT)) #define MULTI_MASK 0x7f -- GitLab From 8304859adc213df9f69a86e06164683f76cd5d49 Mon Sep 17 00:00:00 2001 From: Ariel Elior Date: Sun, 13 Nov 2011 04:34:29 +0000 Subject: [PATCH 0224/3089] bnx2x: add fan failure event handling Shut down the device in case of fan failure to prevent HW damage. Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 3 +++ .../net/ethernet/broadcom/bnx2x/bnx2x_main.c | 24 ++++++++++++++++++- 2 files changed, 26 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index 23e91f4c1485b..78613effd7e4b 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -1141,6 +1141,7 @@ struct bnx2x_fw_stats_data { enum { BNX2X_SP_RTNL_SETUP_TC, BNX2X_SP_RTNL_TX_TIMEOUT, + BNX2X_SP_RTNL_FAN_FAILURE, }; @@ -2048,6 +2049,8 @@ static inline u32 reg_poll(struct bnx2x *bp, u32 reg, u32 expected, int ms, #define BNX2X_VPD_LEN 128 #define VENDOR_ID_LEN 4 +int bnx2x_close(struct net_device *dev); + /* Congestion management fairness mode */ #define CMNG_FNS_NONE 0 #define CMNG_FNS_MINMAX 1 diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 967c41b8b1728..33ff60d9fec8e 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -3299,6 +3299,17 @@ static inline void bnx2x_fan_failure(struct bnx2x *bp) netdev_err(bp->dev, "Fan Failure on Network Controller has caused" " the driver to shutdown the card to prevent permanent" " damage. Please contact OEM Support for assistance\n"); + + /* + * Scheudle device reset (unload) + * This is due to some boards consuming sufficient power when driver is + * up to overheat if fan fails. + */ + smp_mb__before_clear_bit(); + set_bit(BNX2X_SP_RTNL_FAN_FAILURE, &bp->sp_rtnl_state); + smp_mb__after_clear_bit(); + schedule_delayed_work(&bp->sp_rtnl_task, 0); + } static inline void bnx2x_attn_int_deasserted0(struct bnx2x *bp, u32 attn) @@ -8503,6 +8514,17 @@ sp_rtnl_not_reset: if (test_and_clear_bit(BNX2X_SP_RTNL_SETUP_TC, &bp->sp_rtnl_state)) bnx2x_setup_tc(bp->dev, bp->dcbx_port_params.ets.num_of_cos); + /* + * in case of fan failure we need to reset id if the "stop on error" + * debug flag is set, since we trying to prevent permanent overheating + * damage + */ + if (test_and_clear_bit(BNX2X_SP_RTNL_FAN_FAILURE, &bp->sp_rtnl_state)) { + DP(BNX2X_MSG_SP, "fan failure detected. Unloading driver"); + netif_device_detach(bp->dev); + bnx2x_close(bp->dev); + } + sp_rtnl_exit: rtnl_unlock(); } @@ -9969,7 +9991,7 @@ static int bnx2x_open(struct net_device *dev) } /* called with rtnl_lock */ -static int bnx2x_close(struct net_device *dev) +int bnx2x_close(struct net_device *dev) { struct bnx2x *bp = netdev_priv(dev); -- GitLab From 4a025f49d3f2f2f39b474af360c81a5587b41657 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Sun, 13 Nov 2011 04:34:30 +0000 Subject: [PATCH 0225/3089] bnx2x: prevent race in statistics flow The race may cause access of registers while MAC hw block is in reset state. As a result syslog will show error messages. We can prevent this by using state from local variable. Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c index 02ac6a771bf99..3034f0e319387 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_stats.c @@ -1349,12 +1349,14 @@ void bnx2x_stats_handle(struct bnx2x *bp, enum bnx2x_stats_event event) enum bnx2x_stats_state state; if (unlikely(bp->panic)) return; - bnx2x_stats_stm[bp->stats_state][event].action(bp); + spin_lock_bh(&bp->stats_lock); state = bp->stats_state; bp->stats_state = bnx2x_stats_stm[state][event].next_state; spin_unlock_bh(&bp->stats_lock); + bnx2x_stats_stm[state][event].action(bp); + if ((event != STATS_EVENT_UPDATE) || netif_msg_timer(bp)) DP(BNX2X_MSG_STATS, "state %d -> event %d -> state %d\n", state, event, bp->stats_state); -- GitLab From 72754080d14feef1ca0b3ae383ddfdc5d9a71b1a Mon Sep 17 00:00:00 2001 From: Ariel Elior Date: Sun, 13 Nov 2011 04:34:31 +0000 Subject: [PATCH 0226/3089] bnx2x: Remove on-stack napi struct variable Signed-off-by: Ariel Elior Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index e9a91a347c759..13dad9230dbc8 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -79,19 +79,21 @@ static inline void bnx2x_bz_fp(struct bnx2x *bp, int index) * @to: destination FP index * * Makes sure the contents of the bp->fp[to].napi is kept - * intact. + * intact. This is done by first copying the napi struct from + * the target to the source, and then mem copying the entire + * source onto the target */ static inline void bnx2x_move_fp(struct bnx2x *bp, int from, int to) { struct bnx2x_fastpath *from_fp = &bp->fp[from]; struct bnx2x_fastpath *to_fp = &bp->fp[to]; - struct napi_struct orig_napi = to_fp->napi; + + /* Copy the NAPI object as it has been already initialized */ + from_fp->napi = to_fp->napi; + /* Move bnx2x_fastpath contents */ memcpy(to_fp, from_fp, sizeof(*to_fp)); to_fp->index = to; - - /* Restore the NAPI object as it has been already initialized */ - to_fp->napi = orig_napi; } int load_count[2][3] = { {0} }; /* per-path: 0-common, 1-port0, 2-port1 */ -- GitLab From 5d70b88cd41ef0f2ac0caaab4fd492dd686feee6 Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Sun, 13 Nov 2011 04:34:32 +0000 Subject: [PATCH 0227/3089] bnx2x: update driver version to 1.70.35-0 Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index 78613effd7e4b..4b90b518d6a6d 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -23,8 +23,8 @@ * (you will need to reboot afterwards) */ /* #define BNX2X_STOP_ON_ERROR */ -#define DRV_MODULE_VERSION "1.70.30-0" -#define DRV_MODULE_RELDATE "2011/10/25" +#define DRV_MODULE_VERSION "1.70.35-0" +#define DRV_MODULE_RELDATE "2011/11/10" #define BNX2X_BC_VER 0x040200 #if defined(CONFIG_DCB) -- GitLab From 3d249d4ca7d0ed6629a135ea1ea21c72286c0d80 Mon Sep 17 00:00:00 2001 From: Jiri Pirko Date: Fri, 11 Nov 2011 22:16:48 +0000 Subject: [PATCH 0228/3089] net: introduce ethernet teaming device This patch introduces new network device called team. It supposes to be very fast, simple, userspace-driven alternative to existing bonding driver. Userspace library called libteam with couple of demo apps is available here: https://github.com/jpirko/libteam Note it's still in its dipers atm. team<->libteam use generic netlink for communication. That and rtnl suppose to be the only way to configure team device, no sysfs etc. Python binding of libteam was recently introduced. Daemon providing arpmon/miimon active-backup functionality will be introduced shortly. All what's necessary is already implemented in kernel team driver. v7->v8: - check ndo_ndo_vlan_rx_[add/kill]_vid functions before calling them. - use dev_kfree_skb_any() instead of dev_kfree_skb() v6->v7: - transmit and receive functions are not checked in hot paths. That also resolves memory leak on transmit when no port is present v5->v6: - changed couple of _rcu calls to non _rcu ones in non-readers v4->v5: - team_change_mtu() uses team->lock while travesing though port list - mac address changes are moved completely to jurisdiction of userspace daemon. This way the daemon can do FOM1, FOM2 and possibly other weird things with mac addresses. Only round-robin mode sets up all ports to bond's address then enslaved. - Extended Kconfig text v3->v4: - remove redundant synchronize_rcu from __team_change_mode() - revert "set and clear of mode_ops happens per pointer, not per byte" - extend comment of function __team_change_mode() v2->v3: - team_change_mtu() uses rcu version of list traversal to unwind - set and clear of mode_ops happens per pointer, not per byte - port hashlist changed to be embedded into team structure - error branch in team_port_enter() does cleanup now - fixed rtln->rtnl v1->v2: - modes are made as modules. Makes team more modular and extendable. - several commenters' nitpicks found on v1 were fixed - several other bugs were fixed. - note I ignored Eric's comment about roundrobin port selector as Eric's way may be easily implemented as another mode (mode "random") in future. Signed-off-by: Jiri Pirko Signed-off-by: David S. Miller --- Documentation/networking/team.txt | 2 + MAINTAINERS | 7 + drivers/net/Kconfig | 2 + drivers/net/Makefile | 1 + drivers/net/team/Kconfig | 43 + drivers/net/team/Makefile | 7 + drivers/net/team/team.c | 1583 +++++++++++++++++++++ drivers/net/team/team_mode_activebackup.c | 137 ++ drivers/net/team/team_mode_roundrobin.c | 107 ++ include/linux/Kbuild | 1 + include/linux/if.h | 1 + include/linux/if_team.h | 242 ++++ 12 files changed, 2133 insertions(+) create mode 100644 Documentation/networking/team.txt create mode 100644 drivers/net/team/Kconfig create mode 100644 drivers/net/team/Makefile create mode 100644 drivers/net/team/team.c create mode 100644 drivers/net/team/team_mode_activebackup.c create mode 100644 drivers/net/team/team_mode_roundrobin.c create mode 100644 include/linux/if_team.h diff --git a/Documentation/networking/team.txt b/Documentation/networking/team.txt new file mode 100644 index 0000000000000..5a013686b9ea1 --- /dev/null +++ b/Documentation/networking/team.txt @@ -0,0 +1,2 @@ +Team devices are driven from userspace via libteam library which is here: + https://github.com/jpirko/libteam diff --git a/MAINTAINERS b/MAINTAINERS index 4808256446f29..8d941692c3947 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -6484,6 +6484,13 @@ W: http://tcp-lp-mod.sourceforge.net/ S: Maintained F: net/ipv4/tcp_lp.c +TEAM DRIVER +M: Jiri Pirko +L: netdev@vger.kernel.org +S: Supported +F: drivers/net/team/ +F: include/linux/if_team.h + TEGRA SUPPORT M: Colin Cross M: Olof Johansson diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 583f66cd5bbd6..b3020bea39e42 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -125,6 +125,8 @@ config IFB 'ifb1' etc. Look at the iproute2 documentation directory for usage etc +source "drivers/net/team/Kconfig" + config MACVLAN tristate "MAC-VLAN support (EXPERIMENTAL)" depends on EXPERIMENTAL diff --git a/drivers/net/Makefile b/drivers/net/Makefile index fa877cd2b1393..4e4ebfe1aa535 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile @@ -17,6 +17,7 @@ obj-$(CONFIG_NET) += Space.o loopback.o obj-$(CONFIG_NETCONSOLE) += netconsole.o obj-$(CONFIG_PHYLIB) += phy/ obj-$(CONFIG_RIONET) += rionet.o +obj-$(CONFIG_NET_TEAM) += team/ obj-$(CONFIG_TUN) += tun.o obj-$(CONFIG_VETH) += veth.o obj-$(CONFIG_VIRTIO_NET) += virtio_net.o diff --git a/drivers/net/team/Kconfig b/drivers/net/team/Kconfig new file mode 100644 index 0000000000000..248a144033cac --- /dev/null +++ b/drivers/net/team/Kconfig @@ -0,0 +1,43 @@ +menuconfig NET_TEAM + tristate "Ethernet team driver support (EXPERIMENTAL)" + depends on EXPERIMENTAL + ---help--- + This allows one to create virtual interfaces that teams together + multiple ethernet devices. + + Team devices can be added using the "ip" command from the + iproute2 package: + + "ip link add link [ address MAC ] [ NAME ] type team" + + To compile this driver as a module, choose M here: the module + will be called team. + +if NET_TEAM + +config NET_TEAM_MODE_ROUNDROBIN + tristate "Round-robin mode support" + depends on NET_TEAM + ---help--- + Basic mode where port used for transmitting packets is selected in + round-robin fashion using packet counter. + + All added ports are setup to have bond's mac address. + + To compile this team mode as a module, choose M here: the module + will be called team_mode_roundrobin. + +config NET_TEAM_MODE_ACTIVEBACKUP + tristate "Active-backup mode support" + depends on NET_TEAM + ---help--- + Only one port is active at a time and the rest of ports are used + for backup. + + Mac addresses of ports are not modified. Userspace is responsible + to do so. + + To compile this team mode as a module, choose M here: the module + will be called team_mode_activebackup. + +endif # NET_TEAM diff --git a/drivers/net/team/Makefile b/drivers/net/team/Makefile new file mode 100644 index 0000000000000..85f2028a87afb --- /dev/null +++ b/drivers/net/team/Makefile @@ -0,0 +1,7 @@ +# +# Makefile for the network team driver +# + +obj-$(CONFIG_NET_TEAM) += team.o +obj-$(CONFIG_NET_TEAM_MODE_ROUNDROBIN) += team_mode_roundrobin.o +obj-$(CONFIG_NET_TEAM_MODE_ACTIVEBACKUP) += team_mode_activebackup.o diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c new file mode 100644 index 0000000000000..60672bb099603 --- /dev/null +++ b/drivers/net/team/team.c @@ -0,0 +1,1583 @@ +/* + * net/drivers/team/team.c - Network team device driver + * Copyright (c) 2011 Jiri Pirko + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define DRV_NAME "team" + + +/********** + * Helpers + **********/ + +#define team_port_exists(dev) (dev->priv_flags & IFF_TEAM_PORT) + +static struct team_port *team_port_get_rcu(const struct net_device *dev) +{ + struct team_port *port = rcu_dereference(dev->rx_handler_data); + + return team_port_exists(dev) ? port : NULL; +} + +static struct team_port *team_port_get_rtnl(const struct net_device *dev) +{ + struct team_port *port = rtnl_dereference(dev->rx_handler_data); + + return team_port_exists(dev) ? port : NULL; +} + +/* + * Since the ability to change mac address for open port device is tested in + * team_port_add, this function can be called without control of return value + */ +static int __set_port_mac(struct net_device *port_dev, + const unsigned char *dev_addr) +{ + struct sockaddr addr; + + memcpy(addr.sa_data, dev_addr, ETH_ALEN); + addr.sa_family = ARPHRD_ETHER; + return dev_set_mac_address(port_dev, &addr); +} + +int team_port_set_orig_mac(struct team_port *port) +{ + return __set_port_mac(port->dev, port->orig.dev_addr); +} + +int team_port_set_team_mac(struct team_port *port) +{ + return __set_port_mac(port->dev, port->team->dev->dev_addr); +} +EXPORT_SYMBOL(team_port_set_team_mac); + + +/******************* + * Options handling + *******************/ + +void team_options_register(struct team *team, struct team_option *option, + size_t option_count) +{ + int i; + + for (i = 0; i < option_count; i++, option++) + list_add_tail(&option->list, &team->option_list); +} +EXPORT_SYMBOL(team_options_register); + +static void __team_options_change_check(struct team *team, + struct team_option *changed_option); + +static void __team_options_unregister(struct team *team, + struct team_option *option, + size_t option_count) +{ + int i; + + for (i = 0; i < option_count; i++, option++) + list_del(&option->list); +} + +void team_options_unregister(struct team *team, struct team_option *option, + size_t option_count) +{ + __team_options_unregister(team, option, option_count); + __team_options_change_check(team, NULL); +} +EXPORT_SYMBOL(team_options_unregister); + +static int team_option_get(struct team *team, struct team_option *option, + void *arg) +{ + return option->getter(team, arg); +} + +static int team_option_set(struct team *team, struct team_option *option, + void *arg) +{ + int err; + + err = option->setter(team, arg); + if (err) + return err; + + __team_options_change_check(team, option); + return err; +} + +/**************** + * Mode handling + ****************/ + +static LIST_HEAD(mode_list); +static DEFINE_SPINLOCK(mode_list_lock); + +static struct team_mode *__find_mode(const char *kind) +{ + struct team_mode *mode; + + list_for_each_entry(mode, &mode_list, list) { + if (strcmp(mode->kind, kind) == 0) + return mode; + } + return NULL; +} + +static bool is_good_mode_name(const char *name) +{ + while (*name != '\0') { + if (!isalpha(*name) && !isdigit(*name) && *name != '_') + return false; + name++; + } + return true; +} + +int team_mode_register(struct team_mode *mode) +{ + int err = 0; + + if (!is_good_mode_name(mode->kind) || + mode->priv_size > TEAM_MODE_PRIV_SIZE) + return -EINVAL; + spin_lock(&mode_list_lock); + if (__find_mode(mode->kind)) { + err = -EEXIST; + goto unlock; + } + list_add_tail(&mode->list, &mode_list); +unlock: + spin_unlock(&mode_list_lock); + return err; +} +EXPORT_SYMBOL(team_mode_register); + +int team_mode_unregister(struct team_mode *mode) +{ + spin_lock(&mode_list_lock); + list_del_init(&mode->list); + spin_unlock(&mode_list_lock); + return 0; +} +EXPORT_SYMBOL(team_mode_unregister); + +static struct team_mode *team_mode_get(const char *kind) +{ + struct team_mode *mode; + + spin_lock(&mode_list_lock); + mode = __find_mode(kind); + if (!mode) { + spin_unlock(&mode_list_lock); + request_module("team-mode-%s", kind); + spin_lock(&mode_list_lock); + mode = __find_mode(kind); + } + if (mode) + if (!try_module_get(mode->owner)) + mode = NULL; + + spin_unlock(&mode_list_lock); + return mode; +} + +static void team_mode_put(const struct team_mode *mode) +{ + module_put(mode->owner); +} + +static bool team_dummy_transmit(struct team *team, struct sk_buff *skb) +{ + dev_kfree_skb_any(skb); + return false; +} + +rx_handler_result_t team_dummy_receive(struct team *team, + struct team_port *port, + struct sk_buff *skb) +{ + return RX_HANDLER_ANOTHER; +} + +static void team_adjust_ops(struct team *team) +{ + /* + * To avoid checks in rx/tx skb paths, ensure here that non-null and + * correct ops are always set. + */ + + if (list_empty(&team->port_list) || + !team->mode || !team->mode->ops->transmit) + team->ops.transmit = team_dummy_transmit; + else + team->ops.transmit = team->mode->ops->transmit; + + if (list_empty(&team->port_list) || + !team->mode || !team->mode->ops->receive) + team->ops.receive = team_dummy_receive; + else + team->ops.receive = team->mode->ops->receive; +} + +/* + * We can benefit from the fact that it's ensured no port is present + * at the time of mode change. Therefore no packets are in fly so there's no + * need to set mode operations in any special way. + */ +static int __team_change_mode(struct team *team, + const struct team_mode *new_mode) +{ + /* Check if mode was previously set and do cleanup if so */ + if (team->mode) { + void (*exit_op)(struct team *team) = team->ops.exit; + + /* Clear ops area so no callback is called any longer */ + memset(&team->ops, 0, sizeof(struct team_mode_ops)); + team_adjust_ops(team); + + if (exit_op) + exit_op(team); + team_mode_put(team->mode); + team->mode = NULL; + /* zero private data area */ + memset(&team->mode_priv, 0, + sizeof(struct team) - offsetof(struct team, mode_priv)); + } + + if (!new_mode) + return 0; + + if (new_mode->ops->init) { + int err; + + err = new_mode->ops->init(team); + if (err) + return err; + } + + team->mode = new_mode; + memcpy(&team->ops, new_mode->ops, sizeof(struct team_mode_ops)); + team_adjust_ops(team); + + return 0; +} + +static int team_change_mode(struct team *team, const char *kind) +{ + struct team_mode *new_mode; + struct net_device *dev = team->dev; + int err; + + if (!list_empty(&team->port_list)) { + netdev_err(dev, "No ports can be present during mode change\n"); + return -EBUSY; + } + + if (team->mode && strcmp(team->mode->kind, kind) == 0) { + netdev_err(dev, "Unable to change to the same mode the team is in\n"); + return -EINVAL; + } + + new_mode = team_mode_get(kind); + if (!new_mode) { + netdev_err(dev, "Mode \"%s\" not found\n", kind); + return -EINVAL; + } + + err = __team_change_mode(team, new_mode); + if (err) { + netdev_err(dev, "Failed to change to mode \"%s\"\n", kind); + team_mode_put(new_mode); + return err; + } + + netdev_info(dev, "Mode changed to \"%s\"\n", kind); + return 0; +} + + +/************************ + * Rx path frame handler + ************************/ + +/* note: already called with rcu_read_lock */ +static rx_handler_result_t team_handle_frame(struct sk_buff **pskb) +{ + struct sk_buff *skb = *pskb; + struct team_port *port; + struct team *team; + rx_handler_result_t res; + + skb = skb_share_check(skb, GFP_ATOMIC); + if (!skb) + return RX_HANDLER_CONSUMED; + + *pskb = skb; + + port = team_port_get_rcu(skb->dev); + team = port->team; + + res = team->ops.receive(team, port, skb); + if (res == RX_HANDLER_ANOTHER) { + struct team_pcpu_stats *pcpu_stats; + + pcpu_stats = this_cpu_ptr(team->pcpu_stats); + u64_stats_update_begin(&pcpu_stats->syncp); + pcpu_stats->rx_packets++; + pcpu_stats->rx_bytes += skb->len; + if (skb->pkt_type == PACKET_MULTICAST) + pcpu_stats->rx_multicast++; + u64_stats_update_end(&pcpu_stats->syncp); + + skb->dev = team->dev; + } else { + this_cpu_inc(team->pcpu_stats->rx_dropped); + } + + return res; +} + + +/**************** + * Port handling + ****************/ + +static bool team_port_find(const struct team *team, + const struct team_port *port) +{ + struct team_port *cur; + + list_for_each_entry(cur, &team->port_list, list) + if (cur == port) + return true; + return false; +} + +/* + * Add/delete port to the team port list. Write guarded by rtnl_lock. + * Takes care of correct port->index setup (might be racy). + */ +static void team_port_list_add_port(struct team *team, + struct team_port *port) +{ + port->index = team->port_count++; + hlist_add_head_rcu(&port->hlist, + team_port_index_hash(team, port->index)); + list_add_tail_rcu(&port->list, &team->port_list); +} + +static void __reconstruct_port_hlist(struct team *team, int rm_index) +{ + int i; + struct team_port *port; + + for (i = rm_index + 1; i < team->port_count; i++) { + port = team_get_port_by_index(team, i); + hlist_del_rcu(&port->hlist); + port->index--; + hlist_add_head_rcu(&port->hlist, + team_port_index_hash(team, port->index)); + } +} + +static void team_port_list_del_port(struct team *team, + struct team_port *port) +{ + int rm_index = port->index; + + hlist_del_rcu(&port->hlist); + list_del_rcu(&port->list); + __reconstruct_port_hlist(team, rm_index); + team->port_count--; +} + +#define TEAM_VLAN_FEATURES (NETIF_F_ALL_CSUM | NETIF_F_SG | \ + NETIF_F_FRAGLIST | NETIF_F_ALL_TSO | \ + NETIF_F_HIGHDMA | NETIF_F_LRO) + +static void __team_compute_features(struct team *team) +{ + struct team_port *port; + u32 vlan_features = TEAM_VLAN_FEATURES; + unsigned short max_hard_header_len = ETH_HLEN; + + list_for_each_entry(port, &team->port_list, list) { + vlan_features = netdev_increment_features(vlan_features, + port->dev->vlan_features, + TEAM_VLAN_FEATURES); + + if (port->dev->hard_header_len > max_hard_header_len) + max_hard_header_len = port->dev->hard_header_len; + } + + team->dev->vlan_features = vlan_features; + team->dev->hard_header_len = max_hard_header_len; + + netdev_change_features(team->dev); +} + +static void team_compute_features(struct team *team) +{ + spin_lock(&team->lock); + __team_compute_features(team); + spin_unlock(&team->lock); +} + +static int team_port_enter(struct team *team, struct team_port *port) +{ + int err = 0; + + dev_hold(team->dev); + port->dev->priv_flags |= IFF_TEAM_PORT; + if (team->ops.port_enter) { + err = team->ops.port_enter(team, port); + if (err) { + netdev_err(team->dev, "Device %s failed to enter team mode\n", + port->dev->name); + goto err_port_enter; + } + } + + return 0; + +err_port_enter: + port->dev->priv_flags &= ~IFF_TEAM_PORT; + dev_put(team->dev); + + return err; +} + +static void team_port_leave(struct team *team, struct team_port *port) +{ + if (team->ops.port_leave) + team->ops.port_leave(team, port); + port->dev->priv_flags &= ~IFF_TEAM_PORT; + dev_put(team->dev); +} + +static void __team_port_change_check(struct team_port *port, bool linkup); + +static int team_port_add(struct team *team, struct net_device *port_dev) +{ + struct net_device *dev = team->dev; + struct team_port *port; + char *portname = port_dev->name; + int err; + + if (port_dev->flags & IFF_LOOPBACK || + port_dev->type != ARPHRD_ETHER) { + netdev_err(dev, "Device %s is of an unsupported type\n", + portname); + return -EINVAL; + } + + if (team_port_exists(port_dev)) { + netdev_err(dev, "Device %s is already a port " + "of a team device\n", portname); + return -EBUSY; + } + + if (port_dev->flags & IFF_UP) { + netdev_err(dev, "Device %s is up. Set it down before adding it as a team port\n", + portname); + return -EBUSY; + } + + port = kzalloc(sizeof(struct team_port), GFP_KERNEL); + if (!port) + return -ENOMEM; + + port->dev = port_dev; + port->team = team; + + port->orig.mtu = port_dev->mtu; + err = dev_set_mtu(port_dev, dev->mtu); + if (err) { + netdev_dbg(dev, "Error %d calling dev_set_mtu\n", err); + goto err_set_mtu; + } + + memcpy(port->orig.dev_addr, port_dev->dev_addr, ETH_ALEN); + + err = team_port_enter(team, port); + if (err) { + netdev_err(dev, "Device %s failed to enter team mode\n", + portname); + goto err_port_enter; + } + + err = dev_open(port_dev); + if (err) { + netdev_dbg(dev, "Device %s opening failed\n", + portname); + goto err_dev_open; + } + + err = netdev_set_master(port_dev, dev); + if (err) { + netdev_err(dev, "Device %s failed to set master\n", portname); + goto err_set_master; + } + + err = netdev_rx_handler_register(port_dev, team_handle_frame, + port); + if (err) { + netdev_err(dev, "Device %s failed to register rx_handler\n", + portname); + goto err_handler_register; + } + + team_port_list_add_port(team, port); + team_adjust_ops(team); + __team_compute_features(team); + __team_port_change_check(port, !!netif_carrier_ok(port_dev)); + + netdev_info(dev, "Port device %s added\n", portname); + + return 0; + +err_handler_register: + netdev_set_master(port_dev, NULL); + +err_set_master: + dev_close(port_dev); + +err_dev_open: + team_port_leave(team, port); + team_port_set_orig_mac(port); + +err_port_enter: + dev_set_mtu(port_dev, port->orig.mtu); + +err_set_mtu: + kfree(port); + + return err; +} + +static int team_port_del(struct team *team, struct net_device *port_dev) +{ + struct net_device *dev = team->dev; + struct team_port *port; + char *portname = port_dev->name; + + port = team_port_get_rtnl(port_dev); + if (!port || !team_port_find(team, port)) { + netdev_err(dev, "Device %s does not act as a port of this team\n", + portname); + return -ENOENT; + } + + __team_port_change_check(port, false); + team_port_list_del_port(team, port); + team_adjust_ops(team); + netdev_rx_handler_unregister(port_dev); + netdev_set_master(port_dev, NULL); + dev_close(port_dev); + team_port_leave(team, port); + team_port_set_orig_mac(port); + dev_set_mtu(port_dev, port->orig.mtu); + synchronize_rcu(); + kfree(port); + netdev_info(dev, "Port device %s removed\n", portname); + __team_compute_features(team); + + return 0; +} + + +/***************** + * Net device ops + *****************/ + +static const char team_no_mode_kind[] = "*NOMODE*"; + +static int team_mode_option_get(struct team *team, void *arg) +{ + const char **str = arg; + + *str = team->mode ? team->mode->kind : team_no_mode_kind; + return 0; +} + +static int team_mode_option_set(struct team *team, void *arg) +{ + const char **str = arg; + + return team_change_mode(team, *str); +} + +static struct team_option team_options[] = { + { + .name = "mode", + .type = TEAM_OPTION_TYPE_STRING, + .getter = team_mode_option_get, + .setter = team_mode_option_set, + }, +}; + +static int team_init(struct net_device *dev) +{ + struct team *team = netdev_priv(dev); + int i; + + team->dev = dev; + spin_lock_init(&team->lock); + + team->pcpu_stats = alloc_percpu(struct team_pcpu_stats); + if (!team->pcpu_stats) + return -ENOMEM; + + for (i = 0; i < TEAM_PORT_HASHENTRIES; i++) + INIT_HLIST_HEAD(&team->port_hlist[i]); + INIT_LIST_HEAD(&team->port_list); + + team_adjust_ops(team); + + INIT_LIST_HEAD(&team->option_list); + team_options_register(team, team_options, ARRAY_SIZE(team_options)); + netif_carrier_off(dev); + + return 0; +} + +static void team_uninit(struct net_device *dev) +{ + struct team *team = netdev_priv(dev); + struct team_port *port; + struct team_port *tmp; + + spin_lock(&team->lock); + list_for_each_entry_safe(port, tmp, &team->port_list, list) + team_port_del(team, port->dev); + + __team_change_mode(team, NULL); /* cleanup */ + __team_options_unregister(team, team_options, ARRAY_SIZE(team_options)); + spin_unlock(&team->lock); +} + +static void team_destructor(struct net_device *dev) +{ + struct team *team = netdev_priv(dev); + + free_percpu(team->pcpu_stats); + free_netdev(dev); +} + +static int team_open(struct net_device *dev) +{ + netif_carrier_on(dev); + return 0; +} + +static int team_close(struct net_device *dev) +{ + netif_carrier_off(dev); + return 0; +} + +/* + * note: already called with rcu_read_lock + */ +static netdev_tx_t team_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct team *team = netdev_priv(dev); + bool tx_success = false; + unsigned int len = skb->len; + + tx_success = team->ops.transmit(team, skb); + if (tx_success) { + struct team_pcpu_stats *pcpu_stats; + + pcpu_stats = this_cpu_ptr(team->pcpu_stats); + u64_stats_update_begin(&pcpu_stats->syncp); + pcpu_stats->tx_packets++; + pcpu_stats->tx_bytes += len; + u64_stats_update_end(&pcpu_stats->syncp); + } else { + this_cpu_inc(team->pcpu_stats->tx_dropped); + } + + return NETDEV_TX_OK; +} + +static void team_change_rx_flags(struct net_device *dev, int change) +{ + struct team *team = netdev_priv(dev); + struct team_port *port; + int inc; + + rcu_read_lock(); + list_for_each_entry_rcu(port, &team->port_list, list) { + if (change & IFF_PROMISC) { + inc = dev->flags & IFF_PROMISC ? 1 : -1; + dev_set_promiscuity(port->dev, inc); + } + if (change & IFF_ALLMULTI) { + inc = dev->flags & IFF_ALLMULTI ? 1 : -1; + dev_set_allmulti(port->dev, inc); + } + } + rcu_read_unlock(); +} + +static void team_set_rx_mode(struct net_device *dev) +{ + struct team *team = netdev_priv(dev); + struct team_port *port; + + rcu_read_lock(); + list_for_each_entry_rcu(port, &team->port_list, list) { + dev_uc_sync(port->dev, dev); + dev_mc_sync(port->dev, dev); + } + rcu_read_unlock(); +} + +static int team_set_mac_address(struct net_device *dev, void *p) +{ + struct team *team = netdev_priv(dev); + struct team_port *port; + struct sockaddr *addr = p; + + memcpy(dev->dev_addr, addr->sa_data, ETH_ALEN); + rcu_read_lock(); + list_for_each_entry_rcu(port, &team->port_list, list) + if (team->ops.port_change_mac) + team->ops.port_change_mac(team, port); + rcu_read_unlock(); + return 0; +} + +static int team_change_mtu(struct net_device *dev, int new_mtu) +{ + struct team *team = netdev_priv(dev); + struct team_port *port; + int err; + + /* + * Alhough this is reader, it's guarded by team lock. It's not possible + * to traverse list in reverse under rcu_read_lock + */ + spin_lock(&team->lock); + list_for_each_entry(port, &team->port_list, list) { + err = dev_set_mtu(port->dev, new_mtu); + if (err) { + netdev_err(dev, "Device %s failed to change mtu", + port->dev->name); + goto unwind; + } + } + spin_unlock(&team->lock); + + dev->mtu = new_mtu; + + return 0; + +unwind: + list_for_each_entry_continue_reverse(port, &team->port_list, list) + dev_set_mtu(port->dev, dev->mtu); + spin_unlock(&team->lock); + + return err; +} + +static struct rtnl_link_stats64 * +team_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats) +{ + struct team *team = netdev_priv(dev); + struct team_pcpu_stats *p; + u64 rx_packets, rx_bytes, rx_multicast, tx_packets, tx_bytes; + u32 rx_dropped = 0, tx_dropped = 0; + unsigned int start; + int i; + + for_each_possible_cpu(i) { + p = per_cpu_ptr(team->pcpu_stats, i); + do { + start = u64_stats_fetch_begin_bh(&p->syncp); + rx_packets = p->rx_packets; + rx_bytes = p->rx_bytes; + rx_multicast = p->rx_multicast; + tx_packets = p->tx_packets; + tx_bytes = p->tx_bytes; + } while (u64_stats_fetch_retry_bh(&p->syncp, start)); + + stats->rx_packets += rx_packets; + stats->rx_bytes += rx_bytes; + stats->multicast += rx_multicast; + stats->tx_packets += tx_packets; + stats->tx_bytes += tx_bytes; + /* + * rx_dropped & tx_dropped are u32, updated + * without syncp protection. + */ + rx_dropped += p->rx_dropped; + tx_dropped += p->tx_dropped; + } + stats->rx_dropped = rx_dropped; + stats->tx_dropped = tx_dropped; + return stats; +} + +static void team_vlan_rx_add_vid(struct net_device *dev, uint16_t vid) +{ + struct team *team = netdev_priv(dev); + struct team_port *port; + + rcu_read_lock(); + list_for_each_entry_rcu(port, &team->port_list, list) { + const struct net_device_ops *ops = port->dev->netdev_ops; + + if (ops->ndo_vlan_rx_add_vid) + ops->ndo_vlan_rx_add_vid(port->dev, vid); + } + rcu_read_unlock(); +} + +static void team_vlan_rx_kill_vid(struct net_device *dev, uint16_t vid) +{ + struct team *team = netdev_priv(dev); + struct team_port *port; + + rcu_read_lock(); + list_for_each_entry_rcu(port, &team->port_list, list) { + const struct net_device_ops *ops = port->dev->netdev_ops; + + if (ops->ndo_vlan_rx_kill_vid) + ops->ndo_vlan_rx_kill_vid(port->dev, vid); + } + rcu_read_unlock(); +} + +static int team_add_slave(struct net_device *dev, struct net_device *port_dev) +{ + struct team *team = netdev_priv(dev); + int err; + + spin_lock(&team->lock); + err = team_port_add(team, port_dev); + spin_unlock(&team->lock); + return err; +} + +static int team_del_slave(struct net_device *dev, struct net_device *port_dev) +{ + struct team *team = netdev_priv(dev); + int err; + + spin_lock(&team->lock); + err = team_port_del(team, port_dev); + spin_unlock(&team->lock); + return err; +} + +static const struct net_device_ops team_netdev_ops = { + .ndo_init = team_init, + .ndo_uninit = team_uninit, + .ndo_open = team_open, + .ndo_stop = team_close, + .ndo_start_xmit = team_xmit, + .ndo_change_rx_flags = team_change_rx_flags, + .ndo_set_rx_mode = team_set_rx_mode, + .ndo_set_mac_address = team_set_mac_address, + .ndo_change_mtu = team_change_mtu, + .ndo_get_stats64 = team_get_stats64, + .ndo_vlan_rx_add_vid = team_vlan_rx_add_vid, + .ndo_vlan_rx_kill_vid = team_vlan_rx_kill_vid, + .ndo_add_slave = team_add_slave, + .ndo_del_slave = team_del_slave, +}; + + +/*********************** + * rt netlink interface + ***********************/ + +static void team_setup(struct net_device *dev) +{ + ether_setup(dev); + + dev->netdev_ops = &team_netdev_ops; + dev->destructor = team_destructor; + dev->tx_queue_len = 0; + dev->flags |= IFF_MULTICAST; + dev->priv_flags &= ~(IFF_XMIT_DST_RELEASE | IFF_TX_SKB_SHARING); + + /* + * Indicate we support unicast address filtering. That way core won't + * bring us to promisc mode in case a unicast addr is added. + * Let this up to underlay drivers. + */ + dev->priv_flags |= IFF_UNICAST_FLT; + + dev->features |= NETIF_F_LLTX; + dev->features |= NETIF_F_GRO; + dev->hw_features = NETIF_F_HW_VLAN_TX | + NETIF_F_HW_VLAN_RX | + NETIF_F_HW_VLAN_FILTER; + + dev->features |= dev->hw_features; +} + +static int team_newlink(struct net *src_net, struct net_device *dev, + struct nlattr *tb[], struct nlattr *data[]) +{ + int err; + + if (tb[IFLA_ADDRESS] == NULL) + random_ether_addr(dev->dev_addr); + + err = register_netdevice(dev); + if (err) + return err; + + return 0; +} + +static int team_validate(struct nlattr *tb[], struct nlattr *data[]) +{ + if (tb[IFLA_ADDRESS]) { + if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN) + return -EINVAL; + if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS]))) + return -EADDRNOTAVAIL; + } + return 0; +} + +static struct rtnl_link_ops team_link_ops __read_mostly = { + .kind = DRV_NAME, + .priv_size = sizeof(struct team), + .setup = team_setup, + .newlink = team_newlink, + .validate = team_validate, +}; + + +/*********************************** + * Generic netlink custom interface + ***********************************/ + +static struct genl_family team_nl_family = { + .id = GENL_ID_GENERATE, + .name = TEAM_GENL_NAME, + .version = TEAM_GENL_VERSION, + .maxattr = TEAM_ATTR_MAX, + .netnsok = true, +}; + +static const struct nla_policy team_nl_policy[TEAM_ATTR_MAX + 1] = { + [TEAM_ATTR_UNSPEC] = { .type = NLA_UNSPEC, }, + [TEAM_ATTR_TEAM_IFINDEX] = { .type = NLA_U32 }, + [TEAM_ATTR_LIST_OPTION] = { .type = NLA_NESTED }, + [TEAM_ATTR_LIST_PORT] = { .type = NLA_NESTED }, +}; + +static const struct nla_policy +team_nl_option_policy[TEAM_ATTR_OPTION_MAX + 1] = { + [TEAM_ATTR_OPTION_UNSPEC] = { .type = NLA_UNSPEC, }, + [TEAM_ATTR_OPTION_NAME] = { + .type = NLA_STRING, + .len = TEAM_STRING_MAX_LEN, + }, + [TEAM_ATTR_OPTION_CHANGED] = { .type = NLA_FLAG }, + [TEAM_ATTR_OPTION_TYPE] = { .type = NLA_U8 }, + [TEAM_ATTR_OPTION_DATA] = { + .type = NLA_BINARY, + .len = TEAM_STRING_MAX_LEN, + }, +}; + +static int team_nl_cmd_noop(struct sk_buff *skb, struct genl_info *info) +{ + struct sk_buff *msg; + void *hdr; + int err; + + msg = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + if (!msg) + return -ENOMEM; + + hdr = genlmsg_put(msg, info->snd_pid, info->snd_seq, + &team_nl_family, 0, TEAM_CMD_NOOP); + if (IS_ERR(hdr)) { + err = PTR_ERR(hdr); + goto err_msg_put; + } + + genlmsg_end(msg, hdr); + + return genlmsg_unicast(genl_info_net(info), msg, info->snd_pid); + +err_msg_put: + nlmsg_free(msg); + + return err; +} + +/* + * Netlink cmd functions should be locked by following two functions. + * To ensure team_uninit would not be called in between, hold rcu_read_lock + * all the time. + */ +static struct team *team_nl_team_get(struct genl_info *info) +{ + struct net *net = genl_info_net(info); + int ifindex; + struct net_device *dev; + struct team *team; + + if (!info->attrs[TEAM_ATTR_TEAM_IFINDEX]) + return NULL; + + ifindex = nla_get_u32(info->attrs[TEAM_ATTR_TEAM_IFINDEX]); + rcu_read_lock(); + dev = dev_get_by_index_rcu(net, ifindex); + if (!dev || dev->netdev_ops != &team_netdev_ops) { + rcu_read_unlock(); + return NULL; + } + + team = netdev_priv(dev); + spin_lock(&team->lock); + return team; +} + +static void team_nl_team_put(struct team *team) +{ + spin_unlock(&team->lock); + rcu_read_unlock(); +} + +static int team_nl_send_generic(struct genl_info *info, struct team *team, + int (*fill_func)(struct sk_buff *skb, + struct genl_info *info, + int flags, struct team *team)) +{ + struct sk_buff *skb; + int err; + + skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOMEM; + + err = fill_func(skb, info, NLM_F_ACK, team); + if (err < 0) + goto err_fill; + + err = genlmsg_unicast(genl_info_net(info), skb, info->snd_pid); + return err; + +err_fill: + nlmsg_free(skb); + return err; +} + +static int team_nl_fill_options_get_changed(struct sk_buff *skb, + u32 pid, u32 seq, int flags, + struct team *team, + struct team_option *changed_option) +{ + struct nlattr *option_list; + void *hdr; + struct team_option *option; + + hdr = genlmsg_put(skb, pid, seq, &team_nl_family, flags, + TEAM_CMD_OPTIONS_GET); + if (IS_ERR(hdr)) + return PTR_ERR(hdr); + + NLA_PUT_U32(skb, TEAM_ATTR_TEAM_IFINDEX, team->dev->ifindex); + option_list = nla_nest_start(skb, TEAM_ATTR_LIST_OPTION); + if (!option_list) + return -EMSGSIZE; + + list_for_each_entry(option, &team->option_list, list) { + struct nlattr *option_item; + long arg; + + option_item = nla_nest_start(skb, TEAM_ATTR_ITEM_OPTION); + if (!option_item) + goto nla_put_failure; + NLA_PUT_STRING(skb, TEAM_ATTR_OPTION_NAME, option->name); + if (option == changed_option) + NLA_PUT_FLAG(skb, TEAM_ATTR_OPTION_CHANGED); + switch (option->type) { + case TEAM_OPTION_TYPE_U32: + NLA_PUT_U8(skb, TEAM_ATTR_OPTION_TYPE, NLA_U32); + team_option_get(team, option, &arg); + NLA_PUT_U32(skb, TEAM_ATTR_OPTION_DATA, arg); + break; + case TEAM_OPTION_TYPE_STRING: + NLA_PUT_U8(skb, TEAM_ATTR_OPTION_TYPE, NLA_STRING); + team_option_get(team, option, &arg); + NLA_PUT_STRING(skb, TEAM_ATTR_OPTION_DATA, + (char *) arg); + break; + default: + BUG(); + } + nla_nest_end(skb, option_item); + } + + nla_nest_end(skb, option_list); + return genlmsg_end(skb, hdr); + +nla_put_failure: + genlmsg_cancel(skb, hdr); + return -EMSGSIZE; +} + +static int team_nl_fill_options_get(struct sk_buff *skb, + struct genl_info *info, int flags, + struct team *team) +{ + return team_nl_fill_options_get_changed(skb, info->snd_pid, + info->snd_seq, NLM_F_ACK, + team, NULL); +} + +static int team_nl_cmd_options_get(struct sk_buff *skb, struct genl_info *info) +{ + struct team *team; + int err; + + team = team_nl_team_get(info); + if (!team) + return -EINVAL; + + err = team_nl_send_generic(info, team, team_nl_fill_options_get); + + team_nl_team_put(team); + + return err; +} + +static int team_nl_cmd_options_set(struct sk_buff *skb, struct genl_info *info) +{ + struct team *team; + int err = 0; + int i; + struct nlattr *nl_option; + + team = team_nl_team_get(info); + if (!team) + return -EINVAL; + + err = -EINVAL; + if (!info->attrs[TEAM_ATTR_LIST_OPTION]) { + err = -EINVAL; + goto team_put; + } + + nla_for_each_nested(nl_option, info->attrs[TEAM_ATTR_LIST_OPTION], i) { + struct nlattr *mode_attrs[TEAM_ATTR_OPTION_MAX + 1]; + enum team_option_type opt_type; + struct team_option *option; + char *opt_name; + bool opt_found = false; + + if (nla_type(nl_option) != TEAM_ATTR_ITEM_OPTION) { + err = -EINVAL; + goto team_put; + } + err = nla_parse_nested(mode_attrs, TEAM_ATTR_OPTION_MAX, + nl_option, team_nl_option_policy); + if (err) + goto team_put; + if (!mode_attrs[TEAM_ATTR_OPTION_NAME] || + !mode_attrs[TEAM_ATTR_OPTION_TYPE] || + !mode_attrs[TEAM_ATTR_OPTION_DATA]) { + err = -EINVAL; + goto team_put; + } + switch (nla_get_u8(mode_attrs[TEAM_ATTR_OPTION_TYPE])) { + case NLA_U32: + opt_type = TEAM_OPTION_TYPE_U32; + break; + case NLA_STRING: + opt_type = TEAM_OPTION_TYPE_STRING; + break; + default: + goto team_put; + } + + opt_name = nla_data(mode_attrs[TEAM_ATTR_OPTION_NAME]); + list_for_each_entry(option, &team->option_list, list) { + long arg; + struct nlattr *opt_data_attr; + + if (option->type != opt_type || + strcmp(option->name, opt_name)) + continue; + opt_found = true; + opt_data_attr = mode_attrs[TEAM_ATTR_OPTION_DATA]; + switch (opt_type) { + case TEAM_OPTION_TYPE_U32: + arg = nla_get_u32(opt_data_attr); + break; + case TEAM_OPTION_TYPE_STRING: + arg = (long) nla_data(opt_data_attr); + break; + default: + BUG(); + } + err = team_option_set(team, option, &arg); + if (err) + goto team_put; + } + if (!opt_found) { + err = -ENOENT; + goto team_put; + } + } + +team_put: + team_nl_team_put(team); + + return err; +} + +static int team_nl_fill_port_list_get_changed(struct sk_buff *skb, + u32 pid, u32 seq, int flags, + struct team *team, + struct team_port *changed_port) +{ + struct nlattr *port_list; + void *hdr; + struct team_port *port; + + hdr = genlmsg_put(skb, pid, seq, &team_nl_family, flags, + TEAM_CMD_PORT_LIST_GET); + if (IS_ERR(hdr)) + return PTR_ERR(hdr); + + NLA_PUT_U32(skb, TEAM_ATTR_TEAM_IFINDEX, team->dev->ifindex); + port_list = nla_nest_start(skb, TEAM_ATTR_LIST_PORT); + if (!port_list) + return -EMSGSIZE; + + list_for_each_entry(port, &team->port_list, list) { + struct nlattr *port_item; + + port_item = nla_nest_start(skb, TEAM_ATTR_ITEM_PORT); + if (!port_item) + goto nla_put_failure; + NLA_PUT_U32(skb, TEAM_ATTR_PORT_IFINDEX, port->dev->ifindex); + if (port == changed_port) + NLA_PUT_FLAG(skb, TEAM_ATTR_PORT_CHANGED); + if (port->linkup) + NLA_PUT_FLAG(skb, TEAM_ATTR_PORT_LINKUP); + NLA_PUT_U32(skb, TEAM_ATTR_PORT_SPEED, port->speed); + NLA_PUT_U8(skb, TEAM_ATTR_PORT_DUPLEX, port->duplex); + nla_nest_end(skb, port_item); + } + + nla_nest_end(skb, port_list); + return genlmsg_end(skb, hdr); + +nla_put_failure: + genlmsg_cancel(skb, hdr); + return -EMSGSIZE; +} + +static int team_nl_fill_port_list_get(struct sk_buff *skb, + struct genl_info *info, int flags, + struct team *team) +{ + return team_nl_fill_port_list_get_changed(skb, info->snd_pid, + info->snd_seq, NLM_F_ACK, + team, NULL); +} + +static int team_nl_cmd_port_list_get(struct sk_buff *skb, + struct genl_info *info) +{ + struct team *team; + int err; + + team = team_nl_team_get(info); + if (!team) + return -EINVAL; + + err = team_nl_send_generic(info, team, team_nl_fill_port_list_get); + + team_nl_team_put(team); + + return err; +} + +static struct genl_ops team_nl_ops[] = { + { + .cmd = TEAM_CMD_NOOP, + .doit = team_nl_cmd_noop, + .policy = team_nl_policy, + }, + { + .cmd = TEAM_CMD_OPTIONS_SET, + .doit = team_nl_cmd_options_set, + .policy = team_nl_policy, + .flags = GENL_ADMIN_PERM, + }, + { + .cmd = TEAM_CMD_OPTIONS_GET, + .doit = team_nl_cmd_options_get, + .policy = team_nl_policy, + .flags = GENL_ADMIN_PERM, + }, + { + .cmd = TEAM_CMD_PORT_LIST_GET, + .doit = team_nl_cmd_port_list_get, + .policy = team_nl_policy, + .flags = GENL_ADMIN_PERM, + }, +}; + +static struct genl_multicast_group team_change_event_mcgrp = { + .name = TEAM_GENL_CHANGE_EVENT_MC_GRP_NAME, +}; + +static int team_nl_send_event_options_get(struct team *team, + struct team_option *changed_option) +{ + struct sk_buff *skb; + int err; + struct net *net = dev_net(team->dev); + + skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOMEM; + + err = team_nl_fill_options_get_changed(skb, 0, 0, 0, team, + changed_option); + if (err < 0) + goto err_fill; + + err = genlmsg_multicast_netns(net, skb, 0, team_change_event_mcgrp.id, + GFP_KERNEL); + return err; + +err_fill: + nlmsg_free(skb); + return err; +} + +static int team_nl_send_event_port_list_get(struct team_port *port) +{ + struct sk_buff *skb; + int err; + struct net *net = dev_net(port->team->dev); + + skb = nlmsg_new(NLMSG_GOODSIZE, GFP_KERNEL); + if (!skb) + return -ENOMEM; + + err = team_nl_fill_port_list_get_changed(skb, 0, 0, 0, + port->team, port); + if (err < 0) + goto err_fill; + + err = genlmsg_multicast_netns(net, skb, 0, team_change_event_mcgrp.id, + GFP_KERNEL); + return err; + +err_fill: + nlmsg_free(skb); + return err; +} + +static int team_nl_init(void) +{ + int err; + + err = genl_register_family_with_ops(&team_nl_family, team_nl_ops, + ARRAY_SIZE(team_nl_ops)); + if (err) + return err; + + err = genl_register_mc_group(&team_nl_family, &team_change_event_mcgrp); + if (err) + goto err_change_event_grp_reg; + + return 0; + +err_change_event_grp_reg: + genl_unregister_family(&team_nl_family); + + return err; +} + +static void team_nl_fini(void) +{ + genl_unregister_family(&team_nl_family); +} + + +/****************** + * Change checkers + ******************/ + +static void __team_options_change_check(struct team *team, + struct team_option *changed_option) +{ + int err; + + err = team_nl_send_event_options_get(team, changed_option); + if (err) + netdev_warn(team->dev, "Failed to send options change via netlink\n"); +} + +/* rtnl lock is held */ +static void __team_port_change_check(struct team_port *port, bool linkup) +{ + int err; + + if (port->linkup == linkup) + return; + + port->linkup = linkup; + if (linkup) { + struct ethtool_cmd ecmd; + + err = __ethtool_get_settings(port->dev, &ecmd); + if (!err) { + port->speed = ethtool_cmd_speed(&ecmd); + port->duplex = ecmd.duplex; + goto send_event; + } + } + port->speed = 0; + port->duplex = 0; + +send_event: + err = team_nl_send_event_port_list_get(port); + if (err) + netdev_warn(port->team->dev, "Failed to send port change of device %s via netlink\n", + port->dev->name); + +} + +static void team_port_change_check(struct team_port *port, bool linkup) +{ + struct team *team = port->team; + + spin_lock(&team->lock); + __team_port_change_check(port, linkup); + spin_unlock(&team->lock); +} + +/************************************ + * Net device notifier event handler + ************************************/ + +static int team_device_event(struct notifier_block *unused, + unsigned long event, void *ptr) +{ + struct net_device *dev = (struct net_device *) ptr; + struct team_port *port; + + port = team_port_get_rtnl(dev); + if (!port) + return NOTIFY_DONE; + + switch (event) { + case NETDEV_UP: + if (netif_carrier_ok(dev)) + team_port_change_check(port, true); + case NETDEV_DOWN: + team_port_change_check(port, false); + case NETDEV_CHANGE: + if (netif_running(port->dev)) + team_port_change_check(port, + !!netif_carrier_ok(port->dev)); + break; + case NETDEV_UNREGISTER: + team_del_slave(port->team->dev, dev); + break; + case NETDEV_FEAT_CHANGE: + team_compute_features(port->team); + break; + case NETDEV_CHANGEMTU: + /* Forbid to change mtu of underlaying device */ + return NOTIFY_BAD; + case NETDEV_PRE_TYPE_CHANGE: + /* Forbid to change type of underlaying device */ + return NOTIFY_BAD; + } + return NOTIFY_DONE; +} + +static struct notifier_block team_notifier_block __read_mostly = { + .notifier_call = team_device_event, +}; + + +/*********************** + * Module init and exit + ***********************/ + +static int __init team_module_init(void) +{ + int err; + + register_netdevice_notifier(&team_notifier_block); + + err = rtnl_link_register(&team_link_ops); + if (err) + goto err_rtnl_reg; + + err = team_nl_init(); + if (err) + goto err_nl_init; + + return 0; + +err_nl_init: + rtnl_link_unregister(&team_link_ops); + +err_rtnl_reg: + unregister_netdevice_notifier(&team_notifier_block); + + return err; +} + +static void __exit team_module_exit(void) +{ + team_nl_fini(); + rtnl_link_unregister(&team_link_ops); + unregister_netdevice_notifier(&team_notifier_block); +} + +module_init(team_module_init); +module_exit(team_module_exit); + +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Jiri Pirko "); +MODULE_DESCRIPTION("Ethernet team device driver"); +MODULE_ALIAS_RTNL_LINK(DRV_NAME); diff --git a/drivers/net/team/team_mode_activebackup.c b/drivers/net/team/team_mode_activebackup.c new file mode 100644 index 0000000000000..6fe920c440b34 --- /dev/null +++ b/drivers/net/team/team_mode_activebackup.c @@ -0,0 +1,137 @@ +/* + * net/drivers/team/team_mode_activebackup.c - Active-backup mode for team + * Copyright (c) 2011 Jiri Pirko + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +struct ab_priv { + struct team_port __rcu *active_port; +}; + +static struct ab_priv *ab_priv(struct team *team) +{ + return (struct ab_priv *) &team->mode_priv; +} + +static rx_handler_result_t ab_receive(struct team *team, struct team_port *port, + struct sk_buff *skb) { + struct team_port *active_port; + + active_port = rcu_dereference(ab_priv(team)->active_port); + if (active_port != port) + return RX_HANDLER_EXACT; + return RX_HANDLER_ANOTHER; +} + +static bool ab_transmit(struct team *team, struct sk_buff *skb) +{ + struct team_port *active_port; + + active_port = rcu_dereference(ab_priv(team)->active_port); + if (unlikely(!active_port)) + goto drop; + skb->dev = active_port->dev; + if (dev_queue_xmit(skb)) + return false; + return true; + +drop: + dev_kfree_skb_any(skb); + return false; +} + +static void ab_port_leave(struct team *team, struct team_port *port) +{ + if (ab_priv(team)->active_port == port) + rcu_assign_pointer(ab_priv(team)->active_port, NULL); +} + +static int ab_active_port_get(struct team *team, void *arg) +{ + u32 *ifindex = arg; + + *ifindex = 0; + if (ab_priv(team)->active_port) + *ifindex = ab_priv(team)->active_port->dev->ifindex; + return 0; +} + +static int ab_active_port_set(struct team *team, void *arg) +{ + u32 *ifindex = arg; + struct team_port *port; + + list_for_each_entry_rcu(port, &team->port_list, list) { + if (port->dev->ifindex == *ifindex) { + rcu_assign_pointer(ab_priv(team)->active_port, port); + return 0; + } + } + return -ENOENT; +} + +static struct team_option ab_options[] = { + { + .name = "activeport", + .type = TEAM_OPTION_TYPE_U32, + .getter = ab_active_port_get, + .setter = ab_active_port_set, + }, +}; + +int ab_init(struct team *team) +{ + team_options_register(team, ab_options, ARRAY_SIZE(ab_options)); + return 0; +} + +void ab_exit(struct team *team) +{ + team_options_unregister(team, ab_options, ARRAY_SIZE(ab_options)); +} + +static const struct team_mode_ops ab_mode_ops = { + .init = ab_init, + .exit = ab_exit, + .receive = ab_receive, + .transmit = ab_transmit, + .port_leave = ab_port_leave, +}; + +static struct team_mode ab_mode = { + .kind = "activebackup", + .owner = THIS_MODULE, + .priv_size = sizeof(struct ab_priv), + .ops = &ab_mode_ops, +}; + +static int __init ab_init_module(void) +{ + return team_mode_register(&ab_mode); +} + +static void __exit ab_cleanup_module(void) +{ + team_mode_unregister(&ab_mode); +} + +module_init(ab_init_module); +module_exit(ab_cleanup_module); + +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Jiri Pirko "); +MODULE_DESCRIPTION("Active-backup mode for team"); +MODULE_ALIAS("team-mode-activebackup"); diff --git a/drivers/net/team/team_mode_roundrobin.c b/drivers/net/team/team_mode_roundrobin.c new file mode 100644 index 0000000000000..a0e8f806331af --- /dev/null +++ b/drivers/net/team/team_mode_roundrobin.c @@ -0,0 +1,107 @@ +/* + * net/drivers/team/team_mode_roundrobin.c - Round-robin mode for team + * Copyright (c) 2011 Jiri Pirko + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include +#include +#include +#include +#include +#include +#include + +struct rr_priv { + unsigned int sent_packets; +}; + +static struct rr_priv *rr_priv(struct team *team) +{ + return (struct rr_priv *) &team->mode_priv; +} + +static struct team_port *__get_first_port_up(struct team *team, + struct team_port *port) +{ + struct team_port *cur; + + if (port->linkup) + return port; + cur = port; + list_for_each_entry_continue_rcu(cur, &team->port_list, list) + if (cur->linkup) + return cur; + list_for_each_entry_rcu(cur, &team->port_list, list) { + if (cur == port) + break; + if (cur->linkup) + return cur; + } + return NULL; +} + +static bool rr_transmit(struct team *team, struct sk_buff *skb) +{ + struct team_port *port; + int port_index; + + port_index = rr_priv(team)->sent_packets++ % team->port_count; + port = team_get_port_by_index_rcu(team, port_index); + port = __get_first_port_up(team, port); + if (unlikely(!port)) + goto drop; + skb->dev = port->dev; + if (dev_queue_xmit(skb)) + return false; + return true; + +drop: + dev_kfree_skb_any(skb); + return false; +} + +static int rr_port_enter(struct team *team, struct team_port *port) +{ + return team_port_set_team_mac(port); +} + +static void rr_port_change_mac(struct team *team, struct team_port *port) +{ + team_port_set_team_mac(port); +} + +static const struct team_mode_ops rr_mode_ops = { + .transmit = rr_transmit, + .port_enter = rr_port_enter, + .port_change_mac = rr_port_change_mac, +}; + +static struct team_mode rr_mode = { + .kind = "roundrobin", + .owner = THIS_MODULE, + .priv_size = sizeof(struct rr_priv), + .ops = &rr_mode_ops, +}; + +static int __init rr_init_module(void) +{ + return team_mode_register(&rr_mode); +} + +static void __exit rr_cleanup_module(void) +{ + team_mode_unregister(&rr_mode); +} + +module_init(rr_init_module); +module_exit(rr_cleanup_module); + +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Jiri Pirko "); +MODULE_DESCRIPTION("Round-robin mode for team"); +MODULE_ALIAS("team-mode-roundrobin"); diff --git a/include/linux/Kbuild b/include/linux/Kbuild index 619b5657af77b..0b091b32267da 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -185,6 +185,7 @@ header-y += if_pppol2tp.h header-y += if_pppox.h header-y += if_slip.h header-y += if_strip.h +header-y += if_team.h header-y += if_tr.h header-y += if_tun.h header-y += if_tunnel.h diff --git a/include/linux/if.h b/include/linux/if.h index db20bd4fd16b9..06b6ef60c821a 100644 --- a/include/linux/if.h +++ b/include/linux/if.h @@ -79,6 +79,7 @@ #define IFF_TX_SKB_SHARING 0x10000 /* The interface supports sharing * skbs on transmit */ #define IFF_UNICAST_FLT 0x20000 /* Supports unicast filtering */ +#define IFF_TEAM_PORT 0x40000 /* device used as team port */ #define IF_GET_IFACE 0x0001 /* for querying only */ #define IF_GET_PROTO 0x0002 diff --git a/include/linux/if_team.h b/include/linux/if_team.h new file mode 100644 index 0000000000000..14f6388f5460d --- /dev/null +++ b/include/linux/if_team.h @@ -0,0 +1,242 @@ +/* + * include/linux/if_team.h - Network team device driver header + * Copyright (c) 2011 Jiri Pirko + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef _LINUX_IF_TEAM_H_ +#define _LINUX_IF_TEAM_H_ + +#ifdef __KERNEL__ + +struct team_pcpu_stats { + u64 rx_packets; + u64 rx_bytes; + u64 rx_multicast; + u64 tx_packets; + u64 tx_bytes; + struct u64_stats_sync syncp; + u32 rx_dropped; + u32 tx_dropped; +}; + +struct team; + +struct team_port { + struct net_device *dev; + struct hlist_node hlist; /* node in hash list */ + struct list_head list; /* node in ordinary list */ + struct team *team; + int index; + + /* + * A place for storing original values of the device before it + * become a port. + */ + struct { + unsigned char dev_addr[MAX_ADDR_LEN]; + unsigned int mtu; + } orig; + + bool linkup; + u32 speed; + u8 duplex; + + struct rcu_head rcu; +}; + +struct team_mode_ops { + int (*init)(struct team *team); + void (*exit)(struct team *team); + rx_handler_result_t (*receive)(struct team *team, + struct team_port *port, + struct sk_buff *skb); + bool (*transmit)(struct team *team, struct sk_buff *skb); + int (*port_enter)(struct team *team, struct team_port *port); + void (*port_leave)(struct team *team, struct team_port *port); + void (*port_change_mac)(struct team *team, struct team_port *port); +}; + +enum team_option_type { + TEAM_OPTION_TYPE_U32, + TEAM_OPTION_TYPE_STRING, +}; + +struct team_option { + struct list_head list; + const char *name; + enum team_option_type type; + int (*getter)(struct team *team, void *arg); + int (*setter)(struct team *team, void *arg); +}; + +struct team_mode { + struct list_head list; + const char *kind; + struct module *owner; + size_t priv_size; + const struct team_mode_ops *ops; +}; + +#define TEAM_PORT_HASHBITS 4 +#define TEAM_PORT_HASHENTRIES (1 << TEAM_PORT_HASHBITS) + +#define TEAM_MODE_PRIV_LONGS 4 +#define TEAM_MODE_PRIV_SIZE (sizeof(long) * TEAM_MODE_PRIV_LONGS) + +struct team { + struct net_device *dev; /* associated netdevice */ + struct team_pcpu_stats __percpu *pcpu_stats; + + spinlock_t lock; /* used for overall locking, e.g. port lists write */ + + /* + * port lists with port count + */ + int port_count; + struct hlist_head port_hlist[TEAM_PORT_HASHENTRIES]; + struct list_head port_list; + + struct list_head option_list; + + const struct team_mode *mode; + struct team_mode_ops ops; + long mode_priv[TEAM_MODE_PRIV_LONGS]; +}; + +static inline struct hlist_head *team_port_index_hash(struct team *team, + int port_index) +{ + return &team->port_hlist[port_index & (TEAM_PORT_HASHENTRIES - 1)]; +} + +static inline struct team_port *team_get_port_by_index(struct team *team, + int port_index) +{ + struct hlist_node *p; + struct team_port *port; + struct hlist_head *head = team_port_index_hash(team, port_index); + + hlist_for_each_entry(port, p, head, hlist) + if (port->index == port_index) + return port; + return NULL; +} +static inline struct team_port *team_get_port_by_index_rcu(struct team *team, + int port_index) +{ + struct hlist_node *p; + struct team_port *port; + struct hlist_head *head = team_port_index_hash(team, port_index); + + hlist_for_each_entry_rcu(port, p, head, hlist) + if (port->index == port_index) + return port; + return NULL; +} + +extern int team_port_set_team_mac(struct team_port *port); +extern void team_options_register(struct team *team, + struct team_option *option, + size_t option_count); +extern void team_options_unregister(struct team *team, + struct team_option *option, + size_t option_count); +extern int team_mode_register(struct team_mode *mode); +extern int team_mode_unregister(struct team_mode *mode); + +#endif /* __KERNEL__ */ + +#define TEAM_STRING_MAX_LEN 32 + +/********************************** + * NETLINK_GENERIC netlink family. + **********************************/ + +enum { + TEAM_CMD_NOOP, + TEAM_CMD_OPTIONS_SET, + TEAM_CMD_OPTIONS_GET, + TEAM_CMD_PORT_LIST_GET, + + __TEAM_CMD_MAX, + TEAM_CMD_MAX = (__TEAM_CMD_MAX - 1), +}; + +enum { + TEAM_ATTR_UNSPEC, + TEAM_ATTR_TEAM_IFINDEX, /* u32 */ + TEAM_ATTR_LIST_OPTION, /* nest */ + TEAM_ATTR_LIST_PORT, /* nest */ + + __TEAM_ATTR_MAX, + TEAM_ATTR_MAX = __TEAM_ATTR_MAX - 1, +}; + +/* Nested layout of get/set msg: + * + * [TEAM_ATTR_LIST_OPTION] + * [TEAM_ATTR_ITEM_OPTION] + * [TEAM_ATTR_OPTION_*], ... + * [TEAM_ATTR_ITEM_OPTION] + * [TEAM_ATTR_OPTION_*], ... + * ... + * [TEAM_ATTR_LIST_PORT] + * [TEAM_ATTR_ITEM_PORT] + * [TEAM_ATTR_PORT_*], ... + * [TEAM_ATTR_ITEM_PORT] + * [TEAM_ATTR_PORT_*], ... + * ... + */ + +enum { + TEAM_ATTR_ITEM_OPTION_UNSPEC, + TEAM_ATTR_ITEM_OPTION, /* nest */ + + __TEAM_ATTR_ITEM_OPTION_MAX, + TEAM_ATTR_ITEM_OPTION_MAX = __TEAM_ATTR_ITEM_OPTION_MAX - 1, +}; + +enum { + TEAM_ATTR_OPTION_UNSPEC, + TEAM_ATTR_OPTION_NAME, /* string */ + TEAM_ATTR_OPTION_CHANGED, /* flag */ + TEAM_ATTR_OPTION_TYPE, /* u8 */ + TEAM_ATTR_OPTION_DATA, /* dynamic */ + + __TEAM_ATTR_OPTION_MAX, + TEAM_ATTR_OPTION_MAX = __TEAM_ATTR_OPTION_MAX - 1, +}; + +enum { + TEAM_ATTR_ITEM_PORT_UNSPEC, + TEAM_ATTR_ITEM_PORT, /* nest */ + + __TEAM_ATTR_ITEM_PORT_MAX, + TEAM_ATTR_ITEM_PORT_MAX = __TEAM_ATTR_ITEM_PORT_MAX - 1, +}; + +enum { + TEAM_ATTR_PORT_UNSPEC, + TEAM_ATTR_PORT_IFINDEX, /* u32 */ + TEAM_ATTR_PORT_CHANGED, /* flag */ + TEAM_ATTR_PORT_LINKUP, /* flag */ + TEAM_ATTR_PORT_SPEED, /* u32 */ + TEAM_ATTR_PORT_DUPLEX, /* u8 */ + + __TEAM_ATTR_PORT_MAX, + TEAM_ATTR_PORT_MAX = __TEAM_ATTR_PORT_MAX - 1, +}; + +/* + * NETLINK_GENERIC related info + */ +#define TEAM_GENL_NAME "team" +#define TEAM_GENL_VERSION 0x1 +#define TEAM_GENL_CHANGE_EVENT_MC_GRP_NAME "change_event" + +#endif /* _LINUX_IF_TEAM_H_ */ -- GitLab From 9ff03b392fa34f6d549fbb56bf05d8a0483aa818 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Sun, 13 Nov 2011 07:20:04 +0000 Subject: [PATCH 0229/3089] sparc: sigutil: Include This file introduced in 2.6.32.47 currently fails to compile: arch/sparc/kernel/sigutil_64.c: In function 'save_fpu_state': arch/sparc/kernel/sigutil_64.c:25: error: 'EFAULT' undeclared (first use in this function) Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller --- arch/sparc/kernel/sigutil_64.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/sparc/kernel/sigutil_64.c b/arch/sparc/kernel/sigutil_64.c index e7dc508c38eb4..b19570d41a39e 100644 --- a/arch/sparc/kernel/sigutil_64.c +++ b/arch/sparc/kernel/sigutil_64.c @@ -2,6 +2,7 @@ #include #include #include +#include #include #include -- GitLab From 3ed90f766dab55fc9d49638a02704becaec379ae Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Mon, 14 Nov 2011 00:09:21 -0500 Subject: [PATCH 0230/3089] net/packet: remove dead code and unneeded variable from prb_setup_retire_blk_timer() We test for 'tx_ring' being != zero and BUG() if that's the case. So after that check there is no way that 'tx_ring' could be anything _but_ zero, so testing it again is just dead code. Once that dead code is removed, the 'pkc' local variable becomes entirely redundant, so remove that as well. Signed-off-by: Jesper Juhl Signed-off-by: David S. Miller --- net/packet/af_packet.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 82a6f34d39d01..ab10e84eda3af 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -516,13 +516,11 @@ static void prb_init_blk_timer(struct packet_sock *po, static void prb_setup_retire_blk_timer(struct packet_sock *po, int tx_ring) { - struct tpacket_kbdq_core *pkc; - if (tx_ring) BUG(); - pkc = tx_ring ? &po->tx_ring.prb_bdqc : &po->rx_ring.prb_bdqc; - prb_init_blk_timer(po, pkc, prb_retire_rx_blk_timer_expired); + prb_init_blk_timer(po, &po->rx_ring.prb_bdqc, + prb_retire_rx_blk_timer_expired); } static int prb_calc_retire_blk_tmo(struct packet_sock *po, -- GitLab From 77c1c7c4bd4751dbf47cdacd0e73e67f0a1ed316 Mon Sep 17 00:00:00 2001 From: Paul Bolle Date: Sun, 13 Nov 2011 03:36:37 +0000 Subject: [PATCH 0231/3089] rds: drop "select LLIST" Commit 1bc144b625 ("net, rds, Replace xlist in net/rds/xlist.h with llist") added "select LLIST" to the RDS_RDMA Kconfig entry. But there is no Kconfig symbol named LLIST. The select statement for that symbol is a nop. Drop it. lib/llist.o is builtin, so all that's needed to use the llist functionality is to include linux/llist.h, which this commit also did. Signed-off-by: Paul Bolle Signed-off-by: David S. Miller --- net/rds/Kconfig | 1 - 1 file changed, 1 deletion(-) diff --git a/net/rds/Kconfig b/net/rds/Kconfig index 4cf6dc7910e4b..ec753b3ae72ad 100644 --- a/net/rds/Kconfig +++ b/net/rds/Kconfig @@ -9,7 +9,6 @@ config RDS config RDS_RDMA tristate "RDS over Infiniband and iWARP" - select LLIST depends on RDS && INFINIBAND && INFINIBAND_ADDR_TRANS ---help--- Allow RDS to use Infiniband and iWARP as a transport. -- GitLab From 2a24444f8f2bea694003e3eac5c2f8d9a386bdc5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sun, 13 Nov 2011 01:24:04 +0000 Subject: [PATCH 0232/3089] ipv6: reduce percpu needs for icmpv6msg mibs Reading /proc/net/snmp6 on a machine with a lot of cpus is very expensive (can be ~88000 us). This is because ICMPV6MSG MIB uses 4096 bytes per cpu, and folding values for all possible cpus can read 16 Mbytes of memory (32MBytes on non x86 arches) ICMP messages are not considered as fast path on a typical server, and eventually few cpus handle them anyway. We can afford an atomic operation instead of using percpu data. This saves 4096 bytes per cpu and per network namespace. Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller --- include/net/ipv6.h | 15 ++++++++++++--- include/net/netns/mib.h | 2 +- include/net/snmp.h | 2 +- net/ipv6/af_inet6.c | 8 ++++---- net/ipv6/proc.c | 15 +++++---------- 5 files changed, 23 insertions(+), 19 deletions(-) diff --git a/include/net/ipv6.h b/include/net/ipv6.h index a366a8a1fe238..3f0258d2ef01f 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -132,6 +132,15 @@ extern struct ctl_path net_ipv6_ctl_path[]; SNMP_INC_STATS##modifier((net)->mib.statname##_statistics, (field));\ }) +/* per device and per net counters are atomic_long_t */ +#define _DEVINC_ATOMIC_ATOMIC(net, statname, idev, field) \ +({ \ + struct inet6_dev *_idev = (idev); \ + if (likely(_idev != NULL)) \ + SNMP_INC_STATS_ATOMIC_LONG((_idev)->stats.statname##dev, (field)); \ + SNMP_INC_STATS_ATOMIC_LONG((net)->mib.statname##_statistics, (field));\ +}) + #define _DEVADD(net, statname, modifier, idev, field, val) \ ({ \ struct inet6_dev *_idev = (idev); \ @@ -168,11 +177,11 @@ extern struct ctl_path net_ipv6_ctl_path[]; _DEVINCATOMIC(net, icmpv6, _BH, idev, field) #define ICMP6MSGOUT_INC_STATS(net, idev, field) \ - _DEVINCATOMIC(net, icmpv6msg, , idev, field +256) + _DEVINC_ATOMIC_ATOMIC(net, icmpv6msg, idev, field +256) #define ICMP6MSGOUT_INC_STATS_BH(net, idev, field) \ - _DEVINCATOMIC(net, icmpv6msg, _BH, idev, field +256) + _DEVINC_ATOMIC_ATOMIC(net, icmpv6msg, idev, field +256) #define ICMP6MSGIN_INC_STATS_BH(net, idev, field) \ - _DEVINCATOMIC(net, icmpv6msg, _BH, idev, field) + _DEVINC_ATOMIC_ATOMIC(net, icmpv6msg, idev, field) struct ip6_ra_chain { struct ip6_ra_chain *next; diff --git a/include/net/netns/mib.h b/include/net/netns/mib.h index f360135cb69ff..30f6728ee98cb 100644 --- a/include/net/netns/mib.h +++ b/include/net/netns/mib.h @@ -18,7 +18,7 @@ struct netns_mib { DEFINE_SNMP_STAT(struct udp_mib, udplite_stats_in6); DEFINE_SNMP_STAT(struct ipstats_mib, ipv6_statistics); DEFINE_SNMP_STAT(struct icmpv6_mib, icmpv6_statistics); - DEFINE_SNMP_STAT(struct icmpv6msg_mib, icmpv6msg_statistics); + DEFINE_SNMP_STAT_ATOMIC(struct icmpv6msg_mib, icmpv6msg_statistics); #endif #ifdef CONFIG_XFRM_STATISTICS DEFINE_SNMP_STAT(struct linux_xfrm_mib, xfrm_statistics); diff --git a/include/net/snmp.h b/include/net/snmp.h index 0feafa68da01e..2f65e1686fc85 100644 --- a/include/net/snmp.h +++ b/include/net/snmp.h @@ -84,7 +84,7 @@ struct icmpv6_mib_device { #define ICMP6MSG_MIB_MAX __ICMP6MSG_MIB_MAX /* per network ns counters */ struct icmpv6msg_mib { - unsigned long mibs[ICMP6MSG_MIB_MAX]; + atomic_long_t mibs[ICMP6MSG_MIB_MAX]; }; /* per device counters, (shared on all cpus) */ struct icmpv6msg_mib_device { diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 1040424c788fe..282dc7a91f32b 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -985,9 +985,9 @@ static int __net_init ipv6_init_mibs(struct net *net) sizeof(struct icmpv6_mib), __alignof__(struct icmpv6_mib)) < 0) goto err_icmp_mib; - if (snmp_mib_init((void __percpu **)net->mib.icmpv6msg_statistics, - sizeof(struct icmpv6msg_mib), - __alignof__(struct icmpv6msg_mib)) < 0) + net->mib.icmpv6msg_statistics = kzalloc(sizeof(struct icmpv6msg_mib), + GFP_KERNEL); + if (!net->mib.icmpv6msg_statistics) goto err_icmpmsg_mib; return 0; @@ -1008,7 +1008,7 @@ static void ipv6_cleanup_mibs(struct net *net) snmp_mib_free((void __percpu **)net->mib.udplite_stats_in6); snmp_mib_free((void __percpu **)net->mib.ipv6_statistics); snmp_mib_free((void __percpu **)net->mib.icmpv6_statistics); - snmp_mib_free((void __percpu **)net->mib.icmpv6msg_statistics); + kfree(net->mib.icmpv6msg_statistics); } static int __net_init inet6_net_init(struct net *net) diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index 1008ce94bc339..fdeb6d03da812 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -142,11 +142,7 @@ static const struct snmp_mib snmp6_udplite6_list[] = { SNMP_MIB_SENTINEL }; -/* can be called either with percpu mib (pcpumib != NULL), - * or shared one (smib != NULL) - */ -static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void __percpu **pcpumib, - atomic_long_t *smib) +static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, atomic_long_t *smib) { char name[32]; int i; @@ -163,14 +159,14 @@ static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void __percpu **pcpum snprintf(name, sizeof(name), "Icmp6%s%s", i & 0x100 ? "Out" : "In", p); seq_printf(seq, "%-32s\t%lu\n", name, - pcpumib ? snmp_fold_field(pcpumib, i) : atomic_long_read(smib + i)); + atomic_long_read(smib + i)); } /* print by number (nonzero only) - ICMPMsgStat format */ for (i = 0; i < ICMP6MSG_MIB_MAX; i++) { unsigned long val; - val = pcpumib ? snmp_fold_field(pcpumib, i) : atomic_long_read(smib + i); + val = atomic_long_read(smib + i); if (!val) continue; snprintf(name, sizeof(name), "Icmp6%sType%u", @@ -215,8 +211,7 @@ static int snmp6_seq_show(struct seq_file *seq, void *v) snmp6_ipstats_list, offsetof(struct ipstats_mib, syncp)); snmp6_seq_show_item(seq, (void __percpu **)net->mib.icmpv6_statistics, NULL, snmp6_icmp6_list); - snmp6_seq_show_icmpv6msg(seq, - (void __percpu **)net->mib.icmpv6msg_statistics, NULL); + snmp6_seq_show_icmpv6msg(seq, net->mib.icmpv6msg_statistics->mibs); snmp6_seq_show_item(seq, (void __percpu **)net->mib.udp_stats_in6, NULL, snmp6_udp6_list); snmp6_seq_show_item(seq, (void __percpu **)net->mib.udplite_stats_in6, @@ -246,7 +241,7 @@ static int snmp6_dev_seq_show(struct seq_file *seq, void *v) snmp6_ipstats_list); snmp6_seq_show_item(seq, NULL, idev->stats.icmpv6dev->mibs, snmp6_icmp6_list); - snmp6_seq_show_icmpv6msg(seq, NULL, idev->stats.icmpv6msgdev->mibs); + snmp6_seq_show_icmpv6msg(seq, idev->stats.icmpv6msgdev->mibs); return 0; } -- GitLab From 898bdf2cb43eb0a962c397eb4dd1aec2c7211be2 Mon Sep 17 00:00:00 2001 From: david decotigny Date: Fri, 11 Nov 2011 16:27:34 +0000 Subject: [PATCH 0233/3089] forcedeth: fix stats on hardware without extended stats support This change makes sure that tx_packets/rx_bytes ifconfig counters are updated even on NICs that don't provide hardware support for these stats: they are now updated in software. For the sake of consistency, we also now have tx_bytes updated in software (hardware counters include ethernet CRC, and software doesn't account for it). This reverts parts of: - "forcedeth: statistics optimization" (21828163b2) - "forcedeth: Improve stats counters" (0bdfea8ba8) - "forcedeth: remove unneeded stats updates" (4687f3f364) Tested: pktgen + loopback (http://patchwork.ozlabs.org/patch/124698/) reports identical tx_packets/rx_packets and tx_bytes/rx_bytes. Signed-off-by: David Decotigny Signed-off-by: David S. Miller --- drivers/net/ethernet/nvidia/forcedeth.c | 36 ++++++++++++++++++------- 1 file changed, 27 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c index 1dca57013cb2d..1c61d36e65707 100644 --- a/drivers/net/ethernet/nvidia/forcedeth.c +++ b/drivers/net/ethernet/nvidia/forcedeth.c @@ -609,7 +609,7 @@ struct nv_ethtool_str { }; static const struct nv_ethtool_str nv_estats_str[] = { - { "tx_bytes" }, + { "tx_bytes" }, /* includes Ethernet FCS CRC */ { "tx_zero_rexmt" }, { "tx_one_rexmt" }, { "tx_many_rexmt" }, @@ -637,7 +637,7 @@ static const struct nv_ethtool_str nv_estats_str[] = { /* version 2 stats */ { "tx_deferral" }, { "tx_packets" }, - { "rx_bytes" }, + { "rx_bytes" }, /* includes Ethernet FCS CRC */ { "tx_pause" }, { "rx_pause" }, { "rx_drop_frame" }, @@ -649,7 +649,7 @@ static const struct nv_ethtool_str nv_estats_str[] = { }; struct nv_ethtool_stats { - u64 tx_bytes; + u64 tx_bytes; /* should be ifconfig->tx_bytes + 4*tx_packets */ u64 tx_zero_rexmt; u64 tx_one_rexmt; u64 tx_many_rexmt; @@ -670,14 +670,14 @@ struct nv_ethtool_stats { u64 rx_unicast; u64 rx_multicast; u64 rx_broadcast; - u64 rx_packets; + u64 rx_packets; /* should be ifconfig->rx_packets */ u64 rx_errors_total; u64 tx_errors_total; /* version 2 stats */ u64 tx_deferral; - u64 tx_packets; - u64 rx_bytes; + u64 tx_packets; /* should be ifconfig->tx_packets */ + u64 rx_bytes; /* should be ifconfig->rx_bytes + 4*rx_packets */ u64 tx_pause; u64 rx_pause; u64 rx_drop_frame; @@ -1706,10 +1706,17 @@ static struct net_device_stats *nv_get_stats(struct net_device *dev) if (np->driver_data & (DEV_HAS_STATISTICS_V1|DEV_HAS_STATISTICS_V2|DEV_HAS_STATISTICS_V3)) { nv_get_hw_stats(dev); + /* + * Note: because HW stats are not always available and + * for consistency reasons, the following ifconfig + * stats are managed by software: rx_bytes, tx_bytes, + * rx_packets and tx_packets. The related hardware + * stats reported by ethtool should be equivalent to + * these ifconfig stats, with 4 additional bytes per + * packet (Ethernet FCS CRC). + */ + /* copy to net_device stats */ - dev->stats.tx_packets = np->estats.tx_packets; - dev->stats.rx_bytes = np->estats.rx_bytes; - dev->stats.tx_bytes = np->estats.tx_bytes; dev->stats.tx_fifo_errors = np->estats.tx_fifo_errors; dev->stats.tx_carrier_errors = np->estats.tx_carrier_errors; dev->stats.rx_crc_errors = np->estats.rx_crc_errors; @@ -2380,6 +2387,9 @@ static int nv_tx_done(struct net_device *dev, int limit) if (flags & NV_TX_ERROR) { if ((flags & NV_TX_RETRYERROR) && !(flags & NV_TX_RETRYCOUNT_MASK)) nv_legacybackoff_reseed(dev); + } else { + dev->stats.tx_packets++; + dev->stats.tx_bytes += np->get_tx_ctx->skb->len; } dev_kfree_skb_any(np->get_tx_ctx->skb); np->get_tx_ctx->skb = NULL; @@ -2390,6 +2400,9 @@ static int nv_tx_done(struct net_device *dev, int limit) if (flags & NV_TX2_ERROR) { if ((flags & NV_TX2_RETRYERROR) && !(flags & NV_TX2_RETRYCOUNT_MASK)) nv_legacybackoff_reseed(dev); + } else { + dev->stats.tx_packets++; + dev->stats.tx_bytes += np->get_tx_ctx->skb->len; } dev_kfree_skb_any(np->get_tx_ctx->skb); np->get_tx_ctx->skb = NULL; @@ -2429,6 +2442,9 @@ static int nv_tx_done_optimized(struct net_device *dev, int limit) else nv_legacybackoff_reseed(dev); } + } else { + dev->stats.tx_packets++; + dev->stats.tx_bytes += np->get_tx_ctx->skb->len; } dev_kfree_skb_any(np->get_tx_ctx->skb); @@ -2678,6 +2694,7 @@ static int nv_rx_process(struct net_device *dev, int limit) skb->protocol = eth_type_trans(skb, dev); napi_gro_receive(&np->napi, skb); dev->stats.rx_packets++; + dev->stats.rx_bytes += len; next_pkt: if (unlikely(np->get_rx.orig++ == np->last_rx.orig)) np->get_rx.orig = np->first_rx.orig; @@ -2761,6 +2778,7 @@ static int nv_rx_process_optimized(struct net_device *dev, int limit) } napi_gro_receive(&np->napi, skb); dev->stats.rx_packets++; + dev->stats.rx_bytes += len; } else { dev_kfree_skb(skb); } -- GitLab From 719269afbc69ab96339aad6c2d3b32f7d8311146 Mon Sep 17 00:00:00 2001 From: "alex.bluesman.smirnov@gmail.com" Date: Thu, 10 Nov 2011 07:38:38 +0000 Subject: [PATCH 0234/3089] 6LoWPAN: add fragmentation support This patch adds support for frame fragmentation. Signed-off-by: Alexander Smirnov Signed-off-by: David S. Miller --- include/net/ieee802154.h | 6 + net/ieee802154/6lowpan.c | 260 ++++++++++++++++++++++++++++++++++++++- net/ieee802154/6lowpan.h | 18 +++ 3 files changed, 280 insertions(+), 4 deletions(-) diff --git a/include/net/ieee802154.h b/include/net/ieee802154.h index d52685defb11e..ee59f8b188ddf 100644 --- a/include/net/ieee802154.h +++ b/include/net/ieee802154.h @@ -21,11 +21,14 @@ * Maxim Gorbachyov * Maxim Osipov * Dmitry Eremin-Solenikov + * Alexander Smirnov */ #ifndef NET_IEEE802154_H #define NET_IEEE802154_H +#define IEEE802154_MTU 127 + #define IEEE802154_FC_TYPE_BEACON 0x0 /* Frame is beacon */ #define IEEE802154_FC_TYPE_DATA 0x1 /* Frame is data */ #define IEEE802154_FC_TYPE_ACK 0x2 /* Frame is acknowledgment */ @@ -56,6 +59,9 @@ (((x) & IEEE802154_FC_DAMODE_MASK) >> IEEE802154_FC_DAMODE_SHIFT) +/* MAC footer size */ +#define IEEE802154_MFR_SIZE 2 /* 2 octets */ + /* MAC's Command Frames Identifiers */ #define IEEE802154_CMD_ASSOCIATION_REQ 0x01 #define IEEE802154_CMD_ASSOCIATION_RESP 0x02 diff --git a/net/ieee802154/6lowpan.c b/net/ieee802154/6lowpan.c index 19d6aefe97d4b..7d4cb58bbddcb 100644 --- a/net/ieee802154/6lowpan.c +++ b/net/ieee802154/6lowpan.c @@ -113,6 +113,20 @@ struct lowpan_dev_record { struct list_head list; }; +struct lowpan_fragment { + struct sk_buff *skb; /* skb to be assembled */ + spinlock_t lock; /* concurency lock */ + u16 length; /* length to be assemled */ + u32 bytes_rcv; /* bytes received */ + u16 tag; /* current fragment tag */ + struct timer_list timer; /* assembling timer */ + struct list_head list; /* fragments list */ +}; + +static unsigned short fragment_tag; +static LIST_HEAD(lowpan_fragments); +spinlock_t flist_lock; + static inline struct lowpan_dev_info *lowpan_dev_info(const struct net_device *dev) { @@ -244,6 +258,17 @@ static u8 lowpan_fetch_skb_u8(struct sk_buff *skb) return ret; } +static u16 lowpan_fetch_skb_u16(struct sk_buff *skb) +{ + u16 ret; + + BUG_ON(!pskb_may_pull(skb, 2)); + + ret = skb->data[0] | (skb->data[1] << 8); + skb_pull(skb, 2); + return ret; +} + static int lowpan_header_create(struct sk_buff *skb, struct net_device *dev, unsigned short type, const void *_daddr, @@ -467,6 +492,7 @@ static int lowpan_header_create(struct sk_buff *skb, memcpy(&(sa.hwaddr), saddr, 8); mac_cb(skb)->flags = IEEE802154_FC_TYPE_DATA; + return dev_hard_header(skb, lowpan_dev_info(dev)->real_dev, type, (void *)&da, (void *)&sa, skb->len); } @@ -511,6 +537,21 @@ static int lowpan_skb_deliver(struct sk_buff *skb, struct ipv6hdr *hdr) return stat; } +static void lowpan_fragment_timer_expired(unsigned long entry_addr) +{ + struct lowpan_fragment *entry = (struct lowpan_fragment *)entry_addr; + + pr_debug("%s: timer expired for frame with tag %d\n", __func__, + entry->tag); + + spin_lock(&flist_lock); + list_del(&entry->list); + spin_unlock(&flist_lock); + + dev_kfree_skb(entry->skb); + kfree(entry); +} + static int lowpan_process_data(struct sk_buff *skb) { @@ -525,6 +566,107 @@ lowpan_process_data(struct sk_buff *skb) if (skb->len < 2) goto drop; iphc0 = lowpan_fetch_skb_u8(skb); + + /* fragments assembling */ + switch (iphc0 & LOWPAN_DISPATCH_MASK) { + case LOWPAN_DISPATCH_FRAG1: + case LOWPAN_DISPATCH_FRAGN: + { + struct lowpan_fragment *frame; + u8 len, offset; + u16 tag; + bool found = false; + + len = lowpan_fetch_skb_u8(skb); /* frame length */ + tag = lowpan_fetch_skb_u16(skb); + + /* + * check if frame assembling with the same tag is + * already in progress + */ + spin_lock(&flist_lock); + + list_for_each_entry(frame, &lowpan_fragments, list) + if (frame->tag == tag) { + found = true; + break; + } + + /* alloc new frame structure */ + if (!found) { + frame = kzalloc(sizeof(struct lowpan_fragment), + GFP_ATOMIC); + if (!frame) + goto unlock_and_drop; + + INIT_LIST_HEAD(&frame->list); + + frame->length = (iphc0 & 7) | (len << 3); + frame->tag = tag; + + /* allocate buffer for frame assembling */ + frame->skb = alloc_skb(frame->length + + sizeof(struct ipv6hdr), GFP_ATOMIC); + + if (!frame->skb) { + kfree(frame); + goto unlock_and_drop; + } + + frame->skb->priority = skb->priority; + frame->skb->dev = skb->dev; + + /* reserve headroom for uncompressed ipv6 header */ + skb_reserve(frame->skb, sizeof(struct ipv6hdr)); + skb_put(frame->skb, frame->length); + + init_timer(&frame->timer); + /* time out is the same as for ipv6 - 60 sec */ + frame->timer.expires = jiffies + LOWPAN_FRAG_TIMEOUT; + frame->timer.data = (unsigned long)frame; + frame->timer.function = lowpan_fragment_timer_expired; + + add_timer(&frame->timer); + + list_add_tail(&frame->list, &lowpan_fragments); + } + + if ((iphc0 & LOWPAN_DISPATCH_MASK) == LOWPAN_DISPATCH_FRAG1) + goto unlock_and_drop; + + offset = lowpan_fetch_skb_u8(skb); /* fetch offset */ + + /* if payload fits buffer, copy it */ + if (likely((offset * 8 + skb->len) <= frame->length)) + skb_copy_to_linear_data_offset(frame->skb, offset * 8, + skb->data, skb->len); + else + goto unlock_and_drop; + + frame->bytes_rcv += skb->len; + + /* frame assembling complete */ + if ((frame->bytes_rcv == frame->length) && + frame->timer.expires > jiffies) { + /* if timer haven't expired - first of all delete it */ + del_timer(&frame->timer); + list_del(&frame->list); + spin_unlock(&flist_lock); + + dev_kfree_skb(skb); + skb = frame->skb; + kfree(frame); + iphc0 = lowpan_fetch_skb_u8(skb); + break; + } + spin_unlock(&flist_lock); + + return kfree_skb(skb), 0; + } + default: + break; + } + iphc1 = lowpan_fetch_skb_u8(skb); _saddr = mac_cb(skb)->sa.hwaddr; @@ -674,6 +816,9 @@ lowpan_process_data(struct sk_buff *skb) lowpan_raw_dump_table(__func__, "raw header dump", (u8 *)&hdr, sizeof(hdr)); return lowpan_skb_deliver(skb, &hdr); + +unlock_and_drop: + spin_unlock(&flist_lock); drop: kfree_skb(skb); return -EINVAL; @@ -692,18 +837,118 @@ static int lowpan_set_address(struct net_device *dev, void *p) return 0; } +static int lowpan_get_mac_header_length(struct sk_buff *skb) +{ + /* + * Currently long addressing mode is supported only, so the overall + * header size is 21: + * FC SeqNum DPAN DA SA Sec + * 2 + 1 + 2 + 8 + 8 + 0 = 21 + */ + return 21; +} + +static int +lowpan_fragment_xmit(struct sk_buff *skb, u8 *head, + int mlen, int plen, int offset) +{ + struct sk_buff *frag; + int hlen, ret; + + /* if payload length is zero, therefore it's a first fragment */ + hlen = (plen == 0 ? LOWPAN_FRAG1_HEAD_SIZE : LOWPAN_FRAGN_HEAD_SIZE); + + lowpan_raw_dump_inline(__func__, "6lowpan fragment header", head, hlen); + + frag = dev_alloc_skb(hlen + mlen + plen + IEEE802154_MFR_SIZE); + if (!frag) + return -ENOMEM; + + frag->priority = skb->priority; + frag->dev = skb->dev; + + /* copy header, MFR and payload */ + memcpy(skb_put(frag, mlen), skb->data, mlen); + memcpy(skb_put(frag, hlen), head, hlen); + + if (plen) + skb_copy_from_linear_data_offset(skb, offset + mlen, + skb_put(frag, plen), plen); + + lowpan_raw_dump_table(__func__, " raw fragment dump", frag->data, + frag->len); + + ret = dev_queue_xmit(frag); + + if (ret < 0) + dev_kfree_skb(frag); + + return ret; +} + +static int +lowpan_skb_fragmentation(struct sk_buff *skb) +{ + int err, header_length, payload_length, tag, offset = 0; + u8 head[5]; + + header_length = lowpan_get_mac_header_length(skb); + payload_length = skb->len - header_length; + tag = fragment_tag++; + + /* first fragment header */ + head[0] = LOWPAN_DISPATCH_FRAG1 | (payload_length & 0x7); + head[1] = (payload_length >> 3) & 0xff; + head[2] = tag & 0xff; + head[3] = tag >> 8; + + err = lowpan_fragment_xmit(skb, head, header_length, 0, 0); + + /* next fragment header */ + head[0] &= ~LOWPAN_DISPATCH_FRAG1; + head[0] |= LOWPAN_DISPATCH_FRAGN; + + while ((payload_length - offset > 0) && (err >= 0)) { + int len = LOWPAN_FRAG_SIZE; + + head[4] = offset / 8; + + if (payload_length - offset < len) + len = payload_length - offset; + + err = lowpan_fragment_xmit(skb, head, header_length, + len, offset); + offset += len; + } + + return err; +} + static netdev_tx_t lowpan_xmit(struct sk_buff *skb, struct net_device *dev) { - int err = 0; + int err = -1; pr_debug("(%s): package xmit\n", __func__); skb->dev = lowpan_dev_info(dev)->real_dev; if (skb->dev == NULL) { pr_debug("(%s) ERROR: no real wpan device found\n", __func__); - dev_kfree_skb(skb); - } else + goto error; + } + + if (skb->len <= IEEE802154_MTU) { err = dev_queue_xmit(skb); + goto out; + } + + pr_debug("(%s): frame is too big, fragmentation is needed\n", + __func__); + err = lowpan_skb_fragmentation(skb); +error: + dev_kfree_skb(skb); +out: + if (err < 0) + pr_debug("(%s): ERROR: xmit failed\n", __func__); return (err < 0 ? NETDEV_TX_BUSY : NETDEV_TX_OK); } @@ -765,8 +1010,15 @@ static int lowpan_rcv(struct sk_buff *skb, struct net_device *dev, goto drop; /* check that it's our buffer */ - if ((skb->data[0] & 0xe0) == 0x60) + switch (skb->data[0] & 0xe0) { + case LOWPAN_DISPATCH_IPHC: /* ipv6 datagram */ + case LOWPAN_DISPATCH_FRAG1: /* first fragment header */ + case LOWPAN_DISPATCH_FRAGN: /* next fragments headers */ lowpan_process_data(skb); + break; + default: + break; + } return NET_RX_SUCCESS; diff --git a/net/ieee802154/6lowpan.h b/net/ieee802154/6lowpan.h index 5d8cf80b930dc..5d2e5a03742f1 100644 --- a/net/ieee802154/6lowpan.h +++ b/net/ieee802154/6lowpan.h @@ -159,6 +159,24 @@ #define LOWPAN_DISPATCH_FRAG1 0xc0 /* 11000xxx */ #define LOWPAN_DISPATCH_FRAGN 0xe0 /* 11100xxx */ +#define LOWPAN_DISPATCH_MASK 0xf8 /* 11111000 */ + +#define LOWPAN_FRAG_TIMEOUT (HZ * 60) /* time-out 60 sec */ + +#define LOWPAN_FRAG1_HEAD_SIZE 0x4 +#define LOWPAN_FRAGN_HEAD_SIZE 0x5 + +/* + * According IEEE802.15.4 standard: + * - MTU is 127 octets + * - maximum MHR size is 37 octets + * - MFR size is 2 octets + * + * so minimal payload size that we may guarantee is: + * MTU - MHR - MFR = 88 octets + */ +#define LOWPAN_FRAG_SIZE 88 + /* * Values of fields within the IPHC encoding first byte * (C stands for compressed and I for inline) -- GitLab From e86586ba8c570591372bc6cd0435a9bc1364c6d5 Mon Sep 17 00:00:00 2001 From: "alex.bluesman.smirnov@gmail.com" Date: Thu, 10 Nov 2011 07:39:15 +0000 Subject: [PATCH 0235/3089] 6LoWPAN: disable debugging by default This patch disables debug output enabled by default. Signed-off-by: Alexander Smirnov Acked-by: Dmitry Eremin-Solenikov Signed-off-by: David S. Miller --- net/ieee802154/6lowpan.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/net/ieee802154/6lowpan.c b/net/ieee802154/6lowpan.c index 7d4cb58bbddcb..af5553eb1283b 100644 --- a/net/ieee802154/6lowpan.c +++ b/net/ieee802154/6lowpan.c @@ -50,8 +50,6 @@ * SUCH DAMAGE. */ -#define DEBUG - #include #include #include -- GitLab From 4d039f684314b707fdae3c7262faf4a80c548194 Mon Sep 17 00:00:00 2001 From: "alex.bluesman.smirnov@gmail.com" Date: Thu, 10 Nov 2011 07:39:37 +0000 Subject: [PATCH 0236/3089] 6LoWPAN: set proper netdev flags This patch fixes settings for device initialization which makes possible to use NDISC and TCP. Signed-off-by: Alexander Smirnov Acked-by: Dmitry Eremin-Solenikov Signed-off-by: David S. Miller --- net/ieee802154/6lowpan.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/ieee802154/6lowpan.c b/net/ieee802154/6lowpan.c index af5553eb1283b..39ec6e0182273 100644 --- a/net/ieee802154/6lowpan.c +++ b/net/ieee802154/6lowpan.c @@ -973,13 +973,12 @@ static void lowpan_setup(struct net_device *dev) dev->addr_len = IEEE802154_ADDR_LEN; memset(dev->broadcast, 0xff, IEEE802154_ADDR_LEN); dev->type = ARPHRD_IEEE802154; - dev->features = NETIF_F_NO_CSUM; /* Frame Control + Sequence Number + Address fields + Security Header */ dev->hard_header_len = 2 + 1 + 20 + 14; dev->needed_tailroom = 2; /* FCS */ dev->mtu = 1281; dev->tx_queue_len = 0; - dev->flags = IFF_NOARP | IFF_BROADCAST; + dev->flags = IFF_BROADCAST | IFF_MULTICAST; dev->watchdog_timeo = 0; dev->netdev_ops = &lowpan_netdev_ops; -- GitLab From 3bd5b958c2a2dd1a9b4c8d21e75fb47b062fc941 Mon Sep 17 00:00:00 2001 From: "alex.bluesman.smirnov@gmail.com" Date: Thu, 10 Nov 2011 07:40:14 +0000 Subject: [PATCH 0237/3089] 6LoWPAN: UDP header compression This patch adds support for UDP header compression. Derived from Contiki OS. Signed-off-by: Alexander Smirnov Signed-off-by: David S. Miller --- net/ieee802154/6lowpan.c | 51 ++++++++++++++++++++++++++++++++++++---- net/ieee802154/6lowpan.h | 5 ++++ 2 files changed, 52 insertions(+), 4 deletions(-) diff --git a/net/ieee802154/6lowpan.c b/net/ieee802154/6lowpan.c index 39ec6e0182273..9bf82d72bb9e4 100644 --- a/net/ieee802154/6lowpan.c +++ b/net/ieee802154/6lowpan.c @@ -246,6 +246,50 @@ lowpan_uncompress_addr(struct sk_buff *skb, struct in6_addr *ipaddr, return 0; } +static void +lowpan_compress_udp_header(u8 **hc06_ptr, struct sk_buff *skb) +{ + struct udphdr *uh = udp_hdr(skb); + + pr_debug("(%s): UDP header compression\n", __func__); + + if (((uh->source & LOWPAN_NHC_UDP_4BIT_MASK) == + LOWPAN_NHC_UDP_4BIT_PORT) && + ((uh->dest & LOWPAN_NHC_UDP_4BIT_MASK) == + LOWPAN_NHC_UDP_4BIT_PORT)) { + pr_debug("(%s): both ports compression to 4 bits\n", __func__); + **hc06_ptr = LOWPAN_NHC_UDP_CS_P_11; + **(hc06_ptr + 1) = /* subtraction is faster */ + (u8)((uh->dest - LOWPAN_NHC_UDP_4BIT_PORT) + + ((uh->source & LOWPAN_NHC_UDP_4BIT_PORT) << 4)); + *hc06_ptr += 2; + } else if ((uh->dest & LOWPAN_NHC_UDP_8BIT_MASK) == + LOWPAN_NHC_UDP_8BIT_PORT) { + pr_debug("(%s): remove 8 bits of dest\n", __func__); + **hc06_ptr = LOWPAN_NHC_UDP_CS_P_01; + memcpy(*hc06_ptr + 1, &uh->source, 2); + **(hc06_ptr + 3) = (u8)(uh->dest - LOWPAN_NHC_UDP_8BIT_PORT); + *hc06_ptr += 4; + } else if ((uh->source & LOWPAN_NHC_UDP_8BIT_MASK) == + LOWPAN_NHC_UDP_8BIT_PORT) { + pr_debug("(%s): remove 8 bits of source\n", __func__); + **hc06_ptr = LOWPAN_NHC_UDP_CS_P_10; + memcpy(*hc06_ptr + 1, &uh->dest, 2); + **(hc06_ptr + 3) = (u8)(uh->source - LOWPAN_NHC_UDP_8BIT_PORT); + *hc06_ptr += 4; + } else { + pr_debug("(%s): can't compress header\n", __func__); + **hc06_ptr = LOWPAN_NHC_UDP_CS_P_00; + memcpy(*hc06_ptr + 1, &uh->source, 2); + memcpy(*hc06_ptr + 3, &uh->dest, 2); + *hc06_ptr += 5; + } + + /* checksum is always inline */ + memcpy(*hc06_ptr, &uh->check, 2); + *hc06_ptr += 2; +} + static u8 lowpan_fetch_skb_u8(struct sk_buff *skb) { u8 ret; @@ -365,8 +409,6 @@ static int lowpan_header_create(struct sk_buff *skb, if (hdr->nexthdr == UIP_PROTO_UDP) iphc0 |= LOWPAN_IPHC_NH_C; -/* TODO: next header compression */ - if ((iphc0 & LOWPAN_IPHC_NH_C) == 0) { *hc06_ptr = hdr->nexthdr; hc06_ptr += 1; @@ -454,8 +496,9 @@ static int lowpan_header_create(struct sk_buff *skb, } } - /* TODO: UDP header compression */ - /* TODO: Next Header compression */ + /* UDP header compression */ + if (hdr->nexthdr == UIP_PROTO_UDP) + lowpan_compress_udp_header(&hc06_ptr, skb); head[0] = iphc0; head[1] = iphc1; diff --git a/net/ieee802154/6lowpan.h b/net/ieee802154/6lowpan.h index 5d2e5a03742f1..aeff3f3104821 100644 --- a/net/ieee802154/6lowpan.h +++ b/net/ieee802154/6lowpan.h @@ -219,6 +219,11 @@ #define LOWPAN_NHC_UDP_CHECKSUMC 0x04 #define LOWPAN_NHC_UDP_CHECKSUMI 0x00 +#define LOWPAN_NHC_UDP_4BIT_PORT 0xF0B0 +#define LOWPAN_NHC_UDP_4BIT_MASK 0xFFF0 +#define LOWPAN_NHC_UDP_8BIT_PORT 0xF000 +#define LOWPAN_NHC_UDP_8BIT_MASK 0xFF00 + /* values for port compression, _with checksum_ ie bit 5 set to 0 */ #define LOWPAN_NHC_UDP_CS_P_00 0xF0 /* all inline */ #define LOWPAN_NHC_UDP_CS_P_01 0xF1 /* source 16bit inline, -- GitLab From f8b1b5d231c6db03f87e9db195530156fde47c4b Mon Sep 17 00:00:00 2001 From: "alex.bluesman.smirnov@gmail.com" Date: Thu, 10 Nov 2011 07:40:53 +0000 Subject: [PATCH 0238/3089] 6LoWPAN: UDP header decompression This patch provides possibility to decompress UDP headers. Derived from Contiki OS. Signed-off-by: Alexander Smirnov Signed-off-by: David S. Miller --- net/ieee802154/6lowpan.c | 61 +++++++++++++++++++++++++++++++++++++++- 1 file changed, 60 insertions(+), 1 deletion(-) diff --git a/net/ieee802154/6lowpan.c b/net/ieee802154/6lowpan.c index 9bf82d72bb9e4..602f318a8d62f 100644 --- a/net/ieee802154/6lowpan.c +++ b/net/ieee802154/6lowpan.c @@ -311,6 +311,62 @@ static u16 lowpan_fetch_skb_u16(struct sk_buff *skb) return ret; } +static int +lowpan_uncompress_udp_header(struct sk_buff *skb) +{ + struct udphdr *uh = udp_hdr(skb); + u8 tmp; + + tmp = lowpan_fetch_skb_u8(skb); + + if ((tmp & LOWPAN_NHC_UDP_MASK) == LOWPAN_NHC_UDP_ID) { + pr_debug("(%s): UDP header uncompression\n", __func__); + switch (tmp & LOWPAN_NHC_UDP_CS_P_11) { + case LOWPAN_NHC_UDP_CS_P_00: + memcpy(&uh->source, &skb->data[0], 2); + memcpy(&uh->dest, &skb->data[2], 2); + skb_pull(skb, 4); + break; + case LOWPAN_NHC_UDP_CS_P_01: + memcpy(&uh->source, &skb->data[0], 2); + uh->dest = + skb->data[2] + LOWPAN_NHC_UDP_8BIT_PORT; + skb_pull(skb, 3); + break; + case LOWPAN_NHC_UDP_CS_P_10: + uh->source = skb->data[0] + LOWPAN_NHC_UDP_8BIT_PORT; + memcpy(&uh->dest, &skb->data[1], 2); + skb_pull(skb, 3); + break; + case LOWPAN_NHC_UDP_CS_P_11: + uh->source = + LOWPAN_NHC_UDP_4BIT_PORT + (skb->data[0] >> 4); + uh->dest = + LOWPAN_NHC_UDP_4BIT_PORT + (skb->data[0] & 0x0f); + skb_pull(skb, 1); + break; + default: + pr_debug("(%s) ERROR: unknown UDP format\n", __func__); + goto err; + break; + } + + pr_debug("(%s): uncompressed UDP ports: src = %d, dst = %d\n", + __func__, uh->source, uh->dest); + + /* copy checksum */ + memcpy(&uh->check, &skb->data[0], 2); + skb_pull(skb, 2); + } else { + pr_debug("(%s): ERROR: unsupported NH format\n", __func__); + goto err; + } + + return 0; +err: + return -EINVAL; +} + static int lowpan_header_create(struct sk_buff *skb, struct net_device *dev, unsigned short type, const void *_daddr, @@ -842,7 +898,10 @@ lowpan_process_data(struct sk_buff *skb) goto drop; } - /* TODO: UDP header parse */ + /* UDP data uncompression */ + if (iphc0 & LOWPAN_IPHC_NH_C) + if (lowpan_uncompress_udp_header(skb)) + goto drop; /* Not fragmented package */ hdr.payload_len = htons(skb->len); -- GitLab From 63ce40e4fd7d68373127a51dd1facef07c93cf4a Mon Sep 17 00:00:00 2001 From: "alex.bluesman.smirnov@gmail.com" Date: Thu, 10 Nov 2011 07:41:11 +0000 Subject: [PATCH 0239/3089] 6LoWPAN: update documentation This patch adds chapter to documentation which describes how to use 6lowpan technology. Signed-off-by: Alexander Smirnov Signed-off-by: David S. Miller --- Documentation/networking/ieee802154.txt | 27 +++++++++++++++++++++++++ 1 file changed, 27 insertions(+) diff --git a/Documentation/networking/ieee802154.txt b/Documentation/networking/ieee802154.txt index f41ea24052206..1dc1c24a75470 100644 --- a/Documentation/networking/ieee802154.txt +++ b/Documentation/networking/ieee802154.txt @@ -78,3 +78,30 @@ in software. This is currently WIP. See header include/net/mac802154.h and several drivers in drivers/ieee802154/. +6LoWPAN Linux implementation +============================ + +The IEEE 802.15.4 standard specifies an MTU of 128 bytes, yielding about 80 +octets of actual MAC payload once security is turned on, on a wireless link +with a link throughput of 250 kbps or less. The 6LoWPAN adaptation format +[RFC4944] was specified to carry IPv6 datagrams over such constrained links, +taking into account limited bandwidth, memory, or energy resources that are +expected in applications such as wireless Sensor Networks. [RFC4944] defines +a Mesh Addressing header to support sub-IP forwarding, a Fragmentation header +to support the IPv6 minimum MTU requirement [RFC2460], and stateless header +compression for IPv6 datagrams (LOWPAN_HC1 and LOWPAN_HC2) to reduce the +relatively large IPv6 and UDP headers down to (in the best case) several bytes. + +In Semptember 2011 the standard update was published - [RFC6282]. +It deprecates HC1 and HC2 compression and defines IPHC encoding format which is +used in this Linux implementation. + +All the code related to 6lowpan you may find in files: net/ieee802154/6lowpan.* + +To setup 6lowpan interface you need (busybox release > 1.17.0): +1. Add IEEE802.15.4 interface and initialize PANid; +2. Add 6lowpan interface by command like: + # ip link add link wpan0 name lowpan0 type lowpan +3. Set MAC (if needs): + # ip link set lowpan0 address de:ad:be:ef:ca:fe:ba:be +4. Bring up 'lowpan0' interface -- GitLab From e19df76a1113dc57cda696cd78d06f2834f6d6bb Mon Sep 17 00:00:00 2001 From: Sanjay Hortikar Date: Fri, 11 Nov 2011 16:11:21 +0000 Subject: [PATCH 0240/3089] net-forcedeth: Add internal loopback support for forcedeth NICs. Support enabling/disabling/querying internal loopback mode for forcedeth NICs using ethtool. Signed-off-by: Sanjay Hortikar Signed-off-by: Mahesh Bandewar Signed-off-by: David S. Miller --- drivers/net/ethernet/nvidia/forcedeth.c | 156 +++++++++++++++++++++++- 1 file changed, 155 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/nvidia/forcedeth.c b/drivers/net/ethernet/nvidia/forcedeth.c index d24c45bec508b..e8a5ae356407c 100644 --- a/drivers/net/ethernet/nvidia/forcedeth.c +++ b/drivers/net/ethernet/nvidia/forcedeth.c @@ -3003,6 +3003,73 @@ static void nv_update_pause(struct net_device *dev, u32 pause_flags) } } +static void nv_force_linkspeed(struct net_device *dev, int speed, int duplex) +{ + struct fe_priv *np = netdev_priv(dev); + u8 __iomem *base = get_hwbase(dev); + u32 phyreg, txreg; + int mii_status; + + np->linkspeed = NVREG_LINKSPEED_FORCE|speed; + np->duplex = duplex; + + /* see if gigabit phy */ + mii_status = mii_rw(dev, np->phyaddr, MII_BMSR, MII_READ); + if (mii_status & PHY_GIGABIT) { + np->gigabit = PHY_GIGABIT; + phyreg = readl(base + NvRegSlotTime); + phyreg &= ~(0x3FF00); + if ((np->linkspeed & 0xFFF) == NVREG_LINKSPEED_10) + phyreg |= NVREG_SLOTTIME_10_100_FULL; + else if ((np->linkspeed & 0xFFF) == NVREG_LINKSPEED_100) + phyreg |= NVREG_SLOTTIME_10_100_FULL; + else if ((np->linkspeed & 0xFFF) == NVREG_LINKSPEED_1000) + phyreg |= NVREG_SLOTTIME_1000_FULL; + writel(phyreg, base + NvRegSlotTime); + } + + phyreg = readl(base + NvRegPhyInterface); + phyreg &= ~(PHY_HALF|PHY_100|PHY_1000); + if (np->duplex == 0) + phyreg |= PHY_HALF; + if ((np->linkspeed & NVREG_LINKSPEED_MASK) == NVREG_LINKSPEED_100) + phyreg |= PHY_100; + else if ((np->linkspeed & NVREG_LINKSPEED_MASK) == + NVREG_LINKSPEED_1000) + phyreg |= PHY_1000; + writel(phyreg, base + NvRegPhyInterface); + + if (phyreg & PHY_RGMII) { + if ((np->linkspeed & NVREG_LINKSPEED_MASK) == + NVREG_LINKSPEED_1000) + txreg = NVREG_TX_DEFERRAL_RGMII_1000; + else + txreg = NVREG_TX_DEFERRAL_RGMII_10_100; + } else { + txreg = NVREG_TX_DEFERRAL_DEFAULT; + } + writel(txreg, base + NvRegTxDeferral); + + if (np->desc_ver == DESC_VER_1) { + txreg = NVREG_TX_WM_DESC1_DEFAULT; + } else { + if ((np->linkspeed & NVREG_LINKSPEED_MASK) == + NVREG_LINKSPEED_1000) + txreg = NVREG_TX_WM_DESC2_3_1000; + else + txreg = NVREG_TX_WM_DESC2_3_DEFAULT; + } + writel(txreg, base + NvRegTxWatermark); + + writel(NVREG_MISC1_FORCE | (np->duplex ? 0 : NVREG_MISC1_HD), + base + NvRegMisc1); + pci_push(base); + writel(np->linkspeed, base + NvRegLinkSpeed); + pci_push(base); + + return; +} + /** * nv_update_linkspeed: Setup the MAC according to the link partner * @dev: Network device to be configured @@ -3024,11 +3091,25 @@ static int nv_update_linkspeed(struct net_device *dev) int newls = np->linkspeed; int newdup = np->duplex; int mii_status; + u32 bmcr; int retval = 0; u32 control_1000, status_1000, phyreg, pause_flags, txreg; u32 txrxFlags = 0; u32 phy_exp; + /* If device loopback is enabled, set carrier on and enable max link + * speed. + */ + bmcr = mii_rw(dev, np->phyaddr, MII_BMCR, MII_READ); + if (bmcr & BMCR_LOOPBACK) { + if (netif_running(dev)) { + nv_force_linkspeed(dev, NVREG_LINKSPEED_1000, 1); + if (!netif_carrier_ok(dev)) + netif_carrier_on(dev); + } + return 1; + } + /* BMSR_LSTATUS is latched, read it twice: * we want the current value. */ @@ -4455,6 +4536,61 @@ static int nv_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam* return 0; } +static int nv_set_loopback(struct net_device *dev, u32 features) +{ + struct fe_priv *np = netdev_priv(dev); + unsigned long flags; + u32 miicontrol; + int err, retval = 0; + + spin_lock_irqsave(&np->lock, flags); + miicontrol = mii_rw(dev, np->phyaddr, MII_BMCR, MII_READ); + if (features & NETIF_F_LOOPBACK) { + if (miicontrol & BMCR_LOOPBACK) { + spin_unlock_irqrestore(&np->lock, flags); + netdev_info(dev, "Loopback already enabled\n"); + return 0; + } + nv_disable_irq(dev); + /* Turn on loopback mode */ + miicontrol |= BMCR_LOOPBACK | BMCR_FULLDPLX | BMCR_SPEED1000; + err = mii_rw(dev, np->phyaddr, MII_BMCR, miicontrol); + if (err) { + retval = PHY_ERROR; + spin_unlock_irqrestore(&np->lock, flags); + phy_init(dev); + } else { + if (netif_running(dev)) { + /* Force 1000 Mbps full-duplex */ + nv_force_linkspeed(dev, NVREG_LINKSPEED_1000, + 1); + /* Force link up */ + netif_carrier_on(dev); + } + spin_unlock_irqrestore(&np->lock, flags); + netdev_info(dev, + "Internal PHY loopback mode enabled.\n"); + } + } else { + if (!(miicontrol & BMCR_LOOPBACK)) { + spin_unlock_irqrestore(&np->lock, flags); + netdev_info(dev, "Loopback already disabled\n"); + return 0; + } + nv_disable_irq(dev); + /* Turn off loopback */ + spin_unlock_irqrestore(&np->lock, flags); + netdev_info(dev, "Internal PHY loopback mode disabled.\n"); + phy_init(dev); + } + msleep(500); + spin_lock_irqsave(&np->lock, flags); + nv_enable_irq(dev); + spin_unlock_irqrestore(&np->lock, flags); + + return retval; +} + static u32 nv_fix_features(struct net_device *dev, u32 features) { /* vlan is dependent on rx checksum offload */ @@ -4490,6 +4626,13 @@ static int nv_set_features(struct net_device *dev, u32 features) struct fe_priv *np = netdev_priv(dev); u8 __iomem *base = get_hwbase(dev); u32 changed = dev->features ^ features; + int retval; + + if ((changed & NETIF_F_LOOPBACK) && netif_running(dev)) { + retval = nv_set_loopback(dev, features); + if (retval != 0) + return retval; + } if (changed & NETIF_F_RXCSUM) { spin_lock_irq(&np->lock); @@ -5124,6 +5267,12 @@ static int nv_open(struct net_device *dev) spin_unlock_irq(&np->lock); + /* If the loopback feature was set while the device was down, make sure + * that it's set correctly now. + */ + if (dev->features & NETIF_F_LOOPBACK) + nv_set_loopback(dev, dev->features); + return 0; out_drain: nv_drain_rxtx(dev); @@ -5328,6 +5477,9 @@ static int __devinit nv_probe(struct pci_dev *pci_dev, const struct pci_device_i dev->features |= dev->hw_features; + /* Add loopback capability to the device. */ + dev->hw_features |= NETIF_F_LOOPBACK; + np->pause_flags = NV_PAUSEFRAME_RX_CAPABLE | NV_PAUSEFRAME_RX_REQ | NV_PAUSEFRAME_AUTONEG; if ((id->driver_data & DEV_HAS_PAUSEFRAME_TX_V1) || (id->driver_data & DEV_HAS_PAUSEFRAME_TX_V2) || @@ -5603,12 +5755,14 @@ static int __devinit nv_probe(struct pci_dev *pci_dev, const struct pci_device_i dev_info(&pci_dev->dev, "ifname %s, PHY OUI 0x%x @ %d, addr %pM\n", dev->name, np->phy_oui, np->phyaddr, dev->dev_addr); - dev_info(&pci_dev->dev, "%s%s%s%s%s%s%s%s%s%sdesc-v%u\n", + dev_info(&pci_dev->dev, "%s%s%s%s%s%s%s%s%s%s%sdesc-v%u\n", dev->features & NETIF_F_HIGHDMA ? "highdma " : "", dev->features & (NETIF_F_IP_CSUM | NETIF_F_SG) ? "csum " : "", dev->features & (NETIF_F_HW_VLAN_RX | NETIF_F_HW_VLAN_TX) ? "vlan " : "", + dev->features & (NETIF_F_LOOPBACK) ? + "loopback " : "", id->driver_data & DEV_HAS_POWER_CNTRL ? "pwrctl " : "", id->driver_data & DEV_HAS_MGMT_UNIT ? "mgmt " : "", id->driver_data & DEV_NEED_TIMERIRQ ? "timirq " : "", -- GitLab From 731abb9cb27aef6013ce60808a04e04a545f3f4e Mon Sep 17 00:00:00 2001 From: Josh Boyer Date: Thu, 10 Nov 2011 15:10:23 +0000 Subject: [PATCH 0241/3089] ip6_tunnel: copy parms.name after register_netdevice Commit 1c5cae815d removed an explicit call to dev_alloc_name in ip6_tnl_create because register_netdevice will now create a valid name. This works for the net_device itself. However the tunnel keeps a copy of the name in the parms structure for the ip6_tnl associated with the tunnel. parms.name is set by copying the net_device name in ip6_tnl_dev_init_gen. That function is called from ip6_tnl_dev_init in ip6_tnl_create, but it is done before register_netdevice is called so the name is set to a bogus value in the parms.name structure. This shows up if you do a simple tunnel add, followed by a tunnel show: [root@localhost ~]# ip -6 tunnel add remote fec0::100 local fec0::200 [root@localhost ~]# ip -6 tunnel show ip6tnl0: ipv6/ipv6 remote :: local :: encaplimit 0 hoplimit 0 tclass 0x00 flowlabel 0x00000 (flowinfo 0x00000000) ip6tnl%d: ipv6/ipv6 remote fec0::100 local fec0::200 encaplimit 4 hoplimit 64 tclass 0x00 flowlabel 0x00000 (flowinfo 0x00000000) [root@localhost ~]# Fix this by moving the strcpy out of ip6_tnl_dev_init_gen, and calling it after register_netdevice has successfully returned. Cc: stable@vger.kernel.org Signed-off-by: Josh Boyer Signed-off-by: David S. Miller --- net/ipv6/ip6_tunnel.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c index bdc15c9003d78..4e2e9ff67ef29 100644 --- a/net/ipv6/ip6_tunnel.c +++ b/net/ipv6/ip6_tunnel.c @@ -289,6 +289,8 @@ static struct ip6_tnl *ip6_tnl_create(struct net *net, struct ip6_tnl_parm *p) if ((err = register_netdevice(dev)) < 0) goto failed_free; + strcpy(t->parms.name, dev->name); + dev_hold(dev); ip6_tnl_link(ip6n, t); return t; @@ -1407,7 +1409,6 @@ ip6_tnl_dev_init_gen(struct net_device *dev) struct ip6_tnl *t = netdev_priv(dev); t->dev = dev; - strcpy(t->parms.name, dev->name); dev->tstats = alloc_percpu(struct pcpu_tstats); if (!dev->tstats) return -ENOMEM; @@ -1487,6 +1488,7 @@ static void __net_exit ip6_tnl_destroy_tunnels(struct ip6_tnl_net *ip6n) static int __net_init ip6_tnl_init_net(struct net *net) { struct ip6_tnl_net *ip6n = net_generic(net, ip6_tnl_net_id); + struct ip6_tnl *t = NULL; int err; ip6n->tnls[0] = ip6n->tnls_wc; @@ -1507,6 +1509,10 @@ static int __net_init ip6_tnl_init_net(struct net *net) err = register_netdev(ip6n->fb_tnl_dev); if (err < 0) goto err_register; + + t = netdev_priv(ip6n->fb_tnl_dev); + + strcpy(t->parms.name, ip6n->fb_tnl_dev->name); return 0; err_register: -- GitLab From 952c5ca14eab113072716ae075c6fd184f011a89 Mon Sep 17 00:00:00 2001 From: Andy Fleming Date: Fri, 11 Nov 2011 05:10:39 +0000 Subject: [PATCH 0242/3089] fsl_pq_mdio: Clean up tbi address configuration The code for setting the address of the internal TBI PHY was convoluted enough without a maze of ifdefs. Clean it up a bit so we allow the logic to fail down to -ENODEV at the end of the if/else ladder, rather than using ifdefs to repeat the same failure code over and over. Also, remove the support for the auto-configuration. I'm not aware of anyone using it, and it ends up using the bus mutex before it's been initialized. Signed-off-by: Andy Fleming Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/fsl_pq_mdio.c | 53 +++----------------- 1 file changed, 8 insertions(+), 45 deletions(-) diff --git a/drivers/net/ethernet/freescale/fsl_pq_mdio.c b/drivers/net/ethernet/freescale/fsl_pq_mdio.c index 52f4e8ad48e77..4d9f84b8ab977 100644 --- a/drivers/net/ethernet/freescale/fsl_pq_mdio.c +++ b/drivers/net/ethernet/freescale/fsl_pq_mdio.c @@ -183,28 +183,10 @@ void fsl_pq_mdio_bus_name(char *name, struct device_node *np) } EXPORT_SYMBOL_GPL(fsl_pq_mdio_bus_name); -/* Scan the bus in reverse, looking for an empty spot */ -static int fsl_pq_mdio_find_free(struct mii_bus *new_bus) -{ - int i; - - for (i = PHY_MAX_ADDR; i > 0; i--) { - u32 phy_id; - - if (get_phy_id(new_bus, i, &phy_id)) - return -1; - - if (phy_id == 0xffffffff) - break; - } - - return i; -} - -#if defined(CONFIG_GIANFAR) || defined(CONFIG_GIANFAR_MODULE) static u32 __iomem *get_gfar_tbipa(struct fsl_pq_mdio __iomem *regs, struct device_node *np) { +#if defined(CONFIG_GIANFAR) || defined(CONFIG_GIANFAR_MODULE) struct gfar __iomem *enet_regs; /* @@ -220,15 +202,15 @@ static u32 __iomem *get_gfar_tbipa(struct fsl_pq_mdio __iomem *regs, struct devi } else if (of_device_is_compatible(np, "fsl,etsec2-mdio") || of_device_is_compatible(np, "fsl,etsec2-tbi")) { return of_iomap(np, 1); - } else - return NULL; -} + } #endif + return NULL; +} -#if defined(CONFIG_UCC_GETH) || defined(CONFIG_UCC_GETH_MODULE) static int get_ucc_id_for_range(u64 start, u64 end, u32 *ucc_id) { +#if defined(CONFIG_UCC_GETH) || defined(CONFIG_UCC_GETH_MODULE) struct device_node *np = NULL; int err = 0; @@ -261,9 +243,10 @@ static int get_ucc_id_for_range(u64 start, u64 end, u32 *ucc_id) return err; else return -EINVAL; -} +#else + return -ENODEV; #endif - +} static int fsl_pq_mdio_probe(struct platform_device *ofdev) { @@ -339,19 +322,13 @@ static int fsl_pq_mdio_probe(struct platform_device *ofdev) of_device_is_compatible(np, "fsl,etsec2-mdio") || of_device_is_compatible(np, "fsl,etsec2-tbi") || of_device_is_compatible(np, "gianfar")) { -#if defined(CONFIG_GIANFAR) || defined(CONFIG_GIANFAR_MODULE) tbipa = get_gfar_tbipa(regs, np); if (!tbipa) { err = -EINVAL; goto err_free_irqs; } -#else - err = -ENODEV; - goto err_free_irqs; -#endif } else if (of_device_is_compatible(np, "fsl,ucc-mdio") || of_device_is_compatible(np, "ucc_geth_phy")) { -#if defined(CONFIG_UCC_GETH) || defined(CONFIG_UCC_GETH_MODULE) u32 id; static u32 mii_mng_master; @@ -364,10 +341,6 @@ static int fsl_pq_mdio_probe(struct platform_device *ofdev) mii_mng_master = id; ucc_set_qe_mux_mii_mng(id - 1); } -#else - err = -ENODEV; - goto err_free_irqs; -#endif } else { err = -ENODEV; goto err_free_irqs; @@ -386,16 +359,6 @@ static int fsl_pq_mdio_probe(struct platform_device *ofdev) } if (tbiaddr == -1) { - out_be32(tbipa, 0); - - tbiaddr = fsl_pq_mdio_find_free(new_bus); - } - - /* - * We define TBIPA at 0 to be illegal, opting to fail for boards that - * have PHYs at 1-31, rather than change tbipa and rescan. - */ - if (tbiaddr == 0) { err = -EBUSY; goto err_free_irqs; -- GitLab From 22f4cb4bd06db14912f69039917422fe0c5b11fb Mon Sep 17 00:00:00 2001 From: Haojian Zhuang Date: Wed, 9 Nov 2011 10:47:21 +0800 Subject: [PATCH 0243/3089] ARM: mmp: fix build error on gpio MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Parameters of GPIO_REG() should be assigned as volatile. arch/arm/plat-pxa/include/plat/gpio.h: In function ‘gpio_get_value’: arch/arm/plat-pxa/include/plat/gpio.h:12:21: error: invalid operands to binary & (have ‘void *’ and ‘int’) arch/arm/plat-pxa/include/plat/gpio.h: In function ‘gpio_set_value’: arch/arm/plat-pxa/include/plat/gpio.h:21:4: error: lvalue required as left operand of assignment arch/arm/plat-pxa/include/plat/gpio.h:23:4: error: lvalue required as left operand of assignment Signed-off-by: Haojian Zhuang Signed-off-by: Eric Miao --- arch/arm/mach-mmp/include/mach/gpio-pxa.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/mach-mmp/include/mach/gpio-pxa.h b/arch/arm/mach-mmp/include/mach/gpio-pxa.h index d14eeaf163226..99b4ce1b6562c 100644 --- a/arch/arm/mach-mmp/include/mach/gpio-pxa.h +++ b/arch/arm/mach-mmp/include/mach/gpio-pxa.h @@ -7,7 +7,7 @@ #define GPIO_REGS_VIRT (APB_VIRT_BASE + 0x19000) #define BANK_OFF(n) (((n) < 3) ? (n) << 2 : 0x100 + (((n) - 3) << 2)) -#define GPIO_REG(x) (GPIO_REGS_VIRT + (x)) +#define GPIO_REG(x) (*(volatile u32 *)(GPIO_REGS_VIRT + (x))) #define NR_BUILTIN_GPIO IRQ_GPIO_NUM -- GitLab From 23020ab35364f2c91133b099c2b1f7458e29aa96 Mon Sep 17 00:00:00 2001 From: Rick Jones Date: Wed, 9 Nov 2011 09:58:07 +0000 Subject: [PATCH 0244/3089] Sweep additional floors of strcpy in .get_drvinfo routines Perform another round of floor sweeping, converting the .get_drvinfo routines of additional drivers from strcpy to strlcpy along with some conversion of sprintf to snprintf. Signed-off-by: Rick Jones Signed-off-by: David S. Miller --- drivers/net/ethernet/amd/amd8111e.c | 9 +++++---- drivers/net/ethernet/amd/pcnet32.c | 10 ++++++---- drivers/net/ethernet/chelsio/cxgb/cxgb2.c | 9 +++++---- drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c | 9 +++++---- drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 9 +++++---- drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c | 7 ++++--- drivers/net/ethernet/intel/e100.c | 9 +++++---- drivers/net/ethernet/jme.c | 6 +++--- drivers/net/ethernet/micrel/ksz884x.c | 7 ++++--- .../net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c | 9 +++++---- drivers/net/ethernet/sis/sis190.c | 7 ++++--- drivers/net/ethernet/sis/sis900.c | 7 ++++--- drivers/net/ethernet/sun/niu.c | 9 +++++---- drivers/net/ethernet/sun/sungem.c | 6 +++--- drivers/net/ethernet/sun/sunhme.c | 9 +++++---- drivers/net/ethernet/via/via-rhine.c | 6 +++--- drivers/net/ethernet/via/via-velocity.c | 6 +++--- 17 files changed, 74 insertions(+), 60 deletions(-) diff --git a/drivers/net/ethernet/amd/amd8111e.c b/drivers/net/ethernet/amd/amd8111e.c index a9745f4ddbfe1..a388118ee8e6b 100644 --- a/drivers/net/ethernet/amd/amd8111e.c +++ b/drivers/net/ethernet/amd/amd8111e.c @@ -1412,10 +1412,11 @@ static void amd8111e_get_drvinfo(struct net_device* dev, struct ethtool_drvinfo { struct amd8111e_priv *lp = netdev_priv(dev); struct pci_dev *pci_dev = lp->pci_dev; - strcpy (info->driver, MODULE_NAME); - strcpy (info->version, MODULE_VERS); - sprintf(info->fw_version,"%u",chip_version); - strcpy (info->bus_info, pci_name(pci_dev)); + strlcpy(info->driver, MODULE_NAME, sizeof(info->driver)); + strlcpy(info->version, MODULE_VERS, sizeof(info->version)); + snprintf(info->fw_version, sizeof(info->fw_version), + "%u", chip_version); + strlcpy(info->bus_info, pci_name(pci_dev), sizeof(info->bus_info)); } static int amd8111e_get_regs_len(struct net_device *dev) diff --git a/drivers/net/ethernet/amd/pcnet32.c b/drivers/net/ethernet/amd/pcnet32.c index f92bc6e348283..20e6dab0186c7 100644 --- a/drivers/net/ethernet/amd/pcnet32.c +++ b/drivers/net/ethernet/amd/pcnet32.c @@ -711,12 +711,14 @@ static void pcnet32_get_drvinfo(struct net_device *dev, { struct pcnet32_private *lp = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); if (lp->pci_dev) - strcpy(info->bus_info, pci_name(lp->pci_dev)); + strlcpy(info->bus_info, pci_name(lp->pci_dev), + sizeof(info->bus_info)); else - sprintf(info->bus_info, "VLB 0x%lx", dev->base_addr); + snprintf(info->bus_info, sizeof(info->bus_info), + "VLB 0x%lx", dev->base_addr); } static u32 pcnet32_get_link(struct net_device *dev) diff --git a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c index ca26d97171bdd..26d0fd2d9c9d0 100644 --- a/drivers/net/ethernet/chelsio/cxgb/cxgb2.c +++ b/drivers/net/ethernet/chelsio/cxgb/cxgb2.c @@ -434,10 +434,11 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct adapter *adapter = dev->ml_priv; - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->fw_version, "N/A"); - strcpy(info->bus_info, pci_name(adapter->pdev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->fw_version, "N/A", sizeof(info->fw_version)); + strlcpy(info->bus_info, pci_name(adapter->pdev), + sizeof(info->bus_info)); } static int get_sset_count(struct net_device *dev, int sset) diff --git a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c index 4d15c8f99c3b8..053560da63477 100644 --- a/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c +++ b/drivers/net/ethernet/chelsio/cxgb3/cxgb3_main.c @@ -1576,11 +1576,12 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) t3_get_tp_version(adapter, &tp_vers); spin_unlock(&adapter->stats_lock); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(adapter->pdev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(adapter->pdev), + sizeof(info->bus_info)); if (!fw_vers) - strcpy(info->fw_version, "N/A"); + strlcpy(info->fw_version, "N/A", sizeof(info->fw_version)); else { snprintf(info->fw_version, sizeof(info->fw_version), "%s %u.%u.%u TP %u.%u.%u", diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c index 4c8f42afa3c6d..48ffe11d9aa9e 100644 --- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c @@ -1002,12 +1002,13 @@ static void get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct adapter *adapter = netdev2adap(dev); - strcpy(info->driver, KBUILD_MODNAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(adapter->pdev)); + strlcpy(info->driver, KBUILD_MODNAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(adapter->pdev), + sizeof(info->bus_info)); if (!adapter->params.fw_vers) - strcpy(info->fw_version, "N/A"); + strlcpy(info->fw_version, "N/A", sizeof(info->fw_version)); else snprintf(info->fw_version, sizeof(info->fw_version), "%u.%u.%u.%u, TP %u.%u.%u.%u", diff --git a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c index da9072bfca8b8..ee81d8e798ead 100644 --- a/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c +++ b/drivers/net/ethernet/chelsio/cxgb4vf/cxgb4vf_main.c @@ -1203,9 +1203,10 @@ static void cxgb4vf_get_drvinfo(struct net_device *dev, { struct adapter *adapter = netdev2adap(dev); - strcpy(drvinfo->driver, KBUILD_MODNAME); - strcpy(drvinfo->version, DRV_VERSION); - strcpy(drvinfo->bus_info, pci_name(to_pci_dev(dev->dev.parent))); + strlcpy(drvinfo->driver, KBUILD_MODNAME, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version)); + strlcpy(drvinfo->bus_info, pci_name(to_pci_dev(dev->dev.parent)), + sizeof(drvinfo->bus_info)); snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), "%u.%u.%u.%u, TP %u.%u.%u.%u", FW_HDR_FW_VER_MAJOR_GET(adapter->params.dev.fwrev), diff --git a/drivers/net/ethernet/intel/e100.c b/drivers/net/ethernet/intel/e100.c index 5a2fdf7a00c8b..46003278ffa0b 100644 --- a/drivers/net/ethernet/intel/e100.c +++ b/drivers/net/ethernet/intel/e100.c @@ -2376,10 +2376,11 @@ static void e100_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *info) { struct nic *nic = netdev_priv(netdev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->fw_version, "N/A"); - strcpy(info->bus_info, pci_name(nic->pdev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->fw_version, "N/A", sizeof(info->fw_version)); + strlcpy(info->bus_info, pci_name(nic->pdev), + sizeof(info->bus_info)); } #define E100_PHY_REGS 0x1C diff --git a/drivers/net/ethernet/jme.c b/drivers/net/ethernet/jme.c index 7becff1f387d7..7d88c7c28a7cd 100644 --- a/drivers/net/ethernet/jme.c +++ b/drivers/net/ethernet/jme.c @@ -2292,9 +2292,9 @@ jme_get_drvinfo(struct net_device *netdev, { struct jme_adapter *jme = netdev_priv(netdev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(jme->pdev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(jme->pdev), sizeof(info->bus_info)); } static int diff --git a/drivers/net/ethernet/micrel/ksz884x.c b/drivers/net/ethernet/micrel/ksz884x.c index 7ece990381c8f..3b67fe65404a2 100644 --- a/drivers/net/ethernet/micrel/ksz884x.c +++ b/drivers/net/ethernet/micrel/ksz884x.c @@ -6093,9 +6093,10 @@ static void netdev_get_drvinfo(struct net_device *dev, struct dev_priv *priv = netdev_priv(dev); struct dev_info *hw_priv = priv->adapter; - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(hw_priv->pdev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(hw_priv->pdev), + sizeof(info->bus_info)); } /** diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c index 8c8027176bef9..0063194adb9e0 100644 --- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c +++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_ethtool.c @@ -161,10 +161,11 @@ static void pch_gbe_get_drvinfo(struct net_device *netdev, { struct pch_gbe_adapter *adapter = netdev_priv(netdev); - strcpy(drvinfo->driver, KBUILD_MODNAME); - strcpy(drvinfo->version, pch_driver_version); - strcpy(drvinfo->fw_version, "N/A"); - strcpy(drvinfo->bus_info, pci_name(adapter->pdev)); + strlcpy(drvinfo->driver, KBUILD_MODNAME, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, pch_driver_version, sizeof(drvinfo->version)); + strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version)); + strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); drvinfo->regdump_len = pch_gbe_get_regs_len(netdev); } diff --git a/drivers/net/ethernet/sis/sis190.c b/drivers/net/ethernet/sis/sis190.c index 1b4658c99391a..220e98219a3be 100644 --- a/drivers/net/ethernet/sis/sis190.c +++ b/drivers/net/ethernet/sis/sis190.c @@ -1760,9 +1760,10 @@ static void sis190_get_drvinfo(struct net_device *dev, { struct sis190_private *tp = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(tp->pci_dev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(tp->pci_dev), + sizeof(info->bus_info)); } static int sis190_get_regs_len(struct net_device *dev) diff --git a/drivers/net/ethernet/sis/sis900.c b/drivers/net/ethernet/sis/sis900.c index a184abc5ef118..c8efc708c792a 100644 --- a/drivers/net/ethernet/sis/sis900.c +++ b/drivers/net/ethernet/sis/sis900.c @@ -1991,9 +1991,10 @@ static void sis900_get_drvinfo(struct net_device *net_dev, { struct sis900_private *sis_priv = netdev_priv(net_dev); - strcpy (info->driver, SIS900_MODULE_NAME); - strcpy (info->version, SIS900_DRV_VERSION); - strcpy (info->bus_info, pci_name(sis_priv->pci_dev)); + strlcpy(info->driver, SIS900_MODULE_NAME, sizeof(info->driver)); + strlcpy(info->version, SIS900_DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(sis_priv->pci_dev), + sizeof(info->bus_info)); } static u32 sis900_get_msglevel(struct net_device *net_dev) diff --git a/drivers/net/ethernet/sun/niu.c b/drivers/net/ethernet/sun/niu.c index 73c708107a379..3ebeb9d400fbc 100644 --- a/drivers/net/ethernet/sun/niu.c +++ b/drivers/net/ethernet/sun/niu.c @@ -6823,12 +6823,13 @@ static void niu_get_drvinfo(struct net_device *dev, struct niu *np = netdev_priv(dev); struct niu_vpd *vpd = &np->vpd; - strcpy(info->driver, DRV_MODULE_NAME); - strcpy(info->version, DRV_MODULE_VERSION); - sprintf(info->fw_version, "%d.%d", + strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); + snprintf(info->fw_version, sizeof(info->fw_version), "%d.%d", vpd->fcode_major, vpd->fcode_minor); if (np->parent->plat_type != PLAT_TYPE_NIU) - strcpy(info->bus_info, pci_name(np->pdev)); + strlcpy(info->bus_info, pci_name(np->pdev), + sizeof(info->bus_info)); } static int niu_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) diff --git a/drivers/net/ethernet/sun/sungem.c b/drivers/net/ethernet/sun/sungem.c index ceab215bb4a31..31441a870b0b8 100644 --- a/drivers/net/ethernet/sun/sungem.c +++ b/drivers/net/ethernet/sun/sungem.c @@ -2517,9 +2517,9 @@ static void gem_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info { struct gem *gp = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(gp->pdev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(gp->pdev), sizeof(info->bus_info)); } static int gem_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) diff --git a/drivers/net/ethernet/sun/sunhme.c b/drivers/net/ethernet/sun/sunhme.c index cf14ab9db5768..eebd52f103657 100644 --- a/drivers/net/ethernet/sun/sunhme.c +++ b/drivers/net/ethernet/sun/sunhme.c @@ -2457,11 +2457,11 @@ static void hme_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info { struct happy_meal *hp = netdev_priv(dev); - strcpy(info->driver, "sunhme"); - strcpy(info->version, "2.02"); + strlcpy(info->driver, "sunhme", sizeof(info->driver)); + strlcpy(info->version, "2.02", sizeof(info->version)); if (hp->happy_flags & HFLAG_PCI) { struct pci_dev *pdev = hp->happy_dev; - strcpy(info->bus_info, pci_name(pdev)); + strlcpy(info->bus_info, pci_name(pdev), sizeof(info->bus_info)); } #ifdef CONFIG_SBUS else { @@ -2469,7 +2469,8 @@ static void hme_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info struct platform_device *op = hp->happy_dev; regs = of_get_property(op->dev.of_node, "regs", NULL); if (regs) - sprintf(info->bus_info, "SBUS:%d", + snprintf(info->bus_info, sizeof(info->bus_info), + "SBUS:%d", regs->which_io); } #endif diff --git a/drivers/net/ethernet/via/via-rhine.c b/drivers/net/ethernet/via/via-rhine.c index f34dd99fe5791..5587ecdf32e32 100644 --- a/drivers/net/ethernet/via/via-rhine.c +++ b/drivers/net/ethernet/via/via-rhine.c @@ -2009,9 +2009,9 @@ static void netdev_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *i { struct rhine_private *rp = netdev_priv(dev); - strcpy(info->driver, DRV_NAME); - strcpy(info->version, DRV_VERSION); - strcpy(info->bus_info, pci_name(rp->pdev)); + strlcpy(info->driver, DRV_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(rp->pdev), sizeof(info->bus_info)); } static int netdev_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) diff --git a/drivers/net/ethernet/via/via-velocity.c b/drivers/net/ethernet/via/via-velocity.c index 4535d7cc848ed..59bb5fd56afe6 100644 --- a/drivers/net/ethernet/via/via-velocity.c +++ b/drivers/net/ethernet/via/via-velocity.c @@ -3270,9 +3270,9 @@ static int velocity_set_settings(struct net_device *dev, static void velocity_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct velocity_info *vptr = netdev_priv(dev); - strcpy(info->driver, VELOCITY_NAME); - strcpy(info->version, VELOCITY_VERSION); - strcpy(info->bus_info, pci_name(vptr->pdev)); + strlcpy(info->driver, VELOCITY_NAME, sizeof(info->driver)); + strlcpy(info->version, VELOCITY_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(vptr->pdev), sizeof(info->bus_info)); } static void velocity_ethtool_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) -- GitLab From ef5e0d8237287db3a12d84f08fb2483d7a30a943 Mon Sep 17 00:00:00 2001 From: Andrey Vagin Date: Thu, 10 Nov 2011 05:48:03 +0000 Subject: [PATCH 0245/3089] bridge: Fix potential deadlock on br->multicast_lock multicast_lock is taken in softirq context, so we should use spin_lock_bh() in userspace. call-chain in softirq context: run_timer_softirq() br_multicast_query_expired() call-chain in userspace: sysfs_write_file() store_multicast_snooping() br_multicast_toggle() Signed-off-by: Andrew Vagin Signed-off-by: David S. Miller --- net/bridge/br_multicast.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/bridge/br_multicast.c b/net/bridge/br_multicast.c index 995cbe0ac0b2b..2eefe275b3388 100644 --- a/net/bridge/br_multicast.c +++ b/net/bridge/br_multicast.c @@ -1770,7 +1770,7 @@ int br_multicast_toggle(struct net_bridge *br, unsigned long val) int err = 0; struct net_bridge_mdb_htable *mdb; - spin_lock(&br->multicast_lock); + spin_lock_bh(&br->multicast_lock); if (br->multicast_disabled == !val) goto unlock; @@ -1806,7 +1806,7 @@ rollback: } unlock: - spin_unlock(&br->multicast_lock); + spin_unlock_bh(&br->multicast_lock); return err; } -- GitLab From 3ac3546e5f17248d961ef0f4a27e75564bf71578 Mon Sep 17 00:00:00 2001 From: Robert Marklund Date: Tue, 25 Oct 2011 22:05:43 +0000 Subject: [PATCH 0246/3089] net/smsc911x: Always wait for the chip to be ready Wait for the chip to be ready before any access to it. On the Snowball platform we need to enable an external regulator before the chip comes online, and then it happens that the device is not yet ready at probe time, so let's wait for it. Signed-off-by: Robert Marklund Signed-off-by: Linus Walleij Signed-off-by: David S. Miller --- drivers/net/ethernet/smsc/smsc911x.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/drivers/net/ethernet/smsc/smsc911x.c b/drivers/net/ethernet/smsc/smsc911x.c index d2be42aafbef2..8843071fe987b 100644 --- a/drivers/net/ethernet/smsc/smsc911x.c +++ b/drivers/net/ethernet/smsc/smsc911x.c @@ -1937,6 +1937,7 @@ static int __devinit smsc911x_init(struct net_device *dev) { struct smsc911x_data *pdata = netdev_priv(dev); unsigned int byte_test; + unsigned int to = 100; SMSC_TRACE(pdata, probe, "Driver Parameters:"); SMSC_TRACE(pdata, probe, "LAN base: 0x%08lX", @@ -1952,6 +1953,17 @@ static int __devinit smsc911x_init(struct net_device *dev) return -ENODEV; } + /* + * poll the READY bit in PMT_CTRL. Any other access to the device is + * forbidden while this bit isn't set. Try for 100ms + */ + while (!(smsc911x_reg_read(pdata, PMT_CTRL) & PMT_CTRL_READY_) && --to) + udelay(1000); + if (to == 0) { + pr_err("Device not READY in 100ms aborting\n"); + return -ENODEV; + } + /* Check byte ordering */ byte_test = smsc911x_reg_read(pdata, BYTE_TEST); SMSC_TRACE(pdata, probe, "BYTE_TEST: 0x%08X", byte_test); -- GitLab From 292d1398983f3514a0eab13b7606df7f4730b498 Mon Sep 17 00:00:00 2001 From: stephen hemminger Date: Wed, 9 Nov 2011 18:30:08 +0000 Subject: [PATCH 0247/3089] bridge: add NTF_USE support More changes to the recent code to support control of forwarding database via netlink. * Support NTF_USE like neighbour table * Validate state bits from application * Only send notifications (and change bits) if new entry is different. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/bridge/br_fdb.c | 38 +++++++++++++++++++++++++++----------- 1 file changed, 27 insertions(+), 11 deletions(-) diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c index c8e7861b88b02..973813e344283 100644 --- a/net/bridge/br_fdb.c +++ b/net/bridge/br_fdb.c @@ -556,7 +556,7 @@ skip: return skb->len; } -/* Create new static fdb entry */ +/* Update (create or replace) forwarding database entry */ static int fdb_add_entry(struct net_bridge_port *source, const __u8 *addr, __u16 state, __u16 flags) { @@ -575,16 +575,21 @@ static int fdb_add_entry(struct net_bridge_port *source, const __u8 *addr, } else { if (flags & NLM_F_EXCL) return -EEXIST; + } + + if (fdb_to_nud(fdb) != state) { + if (state & NUD_PERMANENT) + fdb->is_local = fdb->is_static = 1; + else if (state & NUD_NOARP) { + fdb->is_local = 0; + fdb->is_static = 1; + } else + fdb->is_local = fdb->is_static = 0; - if (flags & NLM_F_REPLACE) - fdb->updated = fdb->used = jiffies; - fdb->is_local = fdb->is_static = 0; + fdb->updated = fdb->used = jiffies; + fdb_notify(fdb, RTM_NEWNEIGH); } - if (state & NUD_PERMANENT) - fdb->is_local = fdb->is_static = 1; - else if (state & NUD_NOARP) - fdb->is_static = 1; return 0; } @@ -627,6 +632,11 @@ int br_fdb_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) return -EINVAL; } + if (!(ndm->ndm_state & (NUD_PERMANENT|NUD_NOARP|NUD_REACHABLE))) { + pr_info("bridge: RTM_NEWNEIGH with invalid state %#x\n", ndm->ndm_state); + return -EINVAL; + } + p = br_port_get_rtnl(dev); if (p == NULL) { pr_info("bridge: RTM_NEWNEIGH %s not a bridge port\n", @@ -634,9 +644,15 @@ int br_fdb_add(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) return -EINVAL; } - spin_lock_bh(&p->br->hash_lock); - err = fdb_add_entry(p, addr, ndm->ndm_state, nlh->nlmsg_flags); - spin_unlock_bh(&p->br->hash_lock); + if (ndm->ndm_flags & NTF_USE) { + rcu_read_lock(); + br_fdb_update(p->br, p, addr); + rcu_read_unlock(); + } else { + spin_lock_bh(&p->br->hash_lock); + err = fdb_add_entry(p, addr, ndm->ndm_state, nlh->nlmsg_flags); + spin_unlock_bh(&p->br->hash_lock); + } return err; } -- GitLab From 8b5c171bb3dc0686b2647a84e990199c5faa9ef8 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 9 Nov 2011 12:07:14 +0000 Subject: [PATCH 0248/3089] neigh: new unresolved queue limits MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Le mercredi 09 novembre 2011 à 16:21 -0500, David Miller a écrit : > From: David Miller > Date: Wed, 09 Nov 2011 16:16:44 -0500 (EST) > > > From: Eric Dumazet > > Date: Wed, 09 Nov 2011 12:14:09 +0100 > > > >> unres_qlen is the number of frames we are able to queue per unresolved > >> neighbour. Its default value (3) was never changed and is responsible > >> for strange drops, especially if IP fragments are used, or multiple > >> sessions start in parallel. Even a single tcp flow can hit this limit. > > ... > > > > Ok, I've applied this, let's see what happens :-) > > Early answer, build fails. > > Please test build this patch with DECNET enabled and resubmit. The > decnet neigh layer still refers to the removed ->queue_len member. > > Thanks. Ouch, this was fixed on one machine yesterday, but not the other one I used this morning, sorry. [PATCH V5 net-next] neigh: new unresolved queue limits unres_qlen is the number of frames we are able to queue per unresolved neighbour. Its default value (3) was never changed and is responsible for strange drops, especially if IP fragments are used, or multiple sessions start in parallel. Even a single tcp flow can hit this limit. $ arp -d 192.168.20.108 ; ping -c 2 -s 8000 192.168.20.108 PING 192.168.20.108 (192.168.20.108) 8000(8028) bytes of data. 8008 bytes from 192.168.20.108: icmp_seq=2 ttl=64 time=0.322 ms Signed-off-by: David S. Miller --- Documentation/networking/ip-sysctl.txt | 10 ++ include/linux/neighbour.h | 1 + include/net/neighbour.h | 3 +- net/atm/clip.c | 2 +- net/core/neighbour.c | 162 +++++++++++++++++-------- net/decnet/dn_neigh.c | 2 +- net/ipv4/arp.c | 2 +- net/ipv6/ndisc.c | 2 +- 8 files changed, 128 insertions(+), 56 deletions(-) diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index f049a1ca186fb..b8867061fce40 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -31,6 +31,16 @@ neigh/default/gc_thresh3 - INTEGER when using large numbers of interfaces and when communicating with large numbers of directly-connected peers. +neigh/default/unres_qlen_bytes - INTEGER + The maximum number of bytes which may be used by packets + queued for each unresolved address by other network layers. + (added in linux 3.3) + +neigh/default/unres_qlen - INTEGER + The maximum number of packets which may be queued for each + unresolved address by other network layers. + (deprecated in linux 3.3) : use unres_qlen_bytes instead. + mtu_expires - INTEGER Time, in seconds, that cached PMTU information is kept. diff --git a/include/linux/neighbour.h b/include/linux/neighbour.h index a7003b7a695d2..b188f68a08c90 100644 --- a/include/linux/neighbour.h +++ b/include/linux/neighbour.h @@ -116,6 +116,7 @@ enum { NDTPA_PROXY_DELAY, /* u64, msecs */ NDTPA_PROXY_QLEN, /* u32 */ NDTPA_LOCKTIME, /* u64, msecs */ + NDTPA_QUEUE_LENBYTES, /* u32 */ __NDTPA_MAX }; #define NDTPA_MAX (__NDTPA_MAX - 1) diff --git a/include/net/neighbour.h b/include/net/neighbour.h index 2720884287c3e..7ae5acff96e9b 100644 --- a/include/net/neighbour.h +++ b/include/net/neighbour.h @@ -59,7 +59,7 @@ struct neigh_parms { int reachable_time; int delay_probe_time; - int queue_len; + int queue_len_bytes; int ucast_probes; int app_probes; int mcast_probes; @@ -99,6 +99,7 @@ struct neighbour { rwlock_t lock; atomic_t refcnt; struct sk_buff_head arp_queue; + unsigned int arp_queue_len_bytes; struct timer_list timer; unsigned long used; atomic_t probes; diff --git a/net/atm/clip.c b/net/atm/clip.c index 852394072fa15..32c41b8a803e4 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -329,7 +329,7 @@ static struct neigh_table clip_tbl = { .gc_staletime = 60 * HZ, .reachable_time = 30 * HZ, .delay_probe_time = 5 * HZ, - .queue_len = 3, + .queue_len_bytes = 64 * 1024, .ucast_probes = 3, .mcast_probes = 3, .anycast_delay = 1 * HZ, diff --git a/net/core/neighbour.c b/net/core/neighbour.c index 039d51e6c284e..2684794458ca0 100644 --- a/net/core/neighbour.c +++ b/net/core/neighbour.c @@ -238,6 +238,7 @@ static void neigh_flush_dev(struct neigh_table *tbl, struct net_device *dev) it to safe state. */ skb_queue_purge(&n->arp_queue); + n->arp_queue_len_bytes = 0; n->output = neigh_blackhole; if (n->nud_state & NUD_VALID) n->nud_state = NUD_NOARP; @@ -702,6 +703,7 @@ void neigh_destroy(struct neighbour *neigh) printk(KERN_WARNING "Impossible event.\n"); skb_queue_purge(&neigh->arp_queue); + neigh->arp_queue_len_bytes = 0; dev_put(neigh->dev); neigh_parms_put(neigh->parms); @@ -842,6 +844,7 @@ static void neigh_invalidate(struct neighbour *neigh) write_lock(&neigh->lock); } skb_queue_purge(&neigh->arp_queue); + neigh->arp_queue_len_bytes = 0; } static void neigh_probe(struct neighbour *neigh) @@ -980,15 +983,20 @@ int __neigh_event_send(struct neighbour *neigh, struct sk_buff *skb) if (neigh->nud_state == NUD_INCOMPLETE) { if (skb) { - if (skb_queue_len(&neigh->arp_queue) >= - neigh->parms->queue_len) { + while (neigh->arp_queue_len_bytes + skb->truesize > + neigh->parms->queue_len_bytes) { struct sk_buff *buff; + buff = __skb_dequeue(&neigh->arp_queue); + if (!buff) + break; + neigh->arp_queue_len_bytes -= buff->truesize; kfree_skb(buff); NEIGH_CACHE_STAT_INC(neigh->tbl, unres_discards); } skb_dst_force(skb); __skb_queue_tail(&neigh->arp_queue, skb); + neigh->arp_queue_len_bytes += skb->truesize; } rc = 1; } @@ -1175,6 +1183,7 @@ int neigh_update(struct neighbour *neigh, const u8 *lladdr, u8 new, write_lock_bh(&neigh->lock); } skb_queue_purge(&neigh->arp_queue); + neigh->arp_queue_len_bytes = 0; } out: if (update_isrouter) { @@ -1747,7 +1756,11 @@ static int neightbl_fill_parms(struct sk_buff *skb, struct neigh_parms *parms) NLA_PUT_U32(skb, NDTPA_IFINDEX, parms->dev->ifindex); NLA_PUT_U32(skb, NDTPA_REFCNT, atomic_read(&parms->refcnt)); - NLA_PUT_U32(skb, NDTPA_QUEUE_LEN, parms->queue_len); + NLA_PUT_U32(skb, NDTPA_QUEUE_LENBYTES, parms->queue_len_bytes); + /* approximative value for deprecated QUEUE_LEN (in packets) */ + NLA_PUT_U32(skb, NDTPA_QUEUE_LEN, + DIV_ROUND_UP(parms->queue_len_bytes, + SKB_TRUESIZE(ETH_FRAME_LEN))); NLA_PUT_U32(skb, NDTPA_PROXY_QLEN, parms->proxy_qlen); NLA_PUT_U32(skb, NDTPA_APP_PROBES, parms->app_probes); NLA_PUT_U32(skb, NDTPA_UCAST_PROBES, parms->ucast_probes); @@ -1974,7 +1987,11 @@ static int neightbl_set(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) switch (i) { case NDTPA_QUEUE_LEN: - p->queue_len = nla_get_u32(tbp[i]); + p->queue_len_bytes = nla_get_u32(tbp[i]) * + SKB_TRUESIZE(ETH_FRAME_LEN); + break; + case NDTPA_QUEUE_LENBYTES: + p->queue_len_bytes = nla_get_u32(tbp[i]); break; case NDTPA_PROXY_QLEN: p->proxy_qlen = nla_get_u32(tbp[i]); @@ -2635,117 +2652,158 @@ EXPORT_SYMBOL(neigh_app_ns); #ifdef CONFIG_SYSCTL -#define NEIGH_VARS_MAX 19 +static int proc_unres_qlen(ctl_table *ctl, int write, void __user *buffer, + size_t *lenp, loff_t *ppos) +{ + int size, ret; + ctl_table tmp = *ctl; + + tmp.data = &size; + size = DIV_ROUND_UP(*(int *)ctl->data, SKB_TRUESIZE(ETH_FRAME_LEN)); + ret = proc_dointvec(&tmp, write, buffer, lenp, ppos); + if (write && !ret) + *(int *)ctl->data = size * SKB_TRUESIZE(ETH_FRAME_LEN); + return ret; +} + +enum { + NEIGH_VAR_MCAST_PROBE, + NEIGH_VAR_UCAST_PROBE, + NEIGH_VAR_APP_PROBE, + NEIGH_VAR_RETRANS_TIME, + NEIGH_VAR_BASE_REACHABLE_TIME, + NEIGH_VAR_DELAY_PROBE_TIME, + NEIGH_VAR_GC_STALETIME, + NEIGH_VAR_QUEUE_LEN, + NEIGH_VAR_QUEUE_LEN_BYTES, + NEIGH_VAR_PROXY_QLEN, + NEIGH_VAR_ANYCAST_DELAY, + NEIGH_VAR_PROXY_DELAY, + NEIGH_VAR_LOCKTIME, + NEIGH_VAR_RETRANS_TIME_MS, + NEIGH_VAR_BASE_REACHABLE_TIME_MS, + NEIGH_VAR_GC_INTERVAL, + NEIGH_VAR_GC_THRESH1, + NEIGH_VAR_GC_THRESH2, + NEIGH_VAR_GC_THRESH3, + NEIGH_VAR_MAX +}; static struct neigh_sysctl_table { struct ctl_table_header *sysctl_header; - struct ctl_table neigh_vars[NEIGH_VARS_MAX]; + struct ctl_table neigh_vars[NEIGH_VAR_MAX + 1]; char *dev_name; } neigh_sysctl_template __read_mostly = { .neigh_vars = { - { + [NEIGH_VAR_MCAST_PROBE] = { .procname = "mcast_solicit", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec, }, - { + [NEIGH_VAR_UCAST_PROBE] = { .procname = "ucast_solicit", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec, }, - { + [NEIGH_VAR_APP_PROBE] = { .procname = "app_solicit", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec, }, - { + [NEIGH_VAR_RETRANS_TIME] = { .procname = "retrans_time", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_userhz_jiffies, }, - { + [NEIGH_VAR_BASE_REACHABLE_TIME] = { .procname = "base_reachable_time", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { + [NEIGH_VAR_DELAY_PROBE_TIME] = { .procname = "delay_first_probe_time", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { + [NEIGH_VAR_GC_STALETIME] = { .procname = "gc_stale_time", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { + [NEIGH_VAR_QUEUE_LEN] = { .procname = "unres_qlen", .maxlen = sizeof(int), .mode = 0644, + .proc_handler = proc_unres_qlen, + }, + [NEIGH_VAR_QUEUE_LEN_BYTES] = { + .procname = "unres_qlen_bytes", + .maxlen = sizeof(int), + .mode = 0644, .proc_handler = proc_dointvec, }, - { + [NEIGH_VAR_PROXY_QLEN] = { .procname = "proxy_qlen", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec, }, - { + [NEIGH_VAR_ANYCAST_DELAY] = { .procname = "anycast_delay", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_userhz_jiffies, }, - { + [NEIGH_VAR_PROXY_DELAY] = { .procname = "proxy_delay", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_userhz_jiffies, }, - { + [NEIGH_VAR_LOCKTIME] = { .procname = "locktime", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_userhz_jiffies, }, - { + [NEIGH_VAR_RETRANS_TIME_MS] = { .procname = "retrans_time_ms", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_ms_jiffies, }, - { + [NEIGH_VAR_BASE_REACHABLE_TIME_MS] = { .procname = "base_reachable_time_ms", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_ms_jiffies, }, - { + [NEIGH_VAR_GC_INTERVAL] = { .procname = "gc_interval", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_jiffies, }, - { + [NEIGH_VAR_GC_THRESH1] = { .procname = "gc_thresh1", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec, }, - { + [NEIGH_VAR_GC_THRESH2] = { .procname = "gc_thresh2", .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec, }, - { + [NEIGH_VAR_GC_THRESH3] = { .procname = "gc_thresh3", .maxlen = sizeof(int), .mode = 0644, @@ -2778,47 +2836,49 @@ int neigh_sysctl_register(struct net_device *dev, struct neigh_parms *p, if (!t) goto err; - t->neigh_vars[0].data = &p->mcast_probes; - t->neigh_vars[1].data = &p->ucast_probes; - t->neigh_vars[2].data = &p->app_probes; - t->neigh_vars[3].data = &p->retrans_time; - t->neigh_vars[4].data = &p->base_reachable_time; - t->neigh_vars[5].data = &p->delay_probe_time; - t->neigh_vars[6].data = &p->gc_staletime; - t->neigh_vars[7].data = &p->queue_len; - t->neigh_vars[8].data = &p->proxy_qlen; - t->neigh_vars[9].data = &p->anycast_delay; - t->neigh_vars[10].data = &p->proxy_delay; - t->neigh_vars[11].data = &p->locktime; - t->neigh_vars[12].data = &p->retrans_time; - t->neigh_vars[13].data = &p->base_reachable_time; + t->neigh_vars[NEIGH_VAR_MCAST_PROBE].data = &p->mcast_probes; + t->neigh_vars[NEIGH_VAR_UCAST_PROBE].data = &p->ucast_probes; + t->neigh_vars[NEIGH_VAR_APP_PROBE].data = &p->app_probes; + t->neigh_vars[NEIGH_VAR_RETRANS_TIME].data = &p->retrans_time; + t->neigh_vars[NEIGH_VAR_BASE_REACHABLE_TIME].data = &p->base_reachable_time; + t->neigh_vars[NEIGH_VAR_DELAY_PROBE_TIME].data = &p->delay_probe_time; + t->neigh_vars[NEIGH_VAR_GC_STALETIME].data = &p->gc_staletime; + t->neigh_vars[NEIGH_VAR_QUEUE_LEN].data = &p->queue_len_bytes; + t->neigh_vars[NEIGH_VAR_QUEUE_LEN_BYTES].data = &p->queue_len_bytes; + t->neigh_vars[NEIGH_VAR_PROXY_QLEN].data = &p->proxy_qlen; + t->neigh_vars[NEIGH_VAR_ANYCAST_DELAY].data = &p->anycast_delay; + t->neigh_vars[NEIGH_VAR_PROXY_DELAY].data = &p->proxy_delay; + t->neigh_vars[NEIGH_VAR_LOCKTIME].data = &p->locktime; + t->neigh_vars[NEIGH_VAR_RETRANS_TIME_MS].data = &p->retrans_time; + t->neigh_vars[NEIGH_VAR_BASE_REACHABLE_TIME_MS].data = &p->base_reachable_time; if (dev) { dev_name_source = dev->name; /* Terminate the table early */ - memset(&t->neigh_vars[14], 0, sizeof(t->neigh_vars[14])); + memset(&t->neigh_vars[NEIGH_VAR_GC_INTERVAL], 0, + sizeof(t->neigh_vars[NEIGH_VAR_GC_INTERVAL])); } else { dev_name_source = neigh_path[NEIGH_CTL_PATH_DEV].procname; - t->neigh_vars[14].data = (int *)(p + 1); - t->neigh_vars[15].data = (int *)(p + 1) + 1; - t->neigh_vars[16].data = (int *)(p + 1) + 2; - t->neigh_vars[17].data = (int *)(p + 1) + 3; + t->neigh_vars[NEIGH_VAR_GC_INTERVAL].data = (int *)(p + 1); + t->neigh_vars[NEIGH_VAR_GC_THRESH1].data = (int *)(p + 1) + 1; + t->neigh_vars[NEIGH_VAR_GC_THRESH2].data = (int *)(p + 1) + 2; + t->neigh_vars[NEIGH_VAR_GC_THRESH3].data = (int *)(p + 1) + 3; } if (handler) { /* RetransTime */ - t->neigh_vars[3].proc_handler = handler; - t->neigh_vars[3].extra1 = dev; + t->neigh_vars[NEIGH_VAR_RETRANS_TIME].proc_handler = handler; + t->neigh_vars[NEIGH_VAR_RETRANS_TIME].extra1 = dev; /* ReachableTime */ - t->neigh_vars[4].proc_handler = handler; - t->neigh_vars[4].extra1 = dev; + t->neigh_vars[NEIGH_VAR_BASE_REACHABLE_TIME].proc_handler = handler; + t->neigh_vars[NEIGH_VAR_BASE_REACHABLE_TIME].extra1 = dev; /* RetransTime (in milliseconds)*/ - t->neigh_vars[12].proc_handler = handler; - t->neigh_vars[12].extra1 = dev; + t->neigh_vars[NEIGH_VAR_RETRANS_TIME_MS].proc_handler = handler; + t->neigh_vars[NEIGH_VAR_RETRANS_TIME_MS].extra1 = dev; /* ReachableTime (in milliseconds) */ - t->neigh_vars[13].proc_handler = handler; - t->neigh_vars[13].extra1 = dev; + t->neigh_vars[NEIGH_VAR_BASE_REACHABLE_TIME_MS].proc_handler = handler; + t->neigh_vars[NEIGH_VAR_BASE_REACHABLE_TIME_MS].extra1 = dev; } t->dev_name = kstrdup(dev_name_source, GFP_KERNEL); diff --git a/net/decnet/dn_neigh.c b/net/decnet/dn_neigh.c index 7f0eb087dc116..3532ac64c82db 100644 --- a/net/decnet/dn_neigh.c +++ b/net/decnet/dn_neigh.c @@ -107,7 +107,7 @@ struct neigh_table dn_neigh_table = { .gc_staletime = 60 * HZ, .reachable_time = 30 * HZ, .delay_probe_time = 5 * HZ, - .queue_len = 3, + .queue_len_bytes = 64*1024, .ucast_probes = 0, .app_probes = 0, .mcast_probes = 0, diff --git a/net/ipv4/arp.c b/net/ipv4/arp.c index 96a164aa1367b..d732827b32b9c 100644 --- a/net/ipv4/arp.c +++ b/net/ipv4/arp.c @@ -177,7 +177,7 @@ struct neigh_table arp_tbl = { .gc_staletime = 60 * HZ, .reachable_time = 30 * HZ, .delay_probe_time = 5 * HZ, - .queue_len = 3, + .queue_len_bytes = 64*1024, .ucast_probes = 3, .mcast_probes = 3, .anycast_delay = 1 * HZ, diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c index 44e5b7f2a6c1b..4a2098222625e 100644 --- a/net/ipv6/ndisc.c +++ b/net/ipv6/ndisc.c @@ -141,7 +141,7 @@ struct neigh_table nd_tbl = { .gc_staletime = 60 * HZ, .reachable_time = ND_REACHABLE_TIME, .delay_probe_time = 5 * HZ, - .queue_len = 3, + .queue_len_bytes = 64*1024, .ucast_probes = 3, .mcast_probes = 3, .anycast_delay = 1 * HZ, -- GitLab From 6d74eb9442fb113c97edc88a1c658462db711337 Mon Sep 17 00:00:00 2001 From: Mark Kamichoff Date: Wed, 9 Nov 2011 11:48:10 +0000 Subject: [PATCH 0249/3089] net/usb: Misc. fixes for the LG-VL600 LTE USB modem Add checking for valid magic values (needed for stability in the event corrupted packets are received) and remove some other unneeded checks. Also, fix flagging device as WWAN (Bugzilla bug #39952). Signed-off-by: Mark Kamichoff Signed-off-by: David S. Miller --- drivers/net/usb/cdc_ether.c | 2 +- drivers/net/usb/lg-vl600.c | 25 +++++++++++-------------- 2 files changed, 12 insertions(+), 15 deletions(-) diff --git a/drivers/net/usb/cdc_ether.c b/drivers/net/usb/cdc_ether.c index c924ea2bce07a..99ed6eb4dfaf6 100644 --- a/drivers/net/usb/cdc_ether.c +++ b/drivers/net/usb/cdc_ether.c @@ -567,7 +567,7 @@ static const struct usb_device_id products [] = { { USB_DEVICE_AND_INTERFACE_INFO(0x1004, 0x61aa, USB_CLASS_COMM, USB_CDC_SUBCLASS_ETHERNET, USB_CDC_PROTO_NONE), - .driver_info = (unsigned long)&wwan_info, + .driver_info = 0, }, /* diff --git a/drivers/net/usb/lg-vl600.c b/drivers/net/usb/lg-vl600.c index d43db32f94781..9c26c6390d69a 100644 --- a/drivers/net/usb/lg-vl600.c +++ b/drivers/net/usb/lg-vl600.c @@ -144,10 +144,11 @@ static int vl600_rx_fixup(struct usbnet *dev, struct sk_buff *skb) } frame = (struct vl600_frame_hdr *) buf->data; - /* NOTE: Should check that frame->magic == 0x53544448? - * Otherwise if we receive garbage at the beginning of the frame - * we may end up allocating a huge buffer and saving all the - * future incoming data into it. */ + /* Yes, check that frame->magic == 0x53544448 (or 0x44544d48), + * otherwise we may run out of memory w/a bad packet */ + if (ntohl(frame->magic) != 0x53544448 && + ntohl(frame->magic) != 0x44544d48) + goto error; if (buf->len < sizeof(*frame) || buf->len != le32_to_cpup(&frame->len)) { @@ -296,6 +297,11 @@ encapsulate: * overwrite the remaining fields. */ packet = (struct vl600_pkt_hdr *) skb->data; + /* The VL600 wants IPv6 packets to have an IPv4 ethertype + * Since this modem only supports IPv4 and IPv6, just set all + * frames to 0x0800 (ETH_P_IP) + */ + packet->h_proto = htons(ETH_P_IP); memset(&packet->dummy, 0, sizeof(packet->dummy)); packet->len = cpu_to_le32(orig_len); @@ -308,21 +314,12 @@ encapsulate: if (skb->len < full_len) /* Pad */ skb_put(skb, full_len - skb->len); - /* The VL600 wants IPv6 packets to have an IPv4 ethertype - * Check if this is an IPv6 packet, and set the ethertype - * to 0x800 - */ - if ((skb->data[sizeof(struct vl600_pkt_hdr *) + 0x22] & 0xf0) == 0x60) { - skb->data[sizeof(struct vl600_pkt_hdr *) + 0x20] = 0x08; - skb->data[sizeof(struct vl600_pkt_hdr *) + 0x21] = 0; - } - return skb; } static const struct driver_info vl600_info = { .description = "LG VL600 modem", - .flags = FLAG_ETHER | FLAG_RX_ASSEMBLE, + .flags = FLAG_RX_ASSEMBLE | FLAG_WWAN, .bind = vl600_bind, .unbind = vl600_unbind, .status = usbnet_cdc_status, -- GitLab From 452448f9283e1939408b397e87974a418825b0a8 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Wed, 9 Nov 2011 10:50:49 +0000 Subject: [PATCH 0250/3089] net/can/mscan: add listen only mode This patch adds listen only mode to the mscan controller. Signed-off-by: Marc Kleine-Budde Acked-by: Wolfgang Grandegger Signed-off-by: David S. Miller --- drivers/net/can/mscan/mscan.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/net/can/mscan/mscan.c b/drivers/net/can/mscan/mscan.c index ec4a3119e2c93..74f3b1829e029 100644 --- a/drivers/net/can/mscan/mscan.c +++ b/drivers/net/can/mscan/mscan.c @@ -581,7 +581,10 @@ static int mscan_open(struct net_device *dev) priv->open_time = jiffies; - clrbits8(®s->canctl1, MSCAN_LISTEN); + if (ctrlmode.flags & CAN_CTRLMODE_LISTENONLY) + setbits8(®s->canctl1, MSCAN_LISTEN); + else + clrbits8(®s->canctl1, MSCAN_LISTEN); ret = mscan_start(dev); if (ret) @@ -690,7 +693,8 @@ struct net_device *alloc_mscandev(void) priv->can.bittiming_const = &mscan_bittiming_const; priv->can.do_set_bittiming = mscan_do_set_bittiming; priv->can.do_set_mode = mscan_do_set_mode; - priv->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES; + priv->can.ctrlmode_supported = CAN_CTRLMODE_3_SAMPLES | + CAN_CTRLMODE_LISTENONLY; for (i = 0; i < TX_QUEUE_SIZE; i++) { priv->tx_queue[i].id = i; -- GitLab From 06236ac3726f15124839cf16a9e2730a852dad9b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Maciej=20=C5=BBenczykowski?= Date: Mon, 7 Nov 2011 14:23:11 +0000 Subject: [PATCH 0251/3089] net-netlink: Add a new attribute to expose TCLASS values via netlink MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit commit 3ceca749668a52bd795585e0f71c6f0b04814f7b added a TOS attribute. Unfortunately TOS and TCLASS are both present in a dual-stack v6 socket, furthermore they can have different values. As such one cannot in a sane way expose both through a single attribute. Signed-off-by: Maciej Å»enczyowski CC: Murali Raja CC: Stephen Hemminger CC: Eric Dumazet CC: David S. Miller Signed-off-by: David S. Miller --- include/linux/inet_diag.h | 3 ++- net/ipv4/inet_diag.c | 4 ++-- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/include/linux/inet_diag.h b/include/linux/inet_diag.h index 80b480c97532f..abf5028db9819 100644 --- a/include/linux/inet_diag.h +++ b/include/linux/inet_diag.h @@ -98,9 +98,10 @@ enum { INET_DIAG_VEGASINFO, INET_DIAG_CONG, INET_DIAG_TOS, + INET_DIAG_TCLASS, }; -#define INET_DIAG_MAX INET_DIAG_TOS +#define INET_DIAG_MAX INET_DIAG_TCLASS /* INET_DIAG_MEM */ diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c index f5e2bdaef9495..68e8ac5143834 100644 --- a/net/ipv4/inet_diag.c +++ b/net/ipv4/inet_diag.c @@ -133,8 +133,8 @@ static int inet_csk_diag_fill(struct sock *sk, &np->rcv_saddr); ipv6_addr_copy((struct in6_addr *)r->id.idiag_dst, &np->daddr); - if (ext & (1 << (INET_DIAG_TOS - 1))) - RTA_PUT_U8(skb, INET_DIAG_TOS, np->tclass); + if (ext & (1 << (INET_DIAG_TCLASS - 1))) + RTA_PUT_U8(skb, INET_DIAG_TCLASS, np->tclass); } #endif -- GitLab From c3e072f8a6c5625028531c40ec65f7e301531be2 Mon Sep 17 00:00:00 2001 From: Baruch Siach Date: Mon, 14 Nov 2011 08:21:30 +0200 Subject: [PATCH 0252/3089] net: fsl_pq_mdio: fix non tbi phy access Since 952c5ca1 (fsl_pq_mdio: Clean up tbi address configuration) .probe returns -EBUSY when the "tbi-phy" node is missing. Fix this. Cc: Andy Fleming Signed-off-by: Baruch Siach Signed-off-by: David S. Miller --- drivers/net/ethernet/freescale/fsl_pq_mdio.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/freescale/fsl_pq_mdio.c b/drivers/net/ethernet/freescale/fsl_pq_mdio.c index 4d9f84b8ab977..8dee1aed47e4c 100644 --- a/drivers/net/ethernet/freescale/fsl_pq_mdio.c +++ b/drivers/net/ethernet/freescale/fsl_pq_mdio.c @@ -356,16 +356,16 @@ static int fsl_pq_mdio_probe(struct platform_device *ofdev) if (prop) tbiaddr = *prop; - } - if (tbiaddr == -1) { - err = -EBUSY; + if (tbiaddr == -1) { + err = -EBUSY; - goto err_free_irqs; + goto err_free_irqs; + } else { + out_be32(tbipa, tbiaddr); + } } - out_be32(tbipa, tbiaddr); - err = of_mdiobus_register(new_bus, np); if (err) { printk (KERN_ERR "%s: Cannot register as MDIO bus\n", -- GitLab From d53dab3ae1a3c6b438738c3792c98ac63f0061f4 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 23 Jul 2011 12:45:07 +0200 Subject: [PATCH 0253/3089] drm: Remove utterly bogus preempt_disable() sections commit 27641c3f (drm/vblank: Add support for precise vblank timestamping) adds preempt_disable()/enable() around a spin locked section with the comments: * Disable preemption, so vblank_time_lock is held as short as * possible, even under a kernel with PREEMPT_RT patches. /* Disable preemption while holding vblank_time_lock. Do * it explicitely to guard against PREEMPT_RT kernel. Just that this has never been tested on a RT kernel which would have granted that nonsense with a might_sleep() warning because dev->vblank_time_lock is converted to a "sleeping" spinlock on RT. So this is activly wrong on RT and superflous on mainline. Remove it. Signed-off-by: Thomas Gleixner Acked-by: Mario Kleiner Signed-off-by: Dave Airlie --- drivers/gpu/drm/drm_irq.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c index 68b756253f9f2..44a5d0ad8b7c5 100644 --- a/drivers/gpu/drm/drm_irq.c +++ b/drivers/gpu/drm/drm_irq.c @@ -110,10 +110,7 @@ static void vblank_disable_and_save(struct drm_device *dev, int crtc) /* Prevent vblank irq processing while disabling vblank irqs, * so no updates of timestamps or count can happen after we've * disabled. Needed to prevent races in case of delayed irq's. - * Disable preemption, so vblank_time_lock is held as short as - * possible, even under a kernel with PREEMPT_RT patches. */ - preempt_disable(); spin_lock_irqsave(&dev->vblank_time_lock, irqflags); dev->driver->disable_vblank(dev, crtc); @@ -164,7 +161,6 @@ static void vblank_disable_and_save(struct drm_device *dev, int crtc) clear_vblank_timestamps(dev, crtc); spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags); - preempt_enable(); } static void vblank_disable_fn(unsigned long arg) @@ -889,10 +885,6 @@ int drm_vblank_get(struct drm_device *dev, int crtc) spin_lock_irqsave(&dev->vbl_lock, irqflags); /* Going from 0->1 means we have to enable interrupts again */ if (atomic_add_return(1, &dev->vblank_refcount[crtc]) == 1) { - /* Disable preemption while holding vblank_time_lock. Do - * it explicitely to guard against PREEMPT_RT kernel. - */ - preempt_disable(); spin_lock_irqsave(&dev->vblank_time_lock, irqflags2); if (!dev->vblank_enabled[crtc]) { /* Enable vblank irqs under vblank_time_lock protection. @@ -912,7 +904,6 @@ int drm_vblank_get(struct drm_device *dev, int crtc) } } spin_unlock_irqrestore(&dev->vblank_time_lock, irqflags2); - preempt_enable(); } else { if (!dev->vblank_enabled[crtc]) { atomic_dec(&dev->vblank_refcount[crtc]); -- GitLab From e53de8f00c80fd1a312c95bc5157fdb98d46e070 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Sun, 13 Nov 2011 23:11:50 +0100 Subject: [PATCH 0254/3089] ALSA: hda/realtek: remove redundant semicolon Having just one semicolon after a break statement is enough. Signed-off-by: Jesper Juhl Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_realtek.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 308bb575bc064..336d14eb72af8 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -1452,7 +1452,7 @@ static void alc_apply_fixup(struct hda_codec *codec, int action) switch (fix->type) { case ALC_FIXUP_SKU: if (action != ALC_FIXUP_ACT_PRE_PROBE || !fix->v.sku) - break;; + break; snd_printdd(KERN_INFO "hda_codec: %s: " "Apply sku override for %s\n", codec->chip_name, modelname); -- GitLab From f6252114cbb7edabc1c92bf794153922ee0ea46d Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Mon, 10 Oct 2011 09:29:18 +0100 Subject: [PATCH 0255/3089] vgaarb: a NULL bridge is acceptable for root devices. I assumed all PCI buses had a bridge, but playing with qemu recently, I discovered vgaarb bug where it wasn't detecting both devices shared a bridge at the root level. Don't check for NULL, if two buses have a NULL bridge, assume they share the root bus. Acked-by: Jesse Barnes Signed-off-by: Dave Airlie --- drivers/gpu/vga/vgaarb.c | 44 +++++++++++++++++++--------------------- 1 file changed, 21 insertions(+), 23 deletions(-) diff --git a/drivers/gpu/vga/vgaarb.c b/drivers/gpu/vga/vgaarb.c index c72f1c0b5e637..bdde899af72ed 100644 --- a/drivers/gpu/vga/vgaarb.c +++ b/drivers/gpu/vga/vgaarb.c @@ -465,31 +465,29 @@ static void vga_arbiter_check_bridge_sharing(struct vga_device *vgadev) while (new_bus) { new_bridge = new_bus->self; - if (new_bridge) { - /* go through list of devices already registered */ - list_for_each_entry(same_bridge_vgadev, &vga_list, list) { - bus = same_bridge_vgadev->pdev->bus; - bridge = bus->self; - - /* see if the share a bridge with this device */ - if (new_bridge == bridge) { - /* if their direct parent bridge is the same - as any bridge of this device then it can't be used - for that device */ - same_bridge_vgadev->bridge_has_one_vga = false; - } + /* go through list of devices already registered */ + list_for_each_entry(same_bridge_vgadev, &vga_list, list) { + bus = same_bridge_vgadev->pdev->bus; + bridge = bus->self; + + /* see if the share a bridge with this device */ + if (new_bridge == bridge) { + /* if their direct parent bridge is the same + as any bridge of this device then it can't be used + for that device */ + same_bridge_vgadev->bridge_has_one_vga = false; + } - /* now iterate the previous devices bridge hierarchy */ - /* if the new devices parent bridge is in the other devices - hierarchy then we can't use it to control this device */ - while (bus) { - bridge = bus->self; - if (bridge) { - if (bridge == vgadev->pdev->bus->self) - vgadev->bridge_has_one_vga = false; - } - bus = bus->parent; + /* now iterate the previous devices bridge hierarchy */ + /* if the new devices parent bridge is in the other devices + hierarchy then we can't use it to control this device */ + while (bus) { + bridge = bus->self; + if (bridge) { + if (bridge == vgadev->pdev->bus->self) + vgadev->bridge_has_one_vga = false; } + bus = bus->parent; } } new_bus = new_bus->parent; -- GitLab From 800252976be89611ef86d6d04442a821018ed949 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 14 Nov 2011 11:18:59 +0100 Subject: [PATCH 0256/3089] [S390] wire up process_vm syscalls Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/unistd.h | 4 +++- arch/s390/kernel/compat_wrapper.S | 20 ++++++++++++++++++++ arch/s390/kernel/syscalls.S | 2 ++ 3 files changed, 25 insertions(+), 1 deletion(-) diff --git a/arch/s390/include/asm/unistd.h b/arch/s390/include/asm/unistd.h index 404bdb9671b4f..58de4c91c3333 100644 --- a/arch/s390/include/asm/unistd.h +++ b/arch/s390/include/asm/unistd.h @@ -277,7 +277,9 @@ #define __NR_clock_adjtime 337 #define __NR_syncfs 338 #define __NR_setns 339 -#define NR_syscalls 340 +#define __NR_process_vm_readv 340 +#define __NR_process_vm_writev 341 +#define NR_syscalls 342 /* * There are some system calls that are not present on 64 bit, some diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S index 5006a1d9f5d0e..18c51df9fe06c 100644 --- a/arch/s390/kernel/compat_wrapper.S +++ b/arch/s390/kernel/compat_wrapper.S @@ -1627,3 +1627,23 @@ ENTRY(sys_setns_wrapper) lgfr %r2,%r2 # int lgfr %r3,%r3 # int jg sys_setns + +ENTRY(compat_sys_process_vm_readv_wrapper) + lgfr %r2,%r2 # compat_pid_t + llgtr %r3,%r3 # struct compat_iovec __user * + llgfr %r4,%r4 # unsigned long + llgtr %r5,%r5 # struct compat_iovec __user * + llgfr %r6,%r6 # unsigned long + llgf %r0,164(%r15) # unsigned long + stg %r0,160(%r15) + jg sys_process_vm_readv + +ENTRY(compat_sys_process_vm_writev_wrapper) + lgfr %r2,%r2 # compat_pid_t + llgtr %r3,%r3 # struct compat_iovec __user * + llgfr %r4,%r4 # unsigned long + llgtr %r5,%r5 # struct compat_iovec __user * + llgfr %r6,%r6 # unsigned long + llgf %r0,164(%r15) # unsigned long + stg %r0,160(%r15) + jg sys_process_vm_writev diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S index 73eb08c874fb4..bcab2f04ba581 100644 --- a/arch/s390/kernel/syscalls.S +++ b/arch/s390/kernel/syscalls.S @@ -348,3 +348,5 @@ SYSCALL(sys_open_by_handle_at,sys_open_by_handle_at,compat_sys_open_by_handle_at SYSCALL(sys_clock_adjtime,sys_clock_adjtime,compat_sys_clock_adjtime_wrapper) SYSCALL(sys_syncfs,sys_syncfs,sys_syncfs_wrapper) SYSCALL(sys_setns,sys_setns,sys_setns_wrapper) +SYSCALL(sys_process_vm_readv,sys_process_vm_readv,compat_sys_process_vm_readv_wrapper) /* 340 */ +SYSCALL(sys_process_vm_writev,sys_process_vm_writev,compat_sys_process_vm_writev_wrapper) -- GitLab From 09b538833b85521d937a06faf61e6a3273253cc0 Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Mon, 14 Nov 2011 11:19:00 +0100 Subject: [PATCH 0257/3089] [S390] fix pgste update logic The pgste_update_all / pgste_update_young and pgste_set_pte need to check if the pte entry contains a valid page address before the storage key can be accessed. In addition pgste_set_pte needs to set the access key and fetch protection bit of the new pte entry, not the old entry. Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/pgtable.h | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h index 34ede0ea85a9d..524d23b8610ce 100644 --- a/arch/s390/include/asm/pgtable.h +++ b/arch/s390/include/asm/pgtable.h @@ -593,6 +593,8 @@ static inline pgste_t pgste_update_all(pte_t *ptep, pgste_t pgste) unsigned long address, bits; unsigned char skey; + if (!pte_present(*ptep)) + return pgste; address = pte_val(*ptep) & PAGE_MASK; skey = page_get_storage_key(address); bits = skey & (_PAGE_CHANGED | _PAGE_REFERENCED); @@ -625,6 +627,8 @@ static inline pgste_t pgste_update_young(pte_t *ptep, pgste_t pgste) #ifdef CONFIG_PGSTE int young; + if (!pte_present(*ptep)) + return pgste; young = page_reset_referenced(pte_val(*ptep) & PAGE_MASK); /* Transfer page referenced bit to pte software bit (host view) */ if (young || (pgste_val(pgste) & RCP_HR_BIT)) @@ -638,13 +642,15 @@ static inline pgste_t pgste_update_young(pte_t *ptep, pgste_t pgste) } -static inline void pgste_set_pte(pte_t *ptep, pgste_t pgste) +static inline void pgste_set_pte(pte_t *ptep, pgste_t pgste, pte_t entry) { #ifdef CONFIG_PGSTE unsigned long address; unsigned long okey, nkey; - address = pte_val(*ptep) & PAGE_MASK; + if (!pte_present(entry)) + return; + address = pte_val(entry) & PAGE_MASK; okey = nkey = page_get_storage_key(address); nkey &= ~(_PAGE_ACC_BITS | _PAGE_FP_BIT); /* Set page access key and fetch protection bit from pgste */ @@ -712,7 +718,7 @@ static inline void set_pte_at(struct mm_struct *mm, unsigned long addr, if (mm_has_pgste(mm)) { pgste = pgste_get_lock(ptep); - pgste_set_pte(ptep, pgste); + pgste_set_pte(ptep, pgste, entry); *ptep = entry; pgste_set_unlock(ptep, pgste); } else -- GitLab From fa2fb2f4a599c402bb2670dde27867dbbb7e3c45 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 14 Nov 2011 11:19:01 +0100 Subject: [PATCH 0258/3089] [S390] pfault: ignore leftover completion interrupts Ignore completion interrupts if the initial interrupt hasn't been received and the addressed task is not running. This case can only happen if leftover (pending) completion interrupt gets delivered which wasn't removed with the PFAULT CANCEL operation during cpu hotplug. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/mm/fault.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c index 1766def5bc3fc..a9a301866b3c1 100644 --- a/arch/s390/mm/fault.c +++ b/arch/s390/mm/fault.c @@ -587,8 +587,13 @@ static void pfault_interrupt(unsigned int ext_int_code, } else { /* Completion interrupt was faster than initial * interrupt. Set pfault_wait to -1 so the initial - * interrupt doesn't put the task to sleep. */ - tsk->thread.pfault_wait = -1; + * interrupt doesn't put the task to sleep. + * If the task is not running, ignore the completion + * interrupt since it must be a leftover of a PFAULT + * CANCEL operation which didn't remove all pending + * completion interrupts. */ + if (tsk->state == TASK_RUNNING) + tsk->thread.pfault_wait = -1; } put_task_struct(tsk); } else { -- GitLab From 7a2512b744e72377c3fa5976f06a3f343e155d1f Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Mon, 14 Nov 2011 11:19:02 +0100 Subject: [PATCH 0259/3089] [S390] incorrect note program header 'readelf -n' on the s390 vmlinux file generates lots of warnings about corrupt notes. The reason is that the 'NOTE' program header has incorrect file and memory sizes. The problem is that the section following the NOTES section do not switch to a different phdr and they get added to the NOTE program section. Add a dummy entry to the linker script that switches to the data phdr before the start of the RODATA section. Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/vmlinux.lds.S | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S index 56fe6bc81fee4..e4c79ebb40e62 100644 --- a/arch/s390/kernel/vmlinux.lds.S +++ b/arch/s390/kernel/vmlinux.lds.S @@ -43,6 +43,8 @@ SECTIONS NOTES :text :note + .dummy : { *(.dummy) } :data + RODATA #ifdef CONFIG_SHARED_KERNEL -- GitLab From 96603b505cb6c54782a27599afef65cc108ef5f2 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Mon, 14 Nov 2011 11:19:03 +0100 Subject: [PATCH 0260/3089] [S390] Kconfig: Select CONFIG_KEXEC for CONFIG_CRASH_DUMP The kdump infrastructure is built on top of kexec. Therefore CONFIG_KEXEC has to be enabled when CONFIG_CRASH_DUMP is selected. Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky --- arch/s390/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index a9fbd43395f71..373679b3744a7 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -572,6 +572,7 @@ config KEXEC config CRASH_DUMP bool "kernel crash dumps" depends on 64BIT + select KEXEC help Generate crash dump after being started by kexec. Crash dump kernels are loaded in the main kernel with kexec-tools -- GitLab From bc615deaf35ab06e7fe5672b0efb3c7a0b2dcf1a Mon Sep 17 00:00:00 2001 From: Holger Dengler Date: Mon, 14 Nov 2011 11:19:04 +0100 Subject: [PATCH 0261/3089] [S390] ap: Setup processing for messages in request queue. Setup timer for processing messages in request queue, if sending an AP message returns with reason code AP_RESPONSE_RESET_IN_PROGRESS. Signed-off-by: Holger Dengler Signed-off-by: Martin Schwidefsky --- drivers/s390/crypto/ap_bus.c | 23 +++++++++++++++++------ 1 file changed, 17 insertions(+), 6 deletions(-) diff --git a/drivers/s390/crypto/ap_bus.c b/drivers/s390/crypto/ap_bus.c index b77ae519d79c4..ec94f049e9954 100644 --- a/drivers/s390/crypto/ap_bus.c +++ b/drivers/s390/crypto/ap_bus.c @@ -1271,18 +1271,16 @@ ap_config_timeout(unsigned long ptr) } /** - * ap_schedule_poll_timer(): Schedule poll timer. + * __ap_schedule_poll_timer(): Schedule poll timer. * * Set up the timer to run the poll tasklet */ -static inline void ap_schedule_poll_timer(void) +static inline void __ap_schedule_poll_timer(void) { ktime_t hr_time; spin_lock_bh(&ap_poll_timer_lock); - if (ap_using_interrupts() || ap_suspend_flag) - goto out; - if (hrtimer_is_queued(&ap_poll_timer)) + if (hrtimer_is_queued(&ap_poll_timer) || ap_suspend_flag) goto out; if (ktime_to_ns(hrtimer_expires_remaining(&ap_poll_timer)) <= 0) { hr_time = ktime_set(0, poll_timeout); @@ -1293,6 +1291,18 @@ out: spin_unlock_bh(&ap_poll_timer_lock); } +/** + * ap_schedule_poll_timer(): Schedule poll timer. + * + * Set up the timer to run the poll tasklet + */ +static inline void ap_schedule_poll_timer(void) +{ + if (ap_using_interrupts()) + return; + __ap_schedule_poll_timer(); +} + /** * ap_poll_read(): Receive pending reply messages from an AP device. * @ap_dev: pointer to the AP device @@ -1374,8 +1384,9 @@ static int ap_poll_write(struct ap_device *ap_dev, unsigned long *flags) *flags |= 1; *flags |= 2; break; - case AP_RESPONSE_Q_FULL: case AP_RESPONSE_RESET_IN_PROGRESS: + __ap_schedule_poll_timer(); + case AP_RESPONSE_Q_FULL: *flags |= 2; break; case AP_RESPONSE_MESSAGE_TOO_BIG: -- GitLab From 3f25dc4fcbc371f86a61a6af759003ebd4965908 Mon Sep 17 00:00:00 2001 From: Michael Holzheu Date: Mon, 14 Nov 2011 11:19:05 +0100 Subject: [PATCH 0262/3089] [S390] zfcpdump: Do not initialize zfcpdump in kdump mode When the kernel is started in kdump mode, zfcpdump should not be initialized because both dump methods can't be used at the same time. Signed-off-by: Michael Holzheu Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/setup.c | 4 +++- drivers/s390/char/zcore.c | 2 ++ 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 8ac6bfa2786cb..e58a462949b16 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -211,6 +211,8 @@ static void __init setup_zfcpdump(unsigned int console_devno) if (ipl_info.type != IPL_TYPE_FCP_DUMP) return; + if (OLDMEM_BASE) + return; if (console_devno != -1) sprintf(str, " cio_ignore=all,!0.0.%04x,!0.0.%04x", ipl_info.data.fcp.dev_id.devno, console_devno); @@ -482,7 +484,7 @@ static void __init setup_memory_end(void) #ifdef CONFIG_ZFCPDUMP - if (ipl_info.type == IPL_TYPE_FCP_DUMP) { + if (ipl_info.type == IPL_TYPE_FCP_DUMP && !OLDMEM_BASE) { memory_end = ZFCPDUMP_HSA_SIZE; memory_end_set = 1; } diff --git a/drivers/s390/char/zcore.c b/drivers/s390/char/zcore.c index 43068fbd0baac..1b6d9247fdc78 100644 --- a/drivers/s390/char/zcore.c +++ b/drivers/s390/char/zcore.c @@ -641,6 +641,8 @@ static int __init zcore_init(void) if (ipl_info.type != IPL_TYPE_FCP_DUMP) return -ENODATA; + if (OLDMEM_BASE) + return -ENODATA; zcore_dbf = debug_register("zcore", 4, 1, 4 * sizeof(long)); debug_register_view(zcore_dbf, &debug_sprintf_view); -- GitLab From cfa1e7e1d49c6f5f0b00b2cb890b521e5c0dc7ea Mon Sep 17 00:00:00 2001 From: Jan Glauber Date: Mon, 14 Nov 2011 11:19:06 +0100 Subject: [PATCH 0263/3089] [S390] avoid STCKF if running in ESA mode In ESA mode STCKF is not defined even if the facility bit is enabled. To prevent an illegal operation we must also check if we run a 64 bit kernel. To make the check perform well add the STCKF bit to the machine flags. Signed-off-by: Jan Glauber Signed-off-by: Martin Schwidefsky --- arch/s390/include/asm/setup.h | 3 +++ arch/s390/include/asm/timex.h | 2 +- arch/s390/kernel/early.c | 2 ++ 3 files changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/s390/include/asm/setup.h b/arch/s390/include/asm/setup.h index 5a099714df045..097183c70407a 100644 --- a/arch/s390/include/asm/setup.h +++ b/arch/s390/include/asm/setup.h @@ -82,6 +82,7 @@ extern unsigned int user_mode; #define MACHINE_FLAG_LPAR (1UL << 12) #define MACHINE_FLAG_SPP (1UL << 13) #define MACHINE_FLAG_TOPOLOGY (1UL << 14) +#define MACHINE_FLAG_STCKF (1UL << 15) #define MACHINE_IS_VM (S390_lowcore.machine_flags & MACHINE_FLAG_VM) #define MACHINE_IS_KVM (S390_lowcore.machine_flags & MACHINE_FLAG_KVM) @@ -100,6 +101,7 @@ extern unsigned int user_mode; #define MACHINE_HAS_PFMF (0) #define MACHINE_HAS_SPP (0) #define MACHINE_HAS_TOPOLOGY (0) +#define MACHINE_HAS_STCKF (0) #else /* __s390x__ */ #define MACHINE_HAS_IEEE (1) #define MACHINE_HAS_CSP (1) @@ -111,6 +113,7 @@ extern unsigned int user_mode; #define MACHINE_HAS_PFMF (S390_lowcore.machine_flags & MACHINE_FLAG_PFMF) #define MACHINE_HAS_SPP (S390_lowcore.machine_flags & MACHINE_FLAG_SPP) #define MACHINE_HAS_TOPOLOGY (S390_lowcore.machine_flags & MACHINE_FLAG_TOPOLOGY) +#define MACHINE_HAS_STCKF (S390_lowcore.machine_flags & MACHINE_FLAG_STCKF) #endif /* __s390x__ */ #define ZFCPDUMP_HSA_SIZE (32UL<<20) diff --git a/arch/s390/include/asm/timex.h b/arch/s390/include/asm/timex.h index d610bef9c5e91..c447a27a7fdb1 100644 --- a/arch/s390/include/asm/timex.h +++ b/arch/s390/include/asm/timex.h @@ -90,7 +90,7 @@ static inline unsigned long long get_clock_fast(void) { unsigned long long clk; - if (test_facility(25)) + if (MACHINE_HAS_STCKF) asm volatile(".insn s,0xb27c0000,%0" : "=Q" (clk) : : "cc"); else clk = get_clock(); diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index 37394b3413e27..c9ffe00251971 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -390,6 +390,8 @@ static __init void detect_machine_facilities(void) S390_lowcore.machine_flags |= MACHINE_FLAG_MVCOS; if (test_facility(40)) S390_lowcore.machine_flags |= MACHINE_FLAG_SPP; + if (test_facility(25)) + S390_lowcore.machine_flags |= MACHINE_FLAG_STCKF; #endif } -- GitLab From 6ed54387dc470fc439cb154724a1ac81d251c126 Mon Sep 17 00:00:00 2001 From: Jan Glauber Date: Mon, 14 Nov 2011 11:19:07 +0100 Subject: [PATCH 0264/3089] [S390] crypto: avoid MSA3 and MSA4 instructions in ESA mode MSA3 and MSA4 instructions are only available under CONFIG_64BIT. Bail out before using any of these instructions if the kernel is running in 31 bit mode. Signed-off-by: Jan Glauber Signed-off-by: Martin Schwidefsky --- arch/s390/crypto/crypt_s390.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/arch/s390/crypto/crypt_s390.h b/arch/s390/crypto/crypt_s390.h index 49676771bd66a..ffd1ac255f194 100644 --- a/arch/s390/crypto/crypt_s390.h +++ b/arch/s390/crypto/crypt_s390.h @@ -368,9 +368,12 @@ static inline int crypt_s390_func_available(int func, if (facility_mask & CRYPT_S390_MSA && !test_facility(17)) return 0; - if (facility_mask & CRYPT_S390_MSA3 && !test_facility(76)) + + if (facility_mask & CRYPT_S390_MSA3 && + (!test_facility(2) || !test_facility(76))) return 0; - if (facility_mask & CRYPT_S390_MSA4 && !test_facility(77)) + if (facility_mask & CRYPT_S390_MSA4 && + (!test_facility(2) || !test_facility(77))) return 0; switch (func & CRYPT_S390_OP_MASK) { -- GitLab From f6bf1a8acd2cb3a92a7b7c9ab03e56a32ac5ece5 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Mon, 14 Nov 2011 11:19:08 +0100 Subject: [PATCH 0265/3089] [S390] topology: fix topology on z10 machines Make sure that all cpus in a book on a z10 appear as book siblings and not as core siblings. This fixes some performance regressions that appeared after the book scheduling domain got introduced. Signed-off-by: Heiko Carstens Signed-off-by: Martin Schwidefsky --- arch/s390/kernel/topology.c | 45 +++++++++++++++++++++++++++++++------ 1 file changed, 38 insertions(+), 7 deletions(-) diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c index 77b8942b9a153..fdb5b8cb260f6 100644 --- a/arch/s390/kernel/topology.c +++ b/arch/s390/kernel/topology.c @@ -68,8 +68,10 @@ static cpumask_t cpu_group_map(struct mask_info *info, unsigned int cpu) return mask; } -static void add_cpus_to_mask(struct topology_cpu *tl_cpu, - struct mask_info *book, struct mask_info *core) +static struct mask_info *add_cpus_to_mask(struct topology_cpu *tl_cpu, + struct mask_info *book, + struct mask_info *core, + int z10) { unsigned int cpu; @@ -88,10 +90,16 @@ static void add_cpus_to_mask(struct topology_cpu *tl_cpu, cpu_book_id[lcpu] = book->id; #endif cpumask_set_cpu(lcpu, &core->mask); - cpu_core_id[lcpu] = core->id; + if (z10) { + cpu_core_id[lcpu] = rcpu; + core = core->next; + } else { + cpu_core_id[lcpu] = core->id; + } smp_cpu_polarization[lcpu] = tl_cpu->pp; } } + return core; } static void clear_masks(void) @@ -123,18 +131,41 @@ static void tl_to_cores(struct sysinfo_15_1_x *info) { #ifdef CONFIG_SCHED_BOOK struct mask_info *book = &book_info; + struct cpuid cpu_id; #else struct mask_info *book = NULL; #endif struct mask_info *core = &core_info; union topology_entry *tle, *end; + int z10 = 0; - +#ifdef CONFIG_SCHED_BOOK + get_cpu_id(&cpu_id); + z10 = cpu_id.machine == 0x2097 || cpu_id.machine == 0x2098; +#endif spin_lock_irq(&topology_lock); clear_masks(); tle = info->tle; end = (union topology_entry *)((unsigned long)info + info->length); while (tle < end) { +#ifdef CONFIG_SCHED_BOOK + if (z10) { + switch (tle->nl) { + case 1: + book = book->next; + book->id = tle->container.id; + break; + case 0: + core = add_cpus_to_mask(&tle->cpu, book, core, z10); + break; + default: + clear_masks(); + goto out; + } + tle = next_tle(tle); + continue; + } +#endif switch (tle->nl) { #ifdef CONFIG_SCHED_BOOK case 2: @@ -147,7 +178,7 @@ static void tl_to_cores(struct sysinfo_15_1_x *info) core->id = tle->container.id; break; case 0: - add_cpus_to_mask(&tle->cpu, book, core); + add_cpus_to_mask(&tle->cpu, book, core, z10); break; default: clear_masks(); @@ -328,8 +359,8 @@ void __init s390_init_cpu_topology(void) for (i = 0; i < TOPOLOGY_NR_MAG; i++) printk(" %d", info->mag[i]); printk(" / %d\n", info->mnest); - alloc_masks(info, &core_info, 2); + alloc_masks(info, &core_info, 1); #ifdef CONFIG_SCHED_BOOK - alloc_masks(info, &book_info, 3); + alloc_masks(info, &book_info, 2); #endif } -- GitLab From cf5f0acf3935c91379e709a71ecf68805d366659 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 13 Oct 2011 16:52:28 +0200 Subject: [PATCH 0266/3089] sched: Add a comment to effective_load() since it's a pain Every time I have to stare at this function I need to completely reverse engineer its workings, about time I write a comment explaining the thing. Collected bits and pieces from previous changelogs, mostly: 4be9daaa1b33701f011f4117f22dc1e45a3e6e34 83378269a5fad98f562ebc0f09c349575e6cbfe1 Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1318518057.27731.2.camel@twins Signed-off-by: Ingo Molnar --- kernel/sched_fair.c | 113 +++++++++++++++++++++++++++++++++++++------- 1 file changed, 95 insertions(+), 18 deletions(-) diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 5c9e67923b7cf..aba20f495188f 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -772,19 +772,32 @@ static void update_cfs_load(struct cfs_rq *cfs_rq, int global_update) list_del_leaf_cfs_rq(cfs_rq); } +static inline long calc_tg_weight(struct task_group *tg, struct cfs_rq *cfs_rq) +{ + long tg_weight; + + /* + * Use this CPU's actual weight instead of the last load_contribution + * to gain a more accurate current total weight. See + * update_cfs_rq_load_contribution(). + */ + tg_weight = atomic_read(&tg->load_weight); + tg_weight -= cfs_rq->load_contribution; + tg_weight += cfs_rq->load.weight; + + return tg_weight; +} + static long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg) { - long load_weight, load, shares; + long tg_weight, load, shares; + tg_weight = calc_tg_weight(tg, cfs_rq); load = cfs_rq->load.weight; - load_weight = atomic_read(&tg->load_weight); - load_weight += load; - load_weight -= cfs_rq->load_contribution; - shares = (tg->shares * load); - if (load_weight) - shares /= load_weight; + if (tg_weight) + shares /= tg_weight; if (shares < MIN_SHARES) shares = MIN_SHARES; @@ -2036,36 +2049,100 @@ static void task_waking_fair(struct task_struct *p) * Adding load to a group doesn't make a group heavier, but can cause movement * of group shares between cpus. Assuming the shares were perfectly aligned one * can calculate the shift in shares. + * + * Calculate the effective load difference if @wl is added (subtracted) to @tg + * on this @cpu and results in a total addition (subtraction) of @wg to the + * total group weight. + * + * Given a runqueue weight distribution (rw_i) we can compute a shares + * distribution (s_i) using: + * + * s_i = rw_i / \Sum rw_j (1) + * + * Suppose we have 4 CPUs and our @tg is a direct child of the root group and + * has 7 equal weight tasks, distributed as below (rw_i), with the resulting + * shares distribution (s_i): + * + * rw_i = { 2, 4, 1, 0 } + * s_i = { 2/7, 4/7, 1/7, 0 } + * + * As per wake_affine() we're interested in the load of two CPUs (the CPU the + * task used to run on and the CPU the waker is running on), we need to + * compute the effect of waking a task on either CPU and, in case of a sync + * wakeup, compute the effect of the current task going to sleep. + * + * So for a change of @wl to the local @cpu with an overall group weight change + * of @wl we can compute the new shares distribution (s'_i) using: + * + * s'_i = (rw_i + @wl) / (@wg + \Sum rw_j) (2) + * + * Suppose we're interested in CPUs 0 and 1, and want to compute the load + * differences in waking a task to CPU 0. The additional task changes the + * weight and shares distributions like: + * + * rw'_i = { 3, 4, 1, 0 } + * s'_i = { 3/8, 4/8, 1/8, 0 } + * + * We can then compute the difference in effective weight by using: + * + * dw_i = S * (s'_i - s_i) (3) + * + * Where 'S' is the group weight as seen by its parent. + * + * Therefore the effective change in loads on CPU 0 would be 5/56 (3/8 - 2/7) + * times the weight of the group. The effect on CPU 1 would be -4/56 (4/8 - + * 4/7) times the weight of the group. */ static long effective_load(struct task_group *tg, int cpu, long wl, long wg) { struct sched_entity *se = tg->se[cpu]; - if (!tg->parent) + if (!tg->parent) /* the trivial, non-cgroup case */ return wl; for_each_sched_entity(se) { - long lw, w; + long w, W; tg = se->my_q->tg; - w = se->my_q->load.weight; - /* use this cpu's instantaneous contribution */ - lw = atomic_read(&tg->load_weight); - lw -= se->my_q->load_contribution; - lw += w + wg; + /* + * W = @wg + \Sum rw_j + */ + W = wg + calc_tg_weight(tg, se->my_q); - wl += w; + /* + * w = rw_i + @wl + */ + w = se->my_q->load.weight + wl; - if (lw > 0 && wl < lw) - wl = (wl * tg->shares) / lw; + /* + * wl = S * s'_i; see (2) + */ + if (W > 0 && w < W) + wl = (w * tg->shares) / W; else wl = tg->shares; - /* zero point is MIN_SHARES */ + /* + * Per the above, wl is the new se->load.weight value; since + * those are clipped to [MIN_SHARES, ...) do so now. See + * calc_cfs_shares(). + */ if (wl < MIN_SHARES) wl = MIN_SHARES; + + /* + * wl = dw_i = S * (s'_i - s_i); see (3) + */ wl -= se->load.weight; + + /* + * Recursively apply this logic to all parent groups to compute + * the final effective load change on the root group. Since + * only the @tg group gets extra weight, all parent groups can + * only redistribute existing shares. @wl is the shift in shares + * resulting from this level per the above. + */ wg = 0; } -- GitLab From 461819ac8ee950ce027c72a066156a3df9e60c7e Mon Sep 17 00:00:00 2001 From: Hui Kang Date: Tue, 11 Oct 2011 23:00:59 -0400 Subject: [PATCH 0267/3089] sched_fair: Fix a typo in the comment describing update_sd_lb_stats Signed-off-by: Hui Kang Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1318388459-4427-1-git-send-email-hkang.sunysb@gmail.com Signed-off-by: Ingo Molnar --- kernel/sched_fair.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index aba20f495188f..7e51b5bb27cc1 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -3588,7 +3588,7 @@ static bool update_sd_pick_busiest(struct sched_domain *sd, } /** - * update_sd_lb_stats - Update sched_group's statistics for load balancing. + * update_sd_lb_stats - Update sched_domain's statistics for load balancing. * @sd: sched_domain whose statistics are to be updated. * @this_cpu: Cpu for which load balance is currently performed. * @idle: Idle status of this_cpu -- GitLab From c6dc7f055d333ef35f397b8d7c3abcd1918bf8cb Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Thu, 6 Oct 2011 15:22:46 -0400 Subject: [PATCH 0268/3089] sched: Document wait_for_completion_*() return values The return-value convention for these functions varies depending on whether they're interruptible or can timeout. It can be a little confusing--document it. Signed-off-by: J. Bruce Fields Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20111006192246.GB28026@fieldses.org Signed-off-by: Ingo Molnar --- kernel/sched.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/kernel/sched.c b/kernel/sched.c index 0e9344a71be33..3d2c436959a17 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4810,6 +4810,9 @@ EXPORT_SYMBOL(wait_for_completion); * This waits for either a completion of a specific task to be signaled or for a * specified timeout to expire. The timeout is in jiffies. It is not * interruptible. + * + * The return value is 0 if timed out, and positive (at least 1, or number of + * jiffies left till timeout) if completed. */ unsigned long __sched wait_for_completion_timeout(struct completion *x, unsigned long timeout) @@ -4824,6 +4827,8 @@ EXPORT_SYMBOL(wait_for_completion_timeout); * * This waits for completion of a specific task to be signaled. It is * interruptible. + * + * The return value is -ERESTARTSYS if interrupted, 0 if completed. */ int __sched wait_for_completion_interruptible(struct completion *x) { @@ -4841,6 +4846,9 @@ EXPORT_SYMBOL(wait_for_completion_interruptible); * * This waits for either a completion of a specific task to be signaled or for a * specified timeout to expire. It is interruptible. The timeout is in jiffies. + * + * The return value is -ERESTARTSYS if interrupted, 0 if timed out, + * positive (at least 1, or number of jiffies left till timeout) if completed. */ long __sched wait_for_completion_interruptible_timeout(struct completion *x, @@ -4856,6 +4864,8 @@ EXPORT_SYMBOL(wait_for_completion_interruptible_timeout); * * This waits to be signaled for completion of a specific task. It can be * interrupted by a kill signal. + * + * The return value is -ERESTARTSYS if interrupted, 0 if completed. */ int __sched wait_for_completion_killable(struct completion *x) { @@ -4874,6 +4884,9 @@ EXPORT_SYMBOL(wait_for_completion_killable); * This waits for either a completion of a specific task to be * signaled or for a specified timeout to expire. It can be * interrupted by a kill signal. The timeout is in jiffies. + * + * The return value is -ERESTARTSYS if interrupted, 0 if timed out, + * positive (at least 1, or number of jiffies left till timeout) if completed. */ long __sched wait_for_completion_killable_timeout(struct completion *x, -- GitLab From 4a6184ce7a48c478dee0d8a9ed74c1fa35161858 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 6 Oct 2011 22:39:14 +0200 Subject: [PATCH 0269/3089] sched, rt: Provide means of disabling cross-cpu bandwidth sharing Normally the RT bandwidth scheme will share bandwidth across the entire root_domain. However sometimes its convenient to disable this sharing for debug purposes. Provide a simple feature switch to this end. Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- kernel/sched_features.h | 1 + kernel/sched_rt.c | 3 +++ 2 files changed, 4 insertions(+) diff --git a/kernel/sched_features.h b/kernel/sched_features.h index efa0a7b75dde7..84802245abd25 100644 --- a/kernel/sched_features.h +++ b/kernel/sched_features.h @@ -67,3 +67,4 @@ SCHED_FEAT(NONTASK_POWER, 1) SCHED_FEAT(TTWU_QUEUE, 1) SCHED_FEAT(FORCE_SD_OVERLAP, 0) +SCHED_FEAT(RT_RUNTIME_SHARE, 1) diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 056cbd2e2a27f..583a1368afe6e 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -560,6 +560,9 @@ static int balance_runtime(struct rt_rq *rt_rq) { int more = 0; + if (!sched_feat(RT_RUNTIME_SHARE)) + return more; + if (rt_rq->rt_time > rt_rq->rt_runtime) { raw_spin_unlock(&rt_rq->rt_runtime_lock); more = do_balance_runtime(rt_rq); -- GitLab From f1c6f1a7eed963ed233ba4c8b6fa8addb86c6ddc Mon Sep 17 00:00:00 2001 From: Carsten Emde Date: Wed, 26 Oct 2011 23:14:16 +0200 Subject: [PATCH 0270/3089] sched: Set the command name of the idle tasks in SMP kernels In UP systems, the idle task is initialized using the init_task structure from which the command name is taken (currently "swapper"). In SMP systems, one idle task per CPU is forked by the worker thread from which the task structure is copied. The command name is, therefore, "kworker/0:0" or "kworker/0:1", if not updated. Since such update was lacking, all idle tasks in SMP systems were incorrectly named. This longtime bug was not discovered immediately, because there is no /proc/0 entry - the bug only becomes apparent when tracing is enabled. This patch sets the command name of the idle tasks in SMP systems to the name that is used in the INIT_TASK structure suffixed by a slash and the number of the CPU. Signed-off-by: Carsten Emde Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20111026211708.768925506@osadl.org Signed-off-by: Ingo Molnar --- include/linux/init_task.h | 4 +++- kernel/sched.c | 4 ++++ 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 08ffab01e76c3..b6e5b8b000e01 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -126,6 +126,8 @@ extern struct cred init_cred; # define INIT_PERF_EVENTS(tsk) #endif +#define INIT_TASK_COMM "swapper" + /* * INIT_TASK is used to set up the first task table, touch at * your own risk!. Base=0, limit=0x1fffff (=2MB) @@ -162,7 +164,7 @@ extern struct cred init_cred; .group_leader = &tsk, \ RCU_INIT_POINTER(.real_cred, &init_cred), \ RCU_INIT_POINTER(.cred, &init_cred), \ - .comm = "swapper", \ + .comm = INIT_TASK_COMM, \ .thread = INIT_THREAD, \ .fs = &init_fs, \ .files = &init_files, \ diff --git a/kernel/sched.c b/kernel/sched.c index 3d2c436959a17..d6b149ccf925c 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -71,6 +71,7 @@ #include #include #include +#include #include #include @@ -6112,6 +6113,9 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) */ idle->sched_class = &idle_sched_class; ftrace_graph_init_idle_task(idle, cpu); +#if defined(CONFIG_SMP) + sprintf(idle->comm, "%s/%d", INIT_TASK_COMM, cpu); +#endif } /* -- GitLab From 57d1c0c03c6b48b2b96870d831b9ce6b917f53ac Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 7 Oct 2011 13:36:40 +0200 Subject: [PATCH 0271/3089] perf/x86: Fix PEBS instruction unwind Masami spotted that we always try to decode the instruction stream as 64bit instructions when running a 64bit kernel, this doesn't work for ia32-compat proglets. Use TIF_IA32 to detect if we need to use the 32bit instruction decoder. Reported-by: Masami Hiramatsu Cc: stable@kernel.org Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_intel_ds.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c index c0d238f49db84..73da6b64f5b78 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -493,6 +493,7 @@ static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs) unsigned long from = cpuc->lbr_entries[0].from; unsigned long old_to, to = cpuc->lbr_entries[0].to; unsigned long ip = regs->ip; + int is_64bit = 0; /* * We don't need to fixup if the PEBS assist is fault like @@ -544,7 +545,10 @@ static int intel_pmu_pebs_fixup_ip(struct pt_regs *regs) } else kaddr = (void *)to; - kernel_insn_init(&insn, kaddr); +#ifdef CONFIG_X86_64 + is_64bit = kernel_ip(to) || !test_thread_flag(TIF_IA32); +#endif + insn_init(&insn, kaddr, is_64bit); insn_get_length(&insn); to += insn.length; } while (to < ip); -- GitLab From 1d5f003f5a964711853514b04ddc872eec0fdc7b Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Sun, 23 Oct 2011 19:10:33 +0200 Subject: [PATCH 0272/3089] perf: Do not set task_ctx pointer in cpuctx if there are no events in the context Do not set task_ctx pointer during sched_in if there are no events associated with the context. Otherwise if during task execution total number of events in the system will become zero perf_event_context_sched_out() will not be called and cpuctx->task_ctx will be left with a stale value. Signed-off-by: Gleb Natapov Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20111023171033.GI17571@redhat.com Signed-off-by: Ingo Molnar --- kernel/events/core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 0e8457da6f955..b0c1186fd97bc 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -2173,7 +2173,8 @@ static void perf_event_context_sched_in(struct perf_event_context *ctx, perf_event_sched_in(cpuctx, ctx, task); - cpuctx->task_ctx = ctx; + if (ctx->nr_events) + cpuctx->task_ctx = ctx; perf_pmu_enable(ctx->pmu); perf_ctx_unlock(cpuctx, ctx); -- GitLab From aa2bc1ade59003a379ffc485d6da2d92ea3370a6 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 9 Nov 2011 17:56:37 +0100 Subject: [PATCH 0273/3089] perf: Don't use -ENOSPC for out of PMU resources People (Linus) objected to using -ENOSPC to signal not having enough resources on the PMU to satisfy the request. Use -EINVAL. Requested-by: Linus Torvalds Cc: Stephane Eranian Cc: Will Deacon Cc: Deng-Cheng Zhu Cc: David Daney Cc: Ralf Baechle Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-xv8geaz2zpbjhlx0svmpp28n@git.kernel.org [ merged to newer kernel, fixed up MIPS impact ] Signed-off-by: Ingo Molnar --- arch/arm/kernel/perf_event.c | 6 +++--- arch/mips/kernel/perf_event_mipsxx.c | 8 ++++---- arch/x86/kernel/cpu/perf_event.c | 10 +++++----- arch/x86/kernel/cpu/perf_event_p4.c | 2 +- 4 files changed, 13 insertions(+), 13 deletions(-) diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 24e2347be6b10..ff17b17b668ee 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -347,15 +347,15 @@ validate_group(struct perf_event *event) memset(&fake_pmu, 0, sizeof(fake_pmu)); if (!validate_event(&fake_pmu, leader)) - return -ENOSPC; + return -EINVAL; list_for_each_entry(sibling, &leader->sibling_list, group_entry) { if (!validate_event(&fake_pmu, sibling)) - return -ENOSPC; + return -EINVAL; } if (!validate_event(&fake_pmu, event)) - return -ENOSPC; + return -EINVAL; return 0; } diff --git a/arch/mips/kernel/perf_event_mipsxx.c b/arch/mips/kernel/perf_event_mipsxx.c index 4f2971bcf8e54..315fc0b250f8f 100644 --- a/arch/mips/kernel/perf_event_mipsxx.c +++ b/arch/mips/kernel/perf_event_mipsxx.c @@ -623,7 +623,7 @@ static int mipspmu_event_init(struct perf_event *event) if (!atomic_inc_not_zero(&active_events)) { if (atomic_read(&active_events) > MIPS_MAX_HWEVENTS) { atomic_dec(&active_events); - return -ENOSPC; + return -EINVAL; } mutex_lock(&pmu_reserve_mutex); @@ -732,15 +732,15 @@ static int validate_group(struct perf_event *event) memset(&fake_cpuc, 0, sizeof(fake_cpuc)); if (!validate_event(&fake_cpuc, leader)) - return -ENOSPC; + return -EINVAL; list_for_each_entry(sibling, &leader->sibling_list, group_entry) { if (!validate_event(&fake_cpuc, sibling)) - return -ENOSPC; + return -EINVAL; } if (!validate_event(&fake_cpuc, event)) - return -ENOSPC; + return -EINVAL; return 0; } diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 640891014b2ae..ff0e8d4987507 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -588,7 +588,7 @@ done: x86_pmu.put_event_constraints(cpuc, cpuc->event_list[i]); } } - return num ? -ENOSPC : 0; + return num ? -EINVAL : 0; } /* @@ -607,7 +607,7 @@ static int collect_events(struct cpu_hw_events *cpuc, struct perf_event *leader, if (is_x86_event(leader)) { if (n >= max_count) - return -ENOSPC; + return -EINVAL; cpuc->event_list[n] = leader; n++; } @@ -620,7 +620,7 @@ static int collect_events(struct cpu_hw_events *cpuc, struct perf_event *leader, continue; if (n >= max_count) - return -ENOSPC; + return -EINVAL; cpuc->event_list[n] = event; n++; @@ -1316,7 +1316,7 @@ static int validate_event(struct perf_event *event) c = x86_pmu.get_event_constraints(fake_cpuc, event); if (!c || !c->weight) - ret = -ENOSPC; + ret = -EINVAL; if (x86_pmu.put_event_constraints) x86_pmu.put_event_constraints(fake_cpuc, event); @@ -1341,7 +1341,7 @@ static int validate_group(struct perf_event *event) { struct perf_event *leader = event->group_leader; struct cpu_hw_events *fake_cpuc; - int ret = -ENOSPC, n; + int ret = -EINVAL, n; fake_cpuc = allocate_fake_cpuc(); if (IS_ERR(fake_cpuc)) diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c index 492bf1358a7c3..ef484d9d0a251 100644 --- a/arch/x86/kernel/cpu/perf_event_p4.c +++ b/arch/x86/kernel/cpu/perf_event_p4.c @@ -1268,7 +1268,7 @@ reserve: } done: - return num ? -ENOSPC : 0; + return num ? -EINVAL : 0; } static __initconst const struct x86_pmu p4_pmu = { -- GitLab From ed13ec58bfe0d5dc95f748e6118432cb0fa283cb Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 14 Nov 2011 10:03:25 +0100 Subject: [PATCH 0274/3089] perf/x86: Enable raw event access to Intel offcore events Now that the core offcore support is fixed up (thanks Stephane) and we have sane generic events utilizing them, re-enable the raw access to the feature as well. Note that it doesn't matter if you use event 0x1b7 or 0x1bb to specify an offcore event, either one works and neither guarantees you'll end up on a particular offcore MSR. Based on original patch from: Vince Weaver . Signed-off-by: Peter Zijlstra Cc: Vince Weaver . Cc: Stephane Eranian Link: http://lkml.kernel.org/r/alpine.DEB.2.00.1108031200390.703@cl320.eecs.utk.edu Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index ff0e8d4987507..2bda212a0010c 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -312,12 +312,8 @@ int x86_setup_perfctr(struct perf_event *event) return -EOPNOTSUPP; } - /* - * Do not allow config1 (extended registers) to propagate, - * there's no sane user-space generalization yet: - */ if (attr->type == PERF_TYPE_RAW) - return 0; + return x86_pmu_extra_regs(event->attr.config, event); if (attr->type == PERF_TYPE_HW_CACHE) return set_ext_hw_attr(hwc, event); -- GitLab From 78345d2edc25e001558f3b7c85906f645d38d23c Mon Sep 17 00:00:00 2001 From: Rabin Vincent Date: Thu, 27 Oct 2011 13:24:32 +0530 Subject: [PATCH 0275/3089] x86: Call stop_machine_text_poke() on all CPUs It appears that stop_machine_text_poke() wants to be called on all CPUs, like it's done from text_poke_smp(). Fix text_poke_smp_batch() to do this. Signed-off-by: Rabin Vincent Acked-by: Masami Hiramatsu Signed-off-by: Peter Zijlstra Cc: Mathieu Desnoyers Cc: Jason Baron Link: http://lkml.kernel.org/r/1319702072-32676-1-git-send-email-rabin@rab.in Signed-off-by: Ingo Molnar --- arch/x86/kernel/alternative.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index c63822816249e..1f84794f07593 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -738,5 +738,5 @@ void __kprobes text_poke_smp_batch(struct text_poke_param *params, int n) atomic_set(&stop_machine_first, 1); wrote_text = 0; - __stop_machine(stop_machine_text_poke, (void *)&tpp, NULL); + __stop_machine(stop_machine_text_poke, (void *)&tpp, cpu_online_mask); } -- GitLab From 9251f904f95175b4a1d8cbc0449e748f9edd7629 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Sun, 16 Oct 2011 17:15:04 +0200 Subject: [PATCH 0276/3089] perf: Carve out callchain functionality Split the callchain code from the perf events core into a new kernel/events/callchain.c file. This simplifies a bit the big core.c Signed-off-by: Borislav Petkov Cc: Arnaldo Carvalho de Melo Cc: Stephane Eranian [keep ctx recursion handling inline and use internal headers] Signed-off-by: Frederic Weisbecker Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1318778104-17152-1-git-send-email-fweisbec@gmail.com Signed-off-by: Ingo Molnar --- kernel/events/Makefile | 2 +- kernel/events/callchain.c | 191 ++++++++++++++++++++++++++++++++++ kernel/events/core.c | 209 -------------------------------------- kernel/events/internal.h | 39 ++++++- 4 files changed, 230 insertions(+), 211 deletions(-) create mode 100644 kernel/events/callchain.c diff --git a/kernel/events/Makefile b/kernel/events/Makefile index 89e5e8aa4c366..22d901f9caf44 100644 --- a/kernel/events/Makefile +++ b/kernel/events/Makefile @@ -2,5 +2,5 @@ ifdef CONFIG_FUNCTION_TRACER CFLAGS_REMOVE_core.o = -pg endif -obj-y := core.o ring_buffer.o +obj-y := core.o ring_buffer.o callchain.o obj-$(CONFIG_HAVE_HW_BREAKPOINT) += hw_breakpoint.o diff --git a/kernel/events/callchain.c b/kernel/events/callchain.c new file mode 100644 index 0000000000000..057e24b665cf7 --- /dev/null +++ b/kernel/events/callchain.c @@ -0,0 +1,191 @@ +/* + * Performance events callchain code, extracted from core.c: + * + * Copyright (C) 2008 Thomas Gleixner + * Copyright (C) 2008-2011 Red Hat, Inc., Ingo Molnar + * Copyright (C) 2008-2011 Red Hat, Inc., Peter Zijlstra + * Copyright © 2009 Paul Mackerras, IBM Corp. + * + * For licensing details see kernel-base/COPYING + */ + +#include +#include +#include "internal.h" + +struct callchain_cpus_entries { + struct rcu_head rcu_head; + struct perf_callchain_entry *cpu_entries[0]; +}; + +static DEFINE_PER_CPU(int, callchain_recursion[PERF_NR_CONTEXTS]); +static atomic_t nr_callchain_events; +static DEFINE_MUTEX(callchain_mutex); +static struct callchain_cpus_entries *callchain_cpus_entries; + + +__weak void perf_callchain_kernel(struct perf_callchain_entry *entry, + struct pt_regs *regs) +{ +} + +__weak void perf_callchain_user(struct perf_callchain_entry *entry, + struct pt_regs *regs) +{ +} + +static void release_callchain_buffers_rcu(struct rcu_head *head) +{ + struct callchain_cpus_entries *entries; + int cpu; + + entries = container_of(head, struct callchain_cpus_entries, rcu_head); + + for_each_possible_cpu(cpu) + kfree(entries->cpu_entries[cpu]); + + kfree(entries); +} + +static void release_callchain_buffers(void) +{ + struct callchain_cpus_entries *entries; + + entries = callchain_cpus_entries; + rcu_assign_pointer(callchain_cpus_entries, NULL); + call_rcu(&entries->rcu_head, release_callchain_buffers_rcu); +} + +static int alloc_callchain_buffers(void) +{ + int cpu; + int size; + struct callchain_cpus_entries *entries; + + /* + * We can't use the percpu allocation API for data that can be + * accessed from NMI. Use a temporary manual per cpu allocation + * until that gets sorted out. + */ + size = offsetof(struct callchain_cpus_entries, cpu_entries[nr_cpu_ids]); + + entries = kzalloc(size, GFP_KERNEL); + if (!entries) + return -ENOMEM; + + size = sizeof(struct perf_callchain_entry) * PERF_NR_CONTEXTS; + + for_each_possible_cpu(cpu) { + entries->cpu_entries[cpu] = kmalloc_node(size, GFP_KERNEL, + cpu_to_node(cpu)); + if (!entries->cpu_entries[cpu]) + goto fail; + } + + rcu_assign_pointer(callchain_cpus_entries, entries); + + return 0; + +fail: + for_each_possible_cpu(cpu) + kfree(entries->cpu_entries[cpu]); + kfree(entries); + + return -ENOMEM; +} + +int get_callchain_buffers(void) +{ + int err = 0; + int count; + + mutex_lock(&callchain_mutex); + + count = atomic_inc_return(&nr_callchain_events); + if (WARN_ON_ONCE(count < 1)) { + err = -EINVAL; + goto exit; + } + + if (count > 1) { + /* If the allocation failed, give up */ + if (!callchain_cpus_entries) + err = -ENOMEM; + goto exit; + } + + err = alloc_callchain_buffers(); + if (err) + release_callchain_buffers(); +exit: + mutex_unlock(&callchain_mutex); + + return err; +} + +void put_callchain_buffers(void) +{ + if (atomic_dec_and_mutex_lock(&nr_callchain_events, &callchain_mutex)) { + release_callchain_buffers(); + mutex_unlock(&callchain_mutex); + } +} + +static struct perf_callchain_entry *get_callchain_entry(int *rctx) +{ + int cpu; + struct callchain_cpus_entries *entries; + + *rctx = get_recursion_context(__get_cpu_var(callchain_recursion)); + if (*rctx == -1) + return NULL; + + entries = rcu_dereference(callchain_cpus_entries); + if (!entries) + return NULL; + + cpu = smp_processor_id(); + + return &entries->cpu_entries[cpu][*rctx]; +} + +static void +put_callchain_entry(int rctx) +{ + put_recursion_context(__get_cpu_var(callchain_recursion), rctx); +} + +struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) +{ + int rctx; + struct perf_callchain_entry *entry; + + + entry = get_callchain_entry(&rctx); + if (rctx == -1) + return NULL; + + if (!entry) + goto exit_put; + + entry->nr = 0; + + if (!user_mode(regs)) { + perf_callchain_store(entry, PERF_CONTEXT_KERNEL); + perf_callchain_kernel(entry, regs); + if (current->mm) + regs = task_pt_regs(current); + else + regs = NULL; + } + + if (regs) { + perf_callchain_store(entry, PERF_CONTEXT_USER); + perf_callchain_user(entry, regs); + } + +exit_put: + put_callchain_entry(rctx); + + return entry; +} diff --git a/kernel/events/core.c b/kernel/events/core.c index 0e8457da6f955..eadac69265fc6 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -2569,215 +2569,6 @@ static u64 perf_event_read(struct perf_event *event) return perf_event_count(event); } -/* - * Callchain support - */ - -struct callchain_cpus_entries { - struct rcu_head rcu_head; - struct perf_callchain_entry *cpu_entries[0]; -}; - -static DEFINE_PER_CPU(int, callchain_recursion[PERF_NR_CONTEXTS]); -static atomic_t nr_callchain_events; -static DEFINE_MUTEX(callchain_mutex); -struct callchain_cpus_entries *callchain_cpus_entries; - - -__weak void perf_callchain_kernel(struct perf_callchain_entry *entry, - struct pt_regs *regs) -{ -} - -__weak void perf_callchain_user(struct perf_callchain_entry *entry, - struct pt_regs *regs) -{ -} - -static void release_callchain_buffers_rcu(struct rcu_head *head) -{ - struct callchain_cpus_entries *entries; - int cpu; - - entries = container_of(head, struct callchain_cpus_entries, rcu_head); - - for_each_possible_cpu(cpu) - kfree(entries->cpu_entries[cpu]); - - kfree(entries); -} - -static void release_callchain_buffers(void) -{ - struct callchain_cpus_entries *entries; - - entries = callchain_cpus_entries; - rcu_assign_pointer(callchain_cpus_entries, NULL); - call_rcu(&entries->rcu_head, release_callchain_buffers_rcu); -} - -static int alloc_callchain_buffers(void) -{ - int cpu; - int size; - struct callchain_cpus_entries *entries; - - /* - * We can't use the percpu allocation API for data that can be - * accessed from NMI. Use a temporary manual per cpu allocation - * until that gets sorted out. - */ - size = offsetof(struct callchain_cpus_entries, cpu_entries[nr_cpu_ids]); - - entries = kzalloc(size, GFP_KERNEL); - if (!entries) - return -ENOMEM; - - size = sizeof(struct perf_callchain_entry) * PERF_NR_CONTEXTS; - - for_each_possible_cpu(cpu) { - entries->cpu_entries[cpu] = kmalloc_node(size, GFP_KERNEL, - cpu_to_node(cpu)); - if (!entries->cpu_entries[cpu]) - goto fail; - } - - rcu_assign_pointer(callchain_cpus_entries, entries); - - return 0; - -fail: - for_each_possible_cpu(cpu) - kfree(entries->cpu_entries[cpu]); - kfree(entries); - - return -ENOMEM; -} - -static int get_callchain_buffers(void) -{ - int err = 0; - int count; - - mutex_lock(&callchain_mutex); - - count = atomic_inc_return(&nr_callchain_events); - if (WARN_ON_ONCE(count < 1)) { - err = -EINVAL; - goto exit; - } - - if (count > 1) { - /* If the allocation failed, give up */ - if (!callchain_cpus_entries) - err = -ENOMEM; - goto exit; - } - - err = alloc_callchain_buffers(); - if (err) - release_callchain_buffers(); -exit: - mutex_unlock(&callchain_mutex); - - return err; -} - -static void put_callchain_buffers(void) -{ - if (atomic_dec_and_mutex_lock(&nr_callchain_events, &callchain_mutex)) { - release_callchain_buffers(); - mutex_unlock(&callchain_mutex); - } -} - -static int get_recursion_context(int *recursion) -{ - int rctx; - - if (in_nmi()) - rctx = 3; - else if (in_irq()) - rctx = 2; - else if (in_softirq()) - rctx = 1; - else - rctx = 0; - - if (recursion[rctx]) - return -1; - - recursion[rctx]++; - barrier(); - - return rctx; -} - -static inline void put_recursion_context(int *recursion, int rctx) -{ - barrier(); - recursion[rctx]--; -} - -static struct perf_callchain_entry *get_callchain_entry(int *rctx) -{ - int cpu; - struct callchain_cpus_entries *entries; - - *rctx = get_recursion_context(__get_cpu_var(callchain_recursion)); - if (*rctx == -1) - return NULL; - - entries = rcu_dereference(callchain_cpus_entries); - if (!entries) - return NULL; - - cpu = smp_processor_id(); - - return &entries->cpu_entries[cpu][*rctx]; -} - -static void -put_callchain_entry(int rctx) -{ - put_recursion_context(__get_cpu_var(callchain_recursion), rctx); -} - -static struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) -{ - int rctx; - struct perf_callchain_entry *entry; - - - entry = get_callchain_entry(&rctx); - if (rctx == -1) - return NULL; - - if (!entry) - goto exit_put; - - entry->nr = 0; - - if (!user_mode(regs)) { - perf_callchain_store(entry, PERF_CONTEXT_KERNEL); - perf_callchain_kernel(entry, regs); - if (current->mm) - regs = task_pt_regs(current); - else - regs = NULL; - } - - if (regs) { - perf_callchain_store(entry, PERF_CONTEXT_USER); - perf_callchain_user(entry, regs); - } - -exit_put: - put_callchain_entry(rctx); - - return entry; -} - /* * Initialize the perf_event context in a task_struct: */ diff --git a/kernel/events/internal.h b/kernel/events/internal.h index 09097dd8116c0..be4a43f6de4fb 100644 --- a/kernel/events/internal.h +++ b/kernel/events/internal.h @@ -1,6 +1,10 @@ #ifndef _KERNEL_EVENTS_INTERNAL_H #define _KERNEL_EVENTS_INTERNAL_H +#include + +/* Buffer handling */ + #define RING_BUFFER_WRITABLE 0x01 struct ring_buffer { @@ -64,7 +68,7 @@ static inline int page_order(struct ring_buffer *rb) } #endif -static unsigned long perf_data_size(struct ring_buffer *rb) +static inline unsigned long perf_data_size(struct ring_buffer *rb) { return rb->nr_pages << (PAGE_SHIFT + page_order(rb)); } @@ -93,4 +97,37 @@ __output_copy(struct perf_output_handle *handle, } while (len); } +/* Callchain handling */ +extern struct perf_callchain_entry *perf_callchain(struct pt_regs *regs); +extern int get_callchain_buffers(void); +extern void put_callchain_buffers(void); + +static inline int get_recursion_context(int *recursion) +{ + int rctx; + + if (in_nmi()) + rctx = 3; + else if (in_irq()) + rctx = 2; + else if (in_softirq()) + rctx = 1; + else + rctx = 0; + + if (recursion[rctx]) + return -1; + + recursion[rctx]++; + barrier(); + + return rctx; +} + +static inline void put_recursion_context(int *recursion, int rctx) +{ + barrier(); + recursion[rctx]--; +} + #endif /* _KERNEL_EVENTS_INTERNAL_H */ -- GitLab From 5d81e5cfb37a174e8ddc0413e2e70cdf05807ace Mon Sep 17 00:00:00 2001 From: Andrew Vagin Date: Mon, 7 Nov 2011 15:54:12 +0300 Subject: [PATCH 0277/3089] events: Don't divide events if it has field period This patch solves the following problem: Now some samples may be lost due to throttling. The number of samples is restricted by sysctl_perf_event_sample_rate/HZ. A trace event is divided on some samples according to event's period. I don't sure, that we should generate more than one sample on each trace event. I think the better way to use SAMPLE_PERIOD. E.g.: I want to trace when a process sleeps. I created a process, which sleeps for 1ms and for 4ms. perf got 100 events in both cases. swapper 0 [000] 1141.371830: sched_stat_sleep: comm=foo pid=1801 delay=1386750 [ns] swapper 0 [000] 1141.369444: sched_stat_sleep: comm=foo pid=1801 delay=4499585 [ns] In the first case a kernel want to send 4499585 events and in the second case it wants to send 1386750 events. perf-reports shows that process sleeps in both places equal time. It's bug. With this patch kernel generates one event on each "sleep" and the time slice is saved in the field "period". Perf knows how handle it. Signed-off-by: Andrew Vagin Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/1320670457-2633428-3-git-send-email-avagin@openvz.org Signed-off-by: Ingo Molnar --- kernel/events/core.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index eadac69265fc6..8d9dea56c262b 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -4528,7 +4528,6 @@ static void perf_swevent_overflow(struct perf_event *event, u64 overflow, struct hw_perf_event *hwc = &event->hw; int throttle = 0; - data->period = event->hw.last_period; if (!overflow) overflow = perf_swevent_set_period(event); @@ -4562,6 +4561,12 @@ static void perf_swevent_event(struct perf_event *event, u64 nr, if (!is_sampling_event(event)) return; + if ((event->attr.sample_type & PERF_SAMPLE_PERIOD) && !event->attr.freq) { + data->period = nr; + return perf_swevent_overflow(event, 1, data, regs); + } else + data->period = event->hw.last_period; + if (nr == 1 && hwc->sample_period == 1 && !event->attr.freq) return perf_swevent_overflow(event, 1, data, regs); -- GitLab From 94d24fc47219219b5aa23b45956cc37ee5aa5b01 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 7 Jun 2011 11:17:30 +0200 Subject: [PATCH 0278/3089] printk, lockdep: Disable lock debugging on zap_locks() zap_locks() is used by printk() in a last ditch effort to get data out, clearly we cannot trust lock state after this so make it disable lock debugging. Also don't treat printk recursion through lockdep as a normal recursion bug but try hard to get the lockdep splat out. Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/n/tip-kqxwmo4xz37e1s8w0xopvr0q@git.kernel.org Signed-off-by: Ingo Molnar --- include/linux/lockdep.h | 4 ++++ kernel/printk.c | 3 ++- 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index b6a56e37284c4..d36619ead3bae 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -343,6 +343,8 @@ extern void lockdep_trace_alloc(gfp_t mask); #define lockdep_assert_held(l) WARN_ON(debug_locks && !lockdep_is_held(l)) +#define lockdep_recursing(tsk) ((tsk)->lockdep_recursion) + #else /* !LOCKDEP */ static inline void lockdep_off(void) @@ -392,6 +394,8 @@ struct lock_class_key { }; #define lockdep_assert_held(l) do { } while (0) +#define lockdep_recursing(tsk) (0) + #endif /* !LOCKDEP */ #ifdef CONFIG_LOCK_STAT diff --git a/kernel/printk.c b/kernel/printk.c index 1455a0d4eedd4..6d087944e72ab 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -688,6 +688,7 @@ static void zap_locks(void) oops_timestamp = jiffies; + debug_locks_off(); /* If a crash is occurring, make sure we can't deadlock */ raw_spin_lock_init(&logbuf_lock); /* And make sure that we print immediately */ @@ -856,7 +857,7 @@ asmlinkage int vprintk(const char *fmt, va_list args) * recursion and return - but flag the recursion so that * it can be printed at the next appropriate moment: */ - if (!oops_in_progress) { + if (!oops_in_progress && !lockdep_recursing(current)) { recursion_bug = 1; goto out_restore_irqs; } -- GitLab From 47ff5c95db598184122aa634fa3452c0eecea877 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 7 Jun 2011 11:17:30 +0200 Subject: [PATCH 0279/3089] printk, lockdep: Remove superfluous preempt_disable() The raw_lock_irq_{save,restore}() already implies a non-preemptibility. Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- kernel/printk.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/kernel/printk.c b/kernel/printk.c index 6d087944e72ab..ba5ee04358219 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -841,7 +841,6 @@ asmlinkage int vprintk(const char *fmt, va_list args) boot_delay_msec(); printk_delay(); - preempt_disable(); /* This stops the holder of console_sem just where we want him */ raw_local_irq_save(flags); this_cpu = smp_processor_id(); @@ -965,7 +964,6 @@ asmlinkage int vprintk(const char *fmt, va_list args) out_restore_irqs: raw_local_irq_restore(flags); - preempt_enable(); return printed_len; } EXPORT_SYMBOL(printk); -- GitLab From 1a9a8aefa8f0530c97f4606ab7a2fc01fe31e9c1 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 7 Jun 2011 11:17:30 +0200 Subject: [PATCH 0280/3089] printk, lockdep: Switch to tracked irq ops Switch to local_irq_ ops so that the irq state is properly tracked (raw_local_irq_* isn't tracked by lockdep, causing confusion). Possible now that commit dd4e5d3ac4a ("lockdep: Fix trace_[soft,hard]irqs_[on,off]() recursion") cured the reason we needed the raw_ ops. Signed-off-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- kernel/printk.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/printk.c b/kernel/printk.c index ba5ee04358219..dfd8f73dcb765 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -842,7 +842,7 @@ asmlinkage int vprintk(const char *fmt, va_list args) printk_delay(); /* This stops the holder of console_sem just where we want him */ - raw_local_irq_save(flags); + local_irq_save(flags); this_cpu = smp_processor_id(); /* @@ -962,7 +962,7 @@ asmlinkage int vprintk(const char *fmt, va_list args) lockdep_on(); out_restore_irqs: - raw_local_irq_restore(flags); + local_irq_restore(flags); return printed_len; } -- GitLab From 87cdee71166fa107c5dc8e43060eeefa533c6a3b Mon Sep 17 00:00:00 2001 From: Yong Zhang Date: Wed, 9 Nov 2011 16:07:14 +0800 Subject: [PATCH 0281/3089] lockdep: Always try to set ->class_cache in register_lock_class() lockdep_init_map() Commit ["62016250 lockdep: Add improved subclass caching"] tries to improve performance (expecially to reduce the cost of rq->lock) when using lockdep, but it fails due to lockdep_init_map() in which ->class_cache is cleared. The typical caller is lock_set_subclass(), after that class will not be cached anymore. This patch tries to achive the goal of commit 62016250 by always setting ->class_cache in register_lock_class(). === Score comparison of benchmarks === for i in `seq 1 10`; do ./perf bench -f simple sched messaging; done before: min: 0.604, max: 0.660, avg: 0.622 after: min: 0.414, max: 0.473, avg: 0.427 for i in `seq 1 10`; do ./perf bench -f simple sched messaging -g 40; done before: min: 2.347, max: 2.421, avg: 2.391 after: min: 1.652, max: 1.699, avg: 1.671 Signed-off-by: Yong Zhang Cc: Hitoshi Mitake Signed-off-by: Peter Zijlstra Link: http://lkml.kernel.org/r/20111109080714.GC8124@zhy Signed-off-by: Ingo Molnar --- kernel/lockdep.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/lockdep.c b/kernel/lockdep.c index e69434b070da3..103bed8423ff5 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -722,7 +722,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) class = look_up_lock_class(lock, subclass); if (likely(class)) - return class; + goto out_set_class_cache; /* * Debug-check: all keys must be persistent! @@ -807,6 +807,7 @@ out_unlock_set: graph_unlock(); raw_local_irq_restore(flags); +out_set_class_cache: if (!subclass || force) lock->class_cache[0] = class; else if (subclass < NR_LOCKDEP_CACHING_CLASSES) -- GitLab From b05e988ee89d03398bee7b0939a5743cd12fa387 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 14 Nov 2011 12:58:16 +0000 Subject: [PATCH 0282/3089] MAINTAINERS: The MIPS git tree has moved. Signed-off-by: Ralf Baechle --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 4808256446f29..752b731137034 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4329,7 +4329,7 @@ MIPS M: Ralf Baechle L: linux-mips@linux-mips.org W: http://www.linux-mips.org/ -T: git git://git.linux-mips.org/pub/scm/linux.git +T: git git://git.linux-mips.org/pub/scm/ralf/linux.git Q: http://patchwork.linux-mips.org/project/linux-mips/list/ S: Supported F: Documentation/mips/ -- GitLab From 4af92e7a68af7b515d274f9d33b14b8a0804a0f6 Mon Sep 17 00:00:00 2001 From: John Crispin Date: Thu, 10 Nov 2011 21:33:07 +0100 Subject: [PATCH 0283/3089] MIPS: lantiq: use export.h in favour of module.h The code located at arch/mips/lantiq/ included module.h to be able to use the EXPORT_SYMBOL* macros. These can now be directly included using export.h. Signed-off-by: John Crispin Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2937/ Signed-off-by: Ralf Baechle --- arch/mips/lantiq/clk.c | 2 +- arch/mips/lantiq/devices.c | 2 +- arch/mips/lantiq/prom.c | 2 +- arch/mips/lantiq/setup.c | 2 +- arch/mips/lantiq/xway/clk-ase.c | 2 +- arch/mips/lantiq/xway/clk-xway.c | 2 +- arch/mips/lantiq/xway/devices.c | 2 +- arch/mips/lantiq/xway/dma.c | 1 + arch/mips/lantiq/xway/gpio.c | 2 +- arch/mips/lantiq/xway/gpio_ebu.c | 2 +- arch/mips/lantiq/xway/gpio_stp.c | 2 +- arch/mips/lantiq/xway/prom-ase.c | 2 +- arch/mips/lantiq/xway/prom-xway.c | 2 +- arch/mips/lantiq/xway/reset.c | 2 +- arch/mips/pci/pci-lantiq.c | 1 + 15 files changed, 15 insertions(+), 13 deletions(-) diff --git a/arch/mips/lantiq/clk.c b/arch/mips/lantiq/clk.c index 7e9c0ffc11a51..77ed70fc2fe59 100644 --- a/arch/mips/lantiq/clk.c +++ b/arch/mips/lantiq/clk.c @@ -7,7 +7,7 @@ * Copyright (C) 2010 John Crispin */ #include -#include +#include #include #include #include diff --git a/arch/mips/lantiq/devices.c b/arch/mips/lantiq/devices.c index 44a36771c819b..de1cb2bcd79a1 100644 --- a/arch/mips/lantiq/devices.c +++ b/arch/mips/lantiq/devices.c @@ -7,7 +7,7 @@ */ #include -#include +#include #include #include #include diff --git a/arch/mips/lantiq/prom.c b/arch/mips/lantiq/prom.c index 56ba007bf1e59..e34fcfd0d5ca5 100644 --- a/arch/mips/lantiq/prom.c +++ b/arch/mips/lantiq/prom.c @@ -6,7 +6,7 @@ * Copyright (C) 2010 John Crispin */ -#include +#include #include #include #include diff --git a/arch/mips/lantiq/setup.c b/arch/mips/lantiq/setup.c index 9b8af77ed0f98..1ff6c9d6cb93e 100644 --- a/arch/mips/lantiq/setup.c +++ b/arch/mips/lantiq/setup.c @@ -7,7 +7,7 @@ */ #include -#include +#include #include #include #include diff --git a/arch/mips/lantiq/xway/clk-ase.c b/arch/mips/lantiq/xway/clk-ase.c index 22d823acd536a..652258309c9c1 100644 --- a/arch/mips/lantiq/xway/clk-ase.c +++ b/arch/mips/lantiq/xway/clk-ase.c @@ -7,7 +7,7 @@ */ #include -#include +#include #include #include diff --git a/arch/mips/lantiq/xway/clk-xway.c b/arch/mips/lantiq/xway/clk-xway.c index ddd39593c5814..696b1a3e06421 100644 --- a/arch/mips/lantiq/xway/clk-xway.c +++ b/arch/mips/lantiq/xway/clk-xway.c @@ -7,7 +7,7 @@ */ #include -#include +#include #include #include diff --git a/arch/mips/lantiq/xway/devices.c b/arch/mips/lantiq/xway/devices.c index d0e32ab2ea074..d614aa7ff07f4 100644 --- a/arch/mips/lantiq/xway/devices.c +++ b/arch/mips/lantiq/xway/devices.c @@ -7,7 +7,7 @@ */ #include -#include +#include #include #include #include diff --git a/arch/mips/lantiq/xway/dma.c b/arch/mips/lantiq/xway/dma.c index 4278a459d6c48..cbb6ae5747b97 100644 --- a/arch/mips/lantiq/xway/dma.c +++ b/arch/mips/lantiq/xway/dma.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include diff --git a/arch/mips/lantiq/xway/gpio.c b/arch/mips/lantiq/xway/gpio.c index a321451a54554..d2fa98f3c78d8 100644 --- a/arch/mips/lantiq/xway/gpio.c +++ b/arch/mips/lantiq/xway/gpio.c @@ -7,7 +7,7 @@ */ #include -#include +#include #include #include #include diff --git a/arch/mips/lantiq/xway/gpio_ebu.c b/arch/mips/lantiq/xway/gpio_ebu.c index a479355abdb92..b91c7f17f10f0 100644 --- a/arch/mips/lantiq/xway/gpio_ebu.c +++ b/arch/mips/lantiq/xway/gpio_ebu.c @@ -7,7 +7,7 @@ */ #include -#include +#include #include #include #include diff --git a/arch/mips/lantiq/xway/gpio_stp.c b/arch/mips/lantiq/xway/gpio_stp.c index 67d59d6903406..ff9991cddeaa8 100644 --- a/arch/mips/lantiq/xway/gpio_stp.c +++ b/arch/mips/lantiq/xway/gpio_stp.c @@ -9,7 +9,7 @@ #include #include -#include +#include #include #include #include diff --git a/arch/mips/lantiq/xway/prom-ase.c b/arch/mips/lantiq/xway/prom-ase.c index abe49f4db57f1..ae4959ae865c4 100644 --- a/arch/mips/lantiq/xway/prom-ase.c +++ b/arch/mips/lantiq/xway/prom-ase.c @@ -6,7 +6,7 @@ * Copyright (C) 2010 John Crispin */ -#include +#include #include #include #include diff --git a/arch/mips/lantiq/xway/prom-xway.c b/arch/mips/lantiq/xway/prom-xway.c index 1686692ac24d5..2228133ca3566 100644 --- a/arch/mips/lantiq/xway/prom-xway.c +++ b/arch/mips/lantiq/xway/prom-xway.c @@ -6,7 +6,7 @@ * Copyright (C) 2010 John Crispin */ -#include +#include #include #include #include diff --git a/arch/mips/lantiq/xway/reset.c b/arch/mips/lantiq/xway/reset.c index a1be36d0e4904..3d41f0bb5bf73 100644 --- a/arch/mips/lantiq/xway/reset.c +++ b/arch/mips/lantiq/xway/reset.c @@ -10,7 +10,7 @@ #include #include #include -#include +#include #include #include diff --git a/arch/mips/pci/pci-lantiq.c b/arch/mips/pci/pci-lantiq.c index 8656388b34bd4..be1e1afe12c3c 100644 --- a/arch/mips/pci/pci-lantiq.c +++ b/arch/mips/pci/pci-lantiq.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include -- GitLab From e5a44df85e8d78e5c2d3d2e4f59b460905691e2f Mon Sep 17 00:00:00 2001 From: Matthew Garrett Date: Fri, 11 Nov 2011 11:14:23 -0500 Subject: [PATCH 0284/3089] [SCSI] hpsa: Disable ASPM The Windows driver .inf disables ASPM on hpsa devices. Do the same because the selection of a non default ASPM policy can cause the device to hang. Signed-off-by: Matthew Garrett Cc: stable@kernel.org Acked-by: Mike Miller Signed-off-by: James Bottomley --- drivers/scsi/hpsa.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c index e76107b2ade3a..865d452542be9 100644 --- a/drivers/scsi/hpsa.c +++ b/drivers/scsi/hpsa.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include @@ -3922,6 +3923,10 @@ static int __devinit hpsa_pci_init(struct ctlr_info *h) dev_warn(&h->pdev->dev, "controller appears to be disabled\n"); return -ENODEV; } + + pci_disable_link_state(h->pdev, PCIE_LINK_STATE_L0S | + PCIE_LINK_STATE_L1 | PCIE_LINK_STATE_CLKPM); + err = pci_enable_device(h->pdev); if (err) { dev_warn(&h->pdev->dev, "unable to enable PCI device\n"); -- GitLab From 4cac2eb158c6da0c761689345c6cc5df788a6292 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Tue, 23 Aug 2011 10:16:43 -0600 Subject: [PATCH 0285/3089] PCI hotplug: shpchp: don't blindly claim non-AMD 0x7450 device IDs Previously we claimed device ID 0x7450, regardless of the vendor, which is clearly wrong. Now we'll claim that device ID only for AMD. I suspect this was just a typo in the original code, but it's possible this change will break shpchp on non-7450 AMD bridges. If so, we'll have to fix them as we find them. Reference: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=638863 Reported-by: Ralf Jung Cc: Joerg Roedel Signed-off-by: Bjorn Helgaas Signed-off-by: Jesse Barnes --- drivers/pci/hotplug/shpchp_core.c | 4 ++-- drivers/pci/hotplug/shpchp_hpc.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/pci/hotplug/shpchp_core.c b/drivers/pci/hotplug/shpchp_core.c index aca972bbfb4c8..dd7e0c51a33e5 100644 --- a/drivers/pci/hotplug/shpchp_core.c +++ b/drivers/pci/hotplug/shpchp_core.c @@ -278,8 +278,8 @@ static int get_adapter_status (struct hotplug_slot *hotplug_slot, u8 *value) static int is_shpc_capable(struct pci_dev *dev) { - if ((dev->vendor == PCI_VENDOR_ID_AMD) || (dev->device == - PCI_DEVICE_ID_AMD_GOLAM_7450)) + if (dev->vendor == PCI_VENDOR_ID_AMD && + dev->device == PCI_DEVICE_ID_AMD_GOLAM_7450) return 1; if (!pci_find_capability(dev, PCI_CAP_ID_SHPC)) return 0; diff --git a/drivers/pci/hotplug/shpchp_hpc.c b/drivers/pci/hotplug/shpchp_hpc.c index 36547f0ce305e..75ba2311b54f3 100644 --- a/drivers/pci/hotplug/shpchp_hpc.c +++ b/drivers/pci/hotplug/shpchp_hpc.c @@ -944,8 +944,8 @@ int shpc_init(struct controller *ctrl, struct pci_dev *pdev) ctrl->pci_dev = pdev; /* pci_dev of the P2P bridge */ ctrl_dbg(ctrl, "Hotplug Controller:\n"); - if ((pdev->vendor == PCI_VENDOR_ID_AMD) || (pdev->device == - PCI_DEVICE_ID_AMD_GOLAM_7450)) { + if (pdev->vendor == PCI_VENDOR_ID_AMD && + pdev->device == PCI_DEVICE_ID_AMD_GOLAM_7450) { /* amd shpc driver doesn't use Base Offset; assume 0 */ ctrl->mmio_base = pci_resource_start(pdev, 0); ctrl->mmio_size = pci_resource_len(pdev, 0); -- GitLab From b2b5ce9d1ccf1c45f8ac68e5d901112ab76ba199 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 14 Nov 2011 06:03:34 +0000 Subject: [PATCH 0286/3089] net: introduce build_skb() One of the thing we discussed during netdev 2011 conference was the idea to change some network drivers to allocate/populate their skb at RX completion time, right before feeding the skb to network stack. In old days, we allocated skbs when populating the RX ring. This means bringing into cpu cache sk_buff and skb_shared_info cache lines (since we clear/initialize them), then 'queue' skb->data to NIC. By the time NIC fills a frame in skb->data buffer and host can process it, cpu probably threw away the cache lines from its caches, because lot of things happened between the allocation and final use. So the deal would be to allocate only the data buffer for the NIC to populate its RX ring buffer. And use build_skb() at RX completion to attach a data buffer (now filled with an ethernet frame) to a new skb, initialize the skb_shared_info portion, and give the hot skb to network stack. build_skb() is the function to allocate an skb, caller providing the data buffer that should be attached to it. Drivers are expected to call skb_reserve() right after build_skb() to adjust skb->data to the Ethernet frame (usually skipping NET_SKB_PAD and NET_IP_ALIGN, but some drivers might add a hardware provided alignment) Data provided to build_skb() MUST have been allocated by a prior kmalloc() call, with enough room to add SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) bytes at the end of the data without corrupting incoming frame. data = kmalloc(NET_SKB_PAD + NET_IP_ALIGN + 1536 + SKB_DATA_ALIGN(sizeof(struct skb_shared_info)), GFP_ATOMIC); ... skb = build_skb(data); if (!skb) { recycle_data(data); } else { skb_reserve(skb, NET_SKB_PAD + NET_IP_ALIGN); ... } Signed-off-by: Eric Dumazet CC: Eilon Greenstein CC: Ben Hutchings CC: Tom Herbert CC: Jamal Hadi Salim CC: Stephen Hemminger CC: Thomas Graf CC: Herbert Xu CC: Jeff Kirsher Signed-off-by: David S. Miller --- include/linux/skbuff.h | 1 + net/core/skbuff.c | 49 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 50 insertions(+) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index fe864885c1edd..abad8a0941e86 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -540,6 +540,7 @@ extern void consume_skb(struct sk_buff *skb); extern void __kfree_skb(struct sk_buff *skb); extern struct sk_buff *__alloc_skb(unsigned int size, gfp_t priority, int fclone, int node); +extern struct sk_buff *build_skb(void *data); static inline struct sk_buff *alloc_skb(unsigned int size, gfp_t priority) { diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 18a3cebb753d3..8d2c5b32f172c 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -244,6 +244,55 @@ nodata: } EXPORT_SYMBOL(__alloc_skb); +/** + * build_skb - build a network buffer + * @data: data buffer provided by caller + * + * Allocate a new &sk_buff. Caller provides space holding head and + * skb_shared_info. @data must have been allocated by kmalloc() + * The return is the new skb buffer. + * On a failure the return is %NULL, and @data is not freed. + * Notes : + * Before IO, driver allocates only data buffer where NIC put incoming frame + * Driver should add room at head (NET_SKB_PAD) and + * MUST add room at tail (SKB_DATA_ALIGN(skb_shared_info)) + * After IO, driver calls build_skb(), to allocate sk_buff and populate it + * before giving packet to stack. + * RX rings only contains data buffers, not full skbs. + */ +struct sk_buff *build_skb(void *data) +{ + struct skb_shared_info *shinfo; + struct sk_buff *skb; + unsigned int size; + + skb = kmem_cache_alloc(skbuff_head_cache, GFP_ATOMIC); + if (!skb) + return NULL; + + size = ksize(data) - SKB_DATA_ALIGN(sizeof(struct skb_shared_info)); + + memset(skb, 0, offsetof(struct sk_buff, tail)); + skb->truesize = SKB_TRUESIZE(size); + atomic_set(&skb->users, 1); + skb->head = data; + skb->data = data; + skb_reset_tail_pointer(skb); + skb->end = skb->tail + size; +#ifdef NET_SKBUFF_DATA_USES_OFFSET + skb->mac_header = ~0U; +#endif + + /* make sure we initialize shinfo sequentially */ + shinfo = skb_shinfo(skb); + memset(shinfo, 0, offsetof(struct skb_shared_info, dataref)); + atomic_set(&shinfo->dataref, 1); + kmemcheck_annotate_variable(shinfo->destructor_arg); + + return skb; +} +EXPORT_SYMBOL(build_skb); + /** * __netdev_alloc_skb - allocate an skbuff for rx on a specific device * @dev: network device to receive on -- GitLab From e52fcb2462ac484e6dd6e68869536609f0216938 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 14 Nov 2011 06:05:34 +0000 Subject: [PATCH 0287/3089] bnx2x: uses build_skb() in receive path bnx2x uses following formula to compute its rx_buf_sz : dev->mtu + 2*L1_CACHE_BYTES + 14 + 8 + 8 + 2 Then core network adds NET_SKB_PAD and SKB_DATA_ALIGN(sizeof(struct skb_shared_info)) Final allocated size for skb head on x86_64 (L1_CACHE_BYTES = 64, MTU=1500) : 2112 bytes : SLUB/SLAB round this to 4096 bytes. Since skb truesize is then bigger than SK_MEM_QUANTUM, we have lot of false sharing because of mem_reclaim in UDP stack. One possible way to half truesize is to reduce the need by 64 bytes (2112 -> 2048 bytes) Instead of allocating a full cache line at the end of packet for alignment, we can use the fact that skb_shared_info sits at the end of skb->head, and we can use this room, if we convert bnx2x to new build_skb() infrastructure. skb_shared_info will be initialized after hardware finished its transfert, so we can eventually overwrite the final padding. Using build_skb() also reduces cache line misses in the driver, since we use cache hot skb instead of cold ones. Number of in-flight sk_buff structures is lower, they are recycled while still hot. Performance results : (820.000 pps on a rx UDP monothread benchmark, instead of 720.000 pps) Signed-off-by: Eric Dumazet CC: Eilon Greenstein CC: Ben Hutchings CC: Tom Herbert CC: Jamal Hadi Salim CC: Stephen Hemminger CC: Thomas Graf CC: Herbert Xu CC: Jeff Kirsher Acked-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x.h | 30 +- .../net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 267 +++++++++--------- .../net/ethernet/broadcom/bnx2x/bnx2x_cmn.h | 33 +-- .../ethernet/broadcom/bnx2x/bnx2x_ethtool.c | 6 +- .../net/ethernet/broadcom/bnx2x/bnx2x_main.c | 4 +- 5 files changed, 171 insertions(+), 169 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h index 4b90b518d6a6d..0f7b7a463eba0 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x.h @@ -293,8 +293,13 @@ enum { #define FCOE_TXQ_IDX(bp) (MAX_ETH_TXQ_IDX(bp)) /* fast path */ +/* + * This driver uses new build_skb() API : + * RX ring buffer contains pointer to kmalloc() data only, + * skb are built only after Hardware filled the frame. + */ struct sw_rx_bd { - struct sk_buff *skb; + u8 *data; DEFINE_DMA_UNMAP_ADDR(mapping); }; @@ -424,8 +429,8 @@ union host_hc_status_block { struct bnx2x_agg_info { /* - * First aggregation buffer is an skb, the following - are pages. - * We will preallocate the skbs for each aggregation when + * First aggregation buffer is a data buffer, the following - are pages. + * We will preallocate the data buffer for each aggregation when * we open the interface and will replace the BD at the consumer * with this one when we receive the TPA_START CQE in order to * keep the Rx BD ring consistent. @@ -439,6 +444,7 @@ struct bnx2x_agg_info { u16 parsing_flags; u16 vlan_tag; u16 len_on_bd; + u32 rxhash; }; #define Q_STATS_OFFSET32(stat_name) \ @@ -1187,10 +1193,20 @@ struct bnx2x { #define ETH_MAX_JUMBO_PACKET_SIZE 9600 /* Max supported alignment is 256 (8 shift) */ -#define BNX2X_RX_ALIGN_SHIFT ((L1_CACHE_SHIFT < 8) ? \ - L1_CACHE_SHIFT : 8) - /* FW use 2 Cache lines Alignment for start packet and size */ -#define BNX2X_FW_RX_ALIGN (2 << BNX2X_RX_ALIGN_SHIFT) +#define BNX2X_RX_ALIGN_SHIFT min(8, L1_CACHE_SHIFT) + + /* FW uses 2 Cache lines Alignment for start packet and size + * + * We assume skb_build() uses sizeof(struct skb_shared_info) bytes + * at the end of skb->data, to avoid wasting a full cache line. + * This reduces memory use (skb->truesize). + */ +#define BNX2X_FW_RX_ALIGN_START (1UL << BNX2X_RX_ALIGN_SHIFT) + +#define BNX2X_FW_RX_ALIGN_END \ + max(1UL << BNX2X_RX_ALIGN_SHIFT, \ + SKB_DATA_ALIGN(sizeof(struct skb_shared_info))) + #define BNX2X_PXP_DRAM_ALIGN (BNX2X_RX_ALIGN_SHIFT - 5) struct host_sp_status_block *def_status_blk; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c index 13dad9230dbc8..0d60b9e633ad1 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c @@ -294,8 +294,21 @@ static void bnx2x_update_sge_prod(struct bnx2x_fastpath *fp, fp->last_max_sge, fp->rx_sge_prod); } +/* Set Toeplitz hash value in the skb using the value from the + * CQE (calculated by HW). + */ +static u32 bnx2x_get_rxhash(const struct bnx2x *bp, + const struct eth_fast_path_rx_cqe *cqe) +{ + /* Set Toeplitz hash from CQE */ + if ((bp->dev->features & NETIF_F_RXHASH) && + (cqe->status_flags & ETH_FAST_PATH_RX_CQE_RSS_HASH_FLG)) + return le32_to_cpu(cqe->rss_hash_result); + return 0; +} + static void bnx2x_tpa_start(struct bnx2x_fastpath *fp, u16 queue, - struct sk_buff *skb, u16 cons, u16 prod, + u16 cons, u16 prod, struct eth_fast_path_rx_cqe *cqe) { struct bnx2x *bp = fp->bp; @@ -310,9 +323,9 @@ static void bnx2x_tpa_start(struct bnx2x_fastpath *fp, u16 queue, if (tpa_info->tpa_state != BNX2X_TPA_STOP) BNX2X_ERR("start of bin not in stop [%d]\n", queue); - /* Try to map an empty skb from the aggregation info */ + /* Try to map an empty data buffer from the aggregation info */ mapping = dma_map_single(&bp->pdev->dev, - first_buf->skb->data, + first_buf->data + NET_SKB_PAD, fp->rx_buf_size, DMA_FROM_DEVICE); /* * ...if it fails - move the skb from the consumer to the producer @@ -322,15 +335,15 @@ static void bnx2x_tpa_start(struct bnx2x_fastpath *fp, u16 queue, if (unlikely(dma_mapping_error(&bp->pdev->dev, mapping))) { /* Move the BD from the consumer to the producer */ - bnx2x_reuse_rx_skb(fp, cons, prod); + bnx2x_reuse_rx_data(fp, cons, prod); tpa_info->tpa_state = BNX2X_TPA_ERROR; return; } - /* move empty skb from pool to prod */ - prod_rx_buf->skb = first_buf->skb; + /* move empty data from pool to prod */ + prod_rx_buf->data = first_buf->data; dma_unmap_addr_set(prod_rx_buf, mapping, mapping); - /* point prod_bd to new skb */ + /* point prod_bd to new data */ prod_bd->addr_hi = cpu_to_le32(U64_HI(mapping)); prod_bd->addr_lo = cpu_to_le32(U64_LO(mapping)); @@ -344,6 +357,7 @@ static void bnx2x_tpa_start(struct bnx2x_fastpath *fp, u16 queue, tpa_info->tpa_state = BNX2X_TPA_START; tpa_info->len_on_bd = le16_to_cpu(cqe->len_on_bd); tpa_info->placement_offset = cqe->placement_offset; + tpa_info->rxhash = bnx2x_get_rxhash(bp, cqe); #ifdef BNX2X_STOP_ON_ERROR fp->tpa_queue_used |= (1 << queue); @@ -471,11 +485,12 @@ static void bnx2x_tpa_stop(struct bnx2x *bp, struct bnx2x_fastpath *fp, { struct bnx2x_agg_info *tpa_info = &fp->tpa_info[queue]; struct sw_rx_bd *rx_buf = &tpa_info->first_buf; - u8 pad = tpa_info->placement_offset; + u32 pad = tpa_info->placement_offset; u16 len = tpa_info->len_on_bd; - struct sk_buff *skb = rx_buf->skb; + struct sk_buff *skb = NULL; + u8 *data = rx_buf->data; /* alloc new skb */ - struct sk_buff *new_skb; + u8 *new_data; u8 old_tpa_state = tpa_info->tpa_state; tpa_info->tpa_state = BNX2X_TPA_STOP; @@ -486,18 +501,18 @@ static void bnx2x_tpa_stop(struct bnx2x *bp, struct bnx2x_fastpath *fp, if (old_tpa_state == BNX2X_TPA_ERROR) goto drop; - /* Try to allocate the new skb */ - new_skb = netdev_alloc_skb(bp->dev, fp->rx_buf_size); + /* Try to allocate the new data */ + new_data = kmalloc(fp->rx_buf_size + NET_SKB_PAD, GFP_ATOMIC); /* Unmap skb in the pool anyway, as we are going to change pool entry status to BNX2X_TPA_STOP even if new skb allocation fails. */ dma_unmap_single(&bp->pdev->dev, dma_unmap_addr(rx_buf, mapping), fp->rx_buf_size, DMA_FROM_DEVICE); + if (likely(new_data)) + skb = build_skb(data); - if (likely(new_skb)) { - prefetch(skb); - prefetch(((char *)(skb)) + L1_CACHE_BYTES); + if (likely(skb)) { #ifdef BNX2X_STOP_ON_ERROR if (pad + len > fp->rx_buf_size) { @@ -509,8 +524,9 @@ static void bnx2x_tpa_stop(struct bnx2x *bp, struct bnx2x_fastpath *fp, } #endif - skb_reserve(skb, pad); + skb_reserve(skb, pad + NET_SKB_PAD); skb_put(skb, len); + skb->rxhash = tpa_info->rxhash; skb->protocol = eth_type_trans(skb, bp->dev); skb->ip_summed = CHECKSUM_UNNECESSARY; @@ -526,8 +542,8 @@ static void bnx2x_tpa_stop(struct bnx2x *bp, struct bnx2x_fastpath *fp, } - /* put new skb in bin */ - rx_buf->skb = new_skb; + /* put new data in bin */ + rx_buf->data = new_data; return; } @@ -539,19 +555,6 @@ drop: fp->eth_q_stats.rx_skb_alloc_failed++; } -/* Set Toeplitz hash value in the skb using the value from the - * CQE (calculated by HW). - */ -static inline void bnx2x_set_skb_rxhash(struct bnx2x *bp, union eth_rx_cqe *cqe, - struct sk_buff *skb) -{ - /* Set Toeplitz hash from CQE */ - if ((bp->dev->features & NETIF_F_RXHASH) && - (cqe->fast_path_cqe.status_flags & - ETH_FAST_PATH_RX_CQE_RSS_HASH_FLG)) - skb->rxhash = - le32_to_cpu(cqe->fast_path_cqe.rss_hash_result); -} int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget) { @@ -594,6 +597,7 @@ int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget) u8 cqe_fp_flags; enum eth_rx_cqe_type cqe_fp_type; u16 len, pad; + u8 *data; #ifdef BNX2X_STOP_ON_ERROR if (unlikely(bp->panic)) @@ -604,13 +608,6 @@ int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget) bd_prod = RX_BD(bd_prod); bd_cons = RX_BD(bd_cons); - /* Prefetch the page containing the BD descriptor - at producer's index. It will be needed when new skb is - allocated */ - prefetch((void *)(PAGE_ALIGN((unsigned long) - (&fp->rx_desc_ring[bd_prod])) - - PAGE_SIZE + 1)); - cqe = &fp->rx_comp_ring[comp_ring_cons]; cqe_fp = &cqe->fast_path_cqe; cqe_fp_flags = cqe_fp->type_error_flags; @@ -626,125 +623,110 @@ int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget) if (unlikely(CQE_TYPE_SLOW(cqe_fp_type))) { bnx2x_sp_event(fp, cqe); goto next_cqe; + } + rx_buf = &fp->rx_buf_ring[bd_cons]; + data = rx_buf->data; - /* this is an rx packet */ - } else { - rx_buf = &fp->rx_buf_ring[bd_cons]; - skb = rx_buf->skb; - prefetch(skb); - - if (!CQE_TYPE_FAST(cqe_fp_type)) { + if (!CQE_TYPE_FAST(cqe_fp_type)) { #ifdef BNX2X_STOP_ON_ERROR - /* sanity check */ - if (fp->disable_tpa && - (CQE_TYPE_START(cqe_fp_type) || - CQE_TYPE_STOP(cqe_fp_type))) - BNX2X_ERR("START/STOP packet while " - "disable_tpa type %x\n", - CQE_TYPE(cqe_fp_type)); + /* sanity check */ + if (fp->disable_tpa && + (CQE_TYPE_START(cqe_fp_type) || + CQE_TYPE_STOP(cqe_fp_type))) + BNX2X_ERR("START/STOP packet while " + "disable_tpa type %x\n", + CQE_TYPE(cqe_fp_type)); #endif - if (CQE_TYPE_START(cqe_fp_type)) { - u16 queue = cqe_fp->queue_index; - DP(NETIF_MSG_RX_STATUS, - "calling tpa_start on queue %d\n", - queue); - - bnx2x_tpa_start(fp, queue, skb, - bd_cons, bd_prod, - cqe_fp); - - /* Set Toeplitz hash for LRO skb */ - bnx2x_set_skb_rxhash(bp, cqe, skb); - - goto next_rx; - - } else { - u16 queue = - cqe->end_agg_cqe.queue_index; - DP(NETIF_MSG_RX_STATUS, - "calling tpa_stop on queue %d\n", - queue); + if (CQE_TYPE_START(cqe_fp_type)) { + u16 queue = cqe_fp->queue_index; + DP(NETIF_MSG_RX_STATUS, + "calling tpa_start on queue %d\n", + queue); - bnx2x_tpa_stop(bp, fp, queue, - &cqe->end_agg_cqe, - comp_ring_cons); + bnx2x_tpa_start(fp, queue, + bd_cons, bd_prod, + cqe_fp); + goto next_rx; + } else { + u16 queue = + cqe->end_agg_cqe.queue_index; + DP(NETIF_MSG_RX_STATUS, + "calling tpa_stop on queue %d\n", + queue); + + bnx2x_tpa_stop(bp, fp, queue, + &cqe->end_agg_cqe, + comp_ring_cons); #ifdef BNX2X_STOP_ON_ERROR - if (bp->panic) - return 0; + if (bp->panic) + return 0; #endif - bnx2x_update_sge_prod(fp, cqe_fp); - goto next_cqe; - } + bnx2x_update_sge_prod(fp, cqe_fp); + goto next_cqe; } - /* non TPA */ - len = le16_to_cpu(cqe_fp->pkt_len); - pad = cqe_fp->placement_offset; - dma_sync_single_for_cpu(&bp->pdev->dev, + } + /* non TPA */ + len = le16_to_cpu(cqe_fp->pkt_len); + pad = cqe_fp->placement_offset; + dma_sync_single_for_cpu(&bp->pdev->dev, dma_unmap_addr(rx_buf, mapping), - pad + RX_COPY_THRESH, - DMA_FROM_DEVICE); - prefetch(((char *)(skb)) + L1_CACHE_BYTES); + pad + RX_COPY_THRESH, + DMA_FROM_DEVICE); + pad += NET_SKB_PAD; + prefetch(data + pad); /* speedup eth_type_trans() */ + /* is this an error packet? */ + if (unlikely(cqe_fp_flags & ETH_RX_ERROR_FALGS)) { + DP(NETIF_MSG_RX_ERR, + "ERROR flags %x rx packet %u\n", + cqe_fp_flags, sw_comp_cons); + fp->eth_q_stats.rx_err_discard_pkt++; + goto reuse_rx; + } - /* is this an error packet? */ - if (unlikely(cqe_fp_flags & ETH_RX_ERROR_FALGS)) { + /* Since we don't have a jumbo ring + * copy small packets if mtu > 1500 + */ + if ((bp->dev->mtu > ETH_MAX_PACKET_SIZE) && + (len <= RX_COPY_THRESH)) { + skb = netdev_alloc_skb_ip_align(bp->dev, len); + if (skb == NULL) { DP(NETIF_MSG_RX_ERR, - "ERROR flags %x rx packet %u\n", - cqe_fp_flags, sw_comp_cons); - fp->eth_q_stats.rx_err_discard_pkt++; + "ERROR packet dropped because of alloc failure\n"); + fp->eth_q_stats.rx_skb_alloc_failed++; goto reuse_rx; } - - /* Since we don't have a jumbo ring - * copy small packets if mtu > 1500 - */ - if ((bp->dev->mtu > ETH_MAX_PACKET_SIZE) && - (len <= RX_COPY_THRESH)) { - struct sk_buff *new_skb; - - new_skb = netdev_alloc_skb(bp->dev, len + pad); - if (new_skb == NULL) { - DP(NETIF_MSG_RX_ERR, - "ERROR packet dropped " - "because of alloc failure\n"); - fp->eth_q_stats.rx_skb_alloc_failed++; - goto reuse_rx; - } - - /* aligned copy */ - skb_copy_from_linear_data_offset(skb, pad, - new_skb->data + pad, len); - skb_reserve(new_skb, pad); - skb_put(new_skb, len); - - bnx2x_reuse_rx_skb(fp, bd_cons, bd_prod); - - skb = new_skb; - - } else - if (likely(bnx2x_alloc_rx_skb(bp, fp, bd_prod) == 0)) { + memcpy(skb->data, data + pad, len); + bnx2x_reuse_rx_data(fp, bd_cons, bd_prod); + } else { + if (likely(bnx2x_alloc_rx_data(bp, fp, bd_prod) == 0)) { dma_unmap_single(&bp->pdev->dev, - dma_unmap_addr(rx_buf, mapping), + dma_unmap_addr(rx_buf, mapping), fp->rx_buf_size, DMA_FROM_DEVICE); + skb = build_skb(data); + if (unlikely(!skb)) { + kfree(data); + fp->eth_q_stats.rx_skb_alloc_failed++; + goto next_rx; + } skb_reserve(skb, pad); - skb_put(skb, len); - } else { DP(NETIF_MSG_RX_ERR, "ERROR packet dropped because " "of alloc failure\n"); fp->eth_q_stats.rx_skb_alloc_failed++; reuse_rx: - bnx2x_reuse_rx_skb(fp, bd_cons, bd_prod); + bnx2x_reuse_rx_data(fp, bd_cons, bd_prod); goto next_rx; } + skb_put(skb, len); skb->protocol = eth_type_trans(skb, bp->dev); /* Set Toeplitz hash for a none-LRO skb */ - bnx2x_set_skb_rxhash(bp, cqe, skb); + skb->rxhash = bnx2x_get_rxhash(bp, cqe_fp); skb_checksum_none_assert(skb); @@ -767,7 +749,7 @@ reuse_rx: next_rx: - rx_buf->skb = NULL; + rx_buf->data = NULL; bd_cons = NEXT_RX_IDX(bd_cons); bd_prod = NEXT_RX_IDX(bd_prod); @@ -1013,9 +995,9 @@ void bnx2x_init_rx_rings(struct bnx2x *bp) struct sw_rx_bd *first_buf = &tpa_info->first_buf; - first_buf->skb = netdev_alloc_skb(bp->dev, - fp->rx_buf_size); - if (!first_buf->skb) { + first_buf->data = kmalloc(fp->rx_buf_size + NET_SKB_PAD, + GFP_ATOMIC); + if (!first_buf->data) { BNX2X_ERR("Failed to allocate TPA " "skb pool for queue[%d] - " "disabling TPA on this " @@ -1118,16 +1100,16 @@ static void bnx2x_free_rx_bds(struct bnx2x_fastpath *fp) for (i = 0; i < NUM_RX_BD; i++) { struct sw_rx_bd *rx_buf = &fp->rx_buf_ring[i]; - struct sk_buff *skb = rx_buf->skb; + u8 *data = rx_buf->data; - if (skb == NULL) + if (data == NULL) continue; dma_unmap_single(&bp->pdev->dev, dma_unmap_addr(rx_buf, mapping), fp->rx_buf_size, DMA_FROM_DEVICE); - rx_buf->skb = NULL; - dev_kfree_skb(skb); + rx_buf->data = NULL; + kfree(data); } } @@ -1509,6 +1491,7 @@ static inline void bnx2x_set_rx_buf_size(struct bnx2x *bp) for_each_queue(bp, i) { struct bnx2x_fastpath *fp = &bp->fp[i]; + u32 mtu; /* Always use a mini-jumbo MTU for the FCoE L2 ring */ if (IS_FCOE_IDX(i)) @@ -1518,13 +1501,15 @@ static inline void bnx2x_set_rx_buf_size(struct bnx2x *bp) * IP_HEADER_ALIGNMENT_PADDING to prevent a buffer * overrun attack. */ - fp->rx_buf_size = - BNX2X_FCOE_MINI_JUMBO_MTU + ETH_OVREHEAD + - BNX2X_FW_RX_ALIGN + IP_HEADER_ALIGNMENT_PADDING; + mtu = BNX2X_FCOE_MINI_JUMBO_MTU; else - fp->rx_buf_size = - bp->dev->mtu + ETH_OVREHEAD + - BNX2X_FW_RX_ALIGN + IP_HEADER_ALIGNMENT_PADDING; + mtu = bp->dev->mtu; + fp->rx_buf_size = BNX2X_FW_RX_ALIGN_START + + IP_HEADER_ALIGNMENT_PADDING + + ETH_OVREHEAD + + mtu + + BNX2X_FW_RX_ALIGN_END; + /* Note : rx_buf_size doesnt take into account NET_SKB_PAD */ } } diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h index 260226d099168..41eb17e7720f2 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.h @@ -910,26 +910,27 @@ static inline int bnx2x_alloc_rx_sge(struct bnx2x *bp, return 0; } -static inline int bnx2x_alloc_rx_skb(struct bnx2x *bp, - struct bnx2x_fastpath *fp, u16 index) +static inline int bnx2x_alloc_rx_data(struct bnx2x *bp, + struct bnx2x_fastpath *fp, u16 index) { - struct sk_buff *skb; + u8 *data; struct sw_rx_bd *rx_buf = &fp->rx_buf_ring[index]; struct eth_rx_bd *rx_bd = &fp->rx_desc_ring[index]; dma_addr_t mapping; - skb = netdev_alloc_skb(bp->dev, fp->rx_buf_size); - if (unlikely(skb == NULL)) + data = kmalloc(fp->rx_buf_size + NET_SKB_PAD, GFP_ATOMIC); + if (unlikely(data == NULL)) return -ENOMEM; - mapping = dma_map_single(&bp->pdev->dev, skb->data, fp->rx_buf_size, + mapping = dma_map_single(&bp->pdev->dev, data + NET_SKB_PAD, + fp->rx_buf_size, DMA_FROM_DEVICE); if (unlikely(dma_mapping_error(&bp->pdev->dev, mapping))) { - dev_kfree_skb_any(skb); + kfree(data); return -ENOMEM; } - rx_buf->skb = skb; + rx_buf->data = data; dma_unmap_addr_set(rx_buf, mapping, mapping); rx_bd->addr_hi = cpu_to_le32(U64_HI(mapping)); @@ -938,12 +939,12 @@ static inline int bnx2x_alloc_rx_skb(struct bnx2x *bp, return 0; } -/* note that we are not allocating a new skb, +/* note that we are not allocating a new buffer, * we are just moving one from cons to prod * we are not creating a new mapping, * so there is no need to check for dma_mapping_error(). */ -static inline void bnx2x_reuse_rx_skb(struct bnx2x_fastpath *fp, +static inline void bnx2x_reuse_rx_data(struct bnx2x_fastpath *fp, u16 cons, u16 prod) { struct sw_rx_bd *cons_rx_buf = &fp->rx_buf_ring[cons]; @@ -953,7 +954,7 @@ static inline void bnx2x_reuse_rx_skb(struct bnx2x_fastpath *fp, dma_unmap_addr_set(prod_rx_buf, mapping, dma_unmap_addr(cons_rx_buf, mapping)); - prod_rx_buf->skb = cons_rx_buf->skb; + prod_rx_buf->data = cons_rx_buf->data; *prod_bd = *cons_bd; } @@ -1029,9 +1030,9 @@ static inline void bnx2x_free_tpa_pool(struct bnx2x *bp, for (i = 0; i < last; i++) { struct bnx2x_agg_info *tpa_info = &fp->tpa_info[i]; struct sw_rx_bd *first_buf = &tpa_info->first_buf; - struct sk_buff *skb = first_buf->skb; + u8 *data = first_buf->data; - if (skb == NULL) { + if (data == NULL) { DP(NETIF_MSG_IFDOWN, "tpa bin %d empty on free\n", i); continue; } @@ -1039,8 +1040,8 @@ static inline void bnx2x_free_tpa_pool(struct bnx2x *bp, dma_unmap_single(&bp->pdev->dev, dma_unmap_addr(first_buf, mapping), fp->rx_buf_size, DMA_FROM_DEVICE); - dev_kfree_skb(skb); - first_buf->skb = NULL; + kfree(data); + first_buf->data = NULL; } } @@ -1148,7 +1149,7 @@ static inline int bnx2x_alloc_rx_bds(struct bnx2x_fastpath *fp, * fp->eth_q_stats.rx_skb_alloc_failed = 0 */ for (i = 0; i < rx_ring_size; i++) { - if (bnx2x_alloc_rx_skb(bp, fp, ring_prod) < 0) { + if (bnx2x_alloc_rx_data(bp, fp, ring_prod) < 0) { fp->eth_q_stats.rx_skb_alloc_failed++; continue; } diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c index f6402fac26fd0..ec318711f483b 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c @@ -1740,6 +1740,7 @@ static int bnx2x_run_loopback(struct bnx2x *bp, int loopback_mode) struct sw_rx_bd *rx_buf; u16 len; int rc = -ENODEV; + u8 *data; /* check the loopback mode */ switch (loopback_mode) { @@ -1865,10 +1866,9 @@ static int bnx2x_run_loopback(struct bnx2x *bp, int loopback_mode) dma_sync_single_for_cpu(&bp->pdev->dev, dma_unmap_addr(rx_buf, mapping), fp_rx->rx_buf_size, DMA_FROM_DEVICE); - skb = rx_buf->skb; - skb_reserve(skb, cqe->fast_path_cqe.placement_offset); + data = rx_buf->data + NET_SKB_PAD + cqe->fast_path_cqe.placement_offset; for (i = ETH_HLEN; i < pkt_size; i++) - if (*(skb->data + i) != (unsigned char) (i & 0xff)) + if (*(data + i) != (unsigned char) (i & 0xff)) goto test_loopback_rx_exit; rc = 0; diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 33ff60d9fec8e..80fb9b5327116 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -2789,8 +2789,8 @@ static void bnx2x_pf_rx_q_prep(struct bnx2x *bp, /* This should be a maximum number of data bytes that may be * placed on the BD (not including paddings). */ - rxq_init->buf_sz = fp->rx_buf_size - BNX2X_FW_RX_ALIGN - - IP_HEADER_ALIGNMENT_PADDING; + rxq_init->buf_sz = fp->rx_buf_size - BNX2X_FW_RX_ALIGN_START - + BNX2X_FW_RX_ALIGN_END - IP_HEADER_ALIGNMENT_PADDING; rxq_init->cl_qzone_id = fp->cl_qzone_id; rxq_init->tpa_agg_sz = tpa_agg_size; -- GitLab From 612a94d6f24eb2427eabf554392080302da664dd Mon Sep 17 00:00:00 2001 From: Rick Jones Date: Mon, 14 Nov 2011 08:13:25 +0000 Subject: [PATCH 0288/3089] Sweep the last of the active .get_drvinfo floors under ethernet/ This round of floor sweeping converts strncpy calls in various .get_drvinfo routines to the preferred strlcpy. It also does a modicum of other cleaning in those routines. Signed-off-by: Rick Jones Signed-off-by: David S. Miller --- drivers/net/ethernet/cisco/enic/enic_main.c | 8 ++++---- drivers/net/ethernet/intel/e1000/e1000_ethtool.c | 13 +++++++------ drivers/net/ethernet/intel/e1000e/ethtool.c | 16 +++++++--------- drivers/net/ethernet/intel/igb/igb_ethtool.c | 15 ++++++--------- drivers/net/ethernet/intel/igbvf/ethtool.c | 12 +++++++----- drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c | 11 +++++++---- drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 16 ++++++---------- drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 11 +++++++---- drivers/net/ethernet/sun/cassini.c | 7 +++---- 9 files changed, 54 insertions(+), 55 deletions(-) diff --git a/drivers/net/ethernet/cisco/enic/enic_main.c b/drivers/net/ethernet/cisco/enic/enic_main.c index c3786fda11dbc..1fe5df0284a65 100644 --- a/drivers/net/ethernet/cisco/enic/enic_main.c +++ b/drivers/net/ethernet/cisco/enic/enic_main.c @@ -217,11 +217,11 @@ static void enic_get_drvinfo(struct net_device *netdev, enic_dev_fw_info(enic, &fw_info); - strncpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver)); - strncpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version)); - strncpy(drvinfo->fw_version, fw_info->fw_version, + strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, DRV_VERSION, sizeof(drvinfo->version)); + strlcpy(drvinfo->fw_version, fw_info->fw_version, sizeof(drvinfo->fw_version)); - strncpy(drvinfo->bus_info, pci_name(enic->pdev), + strlcpy(drvinfo->bus_info, pci_name(enic->pdev), sizeof(drvinfo->bus_info)); } diff --git a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c index 2b223ac99c421..63faec693deb3 100644 --- a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c +++ b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c @@ -515,14 +515,15 @@ static void e1000_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) { struct e1000_adapter *adapter = netdev_priv(netdev); - char firmware_version[32]; - strncpy(drvinfo->driver, e1000_driver_name, 32); - strncpy(drvinfo->version, e1000_driver_version, 32); + strlcpy(drvinfo->driver, e1000_driver_name, + sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, e1000_driver_version, + sizeof(drvinfo->version)); - sprintf(firmware_version, "N/A"); - strncpy(drvinfo->fw_version, firmware_version, 32); - strncpy(drvinfo->bus_info, pci_name(adapter->pdev), 32); + strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version)); + strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); drvinfo->regdump_len = e1000_get_regs_len(netdev); drvinfo->eedump_len = e1000_get_eeprom_len(netdev); } diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c index 69c9d2199140a..6d8f0ed332143 100644 --- a/drivers/net/ethernet/intel/e1000e/ethtool.c +++ b/drivers/net/ethernet/intel/e1000e/ethtool.c @@ -579,26 +579,24 @@ static void e1000_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) { struct e1000_adapter *adapter = netdev_priv(netdev); - char firmware_version[32]; - strncpy(drvinfo->driver, e1000e_driver_name, - sizeof(drvinfo->driver) - 1); + strlcpy(drvinfo->driver, e1000e_driver_name, + sizeof(drvinfo->driver)); strncpy(drvinfo->version, e1000e_driver_version, - sizeof(drvinfo->version) - 1); + sizeof(drvinfo->version)); /* * EEPROM image version # is reported as firmware version # for * PCI-E controllers */ - snprintf(firmware_version, sizeof(firmware_version), "%d.%d-%d", + snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), + "%d.%d-%d", (adapter->eeprom_vers & 0xF000) >> 12, (adapter->eeprom_vers & 0x0FF0) >> 4, (adapter->eeprom_vers & 0x000F)); - strncpy(drvinfo->fw_version, firmware_version, - sizeof(drvinfo->fw_version) - 1); - strncpy(drvinfo->bus_info, pci_name(adapter->pdev), - sizeof(drvinfo->bus_info) - 1); + strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); drvinfo->regdump_len = e1000_get_regs_len(netdev); drvinfo->eedump_len = e1000_get_eeprom_len(netdev); } diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c index 43873eba2f63e..e9335efac386b 100644 --- a/drivers/net/ethernet/intel/igb/igb_ethtool.c +++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c @@ -673,25 +673,22 @@ static void igb_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) { struct igb_adapter *adapter = netdev_priv(netdev); - char firmware_version[32]; u16 eeprom_data; - strncpy(drvinfo->driver, igb_driver_name, sizeof(drvinfo->driver) - 1); - strncpy(drvinfo->version, igb_driver_version, - sizeof(drvinfo->version) - 1); + strlcpy(drvinfo->driver, igb_driver_name, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, igb_driver_version, sizeof(drvinfo->version)); /* EEPROM image version # is reported as firmware version # for * 82575 controllers */ adapter->hw.nvm.ops.read(&adapter->hw, 5, 1, &eeprom_data); - sprintf(firmware_version, "%d.%d-%d", + snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), + "%d.%d-%d", (eeprom_data & 0xF000) >> 12, (eeprom_data & 0x0FF0) >> 4, eeprom_data & 0x000F); - strncpy(drvinfo->fw_version, firmware_version, - sizeof(drvinfo->fw_version) - 1); - strncpy(drvinfo->bus_info, pci_name(adapter->pdev), - sizeof(drvinfo->bus_info) - 1); + strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); drvinfo->n_stats = IGB_STATS_LEN; drvinfo->testinfo_len = IGB_TEST_LEN; drvinfo->regdump_len = igb_get_regs_len(netdev); diff --git a/drivers/net/ethernet/intel/igbvf/ethtool.c b/drivers/net/ethernet/intel/igbvf/ethtool.c index 2c25858cc0ff8..e60f1c6957af9 100644 --- a/drivers/net/ethernet/intel/igbvf/ethtool.c +++ b/drivers/net/ethernet/intel/igbvf/ethtool.c @@ -191,12 +191,14 @@ static void igbvf_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) { struct igbvf_adapter *adapter = netdev_priv(netdev); - char firmware_version[32] = "N/A"; - strncpy(drvinfo->driver, igbvf_driver_name, 32); - strncpy(drvinfo->version, igbvf_driver_version, 32); - strncpy(drvinfo->fw_version, firmware_version, 32); - strncpy(drvinfo->bus_info, pci_name(adapter->pdev), 32); + strlcpy(drvinfo->driver, igbvf_driver_name, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, igbvf_driver_version, + sizeof(drvinfo->version)); + strlcpy(drvinfo->fw_version, "N/A", + sizeof(drvinfo->fw_version)); + strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); drvinfo->regdump_len = igbvf_get_regs_len(netdev); drvinfo->eedump_len = igbvf_get_eeprom_len(netdev); } diff --git a/drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c b/drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c index 9dfce7dff79b8..96fcb0e064503 100644 --- a/drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c +++ b/drivers/net/ethernet/intel/ixgb/ixgb_ethtool.c @@ -473,10 +473,13 @@ ixgb_get_drvinfo(struct net_device *netdev, { struct ixgb_adapter *adapter = netdev_priv(netdev); - strncpy(drvinfo->driver, ixgb_driver_name, 32); - strncpy(drvinfo->version, ixgb_driver_version, 32); - strncpy(drvinfo->fw_version, "N/A", 32); - strncpy(drvinfo->bus_info, pci_name(adapter->pdev), 32); + strlcpy(drvinfo->driver, ixgb_driver_name, + sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, ixgb_driver_version, + sizeof(drvinfo->version)); + strlcpy(drvinfo->fw_version, "N/A", sizeof(drvinfo->fw_version)); + strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); drvinfo->n_stats = IXGB_STATS_LEN; drvinfo->regdump_len = ixgb_get_regs_len(netdev); drvinfo->eedump_len = ixgb_get_eeprom_len(netdev); diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c index 70d58c3849b05..91f871b6b3bae 100644 --- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c +++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c @@ -888,23 +888,19 @@ static void ixgbe_get_drvinfo(struct net_device *netdev, struct ethtool_drvinfo *drvinfo) { struct ixgbe_adapter *adapter = netdev_priv(netdev); - char firmware_version[32]; u32 nvm_track_id; - strncpy(drvinfo->driver, ixgbe_driver_name, - sizeof(drvinfo->driver) - 1); - strncpy(drvinfo->version, ixgbe_driver_version, - sizeof(drvinfo->version) - 1); + strlcpy(drvinfo->driver, ixgbe_driver_name, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, ixgbe_driver_version, + sizeof(drvinfo->version)); nvm_track_id = (adapter->eeprom_verh << 16) | adapter->eeprom_verl; - snprintf(firmware_version, sizeof(firmware_version), "0x%08x", + snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), "0x%08x", nvm_track_id); - strncpy(drvinfo->fw_version, firmware_version, - sizeof(drvinfo->fw_version) - 1); - strncpy(drvinfo->bus_info, pci_name(adapter->pdev), - sizeof(drvinfo->bus_info) - 1); + strlcpy(drvinfo->bus_info, pci_name(adapter->pdev), + sizeof(drvinfo->bus_info)); drvinfo->n_stats = IXGBE_STATS_LEN; drvinfo->testinfo_len = IXGBE_TEST_LEN; drvinfo->regdump_len = ixgbe_get_regs_len(netdev); diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c index 74e2a2a8a02bb..ee637a2009157 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c @@ -45,13 +45,16 @@ mlx4_en_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *drvinfo) struct mlx4_en_priv *priv = netdev_priv(dev); struct mlx4_en_dev *mdev = priv->mdev; - strncpy(drvinfo->driver, DRV_NAME, 32); - strncpy(drvinfo->version, DRV_VERSION " (" DRV_RELDATE ")", 32); - sprintf(drvinfo->fw_version, "%d.%d.%d", + strlcpy(drvinfo->driver, DRV_NAME, sizeof(drvinfo->driver)); + strlcpy(drvinfo->version, DRV_VERSION " (" DRV_RELDATE ")", + sizeof(drvinfo->version)); + snprintf(drvinfo->fw_version, sizeof(drvinfo->fw_version), + "%d.%d.%d", (u16) (mdev->dev->caps.fw_ver >> 32), (u16) ((mdev->dev->caps.fw_ver >> 16) & 0xffff), (u16) (mdev->dev->caps.fw_ver & 0xffff)); - strncpy(drvinfo->bus_info, pci_name(mdev->dev->pdev), 32); + strlcpy(drvinfo->bus_info, pci_name(mdev->dev->pdev), + sizeof(drvinfo->bus_info)); drvinfo->n_stats = 0; drvinfo->regdump_len = 0; drvinfo->eedump_len = 0; diff --git a/drivers/net/ethernet/sun/cassini.c b/drivers/net/ethernet/sun/cassini.c index fd40988c19a66..f10665f594c4f 100644 --- a/drivers/net/ethernet/sun/cassini.c +++ b/drivers/net/ethernet/sun/cassini.c @@ -4532,10 +4532,9 @@ static void cas_set_multicast(struct net_device *dev) static void cas_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { struct cas *cp = netdev_priv(dev); - strncpy(info->driver, DRV_MODULE_NAME, ETHTOOL_BUSINFO_LEN); - strncpy(info->version, DRV_MODULE_VERSION, ETHTOOL_BUSINFO_LEN); - info->fw_version[0] = '\0'; - strncpy(info->bus_info, pci_name(cp->pdev), ETHTOOL_BUSINFO_LEN); + strlcpy(info->driver, DRV_MODULE_NAME, sizeof(info->driver)); + strlcpy(info->version, DRV_MODULE_VERSION, sizeof(info->version)); + strlcpy(info->bus_info, pci_name(cp->pdev), sizeof(info->bus_info)); info->regdump_len = cp->casreg_len < CAS_MAX_REGS ? cp->casreg_len : CAS_MAX_REGS; info->n_stats = CAS_NUM_STAT_KEYS; -- GitLab From 4a5f4dd8595a3d3cdf75db7247b644ae37f5d460 Mon Sep 17 00:00:00 2001 From: Yevgeny Petrilin Date: Mon, 14 Nov 2011 14:25:36 -0500 Subject: [PATCH 0289/3089] mlx4_en: Remove FCS bytes from packet length. When HW doesn't remove FCS bytes they are reported in the completion byte count, we don't need to take them to skb. Signed-off-by: Yevgeny Petrilin Signed-off-by: David S. Miller --- drivers/net/ethernet/mellanox/mlx4/en_rx.c | 6 +++++- drivers/net/ethernet/mellanox/mlx4/mlx4_en.h | 1 + 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx4/en_rx.c b/drivers/net/ethernet/mellanox/mlx4/en_rx.c index b89c36dbf5b3e..c2df6c3586031 100644 --- a/drivers/net/ethernet/mellanox/mlx4/en_rx.c +++ b/drivers/net/ethernet/mellanox/mlx4/en_rx.c @@ -581,6 +581,7 @@ int mlx4_en_process_rx_cq(struct net_device *dev, struct mlx4_en_cq *cq, int bud * Packet is OK - process it. */ length = be32_to_cpu(cqe->byte_cnt); + length -= ring->fcs_del; ring->bytes += length; ring->packets++; @@ -813,8 +814,11 @@ static int mlx4_en_config_rss_qp(struct mlx4_en_priv *priv, int qpn, context->db_rec_addr = cpu_to_be64(ring->wqres.db.dma); /* Cancel FCS removal if FW allows */ - if (mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_FCS_KEEP) + if (mdev->dev->caps.flags & MLX4_DEV_CAP_FLAG_FCS_KEEP) { context->param3 |= cpu_to_be32(1 << 29); + ring->fcs_del = ETH_FCS_LEN; + } else + ring->fcs_del = 0; err = mlx4_qp_to_ready(mdev->dev, &ring->wqres.mtt, context, qp, state); if (err) { diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h index 8fda331c65dfd..207b5add3ca84 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h +++ b/drivers/net/ethernet/mellanox/mlx4/mlx4_en.h @@ -272,6 +272,7 @@ struct mlx4_en_rx_ring { u32 prod; u32 cons; u32 buf_size; + u8 fcs_del; void *buf; void *rx_info; unsigned long bytes; -- GitLab From abbd00b82a2771b0460ba2cffdb1343aa827ccde Mon Sep 17 00:00:00 2001 From: Wolfgang Grandegger Date: Mon, 14 Nov 2011 14:30:05 -0500 Subject: [PATCH 0290/3089] net/can/mscan: Fix buggy listen only mode setting This patch fixes an issue introduced recently with commit 452448f9283e1939408b397e87974a418825b0a8. CC: Marc Kleine-Budde Signed-off-by: Wolfgang Grandegger Signed-off-by: David S. Miller --- drivers/net/can/mscan/mscan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/can/mscan/mscan.c b/drivers/net/can/mscan/mscan.c index 74f3b1829e029..1c82dd8b896e7 100644 --- a/drivers/net/can/mscan/mscan.c +++ b/drivers/net/can/mscan/mscan.c @@ -581,7 +581,7 @@ static int mscan_open(struct net_device *dev) priv->open_time = jiffies; - if (ctrlmode.flags & CAN_CTRLMODE_LISTENONLY) + if (priv->can.ctrlmode & CAN_CTRLMODE_LISTENONLY) setbits8(®s->canctl1, MSCAN_LISTEN); else clrbits8(®s->canctl1, MSCAN_LISTEN); -- GitLab From a32fd63dbf15f70cafd30b306c31f117129c57f2 Mon Sep 17 00:00:00 2001 From: John Crispin Date: Mon, 14 Nov 2011 05:01:18 +0000 Subject: [PATCH 0291/3089] NET: MIPS: lantiq: fix etop compile error The Lantiq ETOP ethernet driver fails to build in 3.2-rc1 due to 2 missing header files. Signed-off-by: John Crispin Signed-off-by: David S. Miller --- drivers/net/ethernet/lantiq_etop.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/lantiq_etop.c b/drivers/net/ethernet/lantiq_etop.c index 6bb2b9506cadf..0b3567ab81215 100644 --- a/drivers/net/ethernet/lantiq_etop.c +++ b/drivers/net/ethernet/lantiq_etop.c @@ -34,6 +34,8 @@ #include #include #include +#include +#include #include -- GitLab From d71314b4ac88637f9ac2770a9f635babdf6f2ff9 Mon Sep 17 00:00:00 2001 From: Matti Vaittinen Date: Mon, 14 Nov 2011 00:14:49 +0000 Subject: [PATCH 0292/3089] IPv6 routing, NLM_F_* flag support: warn if new route is created without NLM_F_CREATE The support for NLM_F_* flags at IPv6 routing requests. Warn if NLM_F_CREATE flag is not defined for RTM_NEWROUTE request, creating new table. Later NLM_F_CREATE may be required for new route creation. Patch created against linux-3.2-rc1 Signed-off-by: Matti Vaittinen Signed-off-by: David S. Miller --- net/ipv6/route.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 8473016bba4a8..05c89be04c9f0 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1230,9 +1230,18 @@ int ip6_route_add(struct fib6_config *cfg) if (cfg->fc_metric == 0) cfg->fc_metric = IP6_RT_PRIO_USER; - table = fib6_new_table(net, cfg->fc_table); + err = -ENOBUFS; + if (NULL != cfg->fc_nlinfo.nlh && + !(cfg->fc_nlinfo.nlh->nlmsg_flags&NLM_F_CREATE)) { + table = fib6_get_table(net, cfg->fc_table); + if (table == NULL) { + printk(KERN_WARNING "IPv6: NLM_F_CREATE should be specified when creating new route\n"); + table = fib6_new_table(net, cfg->fc_table); + } + } else { + table = fib6_new_table(net, cfg->fc_table); + } if (table == NULL) { - err = -ENOBUFS; goto out; } -- GitLab From 4a287eba2de395713d8b2b2aeaa69fa086832d34 Mon Sep 17 00:00:00 2001 From: Matti Vaittinen Date: Mon, 14 Nov 2011 00:15:14 +0000 Subject: [PATCH 0293/3089] IPv6 routing, NLM_F_* flag support: REPLACE and EXCL flags support, warn about missing CREATE flag The support for NLM_F_* flags at IPv6 routing requests. If NLM_F_CREATE flag is not defined for RTM_NEWROUTE request, warning is printed, but no error is returned. Instead new route is added. Later NLM_F_CREATE may be required for new route creation. Exception is when NLM_F_REPLACE flag is given without NLM_F_CREATE, and no matching route is found. In this case it should be safe to assume that the request issuer is familiar with NLM_F_* flags, and does really not want route to be created. Specifying NLM_F_REPLACE flag will now make the kernel to search for matching route, and replace it with new one. If no route is found and NLM_F_CREATE is specified as well, then new route is created. Also, specifying NLM_F_EXCL will yield returning of error if matching route is found. Patch created against linux-3.2-rc1 Signed-off-by: Matti Vaittinen Signed-off-by: David S. Miller --- net/ipv6/ip6_fib.c | 109 ++++++++++++++++++++++++++++++++++++++------- 1 file changed, 94 insertions(+), 15 deletions(-) diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 93718f3db79b3..9239d559b41b1 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -425,7 +425,8 @@ out: static struct fib6_node * fib6_add_1(struct fib6_node *root, void *addr, int addrlen, int plen, - int offset) + int offset, int allow_create, + int replace_required) { struct fib6_node *fn, *in, *ln; struct fib6_node *pn = NULL; @@ -447,8 +448,12 @@ static struct fib6_node * fib6_add_1(struct fib6_node *root, void *addr, * Prefix match */ if (plen < fn->fn_bit || - !ipv6_prefix_equal(&key->addr, addr, fn->fn_bit)) + !ipv6_prefix_equal(&key->addr, addr, fn->fn_bit)) { + if (!allow_create) + printk(KERN_WARNING + "IPv6: NLM_F_CREATE should be set when creating new route\n"); goto insert_above; + } /* * Exact match ? @@ -477,10 +482,26 @@ static struct fib6_node * fib6_add_1(struct fib6_node *root, void *addr, fn = dir ? fn->right: fn->left; } while (fn); + if (replace_required && !allow_create) { + /* We should not create new node because + * NLM_F_REPLACE was specified without NLM_F_CREATE + * I assume it is safe to require NLM_F_CREATE when + * REPLACE flag is used! Later we may want to remove the + * check for replace_required, because according + * to netlink specification, NLM_F_CREATE + * MUST be specified if new route is created. + * That would keep IPv6 consistent with IPv4 + */ + printk(KERN_WARNING + "IPv6: NLM_F_CREATE should be set when creating new route - ignoring request\n"); + return ERR_PTR(-ENOENT); + } /* * We walked to the bottom of tree. * Create new leaf node without children. */ + if (!allow_create) + printk(KERN_WARNING "IPv6: NLM_F_CREATE should be set when creating new route\n"); ln = node_alloc(); @@ -614,6 +635,12 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt, { struct rt6_info *iter = NULL; struct rt6_info **ins; + int replace = (NULL != info && + NULL != info->nlh && + (info->nlh->nlmsg_flags&NLM_F_REPLACE)); + int add = ((NULL == info || NULL == info->nlh) || + (info->nlh->nlmsg_flags&NLM_F_CREATE)); + int found = 0; ins = &fn->leaf; @@ -626,6 +653,13 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt, /* * Same priority level */ + if (NULL != info->nlh && + (info->nlh->nlmsg_flags&NLM_F_EXCL)) + return -EEXIST; + if (replace) { + found++; + break; + } if (iter->rt6i_dev == rt->rt6i_dev && iter->rt6i_idev == rt->rt6i_idev && @@ -655,17 +689,40 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct rt6_info *rt, /* * insert node */ + if (!replace) { + if (!add) + printk(KERN_WARNING "IPv6: NLM_F_CREATE should be set when creating new route\n"); + +add: + rt->dst.rt6_next = iter; + *ins = rt; + rt->rt6i_node = fn; + atomic_inc(&rt->rt6i_ref); + inet6_rt_notify(RTM_NEWROUTE, rt, info); + info->nl_net->ipv6.rt6_stats->fib_rt_entries++; + + if ((fn->fn_flags & RTN_RTINFO) == 0) { + info->nl_net->ipv6.rt6_stats->fib_route_nodes++; + fn->fn_flags |= RTN_RTINFO; + } - rt->dst.rt6_next = iter; - *ins = rt; - rt->rt6i_node = fn; - atomic_inc(&rt->rt6i_ref); - inet6_rt_notify(RTM_NEWROUTE, rt, info); - info->nl_net->ipv6.rt6_stats->fib_rt_entries++; - - if ((fn->fn_flags & RTN_RTINFO) == 0) { - info->nl_net->ipv6.rt6_stats->fib_route_nodes++; - fn->fn_flags |= RTN_RTINFO; + } else { + if (!found) { + if (add) + goto add; + printk(KERN_WARNING "IPv6: NLM_F_REPLACE set, but no existing node found!\n"); + return -ENOENT; + } + *ins = rt; + rt->rt6i_node = fn; + rt->dst.rt6_next = iter->dst.rt6_next; + atomic_inc(&rt->rt6i_ref); + inet6_rt_notify(RTM_NEWROUTE, rt, info); + rt6_release(iter); + if ((fn->fn_flags & RTN_RTINFO) == 0) { + info->nl_net->ipv6.rt6_stats->fib_route_nodes++; + fn->fn_flags |= RTN_RTINFO; + } } return 0; @@ -696,9 +753,25 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info) { struct fib6_node *fn, *pn = NULL; int err = -ENOMEM; + int allow_create = 1; + int replace_required = 0; + if (NULL != info && NULL != info->nlh) { + if (!(info->nlh->nlmsg_flags&NLM_F_CREATE)) + allow_create = 0; + if ((info->nlh->nlmsg_flags&NLM_F_REPLACE)) + replace_required = 1; + } + if (!allow_create && !replace_required) + printk(KERN_WARNING "IPv6: RTM_NEWROUTE with no NLM_F_CREATE or NLM_F_REPLACE\n"); fn = fib6_add_1(root, &rt->rt6i_dst.addr, sizeof(struct in6_addr), - rt->rt6i_dst.plen, offsetof(struct rt6_info, rt6i_dst)); + rt->rt6i_dst.plen, offsetof(struct rt6_info, rt6i_dst), + allow_create, replace_required); + + if (IS_ERR(fn)) { + err = PTR_ERR(fn); + fn = NULL; + } if (fn == NULL) goto out; @@ -736,7 +809,8 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info) sn = fib6_add_1(sfn, &rt->rt6i_src.addr, sizeof(struct in6_addr), rt->rt6i_src.plen, - offsetof(struct rt6_info, rt6i_src)); + offsetof(struct rt6_info, rt6i_src), + allow_create, replace_required); if (sn == NULL) { /* If it is failed, discard just allocated @@ -753,8 +827,13 @@ int fib6_add(struct fib6_node *root, struct rt6_info *rt, struct nl_info *info) } else { sn = fib6_add_1(fn->subtree, &rt->rt6i_src.addr, sizeof(struct in6_addr), rt->rt6i_src.plen, - offsetof(struct rt6_info, rt6i_src)); + offsetof(struct rt6_info, rt6i_src), + allow_create, replace_required); + if (IS_ERR(sn)) { + err = PTR_ERR(sn); + sn = NULL; + } if (sn == NULL) goto st_failure; } -- GitLab From 5219e4c93c281377700206ae2b3ba4d91653d2ba Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Mon, 14 Nov 2011 14:36:40 -0500 Subject: [PATCH 0294/3089] bnx2x: add endline at end of message Reported-by: Joe Perches Signed-off-by: Dmitry Kravkov Signed-off-by: David S. Miller --- drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c index 80fb9b5327116..9090afc7532b9 100644 --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c @@ -8520,7 +8520,7 @@ sp_rtnl_not_reset: * damage */ if (test_and_clear_bit(BNX2X_SP_RTNL_FAN_FAILURE, &bp->sp_rtnl_state)) { - DP(BNX2X_MSG_SP, "fan failure detected. Unloading driver"); + DP(BNX2X_MSG_SP, "fan failure detected. Unloading driver\n"); netif_device_detach(bp->dev); bnx2x_close(bp->dev); } -- GitLab From ad79eefc42d56cb851a2b28a86e481cf1161005e Mon Sep 17 00:00:00 2001 From: "RongQing.Li" Date: Mon, 14 Nov 2011 14:37:24 -0500 Subject: [PATCH 0295/3089] ipv4: fix a memory leak in ic_bootp_send_if when dev_hard_header() failed, the newly allocated skb should be freed. Signed-off-by: RongQing.Li Signed-off-by: David S. Miller --- net/ipv4/ipconfig.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/net/ipv4/ipconfig.c b/net/ipv4/ipconfig.c index 0da2afc97f32f..7f17ba8b03187 100644 --- a/net/ipv4/ipconfig.c +++ b/net/ipv4/ipconfig.c @@ -822,8 +822,13 @@ static void __init ic_bootp_send_if(struct ic_device *d, unsigned long jiffies_d skb->dev = dev; skb->protocol = htons(ETH_P_IP); if (dev_hard_header(skb, dev, ntohs(skb->protocol), - dev->broadcast, dev->dev_addr, skb->len) < 0 || - dev_queue_xmit(skb) < 0) + dev->broadcast, dev->dev_addr, skb->len) < 0) { + kfree_skb(skb); + printk("E"); + return; + } + + if (dev_queue_xmit(skb) < 0) printk("E"); } -- GitLab From 1d91a96268a0b2d7301c3ee67a784b712f34010f Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Thu, 8 Sep 2011 14:11:17 +0300 Subject: [PATCH 0296/3089] usb: gadget: udc-core: fix bug on soft_connect and srp interfaces We should not be using dev_get_drvdata() because we never call dev_set_drvdata(). Let's use container_of() as all other sysfs attributes. Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/udc-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/gadget/udc-core.c b/drivers/usb/gadget/udc-core.c index 022baeca7c94f..31e410bbe55a3 100644 --- a/drivers/usb/gadget/udc-core.c +++ b/drivers/usb/gadget/udc-core.c @@ -344,7 +344,7 @@ EXPORT_SYMBOL_GPL(usb_gadget_unregister_driver); static ssize_t usb_udc_srp_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t n) { - struct usb_udc *udc = dev_get_drvdata(dev); + struct usb_udc *udc = container_of(dev, struct usb_udc, dev); if (sysfs_streq(buf, "1")) usb_gadget_wakeup(udc->gadget); -- GitLab From 3c2d636a1c6f1f84adf77737ca716f956595ae80 Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Sun, 2 Oct 2011 21:46:47 +0300 Subject: [PATCH 0297/3089] usb: musb: gadget: don't call ->disconnect() on exit that has already being done by udc-core.c. It's unnecessary and might cause issues with some gadget drivers. Tested: Ajay Kumar Gupta Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/musb/musb_gadget.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/drivers/usb/musb/musb_gadget.c b/drivers/usb/musb/musb_gadget.c index ae4a20acef6c6..d51043acfe1ab 100644 --- a/drivers/usb/musb/musb_gadget.c +++ b/drivers/usb/musb/musb_gadget.c @@ -1999,10 +1999,6 @@ static void stop_activity(struct musb *musb, struct usb_gadget_driver *driver) nuke(&hw_ep->ep_out, -ESHUTDOWN); } } - - spin_unlock(&musb->lock); - driver->disconnect(&musb->g); - spin_lock(&musb->lock); } } -- GitLab From 6f39504de55ef69bb8d9d0a645949c5d0407faab Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Mon, 3 Oct 2011 16:39:30 +0300 Subject: [PATCH 0298/3089] usb: gadget: core: fix bug when removing gadget drivers usb_gadget_disconnect() is responsible of removing data pullups. Before doing that we must, first, tell gadget driver we're disconnecting (by calling disconnect method on gadget driver structure), unbind the gadget driver and stop the controller. Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/udc-core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/usb/gadget/udc-core.c b/drivers/usb/gadget/udc-core.c index 31e410bbe55a3..4c5ff14a7c612 100644 --- a/drivers/usb/gadget/udc-core.c +++ b/drivers/usb/gadget/udc-core.c @@ -210,10 +210,10 @@ static void usb_gadget_remove_driver(struct usb_udc *udc) kobject_uevent(&udc->dev.kobj, KOBJ_CHANGE); if (udc_is_newstyle(udc)) { - usb_gadget_disconnect(udc->gadget); + udc->driver->disconnect(udc->gadget); udc->driver->unbind(udc->gadget); usb_gadget_udc_stop(udc->gadget, udc->driver); - + usb_gadget_disconnect(udc->gadget); } else { usb_gadget_stop(udc->gadget, udc->driver); } -- GitLab From 59d81f8139bff0c2a94971ceec41edd42f16ed47 Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Mon, 3 Oct 2011 16:43:56 +0300 Subject: [PATCH 0299/3089] usb: gadget: core: allow everybody to read sysfs attributes Those are simply giving information about the current state of the UDC, nothing really fancy. We can let everybody read those. Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/udc-core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/usb/gadget/udc-core.c b/drivers/usb/gadget/udc-core.c index 4c5ff14a7c612..6939e17f45800 100644 --- a/drivers/usb/gadget/udc-core.c +++ b/drivers/usb/gadget/udc-core.c @@ -378,7 +378,7 @@ static ssize_t usb_udc_speed_show(struct device *dev, return snprintf(buf, PAGE_SIZE, "%s\n", usb_speed_string(udc->gadget->speed)); } -static DEVICE_ATTR(speed, S_IRUSR, usb_udc_speed_show, NULL); +static DEVICE_ATTR(speed, S_IRUGO, usb_udc_speed_show, NULL); #define USB_UDC_ATTR(name) \ ssize_t usb_udc_##name##_show(struct device *dev, \ @@ -389,7 +389,7 @@ ssize_t usb_udc_##name##_show(struct device *dev, \ \ return snprintf(buf, PAGE_SIZE, "%d\n", gadget->name); \ } \ -static DEVICE_ATTR(name, S_IRUSR, usb_udc_##name##_show, NULL) +static DEVICE_ATTR(name, S_IRUGO, usb_udc_##name##_show, NULL) static USB_UDC_ATTR(is_dualspeed); static USB_UDC_ATTR(is_otg); -- GitLab From d06785942de066992a3b6a570829097c23c327aa Mon Sep 17 00:00:00 2001 From: Mian Yousaf Kaukab Date: Tue, 1 Nov 2011 08:37:40 +0100 Subject: [PATCH 0300/3089] usb: musb: remove incorrectly added ARCH_U5500 define ARCH_U8500 covers both MACH_U8500 and MACH_U5500 Reported-by: Paul Bolle Signed-off-by: Mian Yousaf Kaukab Acked-by: Linus Walleij Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/musb/Kconfig | 2 +- drivers/usb/musb/musb_core.c | 3 +-- 2 files changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/usb/musb/Kconfig b/drivers/usb/musb/Kconfig index fc34b8b119105..e1187242a34ef 100644 --- a/drivers/usb/musb/Kconfig +++ b/drivers/usb/musb/Kconfig @@ -60,7 +60,7 @@ config USB_MUSB_BLACKFIN config USB_MUSB_UX500 tristate "U8500 and U5500" - depends on (ARCH_U8500 && AB8500_USB) || (ARCH_U5500) + depends on (ARCH_U8500 && AB8500_USB) endchoice diff --git a/drivers/usb/musb/musb_core.c b/drivers/usb/musb/musb_core.c index 20a28731c338e..c1fa12ec7a9ad 100644 --- a/drivers/usb/musb/musb_core.c +++ b/drivers/usb/musb/musb_core.c @@ -1477,8 +1477,7 @@ static int __init musb_core_init(u16 musb_type, struct musb *musb) /*-------------------------------------------------------------------------*/ #if defined(CONFIG_SOC_OMAP2430) || defined(CONFIG_SOC_OMAP3430) || \ - defined(CONFIG_ARCH_OMAP4) || defined(CONFIG_ARCH_U8500) || \ - defined(CONFIG_ARCH_U5500) + defined(CONFIG_ARCH_OMAP4) || defined(CONFIG_ARCH_U8500) static irqreturn_t generic_interrupt(int irq, void *__hci) { -- GitLab From 0de174b56ba70e5c0a26de9b5d2889260164d666 Mon Sep 17 00:00:00 2001 From: Luciano Coelho Date: Wed, 2 Nov 2011 12:17:59 +0200 Subject: [PATCH 0301/3089] usb: musb: hdrc: fix dependency on USB_GADGET_DUALSPEED in Kconfig USB_MUSB_HDRC depends on USB_GADGET_DUALSPEED. If HDRC is selected but DUALSPEED is not, the kernel oopses: [ 3.132781] Unable to handle kernel NULL pointer dereference at virtual address 00000000 [ 3.141296] pgd = c0004000 [ 3.144134] [00000000] *pgd=00000000 [ 3.147918] Internal error: Oops: 5 [#1] SMP [ 3.152404] Modules linked in: [ 3.155609] CPU: 0 Not tainted (3.1.0-rc9-wl+ #417) [ 3.161132] PC is at composite_setup+0x738/0xbb4 [ 3.165985] LR is at vprintk+0x400/0x47c [ 3.170135] pc : [] lr : [] psr: 60000093 [ 3.170135] sp : c065dd50 ip : dfb1f0fc fp : c065ddbc [ 3.182220] r10: 00000000 r9 : df8fcae8 r8 : df8fcaa0 [ 3.187713] r7 : 00000000 r6 : df8eaa20 r5 : dfae8ea0 r4 : 00000000 [ 3.194580] r3 : df8fcae8 r2 : 00010002 r1 : c065dc40 r0 : 00000047 [ 3.201446] Flags: nZCv IRQs off FIQs on Mode SVC_32 ISA ARM Segment kernel [ 3.209228] Control: 10c53c7d Table: 8000404a DAC: 00000015 [ 3.215270] Process swapper (pid: 0, stack limit = 0xc065c2f8) [ 3.221405] Stack: (0xc065dd50 to 0xc065e000) [...] [ 3.415405] [] (composite_setup+0x738/0xbb4) from [] (musb_g_ep0_irq+0x9d0/0xaf8) [ 3.425109] [] (musb_g_ep0_irq+0x9d0/0xaf8) from [] (musb_interrupt+0xb48/0xc74) [ 3.434722] [] (musb_interrupt+0xb48/0xc74) from [] (generic_interrupt+0x68/0x80) [ 3.444458] [] (generic_interrupt+0x68/0x80) from [] (handle_irq_event_percpu+0x9c/0x234) [ 3.454925] [] (handle_irq_event_percpu+0x9c/0x234) from [] (handle_irq_event+0x4c/0x6c) [ 3.465270] [] (handle_irq_event+0x4c/0x6c) from [] (handle_fasteoi_irq+0xd8/0x110) [ 3.475158] [] (handle_fasteoi_irq+0xd8/0x110) from [] (generic_handle_irq+0x34/0x3c) [ 3.485260] [] (generic_handle_irq+0x34/0x3c) from [] (handle_IRQ+0x88/0xc8) [ 3.494537] [] (handle_IRQ+0x88/0xc8) from [] (asm_do_IRQ+0x18/0x1c) [ 3.503051] [] (asm_do_IRQ+0x18/0x1c) from [] (__irq_svc+0x38/0xc0) This patch changes Kconfig so that USB_GADGET_DUALSPEED is selected automatically by USB_MUSB_HDRC. Signed-off-by: Luciano Coelho Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/musb/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/musb/Kconfig b/drivers/usb/musb/Kconfig index e1187242a34ef..07a03460a598c 100644 --- a/drivers/usb/musb/Kconfig +++ b/drivers/usb/musb/Kconfig @@ -11,6 +11,7 @@ config USB_MUSB_HDRC select TWL4030_USB if MACH_OMAP_3430SDP select TWL6030_USB if MACH_OMAP_4430SDP || MACH_OMAP4_PANDA select USB_OTG_UTILS + select USB_GADGET_DUALSPEED tristate 'Inventra Highspeed Dual Role Controller (TI, ADI, ...)' help Say Y here if your system has a dual role high speed USB -- GitLab From b8cbbf803d5ad0971665dd502d784148464c6d1a Mon Sep 17 00:00:00 2001 From: Yoshihiro Shimoda Date: Mon, 31 Oct 2011 16:01:29 +0900 Subject: [PATCH 0302/3089] usb: gadget: r8a66597-udc: fix for udc-newstyle The udc-newstyle needs device_register in probe() of platform_device. If it doesn't call, kernel panic happens in the sysfs_create_dir() when we run modprobe a gadget driver. [ balbi@ti.com : fix compile warning introduced by this patch ] Signed-off-by: Yoshihiro Shimoda Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/r8a66597-udc.c | 28 +++++++++------------------- 1 file changed, 9 insertions(+), 19 deletions(-) diff --git a/drivers/usb/gadget/r8a66597-udc.c b/drivers/usb/gadget/r8a66597-udc.c index 68a826a1b866e..38f99887b9356 100644 --- a/drivers/usb/gadget/r8a66597-udc.c +++ b/drivers/usb/gadget/r8a66597-udc.c @@ -1742,7 +1742,6 @@ static int r8a66597_start(struct usb_gadget *gadget, struct usb_gadget_driver *driver) { struct r8a66597 *r8a66597 = gadget_to_r8a66597(gadget); - int retval; if (!driver || driver->speed != USB_SPEED_HIGH @@ -1752,16 +1751,7 @@ static int r8a66597_start(struct usb_gadget *gadget, return -ENODEV; /* hook up the driver */ - driver->driver.bus = NULL; r8a66597->driver = driver; - r8a66597->gadget.dev.driver = &driver->driver; - - retval = device_add(&r8a66597->gadget.dev); - if (retval) { - dev_err(r8a66597_to_dev(r8a66597), "device_add error (%d)\n", - retval); - goto error; - } init_controller(r8a66597); r8a66597_bset(r8a66597, VBSE, INTENB0); @@ -1775,12 +1765,6 @@ static int r8a66597_start(struct usb_gadget *gadget, } return 0; - -error: - r8a66597->driver = NULL; - r8a66597->gadget.dev.driver = NULL; - - return retval; } static int r8a66597_stop(struct usb_gadget *gadget, @@ -1794,7 +1778,6 @@ static int r8a66597_stop(struct usb_gadget *gadget, disable_controller(r8a66597); spin_unlock_irqrestore(&r8a66597->lock, flags); - device_del(&r8a66597->gadget.dev); r8a66597->driver = NULL; return 0; } @@ -1845,6 +1828,7 @@ static int __exit r8a66597_remove(struct platform_device *pdev) clk_put(r8a66597->clk); } #endif + device_unregister(&r8a66597->gadget.dev); kfree(r8a66597); return 0; } @@ -1924,13 +1908,17 @@ static int __init r8a66597_probe(struct platform_device *pdev) r8a66597->irq_sense_low = irq_trigger == IRQF_TRIGGER_LOW; r8a66597->gadget.ops = &r8a66597_gadget_ops; - device_initialize(&r8a66597->gadget.dev); dev_set_name(&r8a66597->gadget.dev, "gadget"); r8a66597->gadget.is_dualspeed = 1; r8a66597->gadget.dev.parent = &pdev->dev; r8a66597->gadget.dev.dma_mask = pdev->dev.dma_mask; r8a66597->gadget.dev.release = pdev->dev.release; r8a66597->gadget.name = udc_name; + ret = device_register(&r8a66597->gadget.dev); + if (ret < 0) { + dev_err(&pdev->dev, "device_register failed\n"); + goto clean_up; + } init_timer(&r8a66597->timer); r8a66597->timer.function = r8a66597_timer; @@ -1945,7 +1933,7 @@ static int __init r8a66597_probe(struct platform_device *pdev) dev_err(&pdev->dev, "cannot get clock \"%s\"\n", clk_name); ret = PTR_ERR(r8a66597->clk); - goto clean_up; + goto clean_up_dev; } clk_enable(r8a66597->clk); } @@ -2014,7 +2002,9 @@ clean_up2: clk_disable(r8a66597->clk); clk_put(r8a66597->clk); } +clean_up_dev: #endif + device_unregister(&r8a66597->gadget.dev); clean_up: if (r8a66597) { if (r8a66597->sudmac_reg) -- GitLab From 05bb7013038a2b609aef14ad4e07afe031daec49 Mon Sep 17 00:00:00 2001 From: Yoshihiro Shimoda Date: Mon, 31 Oct 2011 16:01:33 +0900 Subject: [PATCH 0303/3089] usb: gadget: r8a66597-udc: fix flush fifo handling The "BCLR" in CFIFOCTR/DnFIFOCTR can flush the fifo of "CPU side" only. To flush the fifo of "SIE side", we have to use the "ACLRM" in PIPEnCTR. Signed-off-by: Yoshihiro Shimoda Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/r8a66597-udc.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/usb/gadget/r8a66597-udc.c b/drivers/usb/gadget/r8a66597-udc.c index 38f99887b9356..24f84b210ce11 100644 --- a/drivers/usb/gadget/r8a66597-udc.c +++ b/drivers/usb/gadget/r8a66597-udc.c @@ -1718,6 +1718,8 @@ static void r8a66597_fifo_flush(struct usb_ep *_ep) if (list_empty(&ep->queue) && !ep->busy) { pipe_stop(ep->r8a66597, ep->pipenum); r8a66597_bclr(ep->r8a66597, BCLR, ep->fifoctr); + r8a66597_write(ep->r8a66597, ACLRM, ep->pipectr); + r8a66597_write(ep->r8a66597, 0, ep->pipectr); } spin_unlock_irqrestore(&ep->r8a66597->lock, flags); } -- GitLab From 0e042be348b864a57a8a452aaa7ec49be5fb9e59 Mon Sep 17 00:00:00 2001 From: Alexander Aring Date: Fri, 4 Nov 2011 16:16:26 +0100 Subject: [PATCH 0304/3089] usb: gadget: fsl_udc_core: fix compile error. Fix compile error in file drivers/usb/gadget/fsl_udc_core.c. drivers/usb/gadget/fsl_udc_core.c: In function 'portscx_device_speed': drivers/usb/gadget/fsl_udc_core.c:1720: error: 'speed' undeclared (first use in this function) Introduced in commit e538dfdae85244fd2c4231725d82cc1f1bc4942c (usb: Provide usb_speed_string() function) Signed-off-by: Alexander Aring Acked-by: Michal Nazarewicz Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/fsl_udc_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/gadget/fsl_udc_core.c b/drivers/usb/gadget/fsl_udc_core.c index b2c44e1d58134..d786ba31fc07f 100644 --- a/drivers/usb/gadget/fsl_udc_core.c +++ b/drivers/usb/gadget/fsl_udc_core.c @@ -1717,7 +1717,7 @@ static void dtd_complete_irq(struct fsl_udc *udc) static inline enum usb_device_speed portscx_device_speed(u32 reg) { - switch (speed & PORTSCX_PORT_SPEED_MASK) { + switch (reg & PORTSCX_PORT_SPEED_MASK) { case PORTSCX_PORT_SPEED_HIGH: return USB_SPEED_HIGH; case PORTSCX_PORT_SPEED_FULL: -- GitLab From 7fccd480b7fe84a98ee252fa79dd92f7fff5ec2a Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Sun, 23 Oct 2011 22:55:54 -0700 Subject: [PATCH 0305/3089] usb: gadget: renesas_usbhs: fixup struct completion usage Since renesas_usbhs mod_host didn't use struct completion as static object, the warning of lockdep came out. This patch fixup this issue. Signed-off-by: Kuninori Morimoto Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/renesas_usbhs/mod_host.c | 19 +++++-------------- 1 file changed, 5 insertions(+), 14 deletions(-) diff --git a/drivers/usb/renesas_usbhs/mod_host.c b/drivers/usb/renesas_usbhs/mod_host.c index 1a7208a50afc1..9b8886867f4a3 100644 --- a/drivers/usb/renesas_usbhs/mod_host.c +++ b/drivers/usb/renesas_usbhs/mod_host.c @@ -103,7 +103,7 @@ struct usbhsh_hpriv { u32 port_stat; /* USB_PORT_STAT_xxx */ - struct completion *done; + struct completion setup_ack_done; /* see usbhsh_req_alloc/free */ struct list_head ureq_link_active; @@ -549,8 +549,7 @@ static void usbhsh_setup_stage_packet_push(struct usbhsh_hpriv *hpriv, * usbhsh_irq_setup_ack() * usbhsh_irq_setup_err() */ - DECLARE_COMPLETION(done); - hpriv->done = &done; + init_completion(&hpriv->setup_ack_done); /* copy original request */ memcpy(&req, urb->setup_packet, sizeof(struct usb_ctrlrequest)); @@ -572,8 +571,7 @@ static void usbhsh_setup_stage_packet_push(struct usbhsh_hpriv *hpriv, /* * wait setup packet ACK */ - wait_for_completion(&done); - hpriv->done = NULL; + wait_for_completion(&hpriv->setup_ack_done); dev_dbg(dev, "%s done\n", __func__); } @@ -1095,10 +1093,7 @@ static int usbhsh_irq_setup_ack(struct usbhs_priv *priv, dev_dbg(dev, "setup packet OK\n"); - if (unlikely(!hpriv->done)) - dev_err(dev, "setup ack happen without necessary data\n"); - else - complete(hpriv->done); /* see usbhsh_urb_enqueue() */ + complete(&hpriv->setup_ack_done); /* see usbhsh_urb_enqueue() */ return 0; } @@ -1111,10 +1106,7 @@ static int usbhsh_irq_setup_err(struct usbhs_priv *priv, dev_dbg(dev, "setup packet Err\n"); - if (unlikely(!hpriv->done)) - dev_err(dev, "setup err happen without necessary data\n"); - else - complete(hpriv->done); /* see usbhsh_urb_enqueue() */ + complete(&hpriv->setup_ack_done); /* see usbhsh_urb_enqueue() */ return 0; } @@ -1279,7 +1271,6 @@ int __devinit usbhs_mod_host_probe(struct usbhs_priv *priv) hpriv->mod.stop = usbhsh_stop; hpriv->pipe_info = pipe_info; hpriv->pipe_size = pipe_size; - hpriv->done = NULL; usbhsh_req_list_init(hpriv); usbhsh_port_stat_init(hpriv); -- GitLab From c9ae0c91b920a7ba91725d170aa023be8c12db7b Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Wed, 26 Oct 2011 01:21:07 -0700 Subject: [PATCH 0306/3089] usb: gadget: renesas_usbhs: fixup bogus conversion this patch fixup bogus conversion of 8a9775ab71218690ac34bed9e237e2b968857d3a (usb: gadget: renesas_usbhs: fix compile warning) Signed-off-by: Kuninori Morimoto Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/renesas_usbhs/fifo.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/usb/renesas_usbhs/fifo.c b/drivers/usb/renesas_usbhs/fifo.c index 8da685e796d14..ffdf5d15085eb 100644 --- a/drivers/usb/renesas_usbhs/fifo.c +++ b/drivers/usb/renesas_usbhs/fifo.c @@ -820,7 +820,7 @@ static int usbhsf_dma_prepare_push(struct usbhs_pkt *pkt, int *is_done) if (len % 4) /* 32bit alignment */ goto usbhsf_pio_prepare_push; - if ((*(u32 *) pkt->buf + pkt->actual) & 0x7) /* 8byte alignment */ + if ((uintptr_t)(pkt->buf + pkt->actual) & 0x7) /* 8byte alignment */ goto usbhsf_pio_prepare_push; /* get enable DMA fifo */ @@ -897,7 +897,7 @@ static int usbhsf_dma_try_pop(struct usbhs_pkt *pkt, int *is_done) if (!fifo) goto usbhsf_pio_prepare_pop; - if ((*(u32 *) pkt->buf + pkt->actual) & 0x7) /* 8byte alignment */ + if ((uintptr_t)(pkt->buf + pkt->actual) & 0x7) /* 8byte alignment */ goto usbhsf_pio_prepare_pop; ret = usbhsf_fifo_select(pipe, fifo, 0); -- GitLab From b7a8d17db9a86db1040862600cf3a02848f83844 Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Wed, 26 Oct 2011 19:33:49 -0700 Subject: [PATCH 0307/3089] usb: gadget: renesas_usbhs: fixup section mismatch warning Fix up the following section mismatch warnings: WARNING: drivers/usb/renesas_usbhs/renesas_usbhs.o(.text+0xf5d): Section mismatch in reference from the function usbhs_mod_probe() to the function .devinit.text:usbhs_mod_host_probe() The function usbhs_mod_probe() references the function __devinit usbhs_mod_host_probe(). This is often because usbhs_mod_probe lacks a __devinit annotation or the annotation of usbhs_mod_host_probe is wrong. WARNING: drivers/usb/renesas_usbhs/renesas_usbhs.o(.text+0xfd7): Section mismatch in reference from the function usbhs_mod_probe() to the function .devexit.text:usbhs_mod_host_remove() The function usbhs_mod_probe() references a function in an exit section. Often the function usbhs_mod_host_remove() has valid usage outside the exit section and the fix is to remove the __devexit annotation of usbhs_mod_host_remove. WARNING: drivers/usb/renesas_usbhs/renesas_usbhs.o(.text+0x1005): Section mismatch in reference from the function usbhs_mod_remove() to the function .devexit.text:usbhs_mod_host_remove() The function usbhs_mod_remove() references a function in an exit section. Often the function usbhs_mod_host_remove() has valid usage outside the exit section and the fix is to remove the __devexit annotation of usbhs_mod_host_remove. Signed-off-by: Kuninori Morimoto Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/renesas_usbhs/common.c | 2 +- drivers/usb/renesas_usbhs/mod.h | 8 ++++---- drivers/usb/renesas_usbhs/mod_gadget.c | 4 ++-- drivers/usb/renesas_usbhs/mod_host.c | 4 ++-- 4 files changed, 9 insertions(+), 9 deletions(-) diff --git a/drivers/usb/renesas_usbhs/common.c b/drivers/usb/renesas_usbhs/common.c index d2e2efaba658c..08c679c0dde5a 100644 --- a/drivers/usb/renesas_usbhs/common.c +++ b/drivers/usb/renesas_usbhs/common.c @@ -405,7 +405,7 @@ int usbhsc_drvcllbck_notify_hotplug(struct platform_device *pdev) /* * platform functions */ -static int __devinit usbhs_probe(struct platform_device *pdev) +static int usbhs_probe(struct platform_device *pdev) { struct renesas_usbhs_platform_info *info = pdev->dev.platform_data; struct renesas_usbhs_driver_callback *dfunc; diff --git a/drivers/usb/renesas_usbhs/mod.h b/drivers/usb/renesas_usbhs/mod.h index 8ae3733031cdb..6c6875533f019 100644 --- a/drivers/usb/renesas_usbhs/mod.h +++ b/drivers/usb/renesas_usbhs/mod.h @@ -143,8 +143,8 @@ void usbhs_irq_callback_update(struct usbhs_priv *priv, struct usbhs_mod *mod); */ #if defined(CONFIG_USB_RENESAS_USBHS_HCD) || \ defined(CONFIG_USB_RENESAS_USBHS_HCD_MODULE) -extern int __devinit usbhs_mod_host_probe(struct usbhs_priv *priv); -extern int __devexit usbhs_mod_host_remove(struct usbhs_priv *priv); +extern int usbhs_mod_host_probe(struct usbhs_priv *priv); +extern int usbhs_mod_host_remove(struct usbhs_priv *priv); #else static inline int usbhs_mod_host_probe(struct usbhs_priv *priv) { @@ -157,8 +157,8 @@ static inline void usbhs_mod_host_remove(struct usbhs_priv *priv) #if defined(CONFIG_USB_RENESAS_USBHS_UDC) || \ defined(CONFIG_USB_RENESAS_USBHS_UDC_MODULE) -extern int __devinit usbhs_mod_gadget_probe(struct usbhs_priv *priv); -extern void __devexit usbhs_mod_gadget_remove(struct usbhs_priv *priv); +extern int usbhs_mod_gadget_probe(struct usbhs_priv *priv); +extern void usbhs_mod_gadget_remove(struct usbhs_priv *priv); #else static inline int usbhs_mod_gadget_probe(struct usbhs_priv *priv) { diff --git a/drivers/usb/renesas_usbhs/mod_gadget.c b/drivers/usb/renesas_usbhs/mod_gadget.c index 4cc7ee0babc60..d9717e0bc1ff6 100644 --- a/drivers/usb/renesas_usbhs/mod_gadget.c +++ b/drivers/usb/renesas_usbhs/mod_gadget.c @@ -830,7 +830,7 @@ static int usbhsg_stop(struct usbhs_priv *priv) return usbhsg_try_stop(priv, USBHSG_STATUS_STARTED); } -int __devinit usbhs_mod_gadget_probe(struct usbhs_priv *priv) +int usbhs_mod_gadget_probe(struct usbhs_priv *priv) { struct usbhsg_gpriv *gpriv; struct usbhsg_uep *uep; @@ -927,7 +927,7 @@ usbhs_mod_gadget_probe_err_gpriv: return ret; } -void __devexit usbhs_mod_gadget_remove(struct usbhs_priv *priv) +void usbhs_mod_gadget_remove(struct usbhs_priv *priv) { struct usbhsg_gpriv *gpriv = usbhsg_priv_to_gpriv(priv); diff --git a/drivers/usb/renesas_usbhs/mod_host.c b/drivers/usb/renesas_usbhs/mod_host.c index 9b8886867f4a3..b964f25ebdb78 100644 --- a/drivers/usb/renesas_usbhs/mod_host.c +++ b/drivers/usb/renesas_usbhs/mod_host.c @@ -1227,7 +1227,7 @@ static int usbhsh_stop(struct usbhs_priv *priv) return 0; } -int __devinit usbhs_mod_host_probe(struct usbhs_priv *priv) +int usbhs_mod_host_probe(struct usbhs_priv *priv) { struct usbhsh_hpriv *hpriv; struct usb_hcd *hcd; @@ -1290,7 +1290,7 @@ usbhs_mod_host_probe_err: return -ENOMEM; } -int __devexit usbhs_mod_host_remove(struct usbhs_priv *priv) +int usbhs_mod_host_remove(struct usbhs_priv *priv) { struct usbhsh_hpriv *hpriv = usbhsh_priv_to_hpriv(priv); struct usb_hcd *hcd = usbhsh_hpriv_to_hcd(hpriv); -- GitLab From db332bc9b26bbd79a37241721cccc9919489d5a9 Mon Sep 17 00:00:00 2001 From: Paul Zimmerman Date: Thu, 13 Oct 2011 17:46:36 -0700 Subject: [PATCH 0308/3089] usb: gadget: storage: check for valid USB_BULK_GET_MAX_LUN_REQUEST The latest USB-IF CV tester checks for a valid length for this request. Signed-off-by: Paul Zimmerman Acked-by: Alan Stern Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/f_mass_storage.c | 3 ++- drivers/usb/gadget/file_storage.c | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/usb/gadget/f_mass_storage.c b/drivers/usb/gadget/f_mass_storage.c index 52583a2353304..dfd0044cc15a9 100644 --- a/drivers/usb/gadget/f_mass_storage.c +++ b/drivers/usb/gadget/f_mass_storage.c @@ -639,7 +639,8 @@ static int fsg_setup(struct usb_function *f, if (ctrl->bRequestType != (USB_DIR_IN | USB_TYPE_CLASS | USB_RECIP_INTERFACE)) break; - if (w_index != fsg->interface_number || w_value != 0) + if (w_index != fsg->interface_number || w_value != 0 || + w_length != 1) return -EDOM; VDBG(fsg, "get max LUN\n"); *(u8 *)req->buf = fsg->common->nluns - 1; diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c index f7e39b0365cee..4314cf2014209 100644 --- a/drivers/usb/gadget/file_storage.c +++ b/drivers/usb/gadget/file_storage.c @@ -875,7 +875,7 @@ static int class_setup_req(struct fsg_dev *fsg, if (ctrl->bRequestType != (USB_DIR_IN | USB_TYPE_CLASS | USB_RECIP_INTERFACE)) break; - if (w_index != 0 || w_value != 0) { + if (w_index != 0 || w_value != 0 || w_length != 1) { value = -EDOM; break; } -- GitLab From ce7b6121851c72d661134d113a78161095e0ae73 Mon Sep 17 00:00:00 2001 From: Paul Zimmerman Date: Wed, 26 Oct 2011 12:07:54 -0700 Subject: [PATCH 0309/3089] usb: gadget: storage: check for valid USB_BULK_RESET_REQUEST wLength The USB-IF CV compliance tester is getting stricter, and it would be valid for it to fail a mass-storage device that accepts an invalid USB_BULK_RESET_REQUEST request. Although it doesn't do that yet, let's be proactive and fix that now. Suggested by Alan Stern. Signed-off-by: Paul Zimmerman Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/f_mass_storage.c | 3 ++- drivers/usb/gadget/file_storage.c | 2 +- 2 files changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/usb/gadget/f_mass_storage.c b/drivers/usb/gadget/f_mass_storage.c index dfd0044cc15a9..c39d58860fa04 100644 --- a/drivers/usb/gadget/f_mass_storage.c +++ b/drivers/usb/gadget/f_mass_storage.c @@ -624,7 +624,8 @@ static int fsg_setup(struct usb_function *f, if (ctrl->bRequestType != (USB_DIR_OUT | USB_TYPE_CLASS | USB_RECIP_INTERFACE)) break; - if (w_index != fsg->interface_number || w_value != 0) + if (w_index != fsg->interface_number || w_value != 0 || + w_length != 0) return -EDOM; /* diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c index 4314cf2014209..11b5196284aed 100644 --- a/drivers/usb/gadget/file_storage.c +++ b/drivers/usb/gadget/file_storage.c @@ -859,7 +859,7 @@ static int class_setup_req(struct fsg_dev *fsg, if (ctrl->bRequestType != (USB_DIR_OUT | USB_TYPE_CLASS | USB_RECIP_INTERFACE)) break; - if (w_index != 0 || w_value != 0) { + if (w_index != 0 || w_value != 0 || w_length != 0) { value = -EDOM; break; } -- GitLab From 74203de067ae5c71526168b597088022836e31d3 Mon Sep 17 00:00:00 2001 From: Daniel Mack Date: Sat, 15 Oct 2011 13:45:05 +0200 Subject: [PATCH 0310/3089] usb: gadget: fix MIDI gadget jack allocation The dynamic jack allocation of the MIDI gadget currently links all external jacks to one single instance of an embedded jack. According to the spec, this is only valid if these streams always carry the same data stream, as described in the USB MIDI 1.0 spec, chapter 3.3.1. Also, genius Windows 7(tm) terminates it's life cycle instantly with a blue screen of death once a device with more than one input and output port with the current implementation is connected. While at it, and because it grew again by this change, allocate the temporary function pointer list on the heap, not on the stack. Signed-off-by: Daniel Mack Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/f_midi.c | 138 +++++++++++++++++------------------- 1 file changed, 65 insertions(+), 73 deletions(-) diff --git a/drivers/usb/gadget/f_midi.c b/drivers/usb/gadget/f_midi.c index 67b222908cf9f..3797b3d6c622b 100644 --- a/drivers/usb/gadget/f_midi.c +++ b/drivers/usb/gadget/f_midi.c @@ -95,7 +95,6 @@ static void f_midi_transmit(struct f_midi *midi, struct usb_request *req); DECLARE_UAC_AC_HEADER_DESCRIPTOR(1); DECLARE_USB_MIDI_OUT_JACK_DESCRIPTOR(1); -DECLARE_USB_MIDI_OUT_JACK_DESCRIPTOR(16); DECLARE_USB_MS_ENDPOINT_DESCRIPTOR(16); /* B.3.1 Standard AC Interface Descriptor */ @@ -140,26 +139,6 @@ static struct usb_ms_header_descriptor ms_header_desc __initdata = { /* .wTotalLength = DYNAMIC */ }; -/* B.4.3 Embedded MIDI IN Jack Descriptor */ -static struct usb_midi_in_jack_descriptor jack_in_emb_desc = { - .bLength = USB_DT_MIDI_IN_SIZE, - .bDescriptorType = USB_DT_CS_INTERFACE, - .bDescriptorSubtype = USB_MS_MIDI_IN_JACK, - .bJackType = USB_MS_EMBEDDED, - /* .bJackID = DYNAMIC */ -}; - -/* B.4.4 Embedded MIDI OUT Jack Descriptor */ -static struct usb_midi_out_jack_descriptor_16 jack_out_emb_desc = { - /* .bLength = DYNAMIC */ - .bDescriptorType = USB_DT_CS_INTERFACE, - .bDescriptorSubtype = USB_MS_MIDI_OUT_JACK, - .bJackType = USB_MS_EMBEDDED, - /* .bJackID = DYNAMIC */ - /* .bNrInputPins = DYNAMIC */ - /* .pins = DYNAMIC */ -}; - /* B.5.1 Standard Bulk OUT Endpoint Descriptor */ static struct usb_endpoint_descriptor bulk_out_desc = { .bLength = USB_DT_ENDPOINT_AUDIO_SIZE, @@ -758,9 +737,11 @@ fail: static int __init f_midi_bind(struct usb_configuration *c, struct usb_function *f) { - struct usb_descriptor_header *midi_function[(MAX_PORTS * 2) + 12]; + struct usb_descriptor_header **midi_function; struct usb_midi_in_jack_descriptor jack_in_ext_desc[MAX_PORTS]; + struct usb_midi_in_jack_descriptor jack_in_emb_desc[MAX_PORTS]; struct usb_midi_out_jack_descriptor_1 jack_out_ext_desc[MAX_PORTS]; + struct usb_midi_out_jack_descriptor_1 jack_out_emb_desc[MAX_PORTS]; struct usb_composite_dev *cdev = c->cdev; struct f_midi *midi = func_to_midi(f); int status, n, jack = 1, i = 0; @@ -798,6 +779,14 @@ f_midi_bind(struct usb_configuration *c, struct usb_function *f) goto fail; midi->out_ep->driver_data = cdev; /* claim */ + /* allocate temporary function list */ + midi_function = kcalloc((MAX_PORTS * 4) + 9, sizeof(midi_function), + GFP_KERNEL); + if (!midi_function) { + status = -ENOMEM; + goto fail; + } + /* * construct the function's descriptor set. As the number of * input and output MIDI ports is configurable, we have to do @@ -811,73 +800,74 @@ f_midi_bind(struct usb_configuration *c, struct usb_function *f) /* calculate the header's wTotalLength */ n = USB_DT_MS_HEADER_SIZE - + (1 + midi->in_ports) * USB_DT_MIDI_IN_SIZE - + (1 + midi->out_ports) * USB_DT_MIDI_OUT_SIZE(1); + + (midi->in_ports + midi->out_ports) * + (USB_DT_MIDI_IN_SIZE + USB_DT_MIDI_OUT_SIZE(1)); ms_header_desc.wTotalLength = cpu_to_le16(n); midi_function[i++] = (struct usb_descriptor_header *) &ms_header_desc; - /* we have one embedded IN jack */ - jack_in_emb_desc.bJackID = jack++; - midi_function[i++] = (struct usb_descriptor_header *) &jack_in_emb_desc; - - /* and a dynamic amount of external IN jacks */ - for (n = 0; n < midi->in_ports; n++) { - struct usb_midi_in_jack_descriptor *ext = &jack_in_ext_desc[n]; - - ext->bLength = USB_DT_MIDI_IN_SIZE; - ext->bDescriptorType = USB_DT_CS_INTERFACE; - ext->bDescriptorSubtype = USB_MS_MIDI_IN_JACK; - ext->bJackType = USB_MS_EXTERNAL; - ext->bJackID = jack++; - ext->iJack = 0; - - midi_function[i++] = (struct usb_descriptor_header *) ext; - } - - /* one embedded OUT jack ... */ - jack_out_emb_desc.bLength = USB_DT_MIDI_OUT_SIZE(midi->in_ports); - jack_out_emb_desc.bJackID = jack++; - jack_out_emb_desc.bNrInputPins = midi->in_ports; - /* ... which referencess all external IN jacks */ + /* configure the external IN jacks, each linked to an embedded OUT jack */ for (n = 0; n < midi->in_ports; n++) { - jack_out_emb_desc.pins[n].baSourceID = jack_in_ext_desc[n].bJackID; - jack_out_emb_desc.pins[n].baSourcePin = 1; + struct usb_midi_in_jack_descriptor *in_ext = &jack_in_ext_desc[n]; + struct usb_midi_out_jack_descriptor_1 *out_emb = &jack_out_emb_desc[n]; + + in_ext->bLength = USB_DT_MIDI_IN_SIZE; + in_ext->bDescriptorType = USB_DT_CS_INTERFACE; + in_ext->bDescriptorSubtype = USB_MS_MIDI_IN_JACK; + in_ext->bJackType = USB_MS_EXTERNAL; + in_ext->bJackID = jack++; + in_ext->iJack = 0; + midi_function[i++] = (struct usb_descriptor_header *) in_ext; + + out_emb->bLength = USB_DT_MIDI_OUT_SIZE(1); + out_emb->bDescriptorType = USB_DT_CS_INTERFACE; + out_emb->bDescriptorSubtype = USB_MS_MIDI_OUT_JACK; + out_emb->bJackType = USB_MS_EMBEDDED; + out_emb->bJackID = jack++; + out_emb->bNrInputPins = 1; + out_emb->pins[0].baSourcePin = 1; + out_emb->pins[0].baSourceID = in_ext->bJackID; + out_emb->iJack = 0; + midi_function[i++] = (struct usb_descriptor_header *) out_emb; + + /* link it to the endpoint */ + ms_in_desc.baAssocJackID[n] = out_emb->bJackID; } - midi_function[i++] = (struct usb_descriptor_header *) &jack_out_emb_desc; - - /* and multiple external OUT jacks ... */ + /* configure the external OUT jacks, each linked to an embedded IN jack */ for (n = 0; n < midi->out_ports; n++) { - struct usb_midi_out_jack_descriptor_1 *ext = &jack_out_ext_desc[n]; - int m; - - ext->bLength = USB_DT_MIDI_OUT_SIZE(1); - ext->bDescriptorType = USB_DT_CS_INTERFACE; - ext->bDescriptorSubtype = USB_MS_MIDI_OUT_JACK; - ext->bJackType = USB_MS_EXTERNAL; - ext->bJackID = jack++; - ext->bNrInputPins = 1; - ext->iJack = 0; - /* ... which all reference the same embedded IN jack */ - for (m = 0; m < midi->out_ports; m++) { - ext->pins[m].baSourceID = jack_in_emb_desc.bJackID; - ext->pins[m].baSourcePin = 1; - } - - midi_function[i++] = (struct usb_descriptor_header *) ext; + struct usb_midi_in_jack_descriptor *in_emb = &jack_in_emb_desc[n]; + struct usb_midi_out_jack_descriptor_1 *out_ext = &jack_out_ext_desc[n]; + + in_emb->bLength = USB_DT_MIDI_IN_SIZE; + in_emb->bDescriptorType = USB_DT_CS_INTERFACE; + in_emb->bDescriptorSubtype = USB_MS_MIDI_IN_JACK; + in_emb->bJackType = USB_MS_EMBEDDED; + in_emb->bJackID = jack++; + in_emb->iJack = 0; + midi_function[i++] = (struct usb_descriptor_header *) in_emb; + + out_ext->bLength = USB_DT_MIDI_OUT_SIZE(1); + out_ext->bDescriptorType = USB_DT_CS_INTERFACE; + out_ext->bDescriptorSubtype = USB_MS_MIDI_OUT_JACK; + out_ext->bJackType = USB_MS_EXTERNAL; + out_ext->bJackID = jack++; + out_ext->bNrInputPins = 1; + out_ext->iJack = 0; + out_ext->pins[0].baSourceID = in_emb->bJackID; + out_ext->pins[0].baSourcePin = 1; + midi_function[i++] = (struct usb_descriptor_header *) out_ext; + + /* link it to the endpoint */ + ms_out_desc.baAssocJackID[n] = in_emb->bJackID; } /* configure the endpoint descriptors ... */ ms_out_desc.bLength = USB_DT_MS_ENDPOINT_SIZE(midi->in_ports); ms_out_desc.bNumEmbMIDIJack = midi->in_ports; - for (n = 0; n < midi->in_ports; n++) - ms_out_desc.baAssocJackID[n] = jack_in_emb_desc.bJackID; ms_in_desc.bLength = USB_DT_MS_ENDPOINT_SIZE(midi->out_ports); ms_in_desc.bNumEmbMIDIJack = midi->out_ports; - for (n = 0; n < midi->out_ports; n++) - ms_in_desc.baAssocJackID[n] = jack_out_emb_desc.bJackID; /* ... and add them to the list */ midi_function[i++] = (struct usb_descriptor_header *) &bulk_out_desc; @@ -901,6 +891,8 @@ f_midi_bind(struct usb_configuration *c, struct usb_function *f) f->descriptors = usb_copy_descriptors(midi_function); } + kfree(midi_function); + return 0; fail: -- GitLab From 001428e4871d6c62f5e16c62df681624d8b480c1 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Mon, 10 Oct 2011 18:38:05 +0200 Subject: [PATCH 0311/3089] USB: gadgetfs: gadgetfs_disconnect: fix inconsistent lock state Under certain circumstances lockdep finds an inconsistent lock state in gadgetfs. The problem can be reproduced with a hardware using the ci13xxx_udc driver and the gadgetfs test program (needs a patch to support the ci13xxx_udc, though): http://www.linux-usb.org/gadget/usb.c Start the test program, wait to initialize, then press Ctrl+c. This patch fixes the following problem by using spin_lock_irqsave() instead of spin_lock(). ================================= [ INFO: inconsistent lock state ] 3.1.0-rc6+ #158 --------------------------------- inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage. usb/113 [HC0[0]:SC0[0]:HE1:SE1] takes: (&(&dev->lock)->rlock){?.....}, at: [] gadgetfs_disconnect+0x14/0x80 [gadgetfs] {IN-HARDIRQ-W} state was registered at: [] mark_irqflags+0x14c/0x1ac [] __lock_acquire+0x4e0/0x8f0 [] lock_acquire+0x98/0x1a8 [] _raw_spin_lock+0x54/0x8c [] gadgetfs_disconnect+0x14/0x80 [gadgetfs] [] _gadget_stop_activity+0xd4/0x154 [] isr_reset_handler+0x34/0x1c0 [] udc_irq+0x204/0x228 [] handle_irq_event_percpu+0x64/0x3a0 [] handle_irq_event+0x3c/0x5c [] handle_level_irq+0x8c/0x10c [] generic_handle_irq+0x30/0x44 [] handle_IRQ+0x30/0x84 [] __irq_svc+0x38/0x60 [] default_idle+0x30/0x34 [] cpu_idle+0x9c/0xd8 [] start_kernel+0x278/0x2bc irq event stamp: 6412 hardirqs last enabled at (6412): [] _raw_spin_unlock_irqrestore+0x30/0x5c hardirqs last disabled at (6411): [] _raw_spin_lock_irqsave+0x20/0xa0 softirqs last enabled at (6381): [] irq_exit+0xa0/0xa8 softirqs last disabled at (6372): [] irq_exit+0xa0/0xa8 other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(&(&dev->lock)->rlock); lock(&(&dev->lock)->rlock); *** DEADLOCK *** 1 lock held by usb/113: #0: (udc_lock#2){+.+.+.}, at: [] usb_gadget_unregister_driver+0x34/0x88 stack backtrace: [] (unwind_backtrace+0x0/0xf0) from [] (print_usage_bug+0x144/0x1c4) [] (print_usage_bug+0x144/0x1c4) from [] (mark_lock_irq+0x22c/0x274) [] (mark_lock_irq+0x22c/0x274) from [] (mark_lock+0x148/0x3e0) [] (mark_lock+0x148/0x3e0) from [] (mark_irqflags+0xfc/0x1ac) [] (mark_irqflags+0xfc/0x1ac) from [] (__lock_acquire+0x4e0/0x8f0) [] (__lock_acquire+0x4e0/0x8f0) from [] (lock_acquire+0x98/0x1a8) [] (lock_acquire+0x98/0x1a8) from [] (_raw_spin_lock+0x54/0x8c) [] (_raw_spin_lock+0x54/0x8c) from [] (gadgetfs_disconnect+0x14/0x80 [gadgetfs]) [] (gadgetfs_disconnect+0x14/0x80 [gadgetfs]) from [] (_gadget_stop_activity+0xd4/0x154) [] (_gadget_stop_activity+0xd4/0x154) from [] (ci13xxx_stop+0xbc/0x17c) [] (ci13xxx_stop+0xbc/0x17c) from [] (usb_gadget_remove_driver+0x88/0x98) [] (usb_gadget_remove_driver+0x88/0x98) from [] (usb_gadget_unregister_driver+0x68/0x88) [] (usb_gadget_unregister_driver+0x68/0x88) from [] (dev_release+0x14/0x48 [gadgetfs]) [] (dev_release+0x14/0x48 [gadgetfs]) from [] (__fput+0xa4/0x1f0) [] (__fput+0xa4/0x1f0) from [] (filp_close+0x5c/0x74) [] (filp_close+0x5c/0x74) from [] (sys_close+0xa8/0x150) [] (sys_close+0xa8/0x150) from [] (ret_fast_syscall+0x0/0x38) Tested-by: Pavankumar Kondeti Signed-off-by: Marc Kleine-Budde Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/inode.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c index a392ec0d2d517..6ccae2707e596 100644 --- a/drivers/usb/gadget/inode.c +++ b/drivers/usb/gadget/inode.c @@ -1730,8 +1730,9 @@ static void gadgetfs_disconnect (struct usb_gadget *gadget) { struct dev_data *dev = get_gadget_data (gadget); + unsigned long flags; - spin_lock (&dev->lock); + spin_lock_irqsave (&dev->lock, flags); if (dev->state == STATE_DEV_UNCONNECTED) goto exit; dev->state = STATE_DEV_UNCONNECTED; @@ -1740,7 +1741,7 @@ gadgetfs_disconnect (struct usb_gadget *gadget) next_event (dev, GADGETFS_DISCONNECT); ep0_readable (dev); exit: - spin_unlock (&dev->lock); + spin_unlock_irqrestore (&dev->lock, flags); } static void -- GitLab From 954aad8cd18c928e2db5229f6fa71c80d1c3d8b5 Mon Sep 17 00:00:00 2001 From: Michael Grzeschik Date: Mon, 10 Oct 2011 18:38:06 +0200 Subject: [PATCH 0312/3089] USB: ci13xxx_udc: fix logic to mark request dma addresses as invalid The current driver sets the request's dma addr (mReq->req.dma) to 0 to mark the DMA address as not valid. However some gadget drivers (e.g. gadgetfs) set the request's dma addr to DMA_ADDR_INVALID to mark the address as invalid. This leads to bogus data send because the ci13xxx_udc driver assumes the request has already been mapped. This patch fixes the problem, by using DMA_ADDR_INVALID instead of 0 to mark the request's DMA address as invalid. Tested-by: Pavankumar Kondeti Signed-off-by: Michael Grzeschik Signed-off-by: Marc Kleine-Budde Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/ci13xxx_udc.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/usb/gadget/ci13xxx_udc.c b/drivers/usb/gadget/ci13xxx_udc.c index 83428f56253bd..ae6c0010f5e7b 100644 --- a/drivers/usb/gadget/ci13xxx_udc.c +++ b/drivers/usb/gadget/ci13xxx_udc.c @@ -71,6 +71,9 @@ /****************************************************************************** * DEFINE *****************************************************************************/ + +#define DMA_ADDR_INVALID (~(dma_addr_t)0) + /* ctrl register bank access */ static DEFINE_SPINLOCK(udc_lock); @@ -1434,7 +1437,7 @@ static int _hardware_enqueue(struct ci13xxx_ep *mEp, struct ci13xxx_req *mReq) return -EALREADY; mReq->req.status = -EALREADY; - if (length && !mReq->req.dma) { + if (length && mReq->req.dma == DMA_ADDR_INVALID) { mReq->req.dma = \ dma_map_single(mEp->device, mReq->req.buf, length, mEp->dir ? DMA_TO_DEVICE : @@ -1453,7 +1456,7 @@ static int _hardware_enqueue(struct ci13xxx_ep *mEp, struct ci13xxx_req *mReq) dma_unmap_single(mEp->device, mReq->req.dma, length, mEp->dir ? DMA_TO_DEVICE : DMA_FROM_DEVICE); - mReq->req.dma = 0; + mReq->req.dma = DMA_ADDR_INVALID; mReq->map = 0; } return -ENOMEM; @@ -1549,7 +1552,7 @@ static int _hardware_dequeue(struct ci13xxx_ep *mEp, struct ci13xxx_req *mReq) if (mReq->map) { dma_unmap_single(mEp->device, mReq->req.dma, mReq->req.length, mEp->dir ? DMA_TO_DEVICE : DMA_FROM_DEVICE); - mReq->req.dma = 0; + mReq->req.dma = DMA_ADDR_INVALID; mReq->map = 0; } @@ -2189,6 +2192,7 @@ static struct usb_request *ep_alloc_request(struct usb_ep *ep, gfp_t gfp_flags) mReq = kzalloc(sizeof(struct ci13xxx_req), gfp_flags); if (mReq != NULL) { INIT_LIST_HEAD(&mReq->queue); + mReq->req.dma = DMA_ADDR_INVALID; mReq->ptr = dma_pool_alloc(mEp->td_pool, gfp_flags, &mReq->dma); @@ -2328,7 +2332,7 @@ static int ep_dequeue(struct usb_ep *ep, struct usb_request *req) if (mReq->map) { dma_unmap_single(mEp->device, mReq->req.dma, mReq->req.length, mEp->dir ? DMA_TO_DEVICE : DMA_FROM_DEVICE); - mReq->req.dma = 0; + mReq->req.dma = DMA_ADDR_INVALID; mReq->map = 0; } req->status = -ECONNRESET; -- GitLab From fd537c041b7f2cbceb5a21c37946e017006edd1c Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Mon, 10 Oct 2011 18:38:07 +0200 Subject: [PATCH 0313/3089] USB: ci13xxx_udc: fix deadlock during rmmod The inline documentation of _gadget_stop_activity() states that the function should be called holding the udc->lock. This however will result in a deadlock, because _gadget_stop_activity() takes the udc->lock. During normal operation _gadget_stop_activity() is always called unlocked, but in ci13xxx_stop() it's called locked, this results in the following deadlock during rmmod of a gadget driver. This patch fixes the deadlock by calling _gadget_stop_activity() always unlocked, the inline documentation is adjusted accordingly. ============================================= [ INFO: possible recursive locking detected ] 3.1.0-rc6+ #159 --------------------------------------------- rmmod/121 is trying to acquire lock: (udc_lock){-.-...}, at: [] _gadget_stop_activity+0x18/0x154 but task is already holding lock: (udc_lock){-.-...}, at: [] ci13xxx_stop+0x5c/0x164 other info that might help us debug this: Possible unsafe locking scenario: CPU0 ---- lock(udc_lock); lock(udc_lock); *** DEADLOCK *** May be due to missing lock nesting notation 2 locks held by rmmod/121: #0: (udc_lock#2){+.+.+.}, at: [] usb_gadget_unregister_driver+0x34/0x88 #1: (udc_lock){-.-...}, at: [] ci13xxx_stop+0x5c/0x164 stack backtrace: [] (unwind_backtrace+0x0/0xf0) from [] (check_deadlock.clone.24+0x284/0x2c4) [] (check_deadlock.clone.24+0x284/0x2c4) from [] (validate_chain.clone.25+0x430/0x6fc) [] (validate_chain.clone.25+0x430/0x6fc) from [] (__lock_acquire+0x494/0x8f0) [] (__lock_acquire+0x494/0x8f0) from [] (lock_acquire+0x98/0x1a8) [] (lock_acquire+0x98/0x1a8) from [] (_raw_spin_lock_irqsave+0x64/0xa0) [] (_raw_spin_lock_irqsave+0x64/0xa0) from [] (_gadget_stop_activity+0x18/0x154) [] (_gadget_stop_activity+0x18/0x154) from [] (ci13xxx_stop+0xb0/0x164) [] (ci13xxx_stop+0xb0/0x164) from [] (usb_gadget_remove_driver+0x88/0x98) [] (usb_gadget_remove_driver+0x88/0x98) from [] (usb_gadget_unregister_driver+0x68/0x88) [] (usb_gadget_unregister_driver+0x68/0x88) from [] (sys_delete_module+0x1fc/0x26c) [] (sys_delete_module+0x1fc/0x26c) from [] (ret_fast_syscall+0x0/0x38) BUG: spinlock lockup on CPU#0, rmmod/121, c05b1644 [] (unwind_backtrace+0x0/0xf0) from [] (do_raw_spin_lock+0x128/0x144) [] (do_raw_spin_lock+0x128/0x144) from [] (_raw_spin_lock_irqsave+0x88/0xa0) [] (_raw_spin_lock_irqsave+0x88/0xa0) from [] (_gadget_stop_activity+0x18/0x154) [] (_gadget_stop_activity+0x18/0x154) from [] (ci13xxx_stop+0xb0/0x164) [] (ci13xxx_stop+0xb0/0x164) from [] (usb_gadget_remove_driver+0x88/0x98) [] (usb_gadget_remove_driver+0x88/0x98) from [] (usb_gadget_unregister_driver+0x68/0x88) [] (usb_gadget_unregister_driver+0x68/0x88) from [] (sys_delete_module+0x1fc/0x26c) [] (sys_delete_module+0x1fc/0x26c) from [] (ret_fast_syscall+0x0/0x38) Tested-by: Pavankumar Kondeti Signed-off-by: Marc Kleine-Budde Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/ci13xxx_udc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/usb/gadget/ci13xxx_udc.c b/drivers/usb/gadget/ci13xxx_udc.c index ae6c0010f5e7b..4241241128c61 100644 --- a/drivers/usb/gadget/ci13xxx_udc.c +++ b/drivers/usb/gadget/ci13xxx_udc.c @@ -1613,7 +1613,6 @@ __acquires(mEp->lock) * @gadget: gadget * * This function returns an error code - * Caller must hold lock */ static int _gadget_stop_activity(struct usb_gadget *gadget) { @@ -2707,7 +2706,9 @@ static int ci13xxx_stop(struct usb_gadget_driver *driver) if (udc->udc_driver->notify_event) udc->udc_driver->notify_event(udc, CI13XXX_CONTROLLER_STOPPED_EVENT); + spin_unlock_irqrestore(udc->lock, flags); _gadget_stop_activity(&udc->gadget); + spin_lock_irqsave(udc->lock, flags); pm_runtime_put(&udc->gadget.dev); } -- GitLab From 194fa47ef629df07d87064ca4aa973939fa6fb4d Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Mon, 10 Oct 2011 18:38:08 +0200 Subject: [PATCH 0314/3089] USB: ci13xxx_udc: fix debug trace code This patch fixes the following compile errors that show up if switching on the debug trace code: drivers/usb/gadget/ci13xxx_udc.c: In function 'ci13xxx_wakeup': drivers/usb/gadget/ci13xxx_udc.c:2517:3: error: 'dev' undeclared (first use in this function) drivers/usb/gadget/ci13xxx_udc.c:2517:3: note: each undeclared identifier is reported only once for each function it appears in drivers/usb/gadget/ci13xxx_udc.c: In function 'udc_probe': drivers/usb/gadget/ci13xxx_udc.c:2867:2: error: 'name' undeclared (first use in this function) Tested-by: Pavankumar Kondeti Signed-off-by: Marc Kleine-Budde Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/ci13xxx_udc.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/usb/gadget/ci13xxx_udc.c b/drivers/usb/gadget/ci13xxx_udc.c index 4241241128c61..9a0c3979ff43f 100644 --- a/drivers/usb/gadget/ci13xxx_udc.c +++ b/drivers/usb/gadget/ci13xxx_udc.c @@ -2503,12 +2503,12 @@ static int ci13xxx_wakeup(struct usb_gadget *_gadget) spin_lock_irqsave(udc->lock, flags); if (!udc->remote_wakeup) { ret = -EOPNOTSUPP; - dbg_trace("remote wakeup feature is not enabled\n"); + trace("remote wakeup feature is not enabled\n"); goto out; } if (!hw_cread(CAP_PORTSC, PORTSC_SUSP)) { ret = -EINVAL; - dbg_trace("port is not suspended\n"); + trace("port is not suspended\n"); goto out; } hw_cwrite(CAP_PORTSC, PORTSC_FPR, PORTSC_FPR); @@ -2855,7 +2855,7 @@ static int udc_probe(struct ci13xxx_udc_driver *driver, struct device *dev, struct ci13xxx *udc; int retval = 0; - trace("%p, %p, %p", dev, regs, name); + trace("%p, %p, %p", dev, regs, driver->name); if (dev == NULL || regs == NULL || driver == NULL || driver->name == NULL) -- GitLab From 4703d2e9d06cd1e5add773aa6cc5587e6664bf78 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Mon, 10 Oct 2011 18:38:11 +0200 Subject: [PATCH 0315/3089] USB: ci13xxx_msm: add module license Since commit "193ab2a usb: gadget: allow multiple gadgets to be built" the udc controllers can be compiled as a module. The ci13xxx_msm driver is missing the MODULE_LICENSE statement, so loading fails with: ci13xxx_msm: module license 'unspecified' taints kernel. ci13xxx_msm: Unknown symbol dev_set_name (err 0) ci13xxx_msm: Unknown symbol platform_get_irq (err 0) ci13xxx_msm: Unknown symbol device_unregister (err 0) ci13xxx_msm: Unknown symbol usb_add_gadget_udc (err 0) ci13xxx_msm: Unknown symbol put_device (err 0) ci13xxx_msm: Unknown symbol platform_driver_register (err 0) ci13xxx_msm: Unknown symbol platform_get_resource (err 0) ci13xxx_msm: Unknown symbol device_register (err 0) ci13xxx_msm: Unknown symbol usb_del_gadget_udc (err 0) This patch adds the missing MODULE_LICENSE statement with GPL v2 according to the header of the driver. Tested-by: Pavankumar Kondeti Signed-off-by: Marc Kleine-Budde Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/ci13xxx_msm.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/usb/gadget/ci13xxx_msm.c b/drivers/usb/gadget/ci13xxx_msm.c index 4eedfe557154c..1fc612914c52e 100644 --- a/drivers/usb/gadget/ci13xxx_msm.c +++ b/drivers/usb/gadget/ci13xxx_msm.c @@ -122,3 +122,5 @@ static int __init ci13xxx_msm_init(void) return platform_driver_register(&ci13xxx_msm_driver); } module_init(ci13xxx_msm_init); + +MODULE_LICENSE("GPL v2"); -- GitLab From 73ef635a07c0e6a0a159d8beabffb83399429188 Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 24 Oct 2011 02:24:49 -0700 Subject: [PATCH 0316/3089] usb: gadget: renesas_usbhs: bugfix: care pipe direction renesas_usbhs is caring pipe type and its direction. but current usbhs_endpoint_alloc() didn't check direction. this patch modify it. Signed-off-by: Kuninori Morimoto Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/renesas_usbhs/mod_host.c | 29 ++++++++++++++++++++-------- 1 file changed, 21 insertions(+), 8 deletions(-) diff --git a/drivers/usb/renesas_usbhs/mod_host.c b/drivers/usb/renesas_usbhs/mod_host.c index b964f25ebdb78..f8f612c46ea48 100644 --- a/drivers/usb/renesas_usbhs/mod_host.c +++ b/drivers/usb/renesas_usbhs/mod_host.c @@ -355,6 +355,7 @@ static void usbhsh_device_free(struct usbhsh_hpriv *hpriv, struct usbhsh_ep *usbhsh_endpoint_alloc(struct usbhsh_hpriv *hpriv, struct usbhsh_device *udev, struct usb_host_endpoint *ep, + int dir_in_req, gfp_t mem_flags) { struct usbhs_priv *priv = usbhsh_hpriv_to_priv(hpriv); @@ -364,27 +365,38 @@ struct usbhsh_ep *usbhsh_endpoint_alloc(struct usbhsh_hpriv *hpriv, struct usbhs_pipe *pipe, *best_pipe; struct device *dev = usbhsh_hcd_to_dev(hcd); struct usb_endpoint_descriptor *desc = &ep->desc; - int type, i; + int type, i, dir_in; unsigned int min_usr; + dir_in_req = !!dir_in_req; + uep = kzalloc(sizeof(struct usbhsh_ep), mem_flags); if (!uep) { dev_err(dev, "usbhsh_ep alloc fail\n"); return NULL; } - type = usb_endpoint_type(desc); + + if (usb_endpoint_xfer_control(desc)) { + best_pipe = usbhsh_hpriv_to_dcp(hpriv); + goto usbhsh_endpoint_alloc_find_pipe; + } /* * find best pipe for endpoint * see * HARDWARE LIMITATION */ + type = usb_endpoint_type(desc); min_usr = ~0; best_pipe = NULL; - usbhs_for_each_pipe_with_dcp(pipe, priv, i) { + usbhs_for_each_pipe(pipe, priv, i) { if (!usbhs_pipe_type_is(pipe, type)) continue; + dir_in = !!usbhs_pipe_is_dir_in(pipe); + if (0 != (dir_in - dir_in_req)) + continue; + info = usbhsh_pipe_info(pipe); if (min_usr > info->usr_cnt) { @@ -398,7 +410,7 @@ struct usbhsh_ep *usbhsh_endpoint_alloc(struct usbhsh_hpriv *hpriv, kfree(uep); return NULL; } - +usbhsh_endpoint_alloc_find_pipe: /* * init uep */ @@ -430,7 +442,7 @@ struct usbhsh_ep *usbhsh_endpoint_alloc(struct usbhsh_hpriv *hpriv, dev_dbg(dev, "%s [%d-%s](%p)\n", __func__, usbhsh_device_number(hpriv, udev), - usbhs_pipe_name(pipe), uep); + usbhs_pipe_name(uep->pipe), uep); return uep; } @@ -722,11 +734,11 @@ static int usbhsh_urb_enqueue(struct usb_hcd *hcd, struct usbhsh_device *udev, *new_udev = NULL; struct usbhs_pipe *pipe; struct usbhsh_ep *uep; + int is_dir_in = usb_pipein(urb->pipe); int ret; - dev_dbg(dev, "%s (%s)\n", - __func__, usb_pipein(urb->pipe) ? "in" : "out"); + dev_dbg(dev, "%s (%s)\n", __func__, is_dir_in ? "in" : "out"); ret = usb_hcd_link_urb_to_ep(hcd, urb); if (ret) @@ -749,7 +761,8 @@ static int usbhsh_urb_enqueue(struct usb_hcd *hcd, */ uep = usbhsh_ep_to_uep(ep); if (!uep) { - uep = usbhsh_endpoint_alloc(hpriv, udev, ep, mem_flags); + uep = usbhsh_endpoint_alloc(hpriv, udev, ep, + is_dir_in, mem_flags); if (!uep) goto usbhsh_urb_enqueue_error_free_device; } -- GitLab From 146ee50ae51c78fc93d025cb9528883df26ab705 Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 24 Oct 2011 02:25:07 -0700 Subject: [PATCH 0317/3089] usb: gadget: renesas_usbhs: bugfix: disable irq when device stop Signed-off-by: Kuninori Morimoto Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/renesas_usbhs/mod_host.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/drivers/usb/renesas_usbhs/mod_host.c b/drivers/usb/renesas_usbhs/mod_host.c index f8f612c46ea48..228f7b844c1d4 100644 --- a/drivers/usb/renesas_usbhs/mod_host.c +++ b/drivers/usb/renesas_usbhs/mod_host.c @@ -1226,8 +1226,18 @@ static int usbhsh_stop(struct usbhs_priv *priv) { struct usbhsh_hpriv *hpriv = usbhsh_priv_to_hpriv(priv); struct usb_hcd *hcd = usbhsh_hpriv_to_hcd(hpriv); + struct usbhs_mod *mod = usbhs_mod_get_current(priv); struct device *dev = usbhs_priv_to_dev(priv); + /* + * disable irq callback + */ + mod->irq_attch = NULL; + mod->irq_dtch = NULL; + mod->irq_sack = NULL; + mod->irq_sign = NULL; + usbhs_irq_callback_update(priv, mod); + usb_remove_hcd(hcd); /* disable sys */ -- GitLab From d7a00ec1a8debf74317c5110ded9918bad8de772 Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Mon, 24 Oct 2011 02:25:28 -0700 Subject: [PATCH 0318/3089] usb: gadget: renesas_usbhs: bugfix: set DATA0 when usbhsh_endpoint_alloc() new endpoint should start from DATA0, but mod_host didn't care it. This patch fix it. Signed-off-by: Kuninori Morimoto Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/renesas_usbhs/mod_host.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/renesas_usbhs/mod_host.c b/drivers/usb/renesas_usbhs/mod_host.c index 228f7b844c1d4..bade761a1e52b 100644 --- a/drivers/usb/renesas_usbhs/mod_host.c +++ b/drivers/usb/renesas_usbhs/mod_host.c @@ -435,6 +435,7 @@ usbhsh_endpoint_alloc_find_pipe: * see * DCPMAXP/PIPEMAXP */ + usbhs_pipe_sequence_data0(uep->pipe); usbhs_pipe_config_update(uep->pipe, usbhsh_device_number(hpriv, udev), usb_endpoint_num(desc), -- GitLab From 12d36c16bde3ee0643d705caa87723de536dbe49 Mon Sep 17 00:00:00 2001 From: Sebastian Andrzej Siewior Date: Thu, 3 Nov 2011 20:27:50 +0100 Subject: [PATCH 0319/3089] usb: dwc3: gadget: initialize max_streams Without this the gadget will never be able to allocate a stream capable endpoint. The manual says that the stream id is a 16bit id. It does not talk about an upper limit in any other way. So I think 15 is a reasonable limit :) Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/dwc3/gadget.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c index fa824cfdd2eb4..25dbd8614e722 100644 --- a/drivers/usb/dwc3/gadget.c +++ b/drivers/usb/dwc3/gadget.c @@ -1284,6 +1284,7 @@ static int __devinit dwc3_gadget_init_endpoints(struct dwc3 *dwc) int ret; dep->endpoint.maxpacket = 1024; + dep->endpoint.max_streams = 15; dep->endpoint.ops = &dwc3_gadget_ep_ops; list_add_tail(&dep->endpoint.ep_list, &dwc->gadget.ep_list); -- GitLab From ab570da26eec4e840dc1d18f45ab9d64bae49a5d Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Thu, 10 Nov 2011 09:58:04 +0200 Subject: [PATCH 0320/3089] usb: musb: fix compilation breakage introduced by de47725 commit de47725 (include: replace linux/module.h with "struct module" wherever possible) introduced a compilation breaked when it removed from which musb glue layers were (mistakenly) relying on. Include that header to fix the compile error. Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/musb/am35x.c | 1 + drivers/usb/musb/da8xx.c | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/usb/musb/am35x.c b/drivers/usb/musb/am35x.c index 08f1d0b662a37..e233d2b7d3357 100644 --- a/drivers/usb/musb/am35x.c +++ b/drivers/usb/musb/am35x.c @@ -27,6 +27,7 @@ */ #include +#include #include #include #include diff --git a/drivers/usb/musb/da8xx.c b/drivers/usb/musb/da8xx.c index 4da7492ddbdba..2613bfdb09b65 100644 --- a/drivers/usb/musb/da8xx.c +++ b/drivers/usb/musb/da8xx.c @@ -27,6 +27,7 @@ */ #include +#include #include #include #include -- GitLab From 54dc6cabe684375b3cf549c7b0545613d694aba8 Mon Sep 17 00:00:00 2001 From: Johannes Stezenbach Date: Mon, 14 Nov 2011 17:23:16 +0100 Subject: [PATCH 0321/3089] ASoC: sta32x: preserve coefficient RAM The coefficient RAM must be saved in a shadow so it can be restored when the codec is powered on using regulator_bulk_enable(). Signed-off-by: Johannes Stezenbach Signed-off-by: Mark Brown Cc: stable@vger.kernel.org --- sound/soc/codecs/sta32x.c | 63 ++++++++++++++++++++++++++++++++++++++- sound/soc/codecs/sta32x.h | 1 + 2 files changed, 63 insertions(+), 1 deletion(-) diff --git a/sound/soc/codecs/sta32x.c b/sound/soc/codecs/sta32x.c index bb82408ab8e1b..d2f37152f940c 100644 --- a/sound/soc/codecs/sta32x.c +++ b/sound/soc/codecs/sta32x.c @@ -76,6 +76,8 @@ struct sta32x_priv { unsigned int mclk; unsigned int format; + + u32 coef_shadow[STA32X_COEF_COUNT]; }; static const DECLARE_TLV_DB_SCALE(mvol_tlv, -12700, 50, 1); @@ -227,6 +229,7 @@ static int sta32x_coefficient_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol); + struct sta32x_priv *sta32x = snd_soc_codec_get_drvdata(codec); int numcoef = kcontrol->private_value >> 16; int index = kcontrol->private_value & 0xffff; unsigned int cfud; @@ -239,6 +242,11 @@ static int sta32x_coefficient_put(struct snd_kcontrol *kcontrol, snd_soc_write(codec, STA32X_CFUD, cfud); snd_soc_write(codec, STA32X_CFADDR2, index); + for (i = 0; i < numcoef && (index + i < STA32X_COEF_COUNT); i++) + sta32x->coef_shadow[index + i] = + (ucontrol->value.bytes.data[3 * i] << 16) + | (ucontrol->value.bytes.data[3 * i + 1] << 8) + | (ucontrol->value.bytes.data[3 * i + 2]); for (i = 0; i < 3 * numcoef; i++) snd_soc_write(codec, STA32X_B1CF1 + i, ucontrol->value.bytes.data[i]); @@ -252,6 +260,48 @@ static int sta32x_coefficient_put(struct snd_kcontrol *kcontrol, return 0; } +int sta32x_sync_coef_shadow(struct snd_soc_codec *codec) +{ + struct sta32x_priv *sta32x = snd_soc_codec_get_drvdata(codec); + unsigned int cfud; + int i; + + /* preserve reserved bits in STA32X_CFUD */ + cfud = snd_soc_read(codec, STA32X_CFUD) & 0xf0; + + for (i = 0; i < STA32X_COEF_COUNT; i++) { + snd_soc_write(codec, STA32X_CFADDR2, i); + snd_soc_write(codec, STA32X_B1CF1, + (sta32x->coef_shadow[i] >> 16) & 0xff); + snd_soc_write(codec, STA32X_B1CF2, + (sta32x->coef_shadow[i] >> 8) & 0xff); + snd_soc_write(codec, STA32X_B1CF3, + (sta32x->coef_shadow[i]) & 0xff); + /* chip documentation does not say if the bits are + * self-clearing, so do it explicitly */ + snd_soc_write(codec, STA32X_CFUD, cfud); + snd_soc_write(codec, STA32X_CFUD, cfud | 0x01); + } + return 0; +} + +int sta32x_cache_sync(struct snd_soc_codec *codec) +{ + unsigned int mute; + int rc; + + if (!codec->cache_sync) + return 0; + + /* mute during register sync */ + mute = snd_soc_read(codec, STA32X_MMUTE); + snd_soc_write(codec, STA32X_MMUTE, mute | STA32X_MMUTE_MMUTE); + sta32x_sync_coef_shadow(codec); + rc = snd_soc_cache_sync(codec); + snd_soc_write(codec, STA32X_MMUTE, mute); + return rc; +} + #define SINGLE_COEF(xname, index) \ { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = xname, \ .info = sta32x_coefficient_info, \ @@ -661,7 +711,7 @@ static int sta32x_set_bias_level(struct snd_soc_codec *codec, return ret; } - snd_soc_cache_sync(codec); + sta32x_cache_sync(codec); } /* Power up to mute */ @@ -790,6 +840,17 @@ static int sta32x_probe(struct snd_soc_codec *codec) STA32X_CxCFG_OM_MASK, 2 << STA32X_CxCFG_OM_SHIFT); + /* initialize coefficient shadow RAM with reset values */ + for (i = 4; i <= 49; i += 5) + sta32x->coef_shadow[i] = 0x400000; + for (i = 50; i <= 54; i++) + sta32x->coef_shadow[i] = 0x7fffff; + sta32x->coef_shadow[55] = 0x5a9df7; + sta32x->coef_shadow[56] = 0x7fffff; + sta32x->coef_shadow[59] = 0x7fffff; + sta32x->coef_shadow[60] = 0x400000; + sta32x->coef_shadow[61] = 0x400000; + sta32x_set_bias_level(codec, SND_SOC_BIAS_STANDBY); /* Bias level configuration will have done an extra enable */ regulator_bulk_disable(ARRAY_SIZE(sta32x->supplies), sta32x->supplies); diff --git a/sound/soc/codecs/sta32x.h b/sound/soc/codecs/sta32x.h index b97ee5a756673..d8e32a6262ee0 100644 --- a/sound/soc/codecs/sta32x.h +++ b/sound/soc/codecs/sta32x.h @@ -19,6 +19,7 @@ /* STA326 register addresses */ #define STA32X_REGISTER_COUNT 0x2d +#define STA32X_COEF_COUNT 62 #define STA32X_CONFA 0x00 #define STA32X_CONFB 0x01 -- GitLab From 583182ba5f02c8c9be82ea550f2051eaec15b975 Mon Sep 17 00:00:00 2001 From: Bart Hartgers Date: Wed, 26 Oct 2011 13:29:42 +0200 Subject: [PATCH 0322/3089] USB: ark3116 initialisation fix This patch for the usb serial ark3116 driver fixes an initialisation ordering bug that gets triggered on hotplug when using at least recent debian/ubuntu userspace. Without it, ark3116 serial cables don't work. Signed-off-by: Bart Hartgers Tested-by: law_ence.dev@ntlworld.com Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/ark3116.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/usb/serial/ark3116.c b/drivers/usb/serial/ark3116.c index 5cdb9d9122751..18e875b92e001 100644 --- a/drivers/usb/serial/ark3116.c +++ b/drivers/usb/serial/ark3116.c @@ -42,7 +42,7 @@ static int debug; * Version information */ -#define DRIVER_VERSION "v0.6" +#define DRIVER_VERSION "v0.7" #define DRIVER_AUTHOR "Bart Hartgers " #define DRIVER_DESC "USB ARK3116 serial/IrDA driver" #define DRIVER_DEV_DESC "ARK3116 RS232/IrDA" @@ -380,10 +380,6 @@ static int ark3116_open(struct tty_struct *tty, struct usb_serial_port *port) goto err_out; } - /* setup termios */ - if (tty) - ark3116_set_termios(tty, port, NULL); - /* remove any data still left: also clears error state */ ark3116_read_reg(serial, UART_RX, buf); @@ -406,6 +402,10 @@ static int ark3116_open(struct tty_struct *tty, struct usb_serial_port *port) /* enable DMA */ ark3116_write_reg(port->serial, UART_FCR, UART_FCR_DMA_SELECT); + /* setup termios */ + if (tty) + ark3116_set_termios(tty, port, NULL); + err_out: kfree(buf); return result; -- GitLab From 97ff22ee3b4cb3a334f7385e269773141aed702f Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Thu, 27 Oct 2011 11:20:21 -0400 Subject: [PATCH 0323/3089] USB: workaround for bug in old version of GCC This patch (as1491) works around a bug in GCC-3.4.6, which is still supposed to be supported. The number of microseconds in the udelay() call in quirk_usb_disable_ehci() is fixed at 100, but the compiler doesn't understand this and generates a link-time error. So we replace the otherwise unused variable "delta" with a simple constant 100. This same pattern is already used in other delay loops in that source file. Signed-off-by: Alan Stern Reported-by: Konrad Rzepecki Tested-by: Konrad Rzepecki Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/pci-quirks.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/usb/host/pci-quirks.c b/drivers/usb/host/pci-quirks.c index 27a3dec32fa2f..c7fd6ce11904e 100644 --- a/drivers/usb/host/pci-quirks.c +++ b/drivers/usb/host/pci-quirks.c @@ -627,7 +627,7 @@ static void __devinit quirk_usb_disable_ehci(struct pci_dev *pdev) void __iomem *base, *op_reg_base; u32 hcc_params, cap, val; u8 offset, cap_length; - int wait_time, delta, count = 256/4; + int wait_time, count = 256/4; if (!mmio_resource_enabled(pdev, 0)) return; @@ -673,11 +673,10 @@ static void __devinit quirk_usb_disable_ehci(struct pci_dev *pdev) writel(val, op_reg_base + EHCI_USBCMD); wait_time = 2000; - delta = 100; do { writel(0x3f, op_reg_base + EHCI_USBSTS); - udelay(delta); - wait_time -= delta; + udelay(100); + wait_time -= 100; val = readl(op_reg_base + EHCI_USBSTS); if ((val == ~(u32)0) || (val & EHCI_USBSTS_HALTED)) { break; -- GitLab From f69e3120df82391a0ee8118e0a156239a06b2afb Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Thu, 3 Nov 2011 11:37:10 -0400 Subject: [PATCH 0324/3089] USB: XHCI: resume root hubs when the controller resumes This patch (as1494) fixes a problem in xhci-hcd's resume routine. When the controller is runtime-resumed, this can only mean that one of the two root hubs has made a wakeup request and therefore needs to be resumed as well. Rather than try to determine which root hub requires attention (which might be difficult in the case where a new non-SuperSpeed device has been plugged in), the patch simply resumes both root hubs. Without this change, there is a race: The controller might be put back to sleep before it can activate its IRQ line, and the wakeup condition might never get handled. The patch also simplifies the logic in xhci_resume a little, combining some repeated flag settings into a single pair of statements. Signed-off-by: Alan Stern CC: Sarah Sharp Cc: stable Tested-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci.c | 29 ++++++++++++++--------------- 1 file changed, 14 insertions(+), 15 deletions(-) diff --git a/drivers/usb/host/xhci.c b/drivers/usb/host/xhci.c index 747c5ead922bb..aa94c01957919 100644 --- a/drivers/usb/host/xhci.c +++ b/drivers/usb/host/xhci.c @@ -799,7 +799,7 @@ int xhci_resume(struct xhci_hcd *xhci, bool hibernated) u32 command, temp = 0; struct usb_hcd *hcd = xhci_to_hcd(xhci); struct usb_hcd *secondary_hcd; - int retval; + int retval = 0; /* Wait a bit if either of the roothubs need to settle from the * transition into bus suspend. @@ -809,6 +809,9 @@ int xhci_resume(struct xhci_hcd *xhci, bool hibernated) xhci->bus_state[1].next_statechange)) msleep(100); + set_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags); + set_bit(HCD_FLAG_HW_ACCESSIBLE, &xhci->shared_hcd->flags); + spin_lock_irq(&xhci->lock); if (xhci->quirks & XHCI_RESET_ON_RESUME) hibernated = true; @@ -878,20 +881,13 @@ int xhci_resume(struct xhci_hcd *xhci, bool hibernated) return retval; xhci_dbg(xhci, "Start the primary HCD\n"); retval = xhci_run(hcd->primary_hcd); - if (retval) - goto failed_restart; - - xhci_dbg(xhci, "Start the secondary HCD\n"); - retval = xhci_run(secondary_hcd); if (!retval) { - set_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags); - set_bit(HCD_FLAG_HW_ACCESSIBLE, - &xhci->shared_hcd->flags); + xhci_dbg(xhci, "Start the secondary HCD\n"); + retval = xhci_run(secondary_hcd); } -failed_restart: hcd->state = HC_STATE_SUSPENDED; xhci->shared_hcd->state = HC_STATE_SUSPENDED; - return retval; + goto done; } /* step 4: set Run/Stop bit */ @@ -910,11 +906,14 @@ failed_restart: * Running endpoints by ringing their doorbells */ - set_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags); - set_bit(HCD_FLAG_HW_ACCESSIBLE, &xhci->shared_hcd->flags); - spin_unlock_irq(&xhci->lock); - return 0; + + done: + if (retval == 0) { + usb_hcd_resume_root_hub(hcd); + usb_hcd_resume_root_hub(xhci->shared_hcd); + } + return retval; } #endif /* CONFIG_PM */ -- GitLab From 811c926c538f7e8d3c08b630dd5844efd7e000f6 Mon Sep 17 00:00:00 2001 From: Thomas Poussevin Date: Thu, 27 Oct 2011 18:46:48 +0200 Subject: [PATCH 0325/3089] USB: EHCI: fix HUB TT scheduling issue with iso transfer The current TT scheduling doesn't allow to play and then record on a full-speed device connected to a high speed hub. The IN iso stream can only start on the first uframe (0-2 for a 165 us) because of CSPLIT transactions. For the OUT iso stream there no such restriction. uframe 0-5 are possible. The idea of this patch is that the first uframe are precious (for IN TT iso stream) and we should allocate the last uframes first if possible. For that we reverse the order of uframe allocation (last uframe first). Here an example : hid interrupt stream ---------------------------------------------------------------------- uframe | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | ---------------------------------------------------------------------- max_tt_usecs | 125 | 125 | 125 | 125 | 125 | 125 | 30 | 0 | ---------------------------------------------------------------------- used usecs on a frame | 13 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | ---------------------------------------------------------------------- iso OUT stream ---------------------------------------------------------------------- uframe | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | ---------------------------------------------------------------------- max_tt_usecs | 125 | 125 | 125 | 125 | 125 | 125 | 30 | 0 | ---------------------------------------------------------------------- used usecs on a frame | 13 | 125 | 39 | 0 | 0 | 0 | 0 | 0 | ---------------------------------------------------------------------- There no place for iso IN stream (uframe 0-2 are used) and we got "cannot submit datapipe for urb 0, error -28: not enough bandwidth" error. With the patch this become. iso OUT stream ---------------------------------------------------------------------- uframe | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | ---------------------------------------------------------------------- max_tt_usecs | 125 | 125 | 125 | 125 | 125 | 125 | 30 | 0 | ---------------------------------------------------------------------- used usecs on a frame | 13 | 0 | 0 | 0 | 125 | 39 | 0 | 0 | ---------------------------------------------------------------------- iso IN stream ---------------------------------------------------------------------- uframe | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | ---------------------------------------------------------------------- max_tt_usecs | 125 | 125 | 125 | 125 | 125 | 125 | 30 | 0 | ---------------------------------------------------------------------- used usecs on a frame | 13 | 0 | 125 | 40 | 125 | 39 | 0 | 0 | ---------------------------------------------------------------------- Signed-off-by: Matthieu Castet Signed-off-by: Thomas Poussevin Signed-off-by: Alan Stern Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/ehci-sched.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/drivers/usb/host/ehci-sched.c b/drivers/usb/host/ehci-sched.c index 2e829fae64829..56a32033adb34 100644 --- a/drivers/usb/host/ehci-sched.c +++ b/drivers/usb/host/ehci-sched.c @@ -1479,10 +1479,15 @@ iso_stream_schedule ( /* NOTE: assumes URB_ISO_ASAP, to limit complexity/bugs */ - /* find a uframe slot with enough bandwidth */ - next = start + period; - for (; start < next; start++) { - + /* find a uframe slot with enough bandwidth. + * Early uframes are more precious because full-speed + * iso IN transfers can't use late uframes, + * and therefore they should be allocated last. + */ + next = start; + start += period; + do { + start--; /* check schedule: enough space? */ if (stream->highspeed) { if (itd_slot_ok(ehci, mod, start, @@ -1495,7 +1500,7 @@ iso_stream_schedule ( start, sched, period)) break; } - } + } while (start > next); /* no room in the schedule */ if (start == next) { -- GitLab From 60c71ca972a2dd3fd9d0165b405361c8ad48349b Mon Sep 17 00:00:00 2001 From: Josh Boyer Date: Wed, 26 Oct 2011 13:53:17 -0400 Subject: [PATCH 0326/3089] USB: add quirk for Logitech C600 web cam We've had another report of the "chipmunk" sound on a Logitech C600 webcam. This patch resolves the issue. Cc: stable Signed-off-by: Josh Boyer Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/quirks.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/core/quirks.c b/drivers/usb/core/quirks.c index d6a8d8269bfbe..caa19914806a8 100644 --- a/drivers/usb/core/quirks.c +++ b/drivers/usb/core/quirks.c @@ -50,6 +50,9 @@ static const struct usb_device_id usb_quirk_list[] = { /* Logitech Webcam B/C500 */ { USB_DEVICE(0x046d, 0x0807), .driver_info = USB_QUIRK_RESET_RESUME }, + /* Logitech Webcam C600 */ + { USB_DEVICE(0x046d, 0x0808), .driver_info = USB_QUIRK_RESET_RESUME }, + /* Logitech Webcam Pro 9000 */ { USB_DEVICE(0x046d, 0x0809), .driver_info = USB_QUIRK_RESET_RESUME }, -- GitLab From 5dc2470c602da8851907ec18942cd876c3b4ecc1 Mon Sep 17 00:00:00 2001 From: Havard Skinnemoen Date: Wed, 9 Nov 2011 13:47:38 -0800 Subject: [PATCH 0327/3089] USB: cdc-acm: Fix disconnect() vs close() race There's a race between the USB disconnect handler and the TTY close handler which may cause the acm object to be freed while it's still being used. This may lead to things like http://article.gmane.org/gmane.linux.usb.general/54250 and https://lkml.org/lkml/2011/5/29/64 This is the simplest fix I could come up with. Holding on to open_mutex while closing the TTY device prevents acm_disconnect() from freeing the acm object between acm->port.count drops to 0 and the TTY side of the cleanups are finalized. Signed-off-by: Havard Skinnemoen Cc: Oliver Neukum Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/class/cdc-acm.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index 6960715c5063c..e8c564a533469 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -539,7 +539,6 @@ static void acm_port_down(struct acm *acm) { int i; - mutex_lock(&open_mutex); if (acm->dev) { usb_autopm_get_interface(acm->control); acm_set_control(acm, acm->ctrlout = 0); @@ -551,14 +550,15 @@ static void acm_port_down(struct acm *acm) acm->control->needs_remote_wakeup = 0; usb_autopm_put_interface(acm->control); } - mutex_unlock(&open_mutex); } static void acm_tty_hangup(struct tty_struct *tty) { struct acm *acm = tty->driver_data; tty_port_hangup(&acm->port); + mutex_lock(&open_mutex); acm_port_down(acm); + mutex_unlock(&open_mutex); } static void acm_tty_close(struct tty_struct *tty, struct file *filp) @@ -569,8 +569,9 @@ static void acm_tty_close(struct tty_struct *tty, struct file *filp) shutdown */ if (!acm) return; + + mutex_lock(&open_mutex); if (tty_port_close_start(&acm->port, tty, filp) == 0) { - mutex_lock(&open_mutex); if (!acm->dev) { tty_port_tty_set(&acm->port, NULL); acm_tty_unregister(acm); @@ -582,6 +583,7 @@ static void acm_tty_close(struct tty_struct *tty, struct file *filp) acm_port_down(acm); tty_port_close_end(&acm->port, tty); tty_port_tty_set(&acm->port, NULL); + mutex_unlock(&open_mutex); } static int acm_tty_write(struct tty_struct *tty, -- GitLab From bc9f54498eadc5254378e376e8d2111cfd30b229 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 11 Nov 2011 13:10:08 +0100 Subject: [PATCH 0328/3089] PM / shmobile: Avoid restoring the INTCS state during initialization The SH7372 PM domain initialization routine calls pd_power_up() that executes the domain's .resume() callback, if present, and for the A4R domain this callback attepmts to restore the INTCS state from uninitialized data. To avoid that, introduce __pd_power_up() that will only execute the domain's .resume() callback if its second argument is 'true' and make the SH7372 PM domain initialization use it with 'false' as its second argument. Redefine pd_power_up() as a wrapper around __pd_power_up(). Reported-by: Guennadi Liakhovetski Tracked-down-by: Magnus Damm Signed-off-by: Rafael J. Wysocki Acked-by: Magnus Damm --- arch/arm/mach-shmobile/pm-sh7372.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/arch/arm/mach-shmobile/pm-sh7372.c b/arch/arm/mach-shmobile/pm-sh7372.c index 78c98a8474335..1ec35ebe23764 100644 --- a/arch/arm/mach-shmobile/pm-sh7372.c +++ b/arch/arm/mach-shmobile/pm-sh7372.c @@ -107,9 +107,8 @@ static int pd_power_down(struct generic_pm_domain *genpd) return 0; } -static int pd_power_up(struct generic_pm_domain *genpd) +static int __pd_power_up(struct sh7372_pm_domain *sh7372_pd, bool do_resume) { - struct sh7372_pm_domain *sh7372_pd = to_sh7372_pd(genpd); unsigned int mask = 1 << sh7372_pd->bit_shift; unsigned int retry_count; int ret = 0; @@ -138,12 +137,17 @@ static int pd_power_up(struct generic_pm_domain *genpd) mask, __raw_readl(PSTR)); out: - if (ret == 0 && sh7372_pd->resume) + if (ret == 0 && sh7372_pd->resume && do_resume) sh7372_pd->resume(); return ret; } +static int pd_power_up(struct generic_pm_domain *genpd) +{ + return __pd_power_up(to_sh7372_pd(genpd), true); +} + static void sh7372_a4r_suspend(void) { sh7372_intcs_suspend(); @@ -175,7 +179,7 @@ void sh7372_init_pm_domain(struct sh7372_pm_domain *sh7372_pd) genpd->active_wakeup = pd_active_wakeup; genpd->power_off = pd_power_down; genpd->power_on = pd_power_up; - genpd->power_on(&sh7372_pd->genpd); + __pd_power_up(sh7372_pd, false); } void sh7372_add_device_to_domain(struct sh7372_pm_domain *sh7372_pd, -- GitLab From 9f3bdd4f937a75c4589a867dc1f8fefe09c1a618 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Mon, 14 Nov 2011 23:31:29 +0100 Subject: [PATCH 0329/3089] PM / devfreq: fix use after free in devfreq_remove_device In devfreq_remove_device, calling _remove_devfreq will also free devfreq. Don't dereference devfreq->governor->no_central_polling after _remove_devfreq. Signed-off-by: Axel Lin Acked-by: MyungJoo Ham Signed-off-by: Rafael J. Wysocki --- drivers/devfreq/devfreq.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c index d0659253387a3..59d24e9cb8c51 100644 --- a/drivers/devfreq/devfreq.c +++ b/drivers/devfreq/devfreq.c @@ -418,10 +418,14 @@ out: */ int devfreq_remove_device(struct devfreq *devfreq) { + bool central_polling; + if (!devfreq) return -EINVAL; - if (!devfreq->governor->no_central_polling) { + central_polling = !devfreq->governor->no_central_polling; + + if (central_polling) { mutex_lock(&devfreq_list_lock); while (wait_remove_device == devfreq) { mutex_unlock(&devfreq_list_lock); @@ -433,7 +437,7 @@ int devfreq_remove_device(struct devfreq *devfreq) mutex_lock(&devfreq->lock); _remove_devfreq(devfreq, false); /* it unlocks devfreq->lock */ - if (!devfreq->governor->no_central_polling) + if (central_polling) mutex_unlock(&devfreq_list_lock); return 0; -- GitLab From 6c81f90588972b8631bda4d538ffd0199a3e1b41 Mon Sep 17 00:00:00 2001 From: MyungJoo Ham Date: Mon, 14 Nov 2011 23:31:35 +0100 Subject: [PATCH 0330/3089] PM / devfreq: correct Kconfig dependency Devfreq does not depend on OPP. The dependency is removed. Signed-off-by: MyungJoo Ham Signed-off-by: Kyungmin Park Signed-off-by: Rafael J. Wysocki --- drivers/devfreq/Kconfig | 41 +++++++++++++++++------------------------ 1 file changed, 17 insertions(+), 24 deletions(-) diff --git a/drivers/devfreq/Kconfig b/drivers/devfreq/Kconfig index 643b055ed3cdc..8f0491037080d 100644 --- a/drivers/devfreq/Kconfig +++ b/drivers/devfreq/Kconfig @@ -1,36 +1,29 @@ -config ARCH_HAS_DEVFREQ - bool - depends on ARCH_HAS_OPP - help - Denotes that the architecture supports DEVFREQ. If the architecture - supports multiple OPP entries per device and the frequency of the - devices with OPPs may be altered dynamically, the architecture - supports DEVFREQ. - menuconfig PM_DEVFREQ bool "Generic Dynamic Voltage and Frequency Scaling (DVFS) support" - depends on PM_OPP && ARCH_HAS_DEVFREQ help - With OPP support, a device may have a list of frequencies and - voltages available. DEVFREQ, a generic DVFS framework can be - registered for a device with OPP support in order to let the - governor provided to DEVFREQ choose an operating frequency - based on the OPP's list and the policy given with DEVFREQ. + A device may have a list of frequencies and voltages available. + devfreq, a generic DVFS framework can be registered for a device + in order to let the governor provided to devfreq choose an + operating frequency based on the device driver's policy. - Each device may have its own governor and policy. DEVFREQ can + Each device may have its own governor and policy. Devfreq can reevaluate the device state periodically and/or based on the - OPP list changes (each frequency/voltage pair in OPP may be - disabled or enabled). + notification to "nb", a notifier block, of devfreq. - Like some CPUs with CPUFREQ, a device may have multiple clocks. + Like some CPUs with CPUfreq, a device may have multiple clocks. However, because the clock frequencies of a single device are - determined by the single device's state, an instance of DEVFREQ + determined by the single device's state, an instance of devfreq is attached to a single device and returns a "representative" - clock frequency from the OPP of the device, which is also attached - to a device by 1-to-1. The device registering DEVFREQ takes the - responsiblity to "interpret" the frequency listed in OPP and + clock frequency of the device, which is also attached + to a device by 1-to-1. The device registering devfreq takes the + responsiblity to "interpret" the representative frequency and to set its every clock accordingly with the "target" callback - given to DEVFREQ. + given to devfreq. + + When OPP is used with the devfreq device, it is recommended to + register devfreq's nb to the OPP's notifier head. If OPP is + used with the devfreq device, you may use OPP helper + functions defined in devfreq.h. if PM_DEVFREQ -- GitLab From 770f0baaee0c511d0d466082e72815ed861e301f Mon Sep 17 00:00:00 2001 From: Jean-Christophe PLAGNIOL-VILLARD Date: Sat, 12 Nov 2011 16:46:13 +0100 Subject: [PATCH 0331/3089] USB: at91: at91-ohci: fix set/get power in commit aa6e52a35 we introduce the support of overcurrent notification but the set and get of the power without checking if the gpio is valid or not Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD Acked-by: Thomas Petazzoni Acked-by: Nicolas Ferre Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/ohci-at91.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/usb/host/ohci-at91.c b/drivers/usb/host/ohci-at91.c index ba3a46b78b75a..95a9fec38e89b 100644 --- a/drivers/usb/host/ohci-at91.c +++ b/drivers/usb/host/ohci-at91.c @@ -223,6 +223,9 @@ static void ohci_at91_usb_set_power(struct at91_usbh_data *pdata, int port, int if (port < 0 || port >= 2) return; + if (pdata->vbus_pin[port] <= 0) + return; + gpio_set_value(pdata->vbus_pin[port], !pdata->vbus_pin_inverted ^ enable); } @@ -231,6 +234,9 @@ static int ohci_at91_usb_get_power(struct at91_usbh_data *pdata, int port) if (port < 0 || port >= 2) return -EINVAL; + if (pdata->vbus_pin[port] <= 0) + return -EINVAL; + return gpio_get_value(pdata->vbus_pin[port]) ^ !pdata->vbus_pin_inverted; } -- GitLab From 8981d76a2cf4c46333d9a18176a6eee337a71146 Mon Sep 17 00:00:00 2001 From: Shaohui Xie Date: Mon, 7 Nov 2011 16:58:20 +0800 Subject: [PATCH 0332/3089] powerpc/usb: fix type cast for address of ioremap to compatible with 64-bit Below are codes for accessing usb sysif_regs in driver: usb_sys_regs = (struct usb_sys_interface *) ((u32)dr_regs + USB_DR_SYS_OFFSET); these codes work in 32-bit, but in 64-bit, use u32 to type cast the address of ioremap is not right, and accessing members of 'usb_sys_regs' will cause call trace, so use (void *) for both 32-bit and 64-bit. Signed-off-by: Shaohui Xie Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/fsl_udc_core.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/usb/gadget/fsl_udc_core.c b/drivers/usb/gadget/fsl_udc_core.c index d786ba31fc07f..b3b3d83b7c335 100644 --- a/drivers/usb/gadget/fsl_udc_core.c +++ b/drivers/usb/gadget/fsl_udc_core.c @@ -2480,8 +2480,7 @@ static int __init fsl_udc_probe(struct platform_device *pdev) #ifndef CONFIG_ARCH_MXC if (pdata->have_sysif_regs) - usb_sys_regs = (struct usb_sys_interface *) - ((u32)dr_regs + USB_DR_SYS_OFFSET); + usb_sys_regs = (void *)dr_regs + USB_DR_SYS_OFFSET; #endif /* Initialize USB clocks */ -- GitLab From 46e642614b360ae783ce32b43dd885b797352635 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Sun, 30 Oct 2011 07:34:04 -0700 Subject: [PATCH 0333/3089] MAINTAINERS: add CHAR and MISC driver maintainers Yes, we are fools, but humor us. Signed-off-by: Greg Kroah-Hartman Acked-by: Arnd Bergmann --- MAINTAINERS | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/MAINTAINERS b/MAINTAINERS index 4808256446f29..c9c0643bc46ff 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1788,6 +1788,14 @@ F: include/net/cfg80211.h F: net/wireless/* X: net/wireless/wext* +CHAR and MISC DRIVERS +M: Arnd Bergmann +M: Greg Kroah-Hartman +T: git git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc.git +S: Maintained +F: drivers/char/* +F: drivers/misc/* + CHECKPATCH M: Andy Whitcroft S: Supported -- GitLab From 161f14191dc166c4e3f37f68af1bc199c6868b7d Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Mon, 31 Oct 2011 10:20:28 +0800 Subject: [PATCH 0334/3089] pcie-gadget-spear: Add "platform:" prefix for platform modalias Since 43cc71eed1250755986da4c0f9898f9a635cb3bf (platform: prefix MODALIAS with "platform:"), the platform modalias is prefixed with "platform:". Signed-off-by: Axel Lin Acked-by: Pratyush Anand Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/misc/spear13xx_pcie_gadget.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/misc/spear13xx_pcie_gadget.c b/drivers/misc/spear13xx_pcie_gadget.c index cfbddbef11de3..43d073bc1d9c5 100644 --- a/drivers/misc/spear13xx_pcie_gadget.c +++ b/drivers/misc/spear13xx_pcie_gadget.c @@ -903,6 +903,6 @@ static void __exit spear_pcie_gadget_exit(void) } module_exit(spear_pcie_gadget_exit); -MODULE_ALIAS("pcie-gadget-spear"); +MODULE_ALIAS("platform:pcie-gadget-spear"); MODULE_AUTHOR("Pratyush Anand"); MODULE_LICENSE("GPL"); -- GitLab From 0d145d7d4a241c321c832a810bb6edad18e2217b Mon Sep 17 00:00:00 2001 From: sordna Date: Thu, 27 Oct 2011 21:06:26 -0700 Subject: [PATCH 0335/3089] USB: quirks: adding more quirky webcams to avoid squeaky audio The following patch contains additional affected webcam models, on top of the patches commited to linux-next 2394d67e446bf616a0885167d5f0d397bdacfdfc and 5b253d88cc6c65a23cefc457a5a4ef139913c5fc Signed-off-by: sordna Cc: Oliver Neukum Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/quirks.c | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) diff --git a/drivers/usb/core/quirks.c b/drivers/usb/core/quirks.c index caa19914806a8..ecf12e15a7ef4 100644 --- a/drivers/usb/core/quirks.c +++ b/drivers/usb/core/quirks.c @@ -56,12 +56,36 @@ static const struct usb_device_id usb_quirk_list[] = { /* Logitech Webcam Pro 9000 */ { USB_DEVICE(0x046d, 0x0809), .driver_info = USB_QUIRK_RESET_RESUME }, + /* Logitech Webcam C905 */ + { USB_DEVICE(0x046d, 0x080a), .driver_info = USB_QUIRK_RESET_RESUME }, + + /* Logitech Webcam C210 */ + { USB_DEVICE(0x046d, 0x0819), .driver_info = USB_QUIRK_RESET_RESUME }, + + /* Logitech Webcam C260 */ + { USB_DEVICE(0x046d, 0x081a), .driver_info = USB_QUIRK_RESET_RESUME }, + /* Logitech Webcam C310 */ { USB_DEVICE(0x046d, 0x081b), .driver_info = USB_QUIRK_RESET_RESUME }, + /* Logitech Webcam C910 */ + { USB_DEVICE(0x046d, 0x0821), .driver_info = USB_QUIRK_RESET_RESUME }, + + /* Logitech Webcam C160 */ + { USB_DEVICE(0x046d, 0x0824), .driver_info = USB_QUIRK_RESET_RESUME }, + /* Logitech Webcam C270 */ { USB_DEVICE(0x046d, 0x0825), .driver_info = USB_QUIRK_RESET_RESUME }, + /* Logitech Quickcam Pro 9000 */ + { USB_DEVICE(0x046d, 0x0990), .driver_info = USB_QUIRK_RESET_RESUME }, + + /* Logitech Quickcam E3500 */ + { USB_DEVICE(0x046d, 0x09a4), .driver_info = USB_QUIRK_RESET_RESUME }, + + /* Logitech Quickcam Vision Pro */ + { USB_DEVICE(0x046d, 0x09a6), .driver_info = USB_QUIRK_RESET_RESUME }, + /* Logitech Harmony 700-series */ { USB_DEVICE(0x046d, 0xc122), .driver_info = USB_QUIRK_DELAY_INIT }, -- GitLab From 2f640bf4c94324aeaa1b6385c10aab8c5ad1e1cf Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 25 Oct 2011 10:50:58 -0400 Subject: [PATCH 0336/3089] usb-storage: Accept 8020i-protocol commands longer than 12 bytes The 8020i protocol (also 8070i and QIC-157) uses 12-byte commands; shorter commands must be padded. Simon Detheridge reports that his 3-TB USB disk drive claims to use the 8020i protocol (which is normally meant for ATAPI devices like CD drives), and because of its large size, the disk drive requires the use of 16-byte commands. However the usb_stor_pad12_command() routine in usb-storage always sets the command length to 12, making the drive impossible to use. Since the SFF-8020i specification allows for 16-byte commands in future extensions, we may as well accept them. This patch (as1490) changes usb_stor_pad12_command() to leave commands larger than 12 bytes alone rather than truncating them. Signed-off-by: Alan Stern Tested-by: Simon Detheridge CC: Matthew Dharm Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/storage/protocol.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/usb/storage/protocol.c b/drivers/usb/storage/protocol.c index 93c1a4d86f517..82dd834709c78 100644 --- a/drivers/usb/storage/protocol.c +++ b/drivers/usb/storage/protocol.c @@ -59,7 +59,9 @@ void usb_stor_pad12_command(struct scsi_cmnd *srb, struct us_data *us) { - /* Pad the SCSI command with zeros out to 12 bytes + /* + * Pad the SCSI command with zeros out to 12 bytes. If the + * command already is 12 bytes or longer, leave it alone. * * NOTE: This only works because a scsi_cmnd struct field contains * a unsigned char cmnd[16], so we know we have storage available @@ -67,9 +69,6 @@ void usb_stor_pad12_command(struct scsi_cmnd *srb, struct us_data *us) for (; srb->cmd_len<12; srb->cmd_len++) srb->cmnd[srb->cmd_len] = 0; - /* set command length to 12 bytes */ - srb->cmd_len = 12; - /* send the command to the transport layer */ usb_stor_invoke_transport(srb, us); } -- GitLab From eec205719e97c95a422d0aa384d0517f6c74fc0a Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Mon, 14 Nov 2011 14:08:05 +0000 Subject: [PATCH 0337/3089] net/packet: Revert incorrect dead-code changes to prb_setup_retire_blk_timer Signed-off-by: Jesper Juhl Signed-off-by: David S. Miller --- net/packet/af_packet.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index ab10e84eda3af..82a6f34d39d01 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -516,11 +516,13 @@ static void prb_init_blk_timer(struct packet_sock *po, static void prb_setup_retire_blk_timer(struct packet_sock *po, int tx_ring) { + struct tpacket_kbdq_core *pkc; + if (tx_ring) BUG(); - prb_init_blk_timer(po, &po->rx_ring.prb_bdqc, - prb_retire_rx_blk_timer_expired); + pkc = tx_ring ? &po->tx_ring.prb_bdqc : &po->rx_ring.prb_bdqc; + prb_init_blk_timer(po, pkc, prb_retire_rx_blk_timer_expired); } static int prb_calc_retire_blk_tmo(struct packet_sock *po, -- GitLab From 1d299bc7732c34d85bd43ac1a8745f5a2fed2078 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 14 Nov 2011 20:32:16 -0800 Subject: [PATCH 0338/3089] sparc: Fix handling of orig_i0 wrt. debugging when restarting syscalls. Although we provide a proper way for a debugger to control whether syscall restart occurs, we run into problems because orig_i0 is not saved and restored properly. Luckily we can solve this problem without having to make debuggers aware of the issue. Across system calls, several registers are considered volatile and can be safely clobbered. Therefore we use the pt_regs save area of one of those registers, %g2, as a place to save and restore orig_i0. Debuggers transparently will do the right thing because they save and restore this register already. Signed-off-by: David S. Miller --- arch/sparc/kernel/signal32.c | 18 ++++++++++-------- arch/sparc/kernel/signal_32.c | 20 +++++++++++++++----- arch/sparc/kernel/signal_64.c | 32 ++++++++++++++++++-------------- 3 files changed, 43 insertions(+), 27 deletions(-) diff --git a/arch/sparc/kernel/signal32.c b/arch/sparc/kernel/signal32.c index 2caa556db86dc..2e86fd1ddc7b3 100644 --- a/arch/sparc/kernel/signal32.c +++ b/arch/sparc/kernel/signal32.c @@ -822,21 +822,23 @@ static inline void syscall_restart32(unsigned long orig_i0, struct pt_regs *regs * want to handle. Thus you cannot kill init even with a SIGKILL even by * mistake. */ -void do_signal32(sigset_t *oldset, struct pt_regs * regs, - int restart_syscall, unsigned long orig_i0) +void do_signal32(sigset_t *oldset, struct pt_regs * regs) { struct k_sigaction ka; + unsigned long orig_i0; + int restart_syscall; siginfo_t info; int signr; signr = get_signal_to_deliver(&info, &ka, regs, NULL); - /* If the debugger messes with the program counter, it clears - * the "in syscall" bit, directing us to not perform a syscall - * restart. - */ - if (restart_syscall && !pt_regs_is_syscall(regs)) - restart_syscall = 0; + restart_syscall = 0; + orig_i0 = 0; + if (pt_regs_is_syscall(regs) && + (regs->tstate & (TSTATE_XCARRY | TSTATE_ICARRY))) { + restart_syscall = 1; + orig_i0 = regs->u_regs[UREG_G2]; + } if (signr > 0) { if (restart_syscall) diff --git a/arch/sparc/kernel/signal_32.c b/arch/sparc/kernel/signal_32.c index 8ce247ac04cc0..7dfaff64cd6b4 100644 --- a/arch/sparc/kernel/signal_32.c +++ b/arch/sparc/kernel/signal_32.c @@ -519,10 +519,16 @@ static void do_signal(struct pt_regs *regs, unsigned long orig_i0) siginfo_t info; int signr; + /* It's a lot of work and synchronization to add a new ptrace + * register for GDB to save and restore in order to get + * orig_i0 correct for syscall restarts when debugging. + * + * However, we luckily can use the fact that several registers + * are volatile across system calls. One such register is + * %g2, so use that as a place to save away orig_i0. + */ if (pt_regs_is_syscall(regs) && (regs->psr & PSR_C)) - restart_syscall = 1; - else - restart_syscall = 0; + regs->u_regs[UREG_G2] = orig_i0; if (test_thread_flag(TIF_RESTORE_SIGMASK)) oldset = ¤t->saved_sigmask; @@ -535,8 +541,12 @@ static void do_signal(struct pt_regs *regs, unsigned long orig_i0) * the software "in syscall" bit, directing us to not perform * a syscall restart. */ - if (restart_syscall && !pt_regs_is_syscall(regs)) - restart_syscall = 0; + restart_syscall = 0; + if (pt_regs_is_syscall(regs) && (regs->psr & PSR_C)) { + restart_syscall = 1; + orig_i0 = regs->u_regs[UREG_G2]; + } + if (signr > 0) { if (restart_syscall) diff --git a/arch/sparc/kernel/signal_64.c b/arch/sparc/kernel/signal_64.c index a2b81598d9056..1ddf0dedb9290 100644 --- a/arch/sparc/kernel/signal_64.c +++ b/arch/sparc/kernel/signal_64.c @@ -529,11 +529,17 @@ static void do_signal(struct pt_regs *regs, unsigned long orig_i0) siginfo_t info; int signr; + /* It's a lot of work and synchronization to add a new ptrace + * register for GDB to save and restore in order to get + * orig_i0 correct for syscall restarts when debugging. + * + * However, we luckily can use the fact that several registers + * are volatile across system calls. One such register is + * %g2, so use that as a place to save away orig_i0. + */ if (pt_regs_is_syscall(regs) && - (regs->tstate & (TSTATE_XCARRY | TSTATE_ICARRY))) { - restart_syscall = 1; - } else - restart_syscall = 0; + (regs->tstate & (TSTATE_XCARRY | TSTATE_ICARRY))) + regs->u_regs[UREG_G2] = orig_i0; if (current_thread_info()->status & TS_RESTORE_SIGMASK) oldset = ¤t->saved_sigmask; @@ -542,22 +548,20 @@ static void do_signal(struct pt_regs *regs, unsigned long orig_i0) #ifdef CONFIG_COMPAT if (test_thread_flag(TIF_32BIT)) { - extern void do_signal32(sigset_t *, struct pt_regs *, - int restart_syscall, - unsigned long orig_i0); - do_signal32(oldset, regs, restart_syscall, orig_i0); + extern void do_signal32(sigset_t *, struct pt_regs *); + do_signal32(oldset, regs); return; } #endif signr = get_signal_to_deliver(&info, &ka, regs, NULL); - /* If the debugger messes with the program counter, it clears - * the software "in syscall" bit, directing us to not perform - * a syscall restart. - */ - if (restart_syscall && !pt_regs_is_syscall(regs)) - restart_syscall = 0; + restart_syscall = 0; + if (pt_regs_is_syscall(regs) && + (regs->tstate & (TSTATE_XCARRY | TSTATE_ICARRY))) { + restart_syscall = 1; + orig_i0 = regs->u_regs[UREG_G2]; + } if (signr > 0) { if (restart_syscall) -- GitLab From 7db3eba6bf84ab744e39dcce24b7e06d01bab913 Mon Sep 17 00:00:00 2001 From: Seung-Woo Kim Date: Tue, 18 Oct 2011 16:58:05 +0900 Subject: [PATCH 0339/3089] drm/exynos: added kms poll for handling hpd event this patch adds kms poll infrastructure to handle hotplug detection event Signed-off-by: Seung-Woo Kim Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_drv.c | 5 +++++ drivers/gpu/drm/exynos/exynos_drm_fb.c | 12 ++++++++++++ 2 files changed, 17 insertions(+) diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index 83810cbe3c177..53e2216de61dd 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -27,6 +27,7 @@ #include "drmP.h" #include "drm.h" +#include "drm_crtc_helper.h" #include @@ -61,6 +62,9 @@ static int exynos_drm_load(struct drm_device *dev, unsigned long flags) drm_mode_config_init(dev); + /* init kms poll for handling hpd */ + drm_kms_helper_poll_init(dev); + exynos_drm_mode_config_init(dev); /* @@ -116,6 +120,7 @@ static int exynos_drm_unload(struct drm_device *dev) exynos_drm_fbdev_fini(dev); exynos_drm_device_unregister(dev); drm_vblank_cleanup(dev); + drm_kms_helper_poll_fini(dev); drm_mode_config_cleanup(dev); kfree(dev->dev_private); diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c index 48d29cfd52401..7d91a542c7567 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fb.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c @@ -29,7 +29,9 @@ #include "drmP.h" #include "drm_crtc.h" #include "drm_crtc_helper.h" +#include "drm_fb_helper.h" +#include "exynos_drm_drv.h" #include "exynos_drm_fb.h" #include "exynos_drm_buf.h" #include "exynos_drm_gem.h" @@ -238,8 +240,18 @@ struct exynos_drm_buf_entry *exynos_drm_fb_get_buf(struct drm_framebuffer *fb) return entry; } +static void exynos_drm_output_poll_changed(struct drm_device *dev) +{ + struct exynos_drm_private *private = dev->dev_private; + struct drm_fb_helper *fb_helper = private->fb_helper; + + if (fb_helper) + drm_fb_helper_hotplug_event(fb_helper); +} + static struct drm_mode_config_funcs exynos_drm_mode_config_funcs = { .fb_create = exynos_drm_fb_create, + .output_poll_changed = exynos_drm_output_poll_changed, }; void exynos_drm_mode_config_init(struct drm_device *dev) -- GitLab From 1b17b206560c433ae9e8f8409f3f3842949a74c8 Mon Sep 17 00:00:00 2001 From: Seung-Woo Kim Date: Wed, 19 Oct 2011 15:10:10 +0900 Subject: [PATCH 0340/3089] drm/exynos: fixed connector flag with hpd and interlace scan for hdmi hdmi display in exynos supports hotplug event and interlace scan mode Signed-off-by: Seung-Woo Kim Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_connector.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/exynos/exynos_drm_connector.c b/drivers/gpu/drm/exynos/exynos_drm_connector.c index 985d9e7687287..d1a459fdbe92a 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_connector.c +++ b/drivers/gpu/drm/exynos/exynos_drm_connector.c @@ -254,6 +254,8 @@ struct drm_connector *exynos_drm_connector_create(struct drm_device *dev, switch (manager->display->type) { case EXYNOS_DISPLAY_TYPE_HDMI: type = DRM_MODE_CONNECTOR_HDMIA; + connector->interlace_allowed = true; + connector->polled = DRM_CONNECTOR_POLL_HPD; break; default: type = DRM_MODE_CONNECTOR_Unknown; -- GitLab From 8b58dfe0290cb57e3f8601b197f00c23fa39a60d Mon Sep 17 00:00:00 2001 From: Seung-Woo Kim Date: Wed, 19 Oct 2011 15:11:55 +0900 Subject: [PATCH 0341/3089] drm/exynos: fixed converting between display mode and timing missing members are added into converting function between timing and display mode and refresh rate of display mode is calculated by drm mode function. Signed-off-by: Seung-Woo Kim Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_connector.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_connector.c b/drivers/gpu/drm/exynos/exynos_drm_connector.c index d1a459fdbe92a..7ca1274775b77 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_connector.c +++ b/drivers/gpu/drm/exynos/exynos_drm_connector.c @@ -47,6 +47,7 @@ convert_to_display_mode(struct drm_display_mode *mode, DRM_DEBUG_KMS("%s\n", __FILE__); mode->clock = timing->pixclock / 1000; + mode->vrefresh = timing->refresh; mode->hdisplay = timing->xres; mode->hsync_start = mode->hdisplay + timing->left_margin; @@ -57,6 +58,12 @@ convert_to_display_mode(struct drm_display_mode *mode, mode->vsync_start = mode->vdisplay + timing->upper_margin; mode->vsync_end = mode->vsync_start + timing->vsync_len; mode->vtotal = mode->vsync_end + timing->lower_margin; + + if (timing->vmode & FB_VMODE_INTERLACED) + mode->flags |= DRM_MODE_FLAG_INTERLACE; + + if (timing->vmode & FB_VMODE_DOUBLE) + mode->flags |= DRM_MODE_FLAG_DBLSCAN; } /* convert drm_display_mode to exynos_video_timings */ @@ -69,7 +76,7 @@ convert_to_video_timing(struct fb_videomode *timing, memset(timing, 0, sizeof(*timing)); timing->pixclock = mode->clock * 1000; - timing->refresh = mode->vrefresh; + timing->refresh = drm_mode_vrefresh(mode); timing->xres = mode->hdisplay; timing->left_margin = mode->hsync_start - mode->hdisplay; -- GitLab From adb6b1596743e93e50fad2ff26d9604cda4361ab Mon Sep 17 00:00:00 2001 From: Inki Dae Date: Wed, 19 Oct 2011 17:16:55 +0900 Subject: [PATCH 0342/3089] drm/exynos: added manager object to connector connector contains some contents for display controller so the connector also should be able to access controller through manager. Signed-off-by: Inki Dae Signed-off-by: Seung-Woo Kim Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_connector.c | 38 +++++++++++++++---- drivers/gpu/drm/exynos/exynos_drm_encoder.c | 3 +- 2 files changed, 33 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_connector.c b/drivers/gpu/drm/exynos/exynos_drm_connector.c index 7ca1274775b77..d33f8039a882a 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_connector.c +++ b/drivers/gpu/drm/exynos/exynos_drm_connector.c @@ -37,6 +37,8 @@ struct exynos_drm_connector { struct drm_connector drm_connector; + uint32_t encoder_id; + struct exynos_drm_manager *manager; }; /* convert exynos_video_timings to drm_display_mode */ @@ -99,8 +101,9 @@ convert_to_video_timing(struct fb_videomode *timing, static int exynos_drm_connector_get_modes(struct drm_connector *connector) { - struct exynos_drm_manager *manager = - exynos_drm_get_manager(connector->encoder); + struct exynos_drm_connector *exynos_connector = + to_exynos_connector(connector); + struct exynos_drm_manager *manager = exynos_connector->manager; struct exynos_drm_display *display = manager->display; unsigned int count; @@ -169,8 +172,9 @@ static int exynos_drm_connector_get_modes(struct drm_connector *connector) static int exynos_drm_connector_mode_valid(struct drm_connector *connector, struct drm_display_mode *mode) { - struct exynos_drm_manager *manager = - exynos_drm_get_manager(connector->encoder); + struct exynos_drm_connector *exynos_connector = + to_exynos_connector(connector); + struct exynos_drm_manager *manager = exynos_connector->manager; struct exynos_drm_display *display = manager->display; struct fb_videomode timing; int ret = MODE_BAD; @@ -188,9 +192,25 @@ static int exynos_drm_connector_mode_valid(struct drm_connector *connector, struct drm_encoder *exynos_drm_best_encoder(struct drm_connector *connector) { + struct drm_device *dev = connector->dev; + struct exynos_drm_connector *exynos_connector = + to_exynos_connector(connector); + struct drm_mode_object *obj; + struct drm_encoder *encoder; + DRM_DEBUG_KMS("%s\n", __FILE__); - return connector->encoder; + obj = drm_mode_object_find(dev, exynos_connector->encoder_id, + DRM_MODE_OBJECT_ENCODER); + if (!obj) { + DRM_DEBUG_KMS("Unknown ENCODER ID %d\n", + exynos_connector->encoder_id); + return NULL; + } + + encoder = obj_to_encoder(obj); + + return encoder; } static struct drm_connector_helper_funcs exynos_connector_helper_funcs = { @@ -203,8 +223,9 @@ static struct drm_connector_helper_funcs exynos_connector_helper_funcs = { static enum drm_connector_status exynos_drm_connector_detect(struct drm_connector *connector, bool force) { - struct exynos_drm_manager *manager = - exynos_drm_get_manager(connector->encoder); + struct exynos_drm_connector *exynos_connector = + to_exynos_connector(connector); + struct exynos_drm_manager *manager = exynos_connector->manager; struct exynos_drm_display *display = manager->display; enum drm_connector_status status = connector_status_disconnected; @@ -276,7 +297,10 @@ struct drm_connector *exynos_drm_connector_create(struct drm_device *dev, if (err) goto err_connector; + exynos_connector->encoder_id = encoder->base.id; + exynos_connector->manager = manager; connector->encoder = encoder; + err = drm_mode_connector_attach_encoder(connector, encoder); if (err) { DRM_ERROR("failed to attach a connector to a encoder\n"); diff --git a/drivers/gpu/drm/exynos/exynos_drm_encoder.c b/drivers/gpu/drm/exynos/exynos_drm_encoder.c index 7cf6fa86a67ef..4e5535b2b8e38 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_encoder.c +++ b/drivers/gpu/drm/exynos/exynos_drm_encoder.c @@ -58,7 +58,8 @@ static void exynos_drm_encoder_dpms(struct drm_encoder *encoder, int mode) list_for_each_entry(connector, &dev->mode_config.connector_list, head) { if (connector->encoder == encoder) { - struct exynos_drm_display *display = manager->display; + struct exynos_drm_display *display = + manager->display; if (display && display->power_on) display->power_on(manager->dev, mode); -- GitLab From 74ccc539bcebdb24afb74194223f92a96a7285ed Mon Sep 17 00:00:00 2001 From: Inki Dae Date: Wed, 19 Oct 2011 17:23:07 +0900 Subject: [PATCH 0343/3089] drm/exynos: changed exynos_drm_display to exynos_drm_display_ops exynos_drm_display has function pointes so exynos_drm_display_ops is better to describe. Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_connector.c | 29 ++++++++++--------- drivers/gpu/drm/exynos/exynos_drm_drv.h | 4 +-- drivers/gpu/drm/exynos/exynos_drm_encoder.c | 8 ++--- drivers/gpu/drm/exynos/exynos_drm_fimd.c | 4 +-- 4 files changed, 23 insertions(+), 22 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_connector.c b/drivers/gpu/drm/exynos/exynos_drm_connector.c index d33f8039a882a..d620b0784257f 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_connector.c +++ b/drivers/gpu/drm/exynos/exynos_drm_connector.c @@ -104,13 +104,13 @@ static int exynos_drm_connector_get_modes(struct drm_connector *connector) struct exynos_drm_connector *exynos_connector = to_exynos_connector(connector); struct exynos_drm_manager *manager = exynos_connector->manager; - struct exynos_drm_display *display = manager->display; + struct exynos_drm_display_ops *display_ops = manager->display_ops; unsigned int count; DRM_DEBUG_KMS("%s\n", __FILE__); - if (!display) { - DRM_DEBUG_KMS("display is null.\n"); + if (!display_ops) { + DRM_DEBUG_KMS("display_ops is null.\n"); return 0; } @@ -122,7 +122,7 @@ static int exynos_drm_connector_get_modes(struct drm_connector *connector) * P.S. in case of lcd panel, count is always 1 if success * because lcd panel has only one mode. */ - if (display->get_edid) { + if (display_ops->get_edid) { int ret; void *edid; @@ -132,7 +132,7 @@ static int exynos_drm_connector_get_modes(struct drm_connector *connector) return 0; } - ret = display->get_edid(manager->dev, connector, + ret = display_ops->get_edid(manager->dev, connector, edid, MAX_EDID); if (ret < 0) { DRM_ERROR("failed to get edid data.\n"); @@ -150,8 +150,8 @@ static int exynos_drm_connector_get_modes(struct drm_connector *connector) struct drm_display_mode *mode = drm_mode_create(connector->dev); struct fb_videomode *timing; - if (display->get_timing) - timing = display->get_timing(manager->dev); + if (display_ops->get_timing) + timing = display_ops->get_timing(manager->dev); else { drm_mode_destroy(connector->dev, mode); return 0; @@ -175,7 +175,7 @@ static int exynos_drm_connector_mode_valid(struct drm_connector *connector, struct exynos_drm_connector *exynos_connector = to_exynos_connector(connector); struct exynos_drm_manager *manager = exynos_connector->manager; - struct exynos_drm_display *display = manager->display; + struct exynos_drm_display_ops *display_ops = manager->display_ops; struct fb_videomode timing; int ret = MODE_BAD; @@ -183,8 +183,8 @@ static int exynos_drm_connector_mode_valid(struct drm_connector *connector, convert_to_video_timing(&timing, mode); - if (display && display->check_timing) - if (!display->check_timing(manager->dev, (void *)&timing)) + if (display_ops && display_ops->check_timing) + if (!display_ops->check_timing(manager->dev, (void *)&timing)) ret = MODE_OK; return ret; @@ -226,13 +226,14 @@ exynos_drm_connector_detect(struct drm_connector *connector, bool force) struct exynos_drm_connector *exynos_connector = to_exynos_connector(connector); struct exynos_drm_manager *manager = exynos_connector->manager; - struct exynos_drm_display *display = manager->display; + struct exynos_drm_display_ops *display_ops = + manager->display_ops; enum drm_connector_status status = connector_status_disconnected; DRM_DEBUG_KMS("%s\n", __FILE__); - if (display && display->is_connected) { - if (display->is_connected(manager->dev)) + if (display_ops && display_ops->is_connected) { + if (display_ops->is_connected(manager->dev)) status = connector_status_connected; else status = connector_status_disconnected; @@ -279,7 +280,7 @@ struct drm_connector *exynos_drm_connector_create(struct drm_device *dev, connector = &exynos_connector->drm_connector; - switch (manager->display->type) { + switch (manager->display_ops->type) { case EXYNOS_DISPLAY_TYPE_HDMI: type = DRM_MODE_CONNECTOR_HDMIA; connector->interlace_allowed = true; diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h index c03683f2ae72d..1575e5fef51e9 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.h +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h @@ -130,7 +130,7 @@ struct exynos_drm_overlay { * @check_timing: check if timing is valid or not. * @power_on: display device on or off. */ -struct exynos_drm_display { +struct exynos_drm_display_ops { enum exynos_drm_output_type type; bool (*is_connected)(struct device *dev); int (*get_edid)(struct device *dev, struct drm_connector *connector, @@ -178,7 +178,7 @@ struct exynos_drm_manager { int pipe; struct exynos_drm_manager_ops *ops; struct exynos_drm_overlay_ops *overlay_ops; - struct exynos_drm_display *display; + struct exynos_drm_display_ops *display_ops; }; /* diff --git a/drivers/gpu/drm/exynos/exynos_drm_encoder.c b/drivers/gpu/drm/exynos/exynos_drm_encoder.c index 4e5535b2b8e38..0034fa7a58dbe 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_encoder.c +++ b/drivers/gpu/drm/exynos/exynos_drm_encoder.c @@ -58,11 +58,11 @@ static void exynos_drm_encoder_dpms(struct drm_encoder *encoder, int mode) list_for_each_entry(connector, &dev->mode_config.connector_list, head) { if (connector->encoder == encoder) { - struct exynos_drm_display *display = - manager->display; + struct exynos_drm_display_ops *display_ops = + manager->display_ops; - if (display && display->power_on) - display->power_on(manager->dev, mode); + if (display_ops && display_ops->power_on) + display_ops->power_on(manager->dev, mode); } } } diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index 4659c88cdd9bb..f2d883f2999e3 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -124,7 +124,7 @@ static int fimd_display_power_on(struct device *dev, int mode) return 0; } -static struct exynos_drm_display fimd_display = { +static struct exynos_drm_display_ops fimd_display_ops = { .type = EXYNOS_DISPLAY_TYPE_LCD, .is_connected = fimd_display_is_connected, .get_timing = fimd_get_timing, @@ -731,7 +731,7 @@ static int __devinit fimd_probe(struct platform_device *pdev) subdrv->manager.pipe = -1; subdrv->manager.ops = &fimd_manager_ops; subdrv->manager.overlay_ops = &fimd_overlay_ops; - subdrv->manager.display = &fimd_display; + subdrv->manager.display_ops = &fimd_display_ops; subdrv->manager.dev = dev; platform_set_drvdata(pdev, ctx); -- GitLab From 84b46990cb2caf8efe20d5626e1d7e2e40bab832 Mon Sep 17 00:00:00 2001 From: Joonyoung Shim Date: Fri, 4 Nov 2011 13:41:46 +0900 Subject: [PATCH 0344/3089] drm/exynos: restored kernel_fb_list when reiniting fb_helper during recreating exynos_drm_fbdev as a new display device probes, fb_helper is reinitialized but kernel fb is not changed so kernel_fb_list should be restored after fb_helper is reinitialized. Signed-off-by: Joonyoung Shim Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_fbdev.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c index 1f4b3d1a77134..74092a81c1802 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c @@ -405,6 +405,18 @@ int exynos_drm_fbdev_reinit(struct drm_device *dev) fb_helper = private->fb_helper; if (fb_helper) { + struct list_head temp_list; + + INIT_LIST_HEAD(&temp_list); + + /* + * fb_helper is reintialized but kernel fb is reused + * so kernel_fb_list need to be backuped and restored + */ + if (!list_empty(&fb_helper->kernel_fb_list)) + list_replace_init(&fb_helper->kernel_fb_list, + &temp_list); + drm_fb_helper_fini(fb_helper); ret = drm_fb_helper_init(dev, fb_helper, @@ -414,6 +426,9 @@ int exynos_drm_fbdev_reinit(struct drm_device *dev) return ret; } + if (!list_empty(&temp_list)) + list_replace(&temp_list, &fb_helper->kernel_fb_list); + ret = drm_fb_helper_single_add_all_connectors(fb_helper); if (ret < 0) { DRM_ERROR("failed to add fb helper to connectors\n"); -- GitLab From aa6b2b6cd43e4a23c2a220382a8b385b087d8bca Mon Sep 17 00:00:00 2001 From: Seung-Woo Kim Date: Fri, 4 Nov 2011 13:44:38 +0900 Subject: [PATCH 0345/3089] drm/exynos: removed meaningless parameter from fbdev update drm_framebuffer already has width and height so they are meaningless as parameters when updating fb_info. Signed-off-by: Seung-Woo Kim Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_fbdev.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c index 74092a81c1802..0effd77d569b8 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c @@ -85,15 +85,13 @@ static struct fb_ops exynos_drm_fb_ops = { }; static int exynos_drm_fbdev_update(struct drm_fb_helper *helper, - struct drm_framebuffer *fb, - unsigned int fb_width, - unsigned int fb_height) + struct drm_framebuffer *fb) { struct fb_info *fbi = helper->fbdev; struct drm_device *dev = helper->dev; struct exynos_drm_fbdev *exynos_fb = to_exynos_fbdev(helper); struct exynos_drm_buf_entry *entry; - unsigned int size = fb_width * fb_height * (fb->bits_per_pixel >> 3); + unsigned int size = fb->width * fb->height * (fb->bits_per_pixel >> 3); unsigned long offset; DRM_DEBUG_KMS("%s\n", __FILE__); @@ -101,7 +99,7 @@ static int exynos_drm_fbdev_update(struct drm_fb_helper *helper, exynos_fb->fb = fb; drm_fb_helper_fill_fix(fbi, fb->pitch, fb->depth); - drm_fb_helper_fill_var(fbi, helper, fb_width, fb_height); + drm_fb_helper_fill_var(fbi, helper, fb->width, fb->height); entry = exynos_drm_fb_get_buf(fb); if (!entry) { @@ -171,8 +169,7 @@ static int exynos_drm_fbdev_create(struct drm_fb_helper *helper, goto out; } - ret = exynos_drm_fbdev_update(helper, helper->fb, sizes->fb_width, - sizes->fb_height); + ret = exynos_drm_fbdev_update(helper, helper->fb); if (ret < 0) fb_dealloc_cmap(&fbi->cmap); @@ -235,8 +232,7 @@ static int exynos_drm_fbdev_recreate(struct drm_fb_helper *helper, } helper->fb = exynos_fbdev->fb; - return exynos_drm_fbdev_update(helper, helper->fb, sizes->fb_width, - sizes->fb_height); + return exynos_drm_fbdev_update(helper, helper->fb); } static int exynos_drm_fbdev_probe(struct drm_fb_helper *helper, -- GitLab From d2716c896d305fb5d3d0d7f58394c17841ed2967 Mon Sep 17 00:00:00 2001 From: Joonyoung Shim Date: Fri, 4 Nov 2011 17:04:45 +0900 Subject: [PATCH 0346/3089] drm/exynos: added crtc dpms for disable crtc crtc dpms is called as destroying attached fb so dpms off sould be processed. crtc dpms also can be called after crtc is detached from encoder so pipe value of manager is used to find display controller for this case Signed-off-by: Joonyoung Shim Signed-off-by: Seung-Woo Kim Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_crtc.c | 29 +++++++++-- drivers/gpu/drm/exynos/exynos_drm_encoder.c | 54 ++++++++++++++++++--- drivers/gpu/drm/exynos/exynos_drm_encoder.h | 1 + 3 files changed, 73 insertions(+), 11 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c index 9337e5e2dbb62..b1d6526340bb1 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c @@ -85,7 +85,8 @@ static void exynos_drm_crtc_apply(struct drm_crtc *crtc) exynos_drm_fn_encoder(crtc, overlay, exynos_drm_encoder_crtc_mode_set); - exynos_drm_fn_encoder(crtc, NULL, exynos_drm_encoder_crtc_commit); + exynos_drm_fn_encoder(crtc, &exynos_crtc->pipe, + exynos_drm_encoder_crtc_commit); } static int exynos_drm_overlay_update(struct exynos_drm_overlay *overlay, @@ -171,9 +172,26 @@ static int exynos_drm_crtc_update(struct drm_crtc *crtc) static void exynos_drm_crtc_dpms(struct drm_crtc *crtc, int mode) { - DRM_DEBUG_KMS("%s\n", __FILE__); + struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(crtc); - /* TODO */ + DRM_DEBUG_KMS("crtc[%d] mode[%d]\n", crtc->base.id, mode); + + switch (mode) { + case DRM_MODE_DPMS_ON: + exynos_drm_fn_encoder(crtc, &exynos_crtc->pipe, + exynos_drm_encoder_crtc_commit); + break; + case DRM_MODE_DPMS_STANDBY: + case DRM_MODE_DPMS_SUSPEND: + case DRM_MODE_DPMS_OFF: + /* TODO */ + exynos_drm_fn_encoder(crtc, NULL, + exynos_drm_encoder_crtc_disable); + break; + default: + DRM_DEBUG_KMS("unspecified mode %d\n", mode); + break; + } } static void exynos_drm_crtc_prepare(struct drm_crtc *crtc) @@ -185,9 +203,12 @@ static void exynos_drm_crtc_prepare(struct drm_crtc *crtc) static void exynos_drm_crtc_commit(struct drm_crtc *crtc) { + struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(crtc); + DRM_DEBUG_KMS("%s\n", __FILE__); - /* drm framework doesn't check NULL. */ + exynos_drm_fn_encoder(crtc, &exynos_crtc->pipe, + exynos_drm_encoder_crtc_commit); } static bool diff --git a/drivers/gpu/drm/exynos/exynos_drm_encoder.c b/drivers/gpu/drm/exynos/exynos_drm_encoder.c index 0034fa7a58dbe..f15da9581a1e3 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_encoder.c +++ b/drivers/gpu/drm/exynos/exynos_drm_encoder.c @@ -61,6 +61,8 @@ static void exynos_drm_encoder_dpms(struct drm_encoder *encoder, int mode) struct exynos_drm_display_ops *display_ops = manager->display_ops; + DRM_DEBUG_KMS("connector[%d] dpms[%d]\n", + connector->base.id, mode); if (display_ops && display_ops->power_on) display_ops->power_on(manager->dev, mode); } @@ -117,15 +119,11 @@ static void exynos_drm_encoder_commit(struct drm_encoder *encoder) { struct exynos_drm_manager *manager = exynos_drm_get_manager(encoder); struct exynos_drm_manager_ops *manager_ops = manager->ops; - struct exynos_drm_overlay_ops *overlay_ops = manager->overlay_ops; DRM_DEBUG_KMS("%s\n", __FILE__); if (manager_ops && manager_ops->commit) manager_ops->commit(manager->dev); - - if (overlay_ops && overlay_ops->commit) - overlay_ops->commit(manager->dev); } static struct drm_crtc * @@ -209,10 +207,23 @@ void exynos_drm_fn_encoder(struct drm_crtc *crtc, void *data, { struct drm_device *dev = crtc->dev; struct drm_encoder *encoder; + struct exynos_drm_private *private = dev->dev_private; + struct exynos_drm_manager *manager; list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - if (encoder->crtc != crtc) - continue; + /* + * if crtc is detached from encoder, check pipe, + * otherwise check crtc attached to encoder + */ + if (!encoder->crtc) { + manager = to_exynos_encoder(encoder)->manager; + if (manager->pipe < 0 || + private->crtc[manager->pipe] != crtc) + continue; + } else { + if (encoder->crtc != crtc) + continue; + } fn(encoder, data); } @@ -251,8 +262,18 @@ void exynos_drm_encoder_crtc_commit(struct drm_encoder *encoder, void *data) struct exynos_drm_manager *manager = to_exynos_encoder(encoder)->manager; struct exynos_drm_overlay_ops *overlay_ops = manager->overlay_ops; + int crtc = *(int *)data; + + DRM_DEBUG_KMS("%s\n", __FILE__); - overlay_ops->commit(manager->dev); + /* + * when crtc is detached from encoder, this pipe is used + * to select manager operation + */ + manager->pipe = crtc; + + if (overlay_ops && overlay_ops->commit) + overlay_ops->commit(manager->dev); } void exynos_drm_encoder_crtc_mode_set(struct drm_encoder *encoder, void *data) @@ -265,6 +286,25 @@ void exynos_drm_encoder_crtc_mode_set(struct drm_encoder *encoder, void *data) overlay_ops->mode_set(manager->dev, overlay); } +void exynos_drm_encoder_crtc_disable(struct drm_encoder *encoder, void *data) +{ + struct exynos_drm_manager *manager = + to_exynos_encoder(encoder)->manager; + struct exynos_drm_overlay_ops *overlay_ops = manager->overlay_ops; + + DRM_DEBUG_KMS("\n"); + + overlay_ops->disable(manager->dev); + + /* + * crtc is already detached from encoder and last + * function for detaching is properly done, so + * clear pipe from manager to prevent repeated call + */ + if (!encoder->crtc) + manager->pipe = -1; +} + MODULE_AUTHOR("Inki Dae "); MODULE_AUTHOR("Joonyoung Shim "); MODULE_AUTHOR("Seung-Woo Kim "); diff --git a/drivers/gpu/drm/exynos/exynos_drm_encoder.h b/drivers/gpu/drm/exynos/exynos_drm_encoder.h index 5ecd645d06a97..a22acfbf0e4ed 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_encoder.h +++ b/drivers/gpu/drm/exynos/exynos_drm_encoder.h @@ -41,5 +41,6 @@ void exynos_drm_enable_vblank(struct drm_encoder *encoder, void *data); void exynos_drm_disable_vblank(struct drm_encoder *encoder, void *data); void exynos_drm_encoder_crtc_commit(struct drm_encoder *encoder, void *data); void exynos_drm_encoder_crtc_mode_set(struct drm_encoder *encoder, void *data); +void exynos_drm_encoder_crtc_disable(struct drm_encoder *encoder, void *data); #endif -- GitLab From b0e0f85631f9d905095d2896a952430f5eb0aba1 Mon Sep 17 00:00:00 2001 From: Seung-Woo Kim Date: Fri, 4 Nov 2011 17:31:41 +0900 Subject: [PATCH 0347/3089] drm/exynos: checked for null pointer Signed-off-by: Seung-Woo Kim Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_encoder.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_encoder.c b/drivers/gpu/drm/exynos/exynos_drm_encoder.c index f15da9581a1e3..866f419f485b9 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_encoder.c +++ b/drivers/gpu/drm/exynos/exynos_drm_encoder.c @@ -283,7 +283,8 @@ void exynos_drm_encoder_crtc_mode_set(struct drm_encoder *encoder, void *data) struct exynos_drm_overlay_ops *overlay_ops = manager->overlay_ops; struct exynos_drm_overlay *overlay = data; - overlay_ops->mode_set(manager->dev, overlay); + if (overlay_ops && overlay_ops->mode_set) + overlay_ops->mode_set(manager->dev, overlay); } void exynos_drm_encoder_crtc_disable(struct drm_encoder *encoder, void *data) @@ -294,7 +295,8 @@ void exynos_drm_encoder_crtc_disable(struct drm_encoder *encoder, void *data) DRM_DEBUG_KMS("\n"); - overlay_ops->disable(manager->dev); + if (overlay_ops && overlay_ops->disable) + overlay_ops->disable(manager->dev); /* * crtc is already detached from encoder and last -- GitLab From f088d5a9c5dd22b6559fa3f3939973bc374c977b Mon Sep 17 00:00:00 2001 From: Inki Dae Date: Sat, 12 Nov 2011 14:51:23 +0900 Subject: [PATCH 0348/3089] drm/exynos: use gem create function generically this patch addes exynos_drm_gem_init() creating and initialzing a gem. allocation functions could use this function to create new gem and it changes size type of exynos_drm_gem_create structure to 64bit and also corrects comments to exynos_drm_gem_create structure. Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_fb.c | 6 +- drivers/gpu/drm/exynos/exynos_drm_gem.c | 79 +++++++++++++++---------- drivers/gpu/drm/exynos/exynos_drm_gem.h | 6 +- include/drm/exynos_drm.h | 9 ++- 4 files changed, 59 insertions(+), 41 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c index 7d91a542c7567..8d0f662240457 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fb.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c @@ -165,9 +165,9 @@ exynos_drm_fb_init(struct drm_file *file_priv, struct drm_device *dev, goto out; } else { - exynos_gem_obj = exynos_drm_gem_create(file_priv, dev, - size, - &mode_cmd->handle); + exynos_gem_obj = exynos_drm_gem_create(dev, file_priv, + &mode_cmd->handle, + size); if (IS_ERR(exynos_gem_obj)) { ret = PTR_ERR(exynos_gem_obj); goto err_buffer; diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index a8e7a88906ed2..bd6ede83b6846 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -62,40 +62,28 @@ static unsigned int get_gem_mmap_offset(struct drm_gem_object *obj) return (unsigned int)obj->map_list.hash.key << PAGE_SHIFT; } -struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_file *file_priv, - struct drm_device *dev, unsigned int size, - unsigned int *handle) +static struct exynos_drm_gem_obj + *exynos_drm_gem_init(struct drm_device *drm_dev, + struct drm_file *file_priv, unsigned int *handle, + unsigned int size) { struct exynos_drm_gem_obj *exynos_gem_obj; - struct exynos_drm_buf_entry *entry; struct drm_gem_object *obj; int ret; - DRM_DEBUG_KMS("%s\n", __FILE__); - - size = roundup(size, PAGE_SIZE); - exynos_gem_obj = kzalloc(sizeof(*exynos_gem_obj), GFP_KERNEL); if (!exynos_gem_obj) { DRM_ERROR("failed to allocate exynos gem object.\n"); return ERR_PTR(-ENOMEM); } - /* allocate the new buffer object and memory region. */ - entry = exynos_drm_buf_create(dev, size); - if (!entry) { - kfree(exynos_gem_obj); - return ERR_PTR(-ENOMEM); - } - - exynos_gem_obj->entry = entry; - obj = &exynos_gem_obj->base; - ret = drm_gem_object_init(dev, obj, size); + ret = drm_gem_object_init(drm_dev, obj, size); if (ret < 0) { - DRM_ERROR("failed to initailize gem object.\n"); - goto err_obj_init; + DRM_ERROR("failed to initialize gem object.\n"); + ret = -EINVAL; + goto err_object_init; } DRM_DEBUG_KMS("created file object = 0x%x\n", (unsigned int)obj->filp); @@ -127,24 +115,55 @@ err_handle_create: err_create_mmap_offset: drm_gem_object_release(obj); -err_obj_init: - exynos_drm_buf_destroy(dev, exynos_gem_obj->entry); - +err_object_init: kfree(exynos_gem_obj); return ERR_PTR(ret); } +struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev, + struct drm_file *file_priv, + unsigned int *handle, unsigned long size) +{ + + struct exynos_drm_gem_obj *exynos_gem_obj = NULL; + struct exynos_drm_buf_entry *entry; + int ret; + + size = roundup(size, PAGE_SIZE); + + DRM_DEBUG_KMS("%s: size = 0x%lx\n", __FILE__, size); + + entry = exynos_drm_buf_create(dev, size); + if (!entry) + return ERR_PTR(-ENOMEM); + + exynos_gem_obj = exynos_drm_gem_init(dev, file_priv, handle, size); + if (IS_ERR(exynos_gem_obj)) { + ret = PTR_ERR(exynos_gem_obj); + goto err_gem_init; + } + + exynos_gem_obj->entry = entry; + + return exynos_gem_obj; + +err_gem_init: + exynos_drm_buf_destroy(dev, exynos_gem_obj->entry); + + return ERR_PTR(ret); +} + int exynos_drm_gem_create_ioctl(struct drm_device *dev, void *data, - struct drm_file *file_priv) + struct drm_file *file_priv) { struct drm_exynos_gem_create *args = data; - struct exynos_drm_gem_obj *exynos_gem_obj; + struct exynos_drm_gem_obj *exynos_gem_obj = NULL; - DRM_DEBUG_KMS("%s : size = 0x%x\n", __FILE__, args->size); + DRM_DEBUG_KMS("%s\n", __FILE__); - exynos_gem_obj = exynos_drm_gem_create(file_priv, dev, args->size, - &args->handle); + exynos_gem_obj = exynos_drm_gem_create(dev, file_priv, + &args->handle, args->size); if (IS_ERR(exynos_gem_obj)) return PTR_ERR(exynos_gem_obj); @@ -302,8 +321,8 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv, args->pitch = args->width * args->bpp >> 3; args->size = args->pitch * args->height; - exynos_gem_obj = exynos_drm_gem_create(file_priv, dev, args->size, - &args->handle); + exynos_gem_obj = exynos_drm_gem_create(dev, file_priv, &args->handle, + args->size); if (IS_ERR(exynos_gem_obj)) return PTR_ERR(exynos_gem_obj); diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h index e5fc0148277bf..213838d9606e8 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.h +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h @@ -49,9 +49,9 @@ struct exynos_drm_gem_obj { }; /* create a new buffer and get a new gem handle. */ -struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_file *file_priv, - struct drm_device *dev, unsigned int size, - unsigned int *handle); +struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev, + struct drm_file *file_priv, + unsigned int *handle, unsigned long size); /* * request gem object creation and buffer allocation as the size diff --git a/include/drm/exynos_drm.h b/include/drm/exynos_drm.h index 1d161cb3aca5c..12050434d57a2 100644 --- a/include/drm/exynos_drm.h +++ b/include/drm/exynos_drm.h @@ -32,17 +32,16 @@ /** * User-desired buffer creation information structure. * - * @size: requested size for the object. + * @size: user-desired memory allocation size. * - this size value would be page-aligned internally. * @flags: user request for setting memory type or cache attributes. - * @handle: returned handle for the object. - * @pad: just padding to be 64-bit aligned. + * @handle: returned a handle to created gem object. + * - this handle will be set by gem module of kernel side. */ struct drm_exynos_gem_create { - unsigned int size; + uint64_t size; unsigned int flags; unsigned int handle; - unsigned int pad; }; /** -- GitLab From c7493668eeced636afabfed57dfead8329c3d7fa Mon Sep 17 00:00:00 2001 From: Inki Dae Date: Wed, 9 Nov 2011 16:50:30 +0900 Subject: [PATCH 0349/3089] drm/exynos: removed unnecessary variable. Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_buf.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_buf.c b/drivers/gpu/drm/exynos/exynos_drm_buf.c index 6f8afea94fc97..303189c78cf1f 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_buf.c +++ b/drivers/gpu/drm/exynos/exynos_drm_buf.c @@ -29,8 +29,6 @@ #include "exynos_drm_drv.h" #include "exynos_drm_buf.h" -static DEFINE_MUTEX(exynos_drm_buf_lock); - static int lowlevel_buffer_allocate(struct drm_device *dev, struct exynos_drm_buf_entry *entry) { -- GitLab From 2c871127e994a678b82104a4110eb7fcc87f05ad Mon Sep 17 00:00:00 2001 From: Inki Dae Date: Sat, 12 Nov 2011 15:23:32 +0900 Subject: [PATCH 0350/3089] drm/exynos: changed buffer structure. the purpose of this patch is to consider IOMMU support in the future. EXYNOS4 SoC supports IOMMU also so the address for DMA could be physical address with IOMMU or device address with IOMMU. Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_buf.c | 60 ++++++++++++----------- drivers/gpu/drm/exynos/exynos_drm_buf.h | 21 ++------ drivers/gpu/drm/exynos/exynos_drm_crtc.c | 47 ++++++------------ drivers/gpu/drm/exynos/exynos_drm_crtc.h | 25 ++++++++++ drivers/gpu/drm/exynos/exynos_drm_drv.h | 6 +-- drivers/gpu/drm/exynos/exynos_drm_fb.c | 48 +++++++++--------- drivers/gpu/drm/exynos/exynos_drm_fbdev.c | 15 +++--- drivers/gpu/drm/exynos/exynos_drm_fimd.c | 12 ++--- drivers/gpu/drm/exynos/exynos_drm_gem.c | 25 +++++----- drivers/gpu/drm/exynos/exynos_drm_gem.h | 22 +++++++-- 10 files changed, 148 insertions(+), 133 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_buf.c b/drivers/gpu/drm/exynos/exynos_drm_buf.c index 303189c78cf1f..2bb07bca511a1 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_buf.c +++ b/drivers/gpu/drm/exynos/exynos_drm_buf.c @@ -27,80 +27,84 @@ #include "drm.h" #include "exynos_drm_drv.h" +#include "exynos_drm_gem.h" #include "exynos_drm_buf.h" static int lowlevel_buffer_allocate(struct drm_device *dev, - struct exynos_drm_buf_entry *entry) + struct exynos_drm_gem_buf *buffer) { DRM_DEBUG_KMS("%s\n", __FILE__); - entry->vaddr = dma_alloc_writecombine(dev->dev, entry->size, - (dma_addr_t *)&entry->paddr, GFP_KERNEL); - if (!entry->paddr) { + buffer->kvaddr = dma_alloc_writecombine(dev->dev, buffer->size, + &buffer->dma_addr, GFP_KERNEL); + if (!buffer->kvaddr) { DRM_ERROR("failed to allocate buffer.\n"); return -ENOMEM; } - DRM_DEBUG_KMS("allocated : vaddr(0x%x), paddr(0x%x), size(0x%x)\n", - (unsigned int)entry->vaddr, entry->paddr, entry->size); + DRM_DEBUG_KMS("vaddr(0x%lx), dma_addr(0x%lx), size(0x%lx)\n", + (unsigned long)buffer->kvaddr, + (unsigned long)buffer->dma_addr, + buffer->size); return 0; } static void lowlevel_buffer_deallocate(struct drm_device *dev, - struct exynos_drm_buf_entry *entry) + struct exynos_drm_gem_buf *buffer) { DRM_DEBUG_KMS("%s.\n", __FILE__); - if (entry->paddr && entry->vaddr && entry->size) - dma_free_writecombine(dev->dev, entry->size, entry->vaddr, - entry->paddr); + if (buffer->dma_addr && buffer->size) + dma_free_writecombine(dev->dev, buffer->size, buffer->kvaddr, + (dma_addr_t)buffer->dma_addr); else - DRM_DEBUG_KMS("entry data is null.\n"); + DRM_DEBUG_KMS("buffer data are invalid.\n"); } -struct exynos_drm_buf_entry *exynos_drm_buf_create(struct drm_device *dev, +struct exynos_drm_gem_buf *exynos_drm_buf_create(struct drm_device *dev, unsigned int size) { - struct exynos_drm_buf_entry *entry; + struct exynos_drm_gem_buf *buffer; DRM_DEBUG_KMS("%s.\n", __FILE__); + DRM_DEBUG_KMS("desired size = 0x%x\n", size); - entry = kzalloc(sizeof(*entry), GFP_KERNEL); - if (!entry) { - DRM_ERROR("failed to allocate exynos_drm_buf_entry.\n"); + buffer = kzalloc(sizeof(*buffer), GFP_KERNEL); + if (!buffer) { + DRM_ERROR("failed to allocate exynos_drm_gem_buf.\n"); return ERR_PTR(-ENOMEM); } - entry->size = size; + buffer->size = size; /* * allocate memory region with size and set the memory information - * to vaddr and paddr of a entry object. + * to vaddr and dma_addr of a buffer object. */ - if (lowlevel_buffer_allocate(dev, entry) < 0) { - kfree(entry); - entry = NULL; + if (lowlevel_buffer_allocate(dev, buffer) < 0) { + kfree(buffer); + buffer = NULL; return ERR_PTR(-ENOMEM); } - return entry; + return buffer; } void exynos_drm_buf_destroy(struct drm_device *dev, - struct exynos_drm_buf_entry *entry) + struct exynos_drm_gem_buf *buffer) { DRM_DEBUG_KMS("%s.\n", __FILE__); - if (!entry) { - DRM_DEBUG_KMS("entry is null.\n"); + if (!buffer) { + DRM_DEBUG_KMS("buffer is null.\n"); return; } - lowlevel_buffer_deallocate(dev, entry); + lowlevel_buffer_deallocate(dev, buffer); - kfree(entry); - entry = NULL; + kfree(buffer); + buffer = NULL; } MODULE_AUTHOR("Inki Dae "); diff --git a/drivers/gpu/drm/exynos/exynos_drm_buf.h b/drivers/gpu/drm/exynos/exynos_drm_buf.h index 045d59eab01a2..6e91f9caa5dbd 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_buf.h +++ b/drivers/gpu/drm/exynos/exynos_drm_buf.h @@ -26,28 +26,15 @@ #ifndef _EXYNOS_DRM_BUF_H_ #define _EXYNOS_DRM_BUF_H_ -/* - * exynos drm buffer entry structure. - * - * @paddr: physical address of allocated memory. - * @vaddr: kernel virtual address of allocated memory. - * @size: size of allocated memory. - */ -struct exynos_drm_buf_entry { - dma_addr_t paddr; - void __iomem *vaddr; - unsigned int size; -}; - /* allocate physical memory. */ -struct exynos_drm_buf_entry *exynos_drm_buf_create(struct drm_device *dev, +struct exynos_drm_gem_buf *exynos_drm_buf_create(struct drm_device *dev, unsigned int size); -/* get physical memory information of a drm framebuffer. */ -struct exynos_drm_buf_entry *exynos_drm_fb_get_buf(struct drm_framebuffer *fb); +/* get memory information of a drm framebuffer. */ +struct exynos_drm_gem_buf *exynos_drm_fb_get_buf(struct drm_framebuffer *fb); /* remove allocated physical memory. */ void exynos_drm_buf_destroy(struct drm_device *dev, - struct exynos_drm_buf_entry *entry); + struct exynos_drm_gem_buf *buffer); #endif diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c index b1d6526340bb1..ee43cc2208530 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c @@ -29,35 +29,16 @@ #include "drmP.h" #include "drm_crtc_helper.h" +#include "exynos_drm_crtc.h" #include "exynos_drm_drv.h" #include "exynos_drm_fb.h" #include "exynos_drm_encoder.h" +#include "exynos_drm_gem.h" #include "exynos_drm_buf.h" #define to_exynos_crtc(x) container_of(x, struct exynos_drm_crtc,\ drm_crtc) -/* - * Exynos specific crtc postion structure. - * - * @fb_x: offset x on a framebuffer to be displyed - * - the unit is screen coordinates. - * @fb_y: offset y on a framebuffer to be displayed - * - the unit is screen coordinates. - * @crtc_x: offset x on hardware screen. - * @crtc_y: offset y on hardware screen. - * @crtc_w: width of hardware screen. - * @crtc_h: height of hardware screen. - */ -struct exynos_drm_crtc_pos { - unsigned int fb_x; - unsigned int fb_y; - unsigned int crtc_x; - unsigned int crtc_y; - unsigned int crtc_w; - unsigned int crtc_h; -}; - /* * Exynos specific crtc structure. * @@ -89,27 +70,27 @@ static void exynos_drm_crtc_apply(struct drm_crtc *crtc) exynos_drm_encoder_crtc_commit); } -static int exynos_drm_overlay_update(struct exynos_drm_overlay *overlay, - struct drm_framebuffer *fb, - struct drm_display_mode *mode, - struct exynos_drm_crtc_pos *pos) +int exynos_drm_overlay_update(struct exynos_drm_overlay *overlay, + struct drm_framebuffer *fb, + struct drm_display_mode *mode, + struct exynos_drm_crtc_pos *pos) { - struct exynos_drm_buf_entry *entry; + struct exynos_drm_gem_buf *buffer; unsigned int actual_w; unsigned int actual_h; - entry = exynos_drm_fb_get_buf(fb); - if (!entry) { - DRM_LOG_KMS("entry is null.\n"); + buffer = exynos_drm_fb_get_buf(fb); + if (!buffer) { + DRM_LOG_KMS("buffer is null.\n"); return -EFAULT; } - overlay->paddr = entry->paddr; - overlay->vaddr = entry->vaddr; + overlay->dma_addr = buffer->dma_addr; + overlay->vaddr = buffer->kvaddr; - DRM_DEBUG_KMS("vaddr = 0x%lx, paddr = 0x%lx\n", + DRM_DEBUG_KMS("vaddr = 0x%lx, dma_addr = 0x%lx\n", (unsigned long)overlay->vaddr, - (unsigned long)overlay->paddr); + (unsigned long)overlay->dma_addr); actual_w = min((mode->hdisplay - pos->crtc_x), pos->crtc_w); actual_h = min((mode->vdisplay - pos->crtc_y), pos->crtc_h); diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.h b/drivers/gpu/drm/exynos/exynos_drm_crtc.h index c584042d6d2cb..25f72a62cb880 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_crtc.h +++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.h @@ -35,4 +35,29 @@ int exynos_drm_crtc_create(struct drm_device *dev, unsigned int nr); int exynos_drm_crtc_enable_vblank(struct drm_device *dev, int crtc); void exynos_drm_crtc_disable_vblank(struct drm_device *dev, int crtc); +/* + * Exynos specific crtc postion structure. + * + * @fb_x: offset x on a framebuffer to be displyed + * - the unit is screen coordinates. + * @fb_y: offset y on a framebuffer to be displayed + * - the unit is screen coordinates. + * @crtc_x: offset x on hardware screen. + * @crtc_y: offset y on hardware screen. + * @crtc_w: width of hardware screen. + * @crtc_h: height of hardware screen. + */ +struct exynos_drm_crtc_pos { + unsigned int fb_x; + unsigned int fb_y; + unsigned int crtc_x; + unsigned int crtc_y; + unsigned int crtc_w; + unsigned int crtc_h; +}; + +int exynos_drm_overlay_update(struct exynos_drm_overlay *overlay, + struct drm_framebuffer *fb, + struct drm_display_mode *mode, + struct exynos_drm_crtc_pos *pos); #endif diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h index 1575e5fef51e9..38a6f1df75c61 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.h +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h @@ -79,8 +79,8 @@ struct exynos_drm_overlay_ops { * @scan_flag: interlace or progressive way. * (it could be DRM_MODE_FLAG_*) * @bpp: pixel size.(in bit) - * @paddr: bus(accessed by dma) physical memory address to this overlay - * and this is physically continuous. + * @dma_addr: bus(accessed by dma) address to the memory region allocated + * for a overlay. * @vaddr: virtual memory addresss to this overlay. * @default_win: a window to be enabled. * @color_key: color key on or off. @@ -108,7 +108,7 @@ struct exynos_drm_overlay { unsigned int scan_flag; unsigned int bpp; unsigned int pitch; - dma_addr_t paddr; + dma_addr_t dma_addr; void __iomem *vaddr; bool default_win; diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c index 8d0f662240457..5bf4a1ac7f828 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fb.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c @@ -43,14 +43,14 @@ * * @fb: drm framebuffer obejct. * @exynos_gem_obj: exynos specific gem object containing a gem object. - * @entry: pointer to exynos drm buffer entry object. - * - containing only the information to physically continuous memory - * region allocated at default framebuffer creation. + * @buffer: pointer to exynos_drm_gem_buffer object. + * - contain the memory information to memory region allocated + * at default framebuffer creation. */ struct exynos_drm_fb { struct drm_framebuffer fb; struct exynos_drm_gem_obj *exynos_gem_obj; - struct exynos_drm_buf_entry *entry; + struct exynos_drm_gem_buf *buffer; }; static void exynos_drm_fb_destroy(struct drm_framebuffer *fb) @@ -65,8 +65,8 @@ static void exynos_drm_fb_destroy(struct drm_framebuffer *fb) * default framebuffer has no gem object so * a buffer of the default framebuffer should be released at here. */ - if (!exynos_fb->exynos_gem_obj && exynos_fb->entry) - exynos_drm_buf_destroy(fb->dev, exynos_fb->entry); + if (!exynos_fb->exynos_gem_obj && exynos_fb->buffer) + exynos_drm_buf_destroy(fb->dev, exynos_fb->buffer); kfree(exynos_fb); exynos_fb = NULL; @@ -145,23 +145,23 @@ exynos_drm_fb_init(struct drm_file *file_priv, struct drm_device *dev, */ if (!mode_cmd->handle) { if (!file_priv) { - struct exynos_drm_buf_entry *entry; + struct exynos_drm_gem_buf *buffer; /* * in case that file_priv is NULL, it allocates * only buffer and this buffer would be used * for default framebuffer. */ - entry = exynos_drm_buf_create(dev, size); - if (IS_ERR(entry)) { - ret = PTR_ERR(entry); + buffer = exynos_drm_buf_create(dev, size); + if (IS_ERR(buffer)) { + ret = PTR_ERR(buffer); goto err_buffer; } - exynos_fb->entry = entry; + exynos_fb->buffer = buffer; - DRM_LOG_KMS("default fb: paddr = 0x%lx, size = 0x%x\n", - (unsigned long)entry->paddr, size); + DRM_LOG_KMS("default: dma_addr = 0x%lx, size = 0x%x\n", + (unsigned long)buffer->dma_addr, size); goto out; } else { @@ -191,10 +191,10 @@ exynos_drm_fb_init(struct drm_file *file_priv, struct drm_device *dev, * so that default framebuffer has no its own gem object, * only its own buffer object. */ - exynos_fb->entry = exynos_gem_obj->entry; + exynos_fb->buffer = exynos_gem_obj->buffer; - DRM_LOG_KMS("paddr = 0x%lx, size = 0x%x, gem object = 0x%x\n", - (unsigned long)exynos_fb->entry->paddr, size, + DRM_LOG_KMS("dma_addr = 0x%lx, size = 0x%x, gem object = 0x%x\n", + (unsigned long)exynos_fb->buffer->dma_addr, size, (unsigned int)&exynos_gem_obj->base); out: @@ -222,22 +222,22 @@ struct drm_framebuffer *exynos_drm_fb_create(struct drm_device *dev, return exynos_drm_fb_init(file_priv, dev, mode_cmd); } -struct exynos_drm_buf_entry *exynos_drm_fb_get_buf(struct drm_framebuffer *fb) +struct exynos_drm_gem_buf *exynos_drm_fb_get_buf(struct drm_framebuffer *fb) { struct exynos_drm_fb *exynos_fb = to_exynos_fb(fb); - struct exynos_drm_buf_entry *entry; + struct exynos_drm_gem_buf *buffer; DRM_DEBUG_KMS("%s\n", __FILE__); - entry = exynos_fb->entry; - if (!entry) + buffer = exynos_fb->buffer; + if (!buffer) return NULL; - DRM_DEBUG_KMS("vaddr = 0x%lx, paddr = 0x%lx\n", - (unsigned long)entry->vaddr, - (unsigned long)entry->paddr); + DRM_DEBUG_KMS("vaddr = 0x%lx, dma_addr = 0x%lx\n", + (unsigned long)buffer->kvaddr, + (unsigned long)buffer->dma_addr); - return entry; + return buffer; } static void exynos_drm_output_poll_changed(struct drm_device *dev) diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c index 0effd77d569b8..836f410081871 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c @@ -33,6 +33,7 @@ #include "exynos_drm_drv.h" #include "exynos_drm_fb.h" +#include "exynos_drm_gem.h" #include "exynos_drm_buf.h" #define MAX_CONNECTOR 4 @@ -90,7 +91,7 @@ static int exynos_drm_fbdev_update(struct drm_fb_helper *helper, struct fb_info *fbi = helper->fbdev; struct drm_device *dev = helper->dev; struct exynos_drm_fbdev *exynos_fb = to_exynos_fbdev(helper); - struct exynos_drm_buf_entry *entry; + struct exynos_drm_gem_buf *buffer; unsigned int size = fb->width * fb->height * (fb->bits_per_pixel >> 3); unsigned long offset; @@ -101,18 +102,18 @@ static int exynos_drm_fbdev_update(struct drm_fb_helper *helper, drm_fb_helper_fill_fix(fbi, fb->pitch, fb->depth); drm_fb_helper_fill_var(fbi, helper, fb->width, fb->height); - entry = exynos_drm_fb_get_buf(fb); - if (!entry) { - DRM_LOG_KMS("entry is null.\n"); + buffer = exynos_drm_fb_get_buf(fb); + if (!buffer) { + DRM_LOG_KMS("buffer is null.\n"); return -EFAULT; } offset = fbi->var.xoffset * (fb->bits_per_pixel >> 3); offset += fbi->var.yoffset * fb->pitch; - dev->mode_config.fb_base = entry->paddr; - fbi->screen_base = entry->vaddr + offset; - fbi->fix.smem_start = entry->paddr + offset; + dev->mode_config.fb_base = (resource_size_t)buffer->dma_addr; + fbi->screen_base = buffer->kvaddr + offset; + fbi->fix.smem_start = (unsigned long)(buffer->dma_addr + offset); fbi->screen_size = size; fbi->fix.smem_len = size; diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index f2d883f2999e3..f5c8b072e497a 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -64,7 +64,7 @@ struct fimd_win_data { unsigned int fb_width; unsigned int fb_height; unsigned int bpp; - dma_addr_t paddr; + dma_addr_t dma_addr; void __iomem *vaddr; unsigned int buf_offsize; unsigned int line_size; /* bytes */ @@ -251,7 +251,7 @@ static void fimd_win_mode_set(struct device *dev, win_data->ovl_height = overlay->crtc_height; win_data->fb_width = overlay->fb_width; win_data->fb_height = overlay->fb_height; - win_data->paddr = overlay->paddr + offset; + win_data->dma_addr = overlay->dma_addr + offset; win_data->vaddr = overlay->vaddr + offset; win_data->bpp = overlay->bpp; win_data->buf_offsize = (overlay->fb_width - overlay->crtc_width) * @@ -263,7 +263,7 @@ static void fimd_win_mode_set(struct device *dev, DRM_DEBUG_KMS("ovl_width = %d, ovl_height = %d\n", win_data->ovl_width, win_data->ovl_height); DRM_DEBUG_KMS("paddr = 0x%lx, vaddr = 0x%lx\n", - (unsigned long)win_data->paddr, + (unsigned long)win_data->dma_addr, (unsigned long)win_data->vaddr); DRM_DEBUG_KMS("fb_width = %d, crtc_width = %d\n", overlay->fb_width, overlay->crtc_width); @@ -376,16 +376,16 @@ static void fimd_win_commit(struct device *dev) writel(val, ctx->regs + SHADOWCON); /* buffer start address */ - val = win_data->paddr; + val = (unsigned long)win_data->dma_addr; writel(val, ctx->regs + VIDWx_BUF_START(win, 0)); /* buffer end address */ size = win_data->fb_width * win_data->ovl_height * (win_data->bpp >> 3); - val = win_data->paddr + size; + val = (unsigned long)(win_data->dma_addr + size); writel(val, ctx->regs + VIDWx_BUF_END(win, 0)); DRM_DEBUG_KMS("start addr = 0x%lx, end addr = 0x%lx, size = 0x%lx\n", - (unsigned long)win_data->paddr, val, size); + (unsigned long)win_data->dma_addr, val, size); DRM_DEBUG_KMS("ovl_width = %d, ovl_height = %d\n", win_data->ovl_width, win_data->ovl_height); diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index bd6ede83b6846..b1b94b1e44001 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -127,15 +127,15 @@ struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev, { struct exynos_drm_gem_obj *exynos_gem_obj = NULL; - struct exynos_drm_buf_entry *entry; + struct exynos_drm_gem_buf *buffer; int ret; size = roundup(size, PAGE_SIZE); DRM_DEBUG_KMS("%s: size = 0x%lx\n", __FILE__, size); - entry = exynos_drm_buf_create(dev, size); - if (!entry) + buffer = exynos_drm_buf_create(dev, size); + if (!buffer) return ERR_PTR(-ENOMEM); exynos_gem_obj = exynos_drm_gem_init(dev, file_priv, handle, size); @@ -144,12 +144,12 @@ struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev, goto err_gem_init; } - exynos_gem_obj->entry = entry; + exynos_gem_obj->buffer = buffer; return exynos_gem_obj; err_gem_init: - exynos_drm_buf_destroy(dev, exynos_gem_obj->entry); + exynos_drm_buf_destroy(dev, exynos_gem_obj->buffer); return ERR_PTR(ret); } @@ -194,7 +194,7 @@ static int exynos_drm_gem_mmap_buffer(struct file *filp, { struct drm_gem_object *obj = filp->private_data; struct exynos_drm_gem_obj *exynos_gem_obj = to_exynos_gem_obj(obj); - struct exynos_drm_buf_entry *entry; + struct exynos_drm_gem_buf *buffer; unsigned long pfn, vm_size; DRM_DEBUG_KMS("%s\n", __FILE__); @@ -206,20 +206,20 @@ static int exynos_drm_gem_mmap_buffer(struct file *filp, vm_size = vma->vm_end - vma->vm_start; /* - * a entry contains information to physically continuous memory + * a buffer contains information to physically continuous memory * allocated by user request or at framebuffer creation. */ - entry = exynos_gem_obj->entry; + buffer = exynos_gem_obj->buffer; /* check if user-requested size is valid. */ - if (vm_size > entry->size) + if (vm_size > buffer->size) return -EINVAL; /* * get page frame number to physical memory to be mapped * to user space. */ - pfn = exynos_gem_obj->entry->paddr >> PAGE_SHIFT; + pfn = ((unsigned long)exynos_gem_obj->buffer->dma_addr) >> PAGE_SHIFT; DRM_DEBUG_KMS("pfn = 0x%lx\n", pfn); @@ -300,7 +300,7 @@ void exynos_drm_gem_free_object(struct drm_gem_object *gem_obj) exynos_gem_obj = to_exynos_gem_obj(gem_obj); - exynos_drm_buf_destroy(gem_obj->dev, exynos_gem_obj->entry); + exynos_drm_buf_destroy(gem_obj->dev, exynos_gem_obj->buffer); kfree(exynos_gem_obj); } @@ -379,7 +379,8 @@ int exynos_drm_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) mutex_lock(&dev->struct_mutex); - pfn = (exynos_gem_obj->entry->paddr >> PAGE_SHIFT) + page_offset; + pfn = (((unsigned long)exynos_gem_obj->buffer->dma_addr) >> + PAGE_SHIFT) + page_offset; ret = vm_insert_mixed(vma, (unsigned long)vmf->virtual_address, pfn); diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h index 213838d9606e8..ef8797334e6da 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.h +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h @@ -29,14 +29,30 @@ #define to_exynos_gem_obj(x) container_of(x,\ struct exynos_drm_gem_obj, base) +/* + * exynos drm gem buffer structure. + * + * @kvaddr: kernel virtual address to allocated memory region. + * @dma_addr: bus address(accessed by dma) to allocated memory region. + * - this address could be physical address without IOMMU and + * device address with IOMMU. + * @size: size of allocated memory region. + */ +struct exynos_drm_gem_buf { + void __iomem *kvaddr; + dma_addr_t dma_addr; + unsigned long size; +}; + /* * exynos drm buffer structure. * * @base: a gem object. * - a new handle to this gem object would be created * by drm_gem_handle_create(). - * @entry: pointer to exynos drm buffer entry object. - * - containing the information to physically + * @buffer: a pointer to exynos_drm_gem_buffer object. + * - contain the information to memory region allocated + * by user request or at framebuffer creation. * continuous memory region allocated by user request * or at framebuffer creation. * @@ -45,7 +61,7 @@ */ struct exynos_drm_gem_obj { struct drm_gem_object base; - struct exynos_drm_buf_entry *entry; + struct exynos_drm_gem_buf *buffer; }; /* create a new buffer and get a new gem handle. */ -- GitLab From 483b88f86e1682241bfa0848e348aa175257c6e7 Mon Sep 17 00:00:00 2001 From: Inki Dae Date: Fri, 11 Nov 2011 21:28:00 +0900 Subject: [PATCH 0351/3089] drm/exynos: fix vblank bug. In case that vblank_disable_allowed is 1, the problem that manager->pipe could be -1 at vsync interrupt handler could be induced so this patch fixes that. Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_fimd.c | 46 ++++++++++++++++++------ 1 file changed, 36 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index f5c8b072e497a..272c3b53c0627 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -447,7 +447,9 @@ static void fimd_win_commit(struct device *dev) static void fimd_win_disable(struct device *dev) { struct fimd_context *ctx = get_fimd_context(dev); - struct fimd_win_data *win_data; + struct exynos_drm_subdrv *subdrv = &ctx->subdrv; + struct drm_device *drm_dev = subdrv->drm_dev; + struct exynos_drm_manager *manager = &subdrv->manager; int win = ctx->default_win; u32 val; @@ -456,8 +458,6 @@ static void fimd_win_disable(struct device *dev) if (win < 0 || win > WINDOWS_NR) return; - win_data = &ctx->win_data[win]; - /* protect windows */ val = readl(ctx->regs + SHADOWCON); val |= SHADOWCON_WINx_PROTECT(win); @@ -473,6 +473,29 @@ static void fimd_win_disable(struct device *dev) val &= ~SHADOWCON_CHx_ENABLE(win); val &= ~SHADOWCON_WINx_PROTECT(win); writel(val, ctx->regs + SHADOWCON); + + /* fimd dma off. */ + val = readl(ctx->regs + VIDCON0); + val &= ~(VIDCON0_ENVID | VIDCON0_ENVID_F); + writel(val, ctx->regs + VIDCON0); + + /* + * if vblank is enabled status with dma off then + * it disables vsync interrupt. + */ + if (drm_dev->vblank_enabled[manager->pipe] && + atomic_read(&drm_dev->vblank_refcount[manager->pipe])) { + drm_vblank_put(drm_dev, manager->pipe); + + /* + * if vblank_disable_allowed is 0 then disable vsync interrupt + * right now else the vsync interrupt would be disabled by drm + * timer once a current process gives up ownershop of + * vblank event. + */ + if (!drm_dev->vblank_disable_allowed) + drm_vblank_off(drm_dev, manager->pipe); + } } static struct exynos_drm_overlay_ops fimd_overlay_ops = { @@ -528,6 +551,16 @@ static irqreturn_t fimd_irq_handler(int irq, void *dev_id) /* VSYNC interrupt */ writel(VIDINTCON1_INT_FRAME, ctx->regs + VIDINTCON1); + /* + * in case that vblank_disable_allowed is 1, it could induce + * the problem that manager->pipe could be -1 because with + * disable callback, vsync interrupt isn't disabled and at this moment, + * vsync interrupt could occur. the vsync interrupt would be disabled + * by timer handler later. + */ + if (manager->pipe == -1) + return IRQ_HANDLED; + drm_handle_vblank(drm_dev, manager->pipe); fimd_finish_pageflip(drm_dev, manager->pipe); @@ -548,13 +581,6 @@ static int fimd_subdrv_probe(struct drm_device *drm_dev, struct device *dev) */ drm_dev->irq_enabled = 1; - /* - * with vblank_disable_allowed = 1, vblank interrupt will be disabled - * by drm timer once a current process gives up ownership of - * vblank event.(drm_vblank_put function was called) - */ - drm_dev->vblank_disable_allowed = 1; - return 0; } -- GitLab From 4f9eb94f7be3d357e811ec74a53027bd27f1748f Mon Sep 17 00:00:00 2001 From: Inki Dae Date: Sat, 12 Nov 2011 16:57:42 +0900 Subject: [PATCH 0352/3089] drm/exynos: include linux/module.h Signed-off-by: Inki Dae Signed-off-by: Kyungmin Park --- drivers/gpu/drm/exynos/exynos_drm_drv.h | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h index 38a6f1df75c61..2d74f8550e200 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.h +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h @@ -29,6 +29,7 @@ #ifndef _EXYNOS_DRM_DRV_H_ #define _EXYNOS_DRM_DRV_H_ +#include #include "drm.h" #define MAX_CRTC 2 -- GitLab From e13b91c1c47e7be178d2d7914452040ee82c0bd2 Mon Sep 17 00:00:00 2001 From: Guan Xuetao Date: Mon, 13 Jun 2011 11:44:49 +0800 Subject: [PATCH 0353/3089] unicore32: fix build error for find bitops Remove the __uc32_ prefix in find bitops functions. Move find_* macros behind asm-generic/bitops.h inclusion. see commit <19de85ef574c3a2182e3ccad9581805052f14946> bitops: add #ifndef for each of find bitops also see commit <63e424c84429903c92a0f1e9654c31ccaf6694d0> arch: remove CONFIG_GENERIC_FIND_{NEXT_BIT,BIT_LE,LAST_BIT} Signed-off-by: Guan Xuetao Cc: Akinobu Mita Acked-by: Akinobu Mita Cc: Arnd Bergmann --- arch/unicore32/include/asm/bitops.h | 12 ++++++------ arch/unicore32/kernel/ksyms.c | 4 ++-- arch/unicore32/lib/findbit.S | 14 ++++++++------ 3 files changed, 16 insertions(+), 14 deletions(-) diff --git a/arch/unicore32/include/asm/bitops.h b/arch/unicore32/include/asm/bitops.h index 1628a63289946..401f597bc38cf 100644 --- a/arch/unicore32/include/asm/bitops.h +++ b/arch/unicore32/include/asm/bitops.h @@ -13,12 +13,6 @@ #ifndef __UNICORE_BITOPS_H__ #define __UNICORE_BITOPS_H__ -#define find_next_bit __uc32_find_next_bit -#define find_next_zero_bit __uc32_find_next_zero_bit - -#define find_first_bit __uc32_find_first_bit -#define find_first_zero_bit __uc32_find_first_zero_bit - #define _ASM_GENERIC_BITOPS_FLS_H_ #define _ASM_GENERIC_BITOPS___FLS_H_ #define _ASM_GENERIC_BITOPS_FFS_H_ @@ -44,4 +38,10 @@ static inline int fls(int x) #include +/* following definitions: to avoid using codes in lib/find_*.c */ +#define find_next_bit find_next_bit +#define find_next_zero_bit find_next_zero_bit +#define find_first_bit find_first_bit +#define find_first_zero_bit find_first_zero_bit + #endif /* __UNICORE_BITOPS_H__ */ diff --git a/arch/unicore32/kernel/ksyms.c b/arch/unicore32/kernel/ksyms.c index a8970809428a6..d98bd812cae1e 100644 --- a/arch/unicore32/kernel/ksyms.c +++ b/arch/unicore32/kernel/ksyms.c @@ -24,8 +24,8 @@ #include "ksyms.h" -EXPORT_SYMBOL(__uc32_find_next_zero_bit); -EXPORT_SYMBOL(__uc32_find_next_bit); +EXPORT_SYMBOL(find_next_zero_bit); +EXPORT_SYMBOL(find_next_bit); EXPORT_SYMBOL(__backtrace); diff --git a/arch/unicore32/lib/findbit.S b/arch/unicore32/lib/findbit.S index c360ce905d8b7..c77746247d369 100644 --- a/arch/unicore32/lib/findbit.S +++ b/arch/unicore32/lib/findbit.S @@ -17,7 +17,7 @@ * Purpose : Find a 'zero' bit * Prototype: int find_first_zero_bit(void *addr, unsigned int maxbit); */ -__uc32_find_first_zero_bit: +ENTRY(find_first_zero_bit) cxor.a r1, #0 beq 3f mov r2, #0 @@ -29,13 +29,14 @@ __uc32_find_first_zero_bit: bub 1b 3: mov r0, r1 @ no free bits mov pc, lr +ENDPROC(find_first_zero_bit) /* * Purpose : Find next 'zero' bit * Prototype: int find_next_zero_bit * (void *addr, unsigned int maxbit, int offset) */ -ENTRY(__uc32_find_next_zero_bit) +ENTRY(find_next_zero_bit) cxor.a r1, #0 beq 3b and.a ip, r2, #7 @@ -47,14 +48,14 @@ ENTRY(__uc32_find_next_zero_bit) or r2, r2, #7 @ if zero, then no bits here add r2, r2, #1 @ align bit pointer b 2b @ loop for next bit -ENDPROC(__uc32_find_next_zero_bit) +ENDPROC(find_next_zero_bit) /* * Purpose : Find a 'one' bit * Prototype: int find_first_bit * (const unsigned long *addr, unsigned int maxbit); */ -__uc32_find_first_bit: +ENTRY(find_first_bit) cxor.a r1, #0 beq 3f mov r2, #0 @@ -66,13 +67,14 @@ __uc32_find_first_bit: bub 1b 3: mov r0, r1 @ no free bits mov pc, lr +ENDPROC(find_first_bit) /* * Purpose : Find next 'one' bit * Prototype: int find_next_zero_bit * (void *addr, unsigned int maxbit, int offset) */ -ENTRY(__uc32_find_next_bit) +ENTRY(find_next_bit) cxor.a r1, #0 beq 3b and.a ip, r2, #7 @@ -83,7 +85,7 @@ ENTRY(__uc32_find_next_bit) or r2, r2, #7 @ if zero, then no bits here add r2, r2, #1 @ align bit pointer b 2b @ loop for next bit -ENDPROC(__uc32_find_next_bit) +ENDPROC(find_next_bit) /* * One or more bits in the LSB of r3 are assumed to be set. -- GitLab From 63c95fa414032adb56194d026c9ed054488b0783 Mon Sep 17 00:00:00 2001 From: Arnaud Lacombe Date: Mon, 15 Aug 2011 01:07:06 -0400 Subject: [PATCH 0354/3089] arch/unicore32: do not use EXTRA_AFLAGS or EXTRA_CFLAGS Usage of these flags has been deprecated for nearly 4 years by: commit f77bf01425b11947eeb3b5b54685212c302741b8 Author: Sam Ravnborg Date: Mon Oct 15 22:25:06 2007 +0200 kbuild: introduce ccflags-y, asflags-y and ldflags-y Moreover, these flags (at least EXTRA_CFLAGS) have been documented for command line use. By default, gmake(1) do not override command line setting, so this is likely to result in build failure or unexpected behavior. Replace their usage by Kbuild's `{as,cc,ld}flags-y'. Cc: Sam Ravnborg Cc: Guan Xuetao Signed-off-by: Arnaud Lacombe Acked-by: Guan Xuetao --- arch/unicore32/boot/compressed/Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/unicore32/boot/compressed/Makefile b/arch/unicore32/boot/compressed/Makefile index b0954a2d23cfa..950a9afa38f86 100644 --- a/arch/unicore32/boot/compressed/Makefile +++ b/arch/unicore32/boot/compressed/Makefile @@ -10,8 +10,8 @@ # Copyright (C) 2001~2010 GUAN Xue-tao # -EXTRA_CFLAGS := -fpic -fno-builtin -EXTRA_AFLAGS := -Wa,-march=all +ccflags-y := -fpic -fno-builtin +asflags-y := -Wa,-march=all OBJS := misc.o -- GitLab From 66ae45604c90f25397294099cdb214f642471327 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Wed, 5 Oct 2011 22:29:47 +0800 Subject: [PATCH 0355/3089] rtc: rtc-puv3: Add __devinit and __devexit markers for probe and remove Signed-off-by: Axel Lin Acked-by: Guan Xuetao --- drivers/rtc/rtc-puv3.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/rtc/rtc-puv3.c b/drivers/rtc/rtc-puv3.c index b3eba3cddd426..e4b6880aabd05 100644 --- a/drivers/rtc/rtc-puv3.c +++ b/drivers/rtc/rtc-puv3.c @@ -220,7 +220,7 @@ static void puv3_rtc_enable(struct platform_device *pdev, int en) } } -static int puv3_rtc_remove(struct platform_device *dev) +static int __devexit puv3_rtc_remove(struct platform_device *dev) { struct rtc_device *rtc = platform_get_drvdata(dev); @@ -236,7 +236,7 @@ static int puv3_rtc_remove(struct platform_device *dev) return 0; } -static int puv3_rtc_probe(struct platform_device *pdev) +static int __devinit puv3_rtc_probe(struct platform_device *pdev) { struct rtc_device *rtc; struct resource *res; -- GitLab From d159e571cb81a75d50f84fc4b716edee636f6858 Mon Sep 17 00:00:00 2001 From: Paul Bolle Date: Fri, 14 Oct 2011 14:28:11 +0200 Subject: [PATCH 0356/3089] unicore32: drop unused Kconfig symbols Signed-off-by: Paul Bolle Acked-by: Guan Xuetao --- arch/unicore32/Kconfig.debug | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/arch/unicore32/Kconfig.debug b/arch/unicore32/Kconfig.debug index ae2ec334c3c61..1a36262398435 100644 --- a/arch/unicore32/Kconfig.debug +++ b/arch/unicore32/Kconfig.debug @@ -44,18 +44,4 @@ config DEBUG_OCD Say Y here if you want the debug print routines to direct their output to the UniCore On-Chip-Debugger channel using CP #1. -config DEBUG_OCD_BREAKPOINT - bool "Breakpoint support via On-Chip-Debugger" - depends on DEBUG_OCD - -config DEBUG_UART - int "Kernel low-level debugging messages via serial port" - depends on DEBUG_LL - range 0 1 - default "0" - help - Choice for UART for kernel low-level using PKUnity UARTS, - should be between zero and one. The port must have been - initialised by the boot-loader before use. - endmenu -- GitLab From 943ef2ec4f5f14940ea391d25bf93eb99eb2ff2a Mon Sep 17 00:00:00 2001 From: Wim Van Sebroeck Date: Tue, 15 Nov 2011 11:45:16 +0100 Subject: [PATCH 0357/3089] watchdog: adx_wdt.c: remove driver Remove the driver (that was added in v2.6.32) since the architecture has never been merged into mainline. Reported-by: Paul Bolle Acked-by: Thierry Reding Signed-off-by: Wim Van Sebroeck --- drivers/watchdog/Kconfig | 7 - drivers/watchdog/Makefile | 1 - drivers/watchdog/adx_wdt.c | 355 ------------------------------------- 3 files changed, 363 deletions(-) delete mode 100644 drivers/watchdog/adx_wdt.c diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig index 6285867a93568..79fd606b7cd5c 100644 --- a/drivers/watchdog/Kconfig +++ b/drivers/watchdog/Kconfig @@ -314,13 +314,6 @@ config NUC900_WATCHDOG To compile this driver as a module, choose M here: the module will be called nuc900_wdt. -config ADX_WATCHDOG - tristate "Avionic Design Xanthos watchdog" - depends on ARCH_PXA_ADX - help - Say Y here if you want support for the watchdog timer on Avionic - Design Xanthos boards. - config TS72XX_WATCHDOG tristate "TS-72XX SBC Watchdog" depends on MACH_TS72XX diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile index 55bd5740e9100..fe893e91935b6 100644 --- a/drivers/watchdog/Makefile +++ b/drivers/watchdog/Makefile @@ -51,7 +51,6 @@ obj-$(CONFIG_ORION_WATCHDOG) += orion_wdt.o obj-$(CONFIG_COH901327_WATCHDOG) += coh901327_wdt.o obj-$(CONFIG_STMP3XXX_WATCHDOG) += stmp3xxx_wdt.o obj-$(CONFIG_NUC900_WATCHDOG) += nuc900_wdt.o -obj-$(CONFIG_ADX_WATCHDOG) += adx_wdt.o obj-$(CONFIG_TS72XX_WATCHDOG) += ts72xx_wdt.o obj-$(CONFIG_IMX2_WDT) += imx2_wdt.o diff --git a/drivers/watchdog/adx_wdt.c b/drivers/watchdog/adx_wdt.c deleted file mode 100644 index af6e6b16475af..0000000000000 --- a/drivers/watchdog/adx_wdt.c +++ /dev/null @@ -1,355 +0,0 @@ -/* - * Copyright (C) 2008-2009 Avionic Design GmbH - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#define WATCHDOG_NAME "adx-wdt" - -/* register offsets */ -#define ADX_WDT_CONTROL 0x00 -#define ADX_WDT_CONTROL_ENABLE (1 << 0) -#define ADX_WDT_CONTROL_nRESET (1 << 1) -#define ADX_WDT_TIMEOUT 0x08 - -static struct platform_device *adx_wdt_dev; -static unsigned long driver_open; - -#define WDT_STATE_STOP 0 -#define WDT_STATE_START 1 - -struct adx_wdt { - void __iomem *base; - unsigned long timeout; - unsigned int state; - unsigned int wake; - spinlock_t lock; -}; - -static const struct watchdog_info adx_wdt_info = { - .identity = "Avionic Design Xanthos Watchdog", - .options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING, -}; - -static void adx_wdt_start_locked(struct adx_wdt *wdt) -{ - u32 ctrl; - - ctrl = readl(wdt->base + ADX_WDT_CONTROL); - ctrl |= ADX_WDT_CONTROL_ENABLE; - writel(ctrl, wdt->base + ADX_WDT_CONTROL); - wdt->state = WDT_STATE_START; -} - -static void adx_wdt_start(struct adx_wdt *wdt) -{ - unsigned long flags; - - spin_lock_irqsave(&wdt->lock, flags); - adx_wdt_start_locked(wdt); - spin_unlock_irqrestore(&wdt->lock, flags); -} - -static void adx_wdt_stop_locked(struct adx_wdt *wdt) -{ - u32 ctrl; - - ctrl = readl(wdt->base + ADX_WDT_CONTROL); - ctrl &= ~ADX_WDT_CONTROL_ENABLE; - writel(ctrl, wdt->base + ADX_WDT_CONTROL); - wdt->state = WDT_STATE_STOP; -} - -static void adx_wdt_stop(struct adx_wdt *wdt) -{ - unsigned long flags; - - spin_lock_irqsave(&wdt->lock, flags); - adx_wdt_stop_locked(wdt); - spin_unlock_irqrestore(&wdt->lock, flags); -} - -static void adx_wdt_set_timeout(struct adx_wdt *wdt, unsigned long seconds) -{ - unsigned long timeout = seconds * 1000; - unsigned long flags; - unsigned int state; - - spin_lock_irqsave(&wdt->lock, flags); - state = wdt->state; - adx_wdt_stop_locked(wdt); - writel(timeout, wdt->base + ADX_WDT_TIMEOUT); - - if (state == WDT_STATE_START) - adx_wdt_start_locked(wdt); - - wdt->timeout = timeout; - spin_unlock_irqrestore(&wdt->lock, flags); -} - -static void adx_wdt_get_timeout(struct adx_wdt *wdt, unsigned long *seconds) -{ - *seconds = wdt->timeout / 1000; -} - -static void adx_wdt_keepalive(struct adx_wdt *wdt) -{ - unsigned long flags; - - spin_lock_irqsave(&wdt->lock, flags); - writel(wdt->timeout, wdt->base + ADX_WDT_TIMEOUT); - spin_unlock_irqrestore(&wdt->lock, flags); -} - -static int adx_wdt_open(struct inode *inode, struct file *file) -{ - struct adx_wdt *wdt = platform_get_drvdata(adx_wdt_dev); - - if (test_and_set_bit(0, &driver_open)) - return -EBUSY; - - file->private_data = wdt; - adx_wdt_set_timeout(wdt, 30); - adx_wdt_start(wdt); - - return nonseekable_open(inode, file); -} - -static int adx_wdt_release(struct inode *inode, struct file *file) -{ - struct adx_wdt *wdt = file->private_data; - - adx_wdt_stop(wdt); - clear_bit(0, &driver_open); - - return 0; -} - -static long adx_wdt_ioctl(struct file *file, unsigned int cmd, unsigned long arg) -{ - struct adx_wdt *wdt = file->private_data; - void __user *argp = (void __user *)arg; - unsigned long __user *p = argp; - unsigned long seconds = 0; - unsigned int options; - long ret = -EINVAL; - - switch (cmd) { - case WDIOC_GETSUPPORT: - if (copy_to_user(argp, &adx_wdt_info, sizeof(adx_wdt_info))) - return -EFAULT; - else - return 0; - - case WDIOC_GETSTATUS: - case WDIOC_GETBOOTSTATUS: - return put_user(0, p); - - case WDIOC_KEEPALIVE: - adx_wdt_keepalive(wdt); - return 0; - - case WDIOC_SETTIMEOUT: - if (get_user(seconds, p)) - return -EFAULT; - - adx_wdt_set_timeout(wdt, seconds); - - /* fallthrough */ - case WDIOC_GETTIMEOUT: - adx_wdt_get_timeout(wdt, &seconds); - return put_user(seconds, p); - - case WDIOC_SETOPTIONS: - if (copy_from_user(&options, argp, sizeof(options))) - return -EFAULT; - - if (options & WDIOS_DISABLECARD) { - adx_wdt_stop(wdt); - ret = 0; - } - - if (options & WDIOS_ENABLECARD) { - adx_wdt_start(wdt); - ret = 0; - } - - return ret; - - default: - break; - } - - return -ENOTTY; -} - -static ssize_t adx_wdt_write(struct file *file, const char __user *data, - size_t len, loff_t *ppos) -{ - struct adx_wdt *wdt = file->private_data; - - if (len) - adx_wdt_keepalive(wdt); - - return len; -} - -static const struct file_operations adx_wdt_fops = { - .owner = THIS_MODULE, - .llseek = no_llseek, - .open = adx_wdt_open, - .release = adx_wdt_release, - .unlocked_ioctl = adx_wdt_ioctl, - .write = adx_wdt_write, -}; - -static struct miscdevice adx_wdt_miscdev = { - .minor = WATCHDOG_MINOR, - .name = "watchdog", - .fops = &adx_wdt_fops, -}; - -static int __devinit adx_wdt_probe(struct platform_device *pdev) -{ - struct resource *res; - struct adx_wdt *wdt; - int ret = 0; - u32 ctrl; - - wdt = devm_kzalloc(&pdev->dev, sizeof(*wdt), GFP_KERNEL); - if (!wdt) { - dev_err(&pdev->dev, "cannot allocate WDT structure\n"); - return -ENOMEM; - } - - spin_lock_init(&wdt->lock); - - res = platform_get_resource(pdev, IORESOURCE_MEM, 0); - if (!res) { - dev_err(&pdev->dev, "cannot obtain I/O memory region\n"); - return -ENXIO; - } - - res = devm_request_mem_region(&pdev->dev, res->start, - resource_size(res), res->name); - if (!res) { - dev_err(&pdev->dev, "cannot request I/O memory region\n"); - return -ENXIO; - } - - wdt->base = devm_ioremap_nocache(&pdev->dev, res->start, - resource_size(res)); - if (!wdt->base) { - dev_err(&pdev->dev, "cannot remap I/O memory region\n"); - return -ENXIO; - } - - /* disable watchdog and reboot on timeout */ - ctrl = readl(wdt->base + ADX_WDT_CONTROL); - ctrl &= ~ADX_WDT_CONTROL_ENABLE; - ctrl &= ~ADX_WDT_CONTROL_nRESET; - writel(ctrl, wdt->base + ADX_WDT_CONTROL); - - platform_set_drvdata(pdev, wdt); - adx_wdt_dev = pdev; - - ret = misc_register(&adx_wdt_miscdev); - if (ret) { - dev_err(&pdev->dev, "cannot register miscdev on minor %d " - "(err=%d)\n", WATCHDOG_MINOR, ret); - return ret; - } - - return 0; -} - -static int __devexit adx_wdt_remove(struct platform_device *pdev) -{ - struct adx_wdt *wdt = platform_get_drvdata(pdev); - - misc_deregister(&adx_wdt_miscdev); - adx_wdt_stop(wdt); - platform_set_drvdata(pdev, NULL); - - return 0; -} - -static void adx_wdt_shutdown(struct platform_device *pdev) -{ - struct adx_wdt *wdt = platform_get_drvdata(pdev); - adx_wdt_stop(wdt); -} - -#ifdef CONFIG_PM -static int adx_wdt_suspend(struct device *dev) -{ - struct platform_device *pdev = to_platform_device(dev); - struct adx_wdt *wdt = platform_get_drvdata(pdev); - - wdt->wake = (wdt->state == WDT_STATE_START) ? 1 : 0; - adx_wdt_stop(wdt); - - return 0; -} - -static int adx_wdt_resume(struct device *dev) -{ - struct platform_device *pdev = to_platform_device(dev); - struct adx_wdt *wdt = platform_get_drvdata(pdev); - - if (wdt->wake) - adx_wdt_start(wdt); - - return 0; -} - -static const struct dev_pm_ops adx_wdt_pm_ops = { - .suspend = adx_wdt_suspend, - .resume = adx_wdt_resume, -}; - -# define ADX_WDT_PM_OPS (&adx_wdt_pm_ops) -#else -# define ADX_WDT_PM_OPS NULL -#endif - -static struct platform_driver adx_wdt_driver = { - .probe = adx_wdt_probe, - .remove = __devexit_p(adx_wdt_remove), - .shutdown = adx_wdt_shutdown, - .driver = { - .name = WATCHDOG_NAME, - .owner = THIS_MODULE, - .pm = ADX_WDT_PM_OPS, - }, -}; - -static int __init adx_wdt_init(void) -{ - return platform_driver_register(&adx_wdt_driver); -} - -static void __exit adx_wdt_exit(void) -{ - platform_driver_unregister(&adx_wdt_driver); -} - -module_init(adx_wdt_init); -module_exit(adx_wdt_exit); - -MODULE_DESCRIPTION("Avionic Design Xanthos Watchdog Driver"); -MODULE_LICENSE("GPL v2"); -MODULE_AUTHOR("Thierry Reding "); -MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR); -- GitLab From 0f768a7235d3dfb6f4833030a95a06419df089cb Mon Sep 17 00:00:00 2001 From: Timur Tabi Date: Mon, 14 Nov 2011 16:35:26 -0600 Subject: [PATCH 0358/3089] ASoC: fsl_ssi: properly initialize the sysfs attribute object Commit 6992f533 ("sysfs: Use one lockdep class per sysfs attribute") requires 'struct attribute' objects to be initialized with sysfs_attr_init(). Signed-off-by: Timur Tabi Signed-off-by: Mark Brown Cc: stable@kernel.org --- sound/soc/fsl/fsl_ssi.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/fsl/fsl_ssi.c b/sound/soc/fsl/fsl_ssi.c index 0268cf989736f..83c4bd5b2dd76 100644 --- a/sound/soc/fsl/fsl_ssi.c +++ b/sound/soc/fsl/fsl_ssi.c @@ -694,6 +694,7 @@ static int __devinit fsl_ssi_probe(struct platform_device *pdev) /* Initialize the the device_attribute structure */ dev_attr = &ssi_private->dev_attr; + sysfs_attr_init(&dev_attr->attr); dev_attr->attr.name = "statistics"; dev_attr->attr.mode = S_IRUGO; dev_attr->show = fsl_sysfs_ssi_show; -- GitLab From f1ebcc74d5b2159f44c96b479b6eb8afc7829095 Mon Sep 17 00:00:00 2001 From: Liu Bo Date: Mon, 14 Nov 2011 20:48:06 -0500 Subject: [PATCH 0359/3089] Btrfs: fix tree corruption after multi-thread snapshots and inode_cache flush The btrfs snapshotting code requires that once a root has been snapshotted, we don't change it during a commit. But there are two cases to lead to tree corruptions: 1) multi-thread snapshots can commit serveral snapshots in a transaction, and this may change the src root when processing the following pending snapshots, which lead to the former snapshots corruptions; 2) the free inode cache was changing the roots when it root the cache, which lead to corruptions. This fixes things by making sure we force COW the block after we create a snapshot during commiting a transaction, then any changes to the roots will result in COW, and we get all the fs roots and snapshot roots to be consistent. Signed-off-by: Liu Bo Signed-off-by: Miao Xie Signed-off-by: Chris Mason --- fs/btrfs/ctree.c | 17 ++++++++++++++++- fs/btrfs/ctree.h | 2 ++ fs/btrfs/transaction.c | 8 ++++++++ 3 files changed, 26 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index 0fe615e4ea387..dede441bdeee2 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -514,10 +514,25 @@ static inline int should_cow_block(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct extent_buffer *buf) { + /* ensure we can see the force_cow */ + smp_rmb(); + + /* + * We do not need to cow a block if + * 1) this block is not created or changed in this transaction; + * 2) this block does not belong to TREE_RELOC tree; + * 3) the root is not forced COW. + * + * What is forced COW: + * when we create snapshot during commiting the transaction, + * after we've finished coping src root, we must COW the shared + * block to ensure the metadata consistency. + */ if (btrfs_header_generation(buf) == trans->transid && !btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN) && !(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID && - btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC))) + btrfs_header_flag(buf, BTRFS_HEADER_FLAG_RELOC)) && + !root->force_cow) return 0; return 1; } diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index b9ba59ff92925..b1cb3c0524844 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -1271,6 +1271,8 @@ struct btrfs_root { * for stat. It may be used for more later */ dev_t anon_dev; + + int force_cow; }; struct btrfs_ioctl_defrag_range_args { diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 6a0574e923bc6..81376d94cd3c6 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -785,6 +785,10 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans, btrfs_save_ino_cache(root, trans); + /* see comments in should_cow_block() */ + root->force_cow = 0; + smp_wmb(); + if (root->commit_root != root->node) { mutex_lock(&root->fs_commit_mutex); switch_commit_root(root); @@ -947,6 +951,10 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, btrfs_tree_unlock(old); free_extent_buffer(old); + /* see comments in should_cow_block() */ + root->force_cow = 1; + smp_wmb(); + btrfs_set_root_node(new_root_item, tmp); /* record when the snapshot was created in key.offset */ key.offset = trans->transid; -- GitLab From 731ad81e2dd97e3f222361f7b3ff4b35639e46af Mon Sep 17 00:00:00 2001 From: Tomoya MORINAGA Date: Fri, 28 Oct 2011 09:37:34 +0900 Subject: [PATCH 0360/3089] USB: pch_udc: Support new device LAPIS Semiconductor ML7831 IOH ML7831 is companion chip for Intel Atom E6xx series. Signed-off-by: Tomoya MORINAGA Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/Kconfig | 9 +++++---- drivers/usb/gadget/pch_udc.c | 6 ++++++ 2 files changed, 11 insertions(+), 4 deletions(-) diff --git a/drivers/usb/gadget/Kconfig b/drivers/usb/gadget/Kconfig index b21cd376c11af..23a447373c51f 100644 --- a/drivers/usb/gadget/Kconfig +++ b/drivers/usb/gadget/Kconfig @@ -469,7 +469,7 @@ config USB_LANGWELL gadget drivers to also be dynamically linked. config USB_EG20T - tristate "Intel EG20T PCH/OKI SEMICONDUCTOR ML7213 IOH UDC" + tristate "Intel EG20T PCH/LAPIS Semiconductor IOH(ML7213/ML7831) UDC" depends on PCI select USB_GADGET_DUALSPEED help @@ -485,10 +485,11 @@ config USB_EG20T This driver dose not support interrupt transfer or isochronous transfer modes. - This driver also can be used for OKI SEMICONDUCTOR's ML7213 which is + This driver also can be used for LAPIS Semiconductor's ML7213 which is for IVI(In-Vehicle Infotainment) use. - ML7213 is companion chip for Intel Atom E6xx series. - ML7213 is completely compatible for Intel EG20T PCH. + ML7831 is for general purpose use. + ML7213/ML7831 is companion chip for Intel Atom E6xx series. + ML7213/ML7831 is completely compatible for Intel EG20T PCH. config USB_CI13XXX_MSM tristate "MIPS USB CI13xxx for MSM" diff --git a/drivers/usb/gadget/pch_udc.c b/drivers/usb/gadget/pch_udc.c index 550d6dcdf1045..729e974dd6d73 100644 --- a/drivers/usb/gadget/pch_udc.c +++ b/drivers/usb/gadget/pch_udc.c @@ -354,6 +354,7 @@ struct pch_udc_dev { #define PCI_DEVICE_ID_INTEL_EG20T_UDC 0x8808 #define PCI_VENDOR_ID_ROHM 0x10DB #define PCI_DEVICE_ID_ML7213_IOH_UDC 0x801D +#define PCI_DEVICE_ID_ML7831_IOH_UDC 0x8808 static const char ep0_string[] = "ep0in"; static DEFINE_SPINLOCK(udc_stall_spinlock); /* stall spin lock */ @@ -2970,6 +2971,11 @@ static DEFINE_PCI_DEVICE_TABLE(pch_udc_pcidev_id) = { .class = (PCI_CLASS_SERIAL_USB << 8) | 0xfe, .class_mask = 0xffffffff, }, + { + PCI_DEVICE(PCI_VENDOR_ID_ROHM, PCI_DEVICE_ID_ML7831_IOH_UDC), + .class = (PCI_CLASS_SERIAL_USB << 8) | 0xfe, + .class_mask = 0xffffffff, + }, { 0 }, }; -- GitLab From 09b658dcb23efab48af10f85d5003ad3659938bc Mon Sep 17 00:00:00 2001 From: Tomoya MORINAGA Date: Fri, 28 Oct 2011 09:37:35 +0900 Subject: [PATCH 0361/3089] USB: pch_udc: Change company name OKI SEMICONDUCTOR to LAPIS Semiconductor On October 1 in 2011, OKI SEMICONDUCTOR Co., Ltd. changed the company name in to LAPIS Semiconductor Co., Ltd. Signed-off-by: Tomoya MORINAGA Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/pch_udc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/usb/gadget/pch_udc.c b/drivers/usb/gadget/pch_udc.c index 729e974dd6d73..5048a0c07640a 100644 --- a/drivers/usb/gadget/pch_udc.c +++ b/drivers/usb/gadget/pch_udc.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2010 OKI SEMICONDUCTOR CO., LTD. + * Copyright (C) 2011 LAPIS Semiconductor Co., Ltd. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -3005,5 +3005,5 @@ static void __exit pch_udc_pci_exit(void) module_exit(pch_udc_pci_exit); MODULE_DESCRIPTION("Intel EG20T USB Device Controller"); -MODULE_AUTHOR("OKI SEMICONDUCTOR, "); +MODULE_AUTHOR("LAPIS Semiconductor, "); MODULE_LICENSE("GPL"); -- GitLab From 0c16595539b612fe948559433dda08ff96a8bdc7 Mon Sep 17 00:00:00 2001 From: wangyanqing Date: Thu, 10 Nov 2011 14:04:08 +0800 Subject: [PATCH 0362/3089] USB: serial: pl2303: rm duplicate id I get report from customer that his usb-serial converter doesn't work well,it sometimes work, but sometimes it doesn't. The usb-serial converter's id: vendor_id product_id 0x4348 0x5523 Then I search the usb-serial codes, and there are two drivers announce support this device, pl2303 and ch341, commit 026dfaf1 cause it. Through many times to test, ch341 works well with this device, and pl2303 doesn't work quite often(it just work quite little). ch341 works well with this device, so we doesn't need pl2303 to support.I try to revert 026dfaf1 first, but it failed. So I prepare this patch by hand to revert it. Signed-off-by: Wang YanQing Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/pl2303.c | 1 - drivers/usb/serial/pl2303.h | 4 ---- 2 files changed, 5 deletions(-) diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c index 9083d1e616b4b..fc2d66f7f4eb0 100644 --- a/drivers/usb/serial/pl2303.c +++ b/drivers/usb/serial/pl2303.c @@ -91,7 +91,6 @@ static const struct usb_device_id id_table[] = { { USB_DEVICE(SONY_VENDOR_ID, SONY_QN3USB_PRODUCT_ID) }, { USB_DEVICE(SANWA_VENDOR_ID, SANWA_PRODUCT_ID) }, { USB_DEVICE(ADLINK_VENDOR_ID, ADLINK_ND6530_PRODUCT_ID) }, - { USB_DEVICE(WINCHIPHEAD_VENDOR_ID, WINCHIPHEAD_USBSER_PRODUCT_ID) }, { USB_DEVICE(SMART_VENDOR_ID, SMART_PRODUCT_ID) }, { } /* Terminating entry */ }; diff --git a/drivers/usb/serial/pl2303.h b/drivers/usb/serial/pl2303.h index 3d10d7f02072d..c38b8c00c06fd 100644 --- a/drivers/usb/serial/pl2303.h +++ b/drivers/usb/serial/pl2303.h @@ -145,10 +145,6 @@ #define ADLINK_VENDOR_ID 0x0b63 #define ADLINK_ND6530_PRODUCT_ID 0x6530 -/* WinChipHead USB->RS 232 adapter */ -#define WINCHIPHEAD_VENDOR_ID 0x4348 -#define WINCHIPHEAD_USBSER_PRODUCT_ID 0x5523 - /* SMART USB Serial Adapter */ #define SMART_VENDOR_ID 0x0b8c #define SMART_PRODUCT_ID 0x2303 -- GitLab From 5b061623355d8f69327a24838b0aa05e435ae5d5 Mon Sep 17 00:00:00 2001 From: VU Tuan Duc Date: Tue, 15 Nov 2011 14:08:00 +0700 Subject: [PATCH 0363/3089] USB: option: add id for 3G dongle Model VT1000 of Viettel Add VendorID/ProductID for USB 3G dongle Model VT1000 of Viettel. Signed-off-by: VU Tuan Duc Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/option.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index 89ae1f65e1b18..b8da93b3615c6 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -468,6 +468,10 @@ static void option_instat_callback(struct urb *urb); #define YUGA_PRODUCT_CLU528 0x260D #define YUGA_PRODUCT_CLU526 0x260F +/* Viettel products */ +#define VIETTEL_VENDOR_ID 0x2262 +#define VIETTEL_PRODUCT_VT1000 0x0002 + /* some devices interfaces need special handling due to a number of reasons */ enum option_blacklist_reason { OPTION_BLACKLIST_NONE = 0, @@ -1141,6 +1145,7 @@ static const struct usb_device_id option_ids[] = { { USB_DEVICE(YUGA_VENDOR_ID, YUGA_PRODUCT_CLU516) }, { USB_DEVICE(YUGA_VENDOR_ID, YUGA_PRODUCT_CLU528) }, { USB_DEVICE(YUGA_VENDOR_ID, YUGA_PRODUCT_CLU526) }, + { USB_DEVICE_AND_INTERFACE_INFO(VIETTEL_VENDOR_ID, VIETTEL_PRODUCT_VT1000, 0xff, 0xff, 0xff) }, { } /* Terminating entry */ }; MODULE_DEVICE_TABLE(usb, option_ids); -- GitLab From bc985c10f388c0e5fb2cff95ff2220721627fbd8 Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Tue, 15 Nov 2011 09:23:26 +0200 Subject: [PATCH 0364/3089] USB: storage: ene_ub6250: fix compile warnings MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix the following compile warning: | drivers/usb/storage/ene_ub6250.c: In function ‘ms_scsi_write’: | drivers/usb/storage/ene_ub6250.c:1728:6: warning: ‘result’ may \ | be used uninitialized in this function [-Wuninitialized] | drivers/usb/storage/ene_ub6250.c:1795:77: warning: ‘offset’ may \ | be used uninitialized in this function [-Wuninitialized] Signed-off-by: Felipe Balbi Signed-off-by: Greg Kroah-Hartman --- drivers/usb/storage/ene_ub6250.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/usb/storage/ene_ub6250.c b/drivers/usb/storage/ene_ub6250.c index 4dca3ef0668c9..9fbe742343c6c 100644 --- a/drivers/usb/storage/ene_ub6250.c +++ b/drivers/usb/storage/ene_ub6250.c @@ -1762,10 +1762,9 @@ static int ms_scsi_write(struct us_data *us, struct scsi_cmnd *srb) result = ene_send_scsi_cmd(us, FDIR_WRITE, scsi_sglist(srb), 1); } else { void *buf; - int offset; + int offset = 0; u16 PhyBlockAddr; u8 PageNum; - u32 result; u16 len, oldphy, newphy; buf = kmalloc(blenByte, GFP_KERNEL); -- GitLab From db0d4db22a78d31c59087f7057b8f1612fecc35d Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Sat, 12 Nov 2011 16:09:49 +0000 Subject: [PATCH 0365/3089] ARM: gic: allow GIC to support non-banked setups The GIC support code is heavily using the fact that hardware implementations are exposing banked registers. Unfortunately, it looks like at least one GIC implementation (EXYNOS) offers both the distributor and the CPU interfaces at different addresses, depending on the CPU. This problem is solved by allowing the distributor and CPU interface addresses to be per-cpu variables for the platforms that require it. The EXYNOS code is updated not to mess with the GIC internals while handling interrupts, and struct gic_chip_data is back to being private. The DT binding for the gic is updated to allow an optional "cpu-offset" value, which is used to compute the various base addresses. Finally, a new config option (GIC_NON_BANKED) is used to control this feature, so the overhead is only present on kernels compiled with support for EXYNOS. Tested on Origen (EXYNOS4) and Panda (OMAP4). Cc: Kukjin Kim Cc: Will Deacon Cc: Thomas Abraham Acked-by: Rob Herring Signed-off-by: Marc Zyngier --- Documentation/devicetree/bindings/arm/gic.txt | 4 + arch/arm/common/Kconfig | 3 + arch/arm/common/gic.c | 133 +++++++++++++++--- arch/arm/include/asm/hardware/gic.h | 24 ++-- arch/arm/mach-exynos/cpu.c | 16 +-- arch/arm/mach-exynos/platsmp.c | 28 +--- arch/arm/plat-s5p/Kconfig | 1 + 7 files changed, 132 insertions(+), 77 deletions(-) diff --git a/Documentation/devicetree/bindings/arm/gic.txt b/Documentation/devicetree/bindings/arm/gic.txt index 52916b4aa1fe9..9b4b82a721b64 100644 --- a/Documentation/devicetree/bindings/arm/gic.txt +++ b/Documentation/devicetree/bindings/arm/gic.txt @@ -42,6 +42,10 @@ Optional - interrupts : Interrupt source of the parent interrupt controller. Only present on secondary GICs. +- cpu-offset : per-cpu offset within the distributor and cpu interface + regions, used when the GIC doesn't have banked registers. The offset is + cpu-offset * cpu-nr. + Example: intc: interrupt-controller@fff11000 { diff --git a/arch/arm/common/Kconfig b/arch/arm/common/Kconfig index 74df9ca2be316..a3beda1213da0 100644 --- a/arch/arm/common/Kconfig +++ b/arch/arm/common/Kconfig @@ -2,6 +2,9 @@ config ARM_GIC select IRQ_DOMAIN bool +config GIC_NON_BANKED + bool + config ARM_VIC bool diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c index 0e6ae470c94f2..43cb6f1a7cf2e 100644 --- a/arch/arm/common/gic.c +++ b/arch/arm/common/gic.c @@ -43,6 +43,31 @@ #include #include +union gic_base { + void __iomem *common_base; + void __percpu __iomem **percpu_base; +}; + +struct gic_chip_data { + unsigned int irq_offset; + union gic_base dist_base; + union gic_base cpu_base; +#ifdef CONFIG_CPU_PM + u32 saved_spi_enable[DIV_ROUND_UP(1020, 32)]; + u32 saved_spi_conf[DIV_ROUND_UP(1020, 16)]; + u32 saved_spi_target[DIV_ROUND_UP(1020, 4)]; + u32 __percpu *saved_ppi_enable; + u32 __percpu *saved_ppi_conf; +#endif +#ifdef CONFIG_IRQ_DOMAIN + struct irq_domain domain; +#endif + unsigned int gic_irqs; +#ifdef CONFIG_GIC_NON_BANKED + void __iomem *(*get_base)(union gic_base *); +#endif +}; + static DEFINE_RAW_SPINLOCK(irq_controller_lock); /* Address of GIC 0 CPU interface */ @@ -67,16 +92,48 @@ struct irq_chip gic_arch_extn = { static struct gic_chip_data gic_data[MAX_GIC_NR] __read_mostly; +#ifdef CONFIG_GIC_NON_BANKED +static void __iomem *gic_get_percpu_base(union gic_base *base) +{ + return *__this_cpu_ptr(base->percpu_base); +} + +static void __iomem *gic_get_common_base(union gic_base *base) +{ + return base->common_base; +} + +static inline void __iomem *gic_data_dist_base(struct gic_chip_data *data) +{ + return data->get_base(&data->dist_base); +} + +static inline void __iomem *gic_data_cpu_base(struct gic_chip_data *data) +{ + return data->get_base(&data->cpu_base); +} + +static inline void gic_set_base_accessor(struct gic_chip_data *data, + void __iomem *(*f)(union gic_base *)) +{ + data->get_base = f; +} +#else +#define gic_data_dist_base(d) ((d)->dist_base.common_base) +#define gic_data_cpu_base(d) ((d)->cpu_base.common_base) +#define gic_set_base_accessor(d,f) +#endif + static inline void __iomem *gic_dist_base(struct irq_data *d) { struct gic_chip_data *gic_data = irq_data_get_irq_chip_data(d); - return gic_data->dist_base; + return gic_data_dist_base(gic_data); } static inline void __iomem *gic_cpu_base(struct irq_data *d) { struct gic_chip_data *gic_data = irq_data_get_irq_chip_data(d); - return gic_data->cpu_base; + return gic_data_cpu_base(gic_data); } static inline unsigned int gic_irq(struct irq_data *d) @@ -225,7 +282,7 @@ static void gic_handle_cascade_irq(unsigned int irq, struct irq_desc *desc) chained_irq_enter(chip, desc); raw_spin_lock(&irq_controller_lock); - status = readl_relaxed(chip_data->cpu_base + GIC_CPU_INTACK); + status = readl_relaxed(gic_data_cpu_base(chip_data) + GIC_CPU_INTACK); raw_spin_unlock(&irq_controller_lock); gic_irq = (status & 0x3ff); @@ -270,7 +327,7 @@ static void __init gic_dist_init(struct gic_chip_data *gic) u32 cpumask; unsigned int gic_irqs = gic->gic_irqs; struct irq_domain *domain = &gic->domain; - void __iomem *base = gic->dist_base; + void __iomem *base = gic_data_dist_base(gic); u32 cpu = 0; #ifdef CONFIG_SMP @@ -330,8 +387,8 @@ static void __init gic_dist_init(struct gic_chip_data *gic) static void __cpuinit gic_cpu_init(struct gic_chip_data *gic) { - void __iomem *dist_base = gic->dist_base; - void __iomem *base = gic->cpu_base; + void __iomem *dist_base = gic_data_dist_base(gic); + void __iomem *base = gic_data_cpu_base(gic); int i; /* @@ -368,7 +425,7 @@ static void gic_dist_save(unsigned int gic_nr) BUG(); gic_irqs = gic_data[gic_nr].gic_irqs; - dist_base = gic_data[gic_nr].dist_base; + dist_base = gic_data_dist_base(&gic_data[gic_nr]); if (!dist_base) return; @@ -403,7 +460,7 @@ static void gic_dist_restore(unsigned int gic_nr) BUG(); gic_irqs = gic_data[gic_nr].gic_irqs; - dist_base = gic_data[gic_nr].dist_base; + dist_base = gic_data_dist_base(&gic_data[gic_nr]); if (!dist_base) return; @@ -439,8 +496,8 @@ static void gic_cpu_save(unsigned int gic_nr) if (gic_nr >= MAX_GIC_NR) BUG(); - dist_base = gic_data[gic_nr].dist_base; - cpu_base = gic_data[gic_nr].cpu_base; + dist_base = gic_data_dist_base(&gic_data[gic_nr]); + cpu_base = gic_data_cpu_base(&gic_data[gic_nr]); if (!dist_base || !cpu_base) return; @@ -465,8 +522,8 @@ static void gic_cpu_restore(unsigned int gic_nr) if (gic_nr >= MAX_GIC_NR) BUG(); - dist_base = gic_data[gic_nr].dist_base; - cpu_base = gic_data[gic_nr].cpu_base; + dist_base = gic_data_dist_base(&gic_data[gic_nr]); + cpu_base = gic_data_cpu_base(&gic_data[gic_nr]); if (!dist_base || !cpu_base) return; @@ -491,6 +548,11 @@ static int gic_notifier(struct notifier_block *self, unsigned long cmd, void *v) int i; for (i = 0; i < MAX_GIC_NR; i++) { +#ifdef CONFIG_GIC_NON_BANKED + /* Skip over unused GICs */ + if (!gic_data[i].get_base) + continue; +#endif switch (cmd) { case CPU_PM_ENTER: gic_cpu_save(i); @@ -563,8 +625,9 @@ const struct irq_domain_ops gic_irq_domain_ops = { #endif }; -void __init gic_init(unsigned int gic_nr, int irq_start, - void __iomem *dist_base, void __iomem *cpu_base) +void __init gic_init_bases(unsigned int gic_nr, int irq_start, + void __iomem *dist_base, void __iomem *cpu_base, + u32 percpu_offset) { struct gic_chip_data *gic; struct irq_domain *domain; @@ -574,8 +637,36 @@ void __init gic_init(unsigned int gic_nr, int irq_start, gic = &gic_data[gic_nr]; domain = &gic->domain; - gic->dist_base = dist_base; - gic->cpu_base = cpu_base; +#ifdef CONFIG_GIC_NON_BANKED + if (percpu_offset) { /* Frankein-GIC without banked registers... */ + unsigned int cpu; + + gic->dist_base.percpu_base = alloc_percpu(void __iomem *); + gic->cpu_base.percpu_base = alloc_percpu(void __iomem *); + if (WARN_ON(!gic->dist_base.percpu_base || + !gic->cpu_base.percpu_base)) { + free_percpu(gic->dist_base.percpu_base); + free_percpu(gic->cpu_base.percpu_base); + return; + } + + for_each_possible_cpu(cpu) { + unsigned long offset = percpu_offset * cpu_logical_map(cpu); + *per_cpu_ptr(gic->dist_base.percpu_base, cpu) = dist_base + offset; + *per_cpu_ptr(gic->cpu_base.percpu_base, cpu) = cpu_base + offset; + } + + gic_set_base_accessor(gic, gic_get_percpu_base); + } else +#endif + { /* Normal, sane GIC... */ + WARN(percpu_offset, + "GIC_NON_BANKED not enabled, ignoring %08x offset!", + percpu_offset); + gic->dist_base.common_base = dist_base; + gic->cpu_base.common_base = cpu_base; + gic_set_base_accessor(gic, gic_get_common_base); + } /* * For primary GICs, skip over SGIs. @@ -593,7 +684,7 @@ void __init gic_init(unsigned int gic_nr, int irq_start, * Find out how many interrupts are supported. * The GIC only supports up to 1020 interrupt sources. */ - gic_irqs = readl_relaxed(dist_base + GIC_DIST_CTR) & 0x1f; + gic_irqs = readl_relaxed(gic_data_dist_base(gic) + GIC_DIST_CTR) & 0x1f; gic_irqs = (gic_irqs + 1) * 32; if (gic_irqs > 1020) gic_irqs = 1020; @@ -641,7 +732,7 @@ void gic_raise_softirq(const struct cpumask *mask, unsigned int irq) dsb(); /* this always happens on GIC0 */ - writel_relaxed(map << 16 | irq, gic_data[0].dist_base + GIC_DIST_SOFTINT); + writel_relaxed(map << 16 | irq, gic_data_dist_base(&gic_data[0]) + GIC_DIST_SOFTINT); } #endif @@ -652,6 +743,7 @@ int __init gic_of_init(struct device_node *node, struct device_node *parent) { void __iomem *cpu_base; void __iomem *dist_base; + u32 percpu_offset; int irq; struct irq_domain *domain = &gic_data[gic_cnt].domain; @@ -664,9 +756,12 @@ int __init gic_of_init(struct device_node *node, struct device_node *parent) cpu_base = of_iomap(node, 1); WARN(!cpu_base, "unable to map gic cpu registers\n"); + if (of_property_read_u32(node, "cpu-offset", &percpu_offset)) + percpu_offset = 0; + domain->of_node = of_node_get(node); - gic_init(gic_cnt, -1, dist_base, cpu_base); + gic_init_bases(gic_cnt, -1, dist_base, cpu_base, percpu_offset); if (parent) { irq = irq_of_parse_and_map(node, 0); diff --git a/arch/arm/include/asm/hardware/gic.h b/arch/arm/include/asm/hardware/gic.h index 3e91f22046f55..2721d90625e59 100644 --- a/arch/arm/include/asm/hardware/gic.h +++ b/arch/arm/include/asm/hardware/gic.h @@ -39,27 +39,19 @@ struct device_node; extern void __iomem *gic_cpu_base_addr; extern struct irq_chip gic_arch_extn; -void gic_init(unsigned int, int, void __iomem *, void __iomem *); +void gic_init_bases(unsigned int, int, void __iomem *, void __iomem *, + u32 offset); int gic_of_init(struct device_node *node, struct device_node *parent); void gic_secondary_init(unsigned int); void gic_cascade_irq(unsigned int gic_nr, unsigned int irq); void gic_raise_softirq(const struct cpumask *mask, unsigned int irq); -struct gic_chip_data { - void __iomem *dist_base; - void __iomem *cpu_base; -#ifdef CONFIG_CPU_PM - u32 saved_spi_enable[DIV_ROUND_UP(1020, 32)]; - u32 saved_spi_conf[DIV_ROUND_UP(1020, 16)]; - u32 saved_spi_target[DIV_ROUND_UP(1020, 4)]; - u32 __percpu *saved_ppi_enable; - u32 __percpu *saved_ppi_conf; -#endif -#ifdef CONFIG_IRQ_DOMAIN - struct irq_domain domain; -#endif - unsigned int gic_irqs; -}; +static inline void gic_init(unsigned int nr, int start, + void __iomem *dist , void __iomem *cpu) +{ + gic_init_bases(nr, start, dist, cpu, 0); +} + #endif #endif diff --git a/arch/arm/mach-exynos/cpu.c b/arch/arm/mach-exynos/cpu.c index 90ec247f3b375..e92e464bdbba7 100644 --- a/arch/arm/mach-exynos/cpu.c +++ b/arch/arm/mach-exynos/cpu.c @@ -207,27 +207,13 @@ void __init exynos4_init_clocks(int xtal) exynos4_setup_clocks(); } -static void exynos4_gic_irq_fix_base(struct irq_data *d) -{ - struct gic_chip_data *gic_data = irq_data_get_irq_chip_data(d); - - gic_data->cpu_base = S5P_VA_GIC_CPU + - (gic_bank_offset * smp_processor_id()); - - gic_data->dist_base = S5P_VA_GIC_DIST + - (gic_bank_offset * smp_processor_id()); -} - void __init exynos4_init_irq(void) { int irq; gic_bank_offset = soc_is_exynos4412() ? 0x4000 : 0x8000; - gic_init(0, IRQ_PPI(0), S5P_VA_GIC_DIST, S5P_VA_GIC_CPU); - gic_arch_extn.irq_eoi = exynos4_gic_irq_fix_base; - gic_arch_extn.irq_unmask = exynos4_gic_irq_fix_base; - gic_arch_extn.irq_mask = exynos4_gic_irq_fix_base; + gic_init_bases(0, IRQ_PPI(0), S5P_VA_GIC_DIST, S5P_VA_GIC_CPU, gic_bank_offset); for (irq = 0; irq < MAX_COMBINER_NR; irq++) { diff --git a/arch/arm/mach-exynos/platsmp.c b/arch/arm/mach-exynos/platsmp.c index 69ffb2fb38755..60bc45e3e7099 100644 --- a/arch/arm/mach-exynos/platsmp.c +++ b/arch/arm/mach-exynos/platsmp.c @@ -32,7 +32,6 @@ #include -extern unsigned int gic_bank_offset; extern void exynos4_secondary_startup(void); #define CPU1_BOOT_REG (samsung_rev() == EXYNOS4210_REV_1_1 ? \ @@ -65,31 +64,6 @@ static void __iomem *scu_base_addr(void) static DEFINE_SPINLOCK(boot_lock); -static void __cpuinit exynos4_gic_secondary_init(void) -{ - void __iomem *dist_base = S5P_VA_GIC_DIST + - (gic_bank_offset * smp_processor_id()); - void __iomem *cpu_base = S5P_VA_GIC_CPU + - (gic_bank_offset * smp_processor_id()); - int i; - - /* - * Deal with the banked PPI and SGI interrupts - disable all - * PPI interrupts, ensure all SGI interrupts are enabled. - */ - __raw_writel(0xffff0000, dist_base + GIC_DIST_ENABLE_CLEAR); - __raw_writel(0x0000ffff, dist_base + GIC_DIST_ENABLE_SET); - - /* - * Set priority on PPI and SGI interrupts - */ - for (i = 0; i < 32; i += 4) - __raw_writel(0xa0a0a0a0, dist_base + GIC_DIST_PRI + i * 4 / 4); - - __raw_writel(0xf0, cpu_base + GIC_CPU_PRIMASK); - __raw_writel(1, cpu_base + GIC_CPU_CTRL); -} - void __cpuinit platform_secondary_init(unsigned int cpu) { /* @@ -97,7 +71,7 @@ void __cpuinit platform_secondary_init(unsigned int cpu) * core (e.g. timer irq), then they will not have been enabled * for us: do so */ - exynos4_gic_secondary_init(); + gic_secondary_init(0); /* * let the primary processor know we're out of the diff --git a/arch/arm/plat-s5p/Kconfig b/arch/arm/plat-s5p/Kconfig index 9b9968fa8695b..8167ce66188c6 100644 --- a/arch/arm/plat-s5p/Kconfig +++ b/arch/arm/plat-s5p/Kconfig @@ -11,6 +11,7 @@ config PLAT_S5P default y select ARM_VIC if !ARCH_EXYNOS4 select ARM_GIC if ARCH_EXYNOS4 + select GIC_NON_BANKED if ARCH_EXYNOS4 select NO_IOPORT select ARCH_REQUIRE_GPIOLIB select S3C_GPIO_TRACK -- GitLab From abeb24ae4d3e543ecf0104cff08a3af4e7a42479 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 6 Sep 2011 09:23:26 +0100 Subject: [PATCH 0366/3089] ARM: Make global handler and CONFIG_MULTI_IRQ_HANDLER mutually exclusive Even when CONFIG_MULTI_IRQ_HANDLER is selected, the core code requires the arch_irq_handler_default macro to be defined as a fallback. It turns out nobody is using that particular feature as both PXA and shmobile have all their machine descriptors populated with the interrupt handler, leaving unused code (or empty macros) in their entry-macro.S file just to be able to compile entry-armv.S. Make CONFIG_MULTI_IRQ_HANDLER exclusive wrt arch_irq_handler_default, which allows to remove one test from the hot path. Also cleanup both PXA and shmobile entry-macro.S. Cc: Paul Mundt Acked-by: Nicolas Pitre Acked-by: Eric Miao Tested-by: Jamie Iles Signed-off-by: Marc Zyngier --- arch/arm/kernel/entry-armv.S | 7 ++-- arch/arm/mach-pxa/include/mach/entry-macro.S | 36 ------------------- .../mach-shmobile/include/mach/entry-macro.S | 9 ----- 3 files changed, 3 insertions(+), 49 deletions(-) diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index 9ad50c4208aeb..bd49a6a2a17db 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -36,12 +36,11 @@ #ifdef CONFIG_MULTI_IRQ_HANDLER ldr r1, =handle_arch_irq mov r0, sp - ldr r1, [r1] adr lr, BSYM(9997f) - teq r1, #0 - movne pc, r1 -#endif + ldr pc, [r1] +#else arch_irq_handler_default +#endif 9997: .endm diff --git a/arch/arm/mach-pxa/include/mach/entry-macro.S b/arch/arm/mach-pxa/include/mach/entry-macro.S index a73bc86a3c263..260c0c17692a0 100644 --- a/arch/arm/mach-pxa/include/mach/entry-macro.S +++ b/arch/arm/mach-pxa/include/mach/entry-macro.S @@ -7,45 +7,9 @@ * License version 2. This program is licensed "as is" without any * warranty of any kind, whether express or implied. */ -#include -#include .macro disable_fiq .endm - .macro get_irqnr_preamble, base, tmp - .endm - .macro arch_ret_to_user, tmp1, tmp2 .endm - - .macro get_irqnr_and_base, irqnr, irqstat, base, tmp - mrc p15, 0, \tmp, c0, c0, 0 @ CPUID - mov \tmp, \tmp, lsr #13 - and \tmp, \tmp, #0x7 @ Core G - cmp \tmp, #1 - bhi 1002f - - @ Core Generation 1 (PXA25x) - mov \base, #io_p2v(0x40000000) @ IIR Ctl = 0x40d00000 - add \base, \base, #0x00d00000 - ldr \irqstat, [\base, #0] @ ICIP - ldr \irqnr, [\base, #4] @ ICMR - - ands \irqnr, \irqstat, \irqnr - beq 1001f - rsb \irqstat, \irqnr, #0 - and \irqstat, \irqstat, \irqnr - clz \irqnr, \irqstat - rsb \irqnr, \irqnr, #(31 + PXA_IRQ(0)) - b 1001f -1002: - @ Core Generation 2 (PXA27x) or Core Generation 3 (PXA3xx) - mrc p6, 0, \irqstat, c5, c0, 0 @ ICHP - tst \irqstat, #0x80000000 - beq 1001f - bic \irqstat, \irqstat, #0x80000000 - mov \irqnr, \irqstat, lsr #16 - add \irqnr, \irqnr, #(PXA_IRQ(0)) -1001: - .endm diff --git a/arch/arm/mach-shmobile/include/mach/entry-macro.S b/arch/arm/mach-shmobile/include/mach/entry-macro.S index 8d4a416d42859..2a57b2964ee91 100644 --- a/arch/arm/mach-shmobile/include/mach/entry-macro.S +++ b/arch/arm/mach-shmobile/include/mach/entry-macro.S @@ -18,14 +18,5 @@ .macro disable_fiq .endm - .macro get_irqnr_preamble, base, tmp - .endm - - .macro get_irqnr_and_base, irqnr, irqstat, base, tmp - .endm - - .macro test_for_ipi, irqnr, irqstat, base, tmp - .endm - .macro arch_ret_to_user, tmp1, tmp2 .endm -- GitLab From baeeb8229cace91c10c856d91e5ca861d3c44968 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Wed, 2 Nov 2011 15:28:49 +0000 Subject: [PATCH 0367/3089] ARM: mxc: rename gic_handle_irq to avoid name clash Before introducing a global gic_handle_irq(), rename MXC's version to mxc_gic_handle_irq(). This function will be removed altogether in a later patch. Cc: Sascha Hauer Acked-by: Shawn Guo Signed-off-by: Marc Zyngier --- arch/arm/plat-mxc/gic.c | 2 +- arch/arm/plat-mxc/include/mach/common.h | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm/plat-mxc/gic.c b/arch/arm/plat-mxc/gic.c index 12f8f81090104..ea51ea4aba67b 100644 --- a/arch/arm/plat-mxc/gic.c +++ b/arch/arm/plat-mxc/gic.c @@ -18,7 +18,7 @@ #include #endif -asmlinkage void __exception_irq_entry gic_handle_irq(struct pt_regs *regs) +asmlinkage void __exception_irq_entry mxc_gic_handle_irq(struct pt_regs *regs) { u32 irqstat, irqnr; diff --git a/arch/arm/plat-mxc/include/mach/common.h b/arch/arm/plat-mxc/include/mach/common.h index 83b745a5e1b72..d7c290d32e3e3 100644 --- a/arch/arm/plat-mxc/include/mach/common.h +++ b/arch/arm/plat-mxc/include/mach/common.h @@ -90,7 +90,7 @@ extern void imx_print_silicon_rev(const char *cpu, int srev); void avic_handle_irq(struct pt_regs *); void tzic_handle_irq(struct pt_regs *); -void gic_handle_irq(struct pt_regs *); +void mxc_gic_handle_irq(struct pt_regs *); #define imx1_handle_irq avic_handle_irq #define imx21_handle_irq avic_handle_irq @@ -101,7 +101,7 @@ void gic_handle_irq(struct pt_regs *); #define imx50_handle_irq tzic_handle_irq #define imx51_handle_irq tzic_handle_irq #define imx53_handle_irq tzic_handle_irq -#define imx6q_handle_irq gic_handle_irq +#define imx6q_handle_irq mxc_gic_handle_irq extern void imx_enable_cpu(int cpu, bool enable); extern void imx_set_cpu_jump(int cpu, void *jump_addr); -- GitLab From 562e0027d21bf64838178e2f5157df3d5833972e Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 6 Sep 2011 09:56:17 +0100 Subject: [PATCH 0368/3089] ARM: GIC: Add global gic_handle_irq() function Provide the GIC code with a low level handler that can be used by platforms using CONFIG_MULTI_IRQ_HANDLER. Signed-off-by: Marc Zyngier --- arch/arm/common/gic.c | 27 +++++++++++++++++++++++++++ arch/arm/include/asm/hardware/gic.h | 1 + 2 files changed, 28 insertions(+) diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c index 43cb6f1a7cf2e..3c78b7c60691e 100644 --- a/arch/arm/common/gic.c +++ b/arch/arm/common/gic.c @@ -40,6 +40,7 @@ #include #include +#include #include #include @@ -272,6 +273,32 @@ static int gic_set_wake(struct irq_data *d, unsigned int on) #define gic_set_wake NULL #endif +asmlinkage void __exception_irq_entry gic_handle_irq(struct pt_regs *regs) +{ + u32 irqstat, irqnr; + struct gic_chip_data *gic = &gic_data[0]; + void __iomem *cpu_base = gic_data_cpu_base(gic); + + do { + irqstat = readl_relaxed(cpu_base + GIC_CPU_INTACK); + irqnr = irqstat & ~0x1c00; + + if (likely(irqnr > 15 && irqnr < 1021)) { + irqnr = irq_domain_to_irq(&gic->domain, irqnr); + handle_IRQ(irqnr, regs); + continue; + } + if (irqnr < 16) { + writel_relaxed(irqstat, cpu_base + GIC_CPU_EOI); +#ifdef CONFIG_SMP + handle_IPI(irqnr, regs); +#endif + continue; + } + break; + } while (1); +} + static void gic_handle_cascade_irq(unsigned int irq, struct irq_desc *desc) { struct gic_chip_data *chip_data = irq_get_handler_data(irq); diff --git a/arch/arm/include/asm/hardware/gic.h b/arch/arm/include/asm/hardware/gic.h index 2721d90625e59..ecf7c02fa16cf 100644 --- a/arch/arm/include/asm/hardware/gic.h +++ b/arch/arm/include/asm/hardware/gic.h @@ -43,6 +43,7 @@ void gic_init_bases(unsigned int, int, void __iomem *, void __iomem *, u32 offset); int gic_of_init(struct device_node *node, struct device_node *parent); void gic_secondary_init(unsigned int); +void gic_handle_irq(struct pt_regs *regs); void gic_cascade_irq(unsigned int gic_nr, unsigned int irq); void gic_raise_softirq(const struct cpumask *mask, unsigned int irq); -- GitLab From fea9fe8329c74159fb2030f1aacf29fdab9fc6d6 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 6 Sep 2011 10:23:45 +0100 Subject: [PATCH 0369/3089] ARM: imx: convert smp platforms to global gic_handle_irq() Convert the SMP imx platforms to use the global gic_handle_irq() function instead a private function. Cc: Sascha Hauer Acked-by: Shawn Guo Signed-off-by: Marc Zyngier --- arch/arm/plat-mxc/Makefile | 1 - arch/arm/plat-mxc/gic.c | 41 -------------------- arch/arm/plat-mxc/include/mach/common.h | 3 +- arch/arm/plat-mxc/include/mach/entry-macro.S | 11 ------ 4 files changed, 1 insertion(+), 55 deletions(-) delete mode 100644 arch/arm/plat-mxc/gic.c diff --git a/arch/arm/plat-mxc/Makefile b/arch/arm/plat-mxc/Makefile index b9f0f5f499a4e..076db84f3e318 100644 --- a/arch/arm/plat-mxc/Makefile +++ b/arch/arm/plat-mxc/Makefile @@ -5,7 +5,6 @@ # Common support obj-y := clock.o time.o devices.o cpu.o system.o irq-common.o -obj-$(CONFIG_ARM_GIC) += gic.o obj-$(CONFIG_MXC_TZIC) += tzic.o obj-$(CONFIG_MXC_AVIC) += avic.o diff --git a/arch/arm/plat-mxc/gic.c b/arch/arm/plat-mxc/gic.c deleted file mode 100644 index ea51ea4aba67b..0000000000000 --- a/arch/arm/plat-mxc/gic.c +++ /dev/null @@ -1,41 +0,0 @@ -/* - * Copyright 2011 Freescale Semiconductor, Inc. - * Copyright 2011 Linaro Ltd. - * - * The code contained herein is licensed under the GNU General Public - * License. You may obtain a copy of the GNU General Public License - * Version 2 or later at the following locations: - * - * http://www.opensource.org/licenses/gpl-license.html - * http://www.gnu.org/copyleft/gpl.html - */ - -#include -#include -#include -#include -#ifdef CONFIG_SMP -#include -#endif - -asmlinkage void __exception_irq_entry mxc_gic_handle_irq(struct pt_regs *regs) -{ - u32 irqstat, irqnr; - - do { - irqstat = readl_relaxed(gic_cpu_base_addr + GIC_CPU_INTACK); - irqnr = irqstat & 0x3ff; - if (irqnr == 1023) - break; - - if (irqnr > 15 && irqnr < 1021) - handle_IRQ(irqnr, regs); -#ifdef CONFIG_SMP - else { - writel_relaxed(irqstat, gic_cpu_base_addr + - GIC_CPU_EOI); - handle_IPI(irqnr, regs); - } -#endif - } while (1); -} diff --git a/arch/arm/plat-mxc/include/mach/common.h b/arch/arm/plat-mxc/include/mach/common.h index d7c290d32e3e3..14b4703e6e467 100644 --- a/arch/arm/plat-mxc/include/mach/common.h +++ b/arch/arm/plat-mxc/include/mach/common.h @@ -90,7 +90,6 @@ extern void imx_print_silicon_rev(const char *cpu, int srev); void avic_handle_irq(struct pt_regs *); void tzic_handle_irq(struct pt_regs *); -void mxc_gic_handle_irq(struct pt_regs *); #define imx1_handle_irq avic_handle_irq #define imx21_handle_irq avic_handle_irq @@ -101,7 +100,7 @@ void mxc_gic_handle_irq(struct pt_regs *); #define imx50_handle_irq tzic_handle_irq #define imx51_handle_irq tzic_handle_irq #define imx53_handle_irq tzic_handle_irq -#define imx6q_handle_irq mxc_gic_handle_irq +#define imx6q_handle_irq gic_handle_irq extern void imx_enable_cpu(int cpu, bool enable); extern void imx_set_cpu_jump(int cpu, void *jump_addr); diff --git a/arch/arm/plat-mxc/include/mach/entry-macro.S b/arch/arm/plat-mxc/include/mach/entry-macro.S index ca5cf26a04b1c..def5d30cb67e8 100644 --- a/arch/arm/plat-mxc/include/mach/entry-macro.S +++ b/arch/arm/plat-mxc/include/mach/entry-macro.S @@ -9,19 +9,8 @@ * published by the Free Software Foundation. */ -/* Unused, we use CONFIG_MULTI_IRQ_HANDLER */ - .macro disable_fiq .endm - .macro get_irqnr_preamble, base, tmp - .endm - .macro arch_ret_to_user, tmp1, tmp2 .endm - - .macro get_irqnr_and_base, irqnr, irqstat, base, tmp - .endm - - .macro test_for_ipi, irqnr, irqstat, base, tmp - .endm -- GitLab From 7e01799c669c60460dce43556065ca3b66760dcf Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 6 Sep 2011 10:23:45 +0100 Subject: [PATCH 0370/3089] ARM: highbank: convert to CONFIG_MULTI_IRQ_HANDLER Convert the highbank platform to be using the gic_handle_irq function as its primary interrupt handler. Cc: Rob Herring Signed-off-by: Marc Zyngier --- arch/arm/Kconfig | 1 + arch/arm/mach-highbank/highbank.c | 1 + arch/arm/mach-highbank/include/mach/entry-macro.S | 2 -- 3 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 44789eff983f4..5c014c6f056da 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -345,6 +345,7 @@ config ARCH_HIGHBANK select GENERIC_CLOCKEVENTS select HAVE_ARM_SCU select USE_OF + select MULTI_IRQ_HANDLER help Support for the Calxeda Highbank SoC based boards. diff --git a/arch/arm/mach-highbank/highbank.c b/arch/arm/mach-highbank/highbank.c index b82dcf08e747e..89bdf0039f7bf 100644 --- a/arch/arm/mach-highbank/highbank.c +++ b/arch/arm/mach-highbank/highbank.c @@ -140,6 +140,7 @@ DT_MACHINE_START(HIGHBANK, "Highbank") .map_io = highbank_map_io, .init_irq = highbank_init_irq, .timer = &highbank_timer, + .handle_irq = gic_handle_irq, .init_machine = highbank_init, .dt_compat = highbank_match, MACHINE_END diff --git a/arch/arm/mach-highbank/include/mach/entry-macro.S b/arch/arm/mach-highbank/include/mach/entry-macro.S index 73c11297509ed..a14f9e62ca920 100644 --- a/arch/arm/mach-highbank/include/mach/entry-macro.S +++ b/arch/arm/mach-highbank/include/mach/entry-macro.S @@ -1,5 +1,3 @@ -#include - .macro disable_fiq .endm -- GitLab From 1b99d9ccb52d64bfdab0d58915909b10cbbdf9e1 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 6 Sep 2011 10:23:45 +0100 Subject: [PATCH 0371/3089] ARM: RealView: convert to CONFIG_MULTI_IRQ_HANDLER Convert the RealView platforms to be using the gic_handle_irq function as their primary interrupt handler. Signed-off-by: Marc Zyngier --- arch/arm/Kconfig | 1 + arch/arm/mach-realview/include/mach/entry-macro.S | 2 -- arch/arm/mach-realview/realview_eb.c | 1 + arch/arm/mach-realview/realview_pb1176.c | 1 + arch/arm/mach-realview/realview_pb11mp.c | 1 + arch/arm/mach-realview/realview_pba8.c | 1 + arch/arm/mach-realview/realview_pbx.c | 1 + 7 files changed, 6 insertions(+), 2 deletions(-) diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 5c014c6f056da..5f3b7fedc6625 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -278,6 +278,7 @@ config ARCH_REALVIEW select ARM_TIMER_SP804 select GPIO_PL061 if GPIOLIB select NEED_MACH_MEMORY_H + select MULTI_IRQ_HANDLER help This enables support for ARM Ltd RealView boards. diff --git a/arch/arm/mach-realview/include/mach/entry-macro.S b/arch/arm/mach-realview/include/mach/entry-macro.S index 4071164aebaaa..e8a5179c26536 100644 --- a/arch/arm/mach-realview/include/mach/entry-macro.S +++ b/arch/arm/mach-realview/include/mach/entry-macro.S @@ -7,8 +7,6 @@ * License version 2. This program is licensed "as is" without any * warranty of any kind, whether express or implied. */ -#include -#include .macro disable_fiq .endm diff --git a/arch/arm/mach-realview/realview_eb.c b/arch/arm/mach-realview/realview_eb.c index 026c66ad7ec24..10ef2e74cce58 100644 --- a/arch/arm/mach-realview/realview_eb.c +++ b/arch/arm/mach-realview/realview_eb.c @@ -469,6 +469,7 @@ MACHINE_START(REALVIEW_EB, "ARM-RealView EB") .init_early = realview_init_early, .init_irq = gic_init_irq, .timer = &realview_eb_timer, + .handle_irq = gic_handle_irq, .init_machine = realview_eb_init, #ifdef CONFIG_ZONE_DMA .dma_zone_size = SZ_256M, diff --git a/arch/arm/mach-realview/realview_pb1176.c b/arch/arm/mach-realview/realview_pb1176.c index c057540ec776e..bd8fec8b20d96 100644 --- a/arch/arm/mach-realview/realview_pb1176.c +++ b/arch/arm/mach-realview/realview_pb1176.c @@ -392,6 +392,7 @@ MACHINE_START(REALVIEW_PB1176, "ARM-RealView PB1176") .init_early = realview_init_early, .init_irq = gic_init_irq, .timer = &realview_pb1176_timer, + .handle_irq = gic_handle_irq, .init_machine = realview_pb1176_init, #ifdef CONFIG_ZONE_DMA .dma_zone_size = SZ_256M, diff --git a/arch/arm/mach-realview/realview_pb11mp.c b/arch/arm/mach-realview/realview_pb11mp.c index 671ad6d6ff00d..fa73ba81a4493 100644 --- a/arch/arm/mach-realview/realview_pb11mp.c +++ b/arch/arm/mach-realview/realview_pb11mp.c @@ -366,6 +366,7 @@ MACHINE_START(REALVIEW_PB11MP, "ARM-RealView PB11MPCore") .init_early = realview_init_early, .init_irq = gic_init_irq, .timer = &realview_pb11mp_timer, + .handle_irq = gic_handle_irq, .init_machine = realview_pb11mp_init, #ifdef CONFIG_ZONE_DMA .dma_zone_size = SZ_256M, diff --git a/arch/arm/mach-realview/realview_pba8.c b/arch/arm/mach-realview/realview_pba8.c index cbf22df4ad5b2..6e5f2b9ddb7e3 100644 --- a/arch/arm/mach-realview/realview_pba8.c +++ b/arch/arm/mach-realview/realview_pba8.c @@ -316,6 +316,7 @@ MACHINE_START(REALVIEW_PBA8, "ARM-RealView PB-A8") .init_early = realview_init_early, .init_irq = gic_init_irq, .timer = &realview_pba8_timer, + .handle_irq = gic_handle_irq, .init_machine = realview_pba8_init, #ifdef CONFIG_ZONE_DMA .dma_zone_size = SZ_256M, diff --git a/arch/arm/mach-realview/realview_pbx.c b/arch/arm/mach-realview/realview_pbx.c index 63c4114afae92..291bf1a8f2f07 100644 --- a/arch/arm/mach-realview/realview_pbx.c +++ b/arch/arm/mach-realview/realview_pbx.c @@ -399,6 +399,7 @@ MACHINE_START(REALVIEW_PBX, "ARM-RealView PBX") .init_early = realview_init_early, .init_irq = gic_init_irq, .timer = &realview_pbx_timer, + .handle_irq = gic_handle_irq, .init_machine = realview_pbx_init, #ifdef CONFIG_ZONE_DMA .dma_zone_size = SZ_256M, -- GitLab From abd3ca51fba8bc200ae78e99bd1ac19c4e3b6781 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 6 Sep 2011 10:23:45 +0100 Subject: [PATCH 0372/3089] ARM: VExpress: convert to CONFIG_MULTI_IRQ_HANDLER Convert the VExpress platform to be using the gic_handle_irq function as its primary interrupt handler. Signed-off-by: Marc Zyngier --- arch/arm/Kconfig | 1 + arch/arm/mach-vexpress/include/mach/entry-macro.S | 2 -- arch/arm/mach-vexpress/v2m.c | 2 ++ 3 files changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 5f3b7fedc6625..8f39263c0768e 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -311,6 +311,7 @@ config ARCH_VEXPRESS select ICST select PLAT_VERSATILE select PLAT_VERSATILE_CLCD + select MULTI_IRQ_HANDLER help This enables support for the ARM Ltd Versatile Express boards. diff --git a/arch/arm/mach-vexpress/include/mach/entry-macro.S b/arch/arm/mach-vexpress/include/mach/entry-macro.S index 73c11297509ed..a14f9e62ca920 100644 --- a/arch/arm/mach-vexpress/include/mach/entry-macro.S +++ b/arch/arm/mach-vexpress/include/mach/entry-macro.S @@ -1,5 +1,3 @@ -#include - .macro disable_fiq .endm diff --git a/arch/arm/mach-vexpress/v2m.c b/arch/arm/mach-vexpress/v2m.c index 1fafc32446074..7aa07a8ce2323 100644 --- a/arch/arm/mach-vexpress/v2m.c +++ b/arch/arm/mach-vexpress/v2m.c @@ -23,6 +23,7 @@ #include #include #include +#include #include #include @@ -448,5 +449,6 @@ MACHINE_START(VEXPRESS, "ARM-Versatile Express") .init_early = v2m_init_early, .init_irq = v2m_init_irq, .timer = &v2m_timer, + .handle_irq = gic_handle_irq, .init_machine = v2m_init, MACHINE_END -- GitLab From 041f777c93bab29565d999a292b9b9b533fe6f5e Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Tue, 6 Sep 2011 10:23:45 +0100 Subject: [PATCH 0373/3089] ARM: msm: convert SMP platforms to CONFIG_MULTI_IRQ_HANDLER Convert the SMP msm platforms to be using the gic_handle_irq function as their primary interrupt handler. Tested-by: David Brown Acked-by: David Brown Signed-off-by: Marc Zyngier --- arch/arm/mach-msm/Kconfig | 2 + arch/arm/mach-msm/board-msm8960.c | 2 + arch/arm/mach-msm/board-msm8x60.c | 4 ++ .../mach-msm/include/mach/entry-macro-qgic.S | 17 --------- .../mach-msm/include/mach/entry-macro-vic.S | 37 ------------------- arch/arm/mach-msm/include/mach/entry-macro.S | 27 ++++++++++++-- 6 files changed, 31 insertions(+), 58 deletions(-) delete mode 100644 arch/arm/mach-msm/include/mach/entry-macro-qgic.S delete mode 100644 arch/arm/mach-msm/include/mach/entry-macro-vic.S diff --git a/arch/arm/mach-msm/Kconfig b/arch/arm/mach-msm/Kconfig index ebde97f5d5f0d..ba36b74881c63 100644 --- a/arch/arm/mach-msm/Kconfig +++ b/arch/arm/mach-msm/Kconfig @@ -50,6 +50,7 @@ config ARCH_MSM8X60 select GPIO_MSM_V2 select MSM_GPIOMUX select MSM_SCM if SMP + select MULTI_IRQ_HANDLER config ARCH_MSM8960 bool "MSM8960" @@ -60,6 +61,7 @@ config ARCH_MSM8960 select MSM_V2_TLMM select MSM_GPIOMUX select MSM_SCM if SMP + select MULTI_IRQ_HANDLER endchoice diff --git a/arch/arm/mach-msm/board-msm8960.c b/arch/arm/mach-msm/board-msm8960.c index 6dc1cbd2a595b..ed3598128530f 100644 --- a/arch/arm/mach-msm/board-msm8960.c +++ b/arch/arm/mach-msm/board-msm8960.c @@ -99,6 +99,7 @@ MACHINE_START(MSM8960_SIM, "QCT MSM8960 SIMULATOR") .map_io = msm8960_map_io, .init_irq = msm8960_init_irq, .timer = &msm_timer, + .handle_irq = gic_handle_irq, .init_machine = msm8960_sim_init, MACHINE_END @@ -108,6 +109,7 @@ MACHINE_START(MSM8960_RUMI3, "QCT MSM8960 RUMI3") .map_io = msm8960_map_io, .init_irq = msm8960_init_irq, .timer = &msm_timer, + .handle_irq = gic_handle_irq, .init_machine = msm8960_rumi3_init, MACHINE_END diff --git a/arch/arm/mach-msm/board-msm8x60.c b/arch/arm/mach-msm/board-msm8x60.c index 44bf71688373b..0a113424632c4 100644 --- a/arch/arm/mach-msm/board-msm8x60.c +++ b/arch/arm/mach-msm/board-msm8x60.c @@ -108,6 +108,7 @@ MACHINE_START(MSM8X60_RUMI3, "QCT MSM8X60 RUMI3") .reserve = msm8x60_reserve, .map_io = msm8x60_map_io, .init_irq = msm8x60_init_irq, + .handle_irq = gic_handle_irq, .init_machine = msm8x60_init, .timer = &msm_timer, MACHINE_END @@ -117,6 +118,7 @@ MACHINE_START(MSM8X60_SURF, "QCT MSM8X60 SURF") .reserve = msm8x60_reserve, .map_io = msm8x60_map_io, .init_irq = msm8x60_init_irq, + .handle_irq = gic_handle_irq, .init_machine = msm8x60_init, .timer = &msm_timer, MACHINE_END @@ -126,6 +128,7 @@ MACHINE_START(MSM8X60_SIM, "QCT MSM8X60 SIMULATOR") .reserve = msm8x60_reserve, .map_io = msm8x60_map_io, .init_irq = msm8x60_init_irq, + .handle_irq = gic_handle_irq, .init_machine = msm8x60_init, .timer = &msm_timer, MACHINE_END @@ -135,6 +138,7 @@ MACHINE_START(MSM8X60_FFA, "QCT MSM8X60 FFA") .reserve = msm8x60_reserve, .map_io = msm8x60_map_io, .init_irq = msm8x60_init_irq, + .handle_irq = gic_handle_irq, .init_machine = msm8x60_init, .timer = &msm_timer, MACHINE_END diff --git a/arch/arm/mach-msm/include/mach/entry-macro-qgic.S b/arch/arm/mach-msm/include/mach/entry-macro-qgic.S deleted file mode 100644 index 717076f3ca735..0000000000000 --- a/arch/arm/mach-msm/include/mach/entry-macro-qgic.S +++ /dev/null @@ -1,17 +0,0 @@ -/* - * Low-level IRQ helper macros - * - * Copyright (c) 2010, Code Aurora Forum. All rights reserved. - * - * This file is licensed under the terms of the GNU General Public - * License version 2. This program is licensed "as is" without any - * warranty of any kind, whether express or implied. - */ - -#include - - .macro disable_fiq - .endm - - .macro arch_ret_to_user, tmp1, tmp2 - .endm diff --git a/arch/arm/mach-msm/include/mach/entry-macro-vic.S b/arch/arm/mach-msm/include/mach/entry-macro-vic.S deleted file mode 100644 index 70563ed11b363..0000000000000 --- a/arch/arm/mach-msm/include/mach/entry-macro-vic.S +++ /dev/null @@ -1,37 +0,0 @@ -/* - * Copyright (C) 2007 Google, Inc. - * Author: Brian Swetland - * - * This software is licensed under the terms of the GNU General Public - * License version 2, as published by the Free Software Foundation, and - * may be copied, distributed, and modified under those terms. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - */ - -#include - - .macro disable_fiq - .endm - - .macro get_irqnr_preamble, base, tmp - @ enable imprecise aborts - cpsie a - mov \base, #MSM_VIC_BASE - .endm - - .macro arch_ret_to_user, tmp1, tmp2 - .endm - - .macro get_irqnr_and_base, irqnr, irqstat, base, tmp - @ 0xD0 has irq# or old irq# if the irq has been handled - @ 0xD4 has irq# or -1 if none pending *but* if you just - @ read 0xD4 you never get the first irq for some reason - ldr \irqnr, [\base, #0xD0] - ldr \irqnr, [\base, #0xD4] - cmp \irqnr, #0xffffffff - .endm diff --git a/arch/arm/mach-msm/include/mach/entry-macro.S b/arch/arm/mach-msm/include/mach/entry-macro.S index b16f082eeb6f7..41f7003ef34f7 100644 --- a/arch/arm/mach-msm/include/mach/entry-macro.S +++ b/arch/arm/mach-msm/include/mach/entry-macro.S @@ -16,8 +16,27 @@ * */ -#if defined(CONFIG_ARM_GIC) -#include -#else -#include + .macro disable_fiq + .endm + + .macro arch_ret_to_user, tmp1, tmp2 + .endm + +#if !defined(CONFIG_ARM_GIC) +#include + + .macro get_irqnr_preamble, base, tmp + @ enable imprecise aborts + cpsie a + mov \base, #MSM_VIC_BASE + .endm + + .macro get_irqnr_and_base, irqnr, irqstat, base, tmp + @ 0xD0 has irq# or old irq# if the irq has been handled + @ 0xD4 has irq# or -1 if none pending *but* if you just + @ read 0xD4 you never get the first irq for some reason + ldr \irqnr, [\base, #0xD0] + ldr \irqnr, [\base, #0xD4] + cmp \irqnr, #0xffffffff + .endm #endif -- GitLab From 4e44d2cb95bd93abe16a131dbcd4c052ae36665f Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Mon, 30 May 2011 11:04:53 +0100 Subject: [PATCH 0374/3089] ARM: exynos4: convert to CONFIG_MULTI_IRQ_HANDLER Convert the Exynos4 platforms to be using the gic_handle_irq function as their primary interrupt handler. Cc: Ben Dooks Cc: Kukjin Kim Signed-off-by: Marc Zyngier --- arch/arm/Kconfig | 1 + arch/arm/mach-exynos/cpu.c | 4 +- .../mach-exynos/include/mach/entry-macro.S | 75 ------------------- arch/arm/mach-exynos/mach-armlex4210.c | 2 + arch/arm/mach-exynos/mach-nuri.c | 2 + arch/arm/mach-exynos/mach-origen.c | 2 + arch/arm/mach-exynos/mach-smdk4x12.c | 3 + arch/arm/mach-exynos/mach-smdkv310.c | 3 + arch/arm/mach-exynos/mach-universal_c210.c | 4 +- 9 files changed, 18 insertions(+), 78 deletions(-) diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 8f39263c0768e..d897255bcdf5e 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -854,6 +854,7 @@ config ARCH_EXYNOS select HAVE_S3C2410_I2C if I2C select HAVE_S3C2410_WATCHDOG if WATCHDOG select NEED_MACH_MEMORY_H + select MULTI_IRQ_HANDLER help Support for SAMSUNG's EXYNOS SoCs (EXYNOS4/5) diff --git a/arch/arm/mach-exynos/cpu.c b/arch/arm/mach-exynos/cpu.c index e92e464bdbba7..6e34485caa362 100644 --- a/arch/arm/mach-exynos/cpu.c +++ b/arch/arm/mach-exynos/cpu.c @@ -15,6 +15,7 @@ #include #include +#include #include #include @@ -33,8 +34,6 @@ #include #include -unsigned int gic_bank_offset __read_mostly; - extern int combiner_init(unsigned int combiner_nr, void __iomem *base, unsigned int irq_start); extern void combiner_cascade_irq(unsigned int combiner_nr, unsigned int irq); @@ -210,6 +209,7 @@ void __init exynos4_init_clocks(int xtal) void __init exynos4_init_irq(void) { int irq; + unsigned int bank_offset; gic_bank_offset = soc_is_exynos4412() ? 0x4000 : 0x8000; diff --git a/arch/arm/mach-exynos/include/mach/entry-macro.S b/arch/arm/mach-exynos/include/mach/entry-macro.S index f5e9fd8e37b4f..3ba4f547534b0 100644 --- a/arch/arm/mach-exynos/include/mach/entry-macro.S +++ b/arch/arm/mach-exynos/include/mach/entry-macro.S @@ -9,83 +9,8 @@ * warranty of any kind, whether express or implied. */ -#include -#include -#include - .macro disable_fiq .endm - .macro get_irqnr_preamble, base, tmp - mov \tmp, #0 - - mrc p15, 0, \base, c0, c0, 5 - and \base, \base, #3 - cmp \base, #0 - beq 1f - - ldr \tmp, =gic_bank_offset - ldr \tmp, [\tmp] - cmp \base, #1 - beq 1f - - cmp \base, #2 - addeq \tmp, \tmp, \tmp - addne \tmp, \tmp, \tmp, LSL #1 - -1: ldr \base, =gic_cpu_base_addr - ldr \base, [\base] - add \base, \base, \tmp - .endm - .macro arch_ret_to_user, tmp1, tmp2 .endm - - /* - * The interrupt numbering scheme is defined in the - * interrupt controller spec. To wit: - * - * Interrupts 0-15 are IPI - * 16-28 are reserved - * 29-31 are local. We allow 30 to be used for the watchdog. - * 32-1020 are global - * 1021-1022 are reserved - * 1023 is "spurious" (no interrupt) - * - * For now, we ignore all local interrupts so only return an interrupt if it's - * between 30 and 1020. The test_for_ipi routine below will pick up on IPIs. - * - * A simple read from the controller will tell us the number of the highest - * priority enabled interrupt. We then just need to check whether it is in the - * valid range for an IRQ (30-1020 inclusive). - */ - - .macro get_irqnr_and_base, irqnr, irqstat, base, tmp - - ldr \irqstat, [\base, #GIC_CPU_INTACK] /* bits 12-10 = src CPU, 9-0 = int # */ - - ldr \tmp, =1021 - - bic \irqnr, \irqstat, #0x1c00 - - cmp \irqnr, #15 - cmpcc \irqnr, \irqnr - cmpne \irqnr, \tmp - cmpcs \irqnr, \irqnr - addne \irqnr, \irqnr, #32 - - .endm - - /* We assume that irqstat (the raw value of the IRQ acknowledge - * register) is preserved from the macro above. - * If there is an IPI, we immediately signal end of interrupt on the - * controller, since this requires the original irqstat value which - * we won't easily be able to recreate later. - */ - - .macro test_for_ipi, irqnr, irqstat, base, tmp - bic \irqnr, \irqstat, #0x1c00 - cmp \irqnr, #16 - strcc \irqstat, [\base, #GIC_CPU_EOI] - cmpcs \irqnr, \irqnr - .endm diff --git a/arch/arm/mach-exynos/mach-armlex4210.c b/arch/arm/mach-exynos/mach-armlex4210.c index f0ca6c157d292..49da3089249a0 100644 --- a/arch/arm/mach-exynos/mach-armlex4210.c +++ b/arch/arm/mach-exynos/mach-armlex4210.c @@ -16,6 +16,7 @@ #include #include +#include #include #include @@ -210,6 +211,7 @@ MACHINE_START(ARMLEX4210, "ARMLEX4210") .atag_offset = 0x100, .init_irq = exynos4_init_irq, .map_io = armlex4210_map_io, + .handle_irq = gic_handle_irq, .init_machine = armlex4210_machine_init, .timer = &exynos4_timer, MACHINE_END diff --git a/arch/arm/mach-exynos/mach-nuri.c b/arch/arm/mach-exynos/mach-nuri.c index 236bbe1871632..5acec11821a4a 100644 --- a/arch/arm/mach-exynos/mach-nuri.c +++ b/arch/arm/mach-exynos/mach-nuri.c @@ -32,6 +32,7 @@ #include #include +#include #include #include @@ -1333,6 +1334,7 @@ MACHINE_START(NURI, "NURI") .atag_offset = 0x100, .init_irq = exynos4_init_irq, .map_io = nuri_map_io, + .handle_irq = gic_handle_irq, .init_machine = nuri_machine_init, .timer = &exynos4_timer, .reserve = &nuri_reserve, diff --git a/arch/arm/mach-exynos/mach-origen.c b/arch/arm/mach-exynos/mach-origen.c index f80b563f2be78..5561b06c38ece 100644 --- a/arch/arm/mach-exynos/mach-origen.c +++ b/arch/arm/mach-exynos/mach-origen.c @@ -22,6 +22,7 @@ #include #include +#include #include #include