Patch series "mm/memory_hotplug: Shrink zones before removing memory",
v6.

This series fixes the access of uninitialized memmaps when shrinking
zones/nodes and when removing memory.  Also, it contains all fixes for
crashes that can be triggered when removing certain namespace using
memunmap_pages() - ZONE_DEVICE, reported by Aneesh.

We stop trying to shrink ZONE_DEVICE, as it's buggy, fixing it would be
more involved (we don't have SECTION_IS_ONLINE as an indicator), and
shrinking is only of limited use (set_zone_contiguous() cannot detect
the ZONE_DEVICE as contiguous).

We continue shrinking !ZONE_DEVICE zones, however, I reduced the amount
of code to a minimum.  Shrinking is especially necessary to keep
zone->contiguous set where possible, especially, on memory unplug of
DIMMs at zone boundaries.

--------------------------------------------------------------------------

Zones are now properly shrunk when offlining memory blocks or when
onlining failed.  This allows to properly shrink zones on memory unplug
even if the separate memory blocks of a DIMM were onlined to different
zones or re-onlined to a different zone after offlining.

Example:

  :/# cat /proc/zoneinfo
  Node 1, zone  Movable
          spanned  0
          present  0
          managed  0
  :/# echo "online_movable" > /sys/devices/system/memory/memory41/state
  :/# echo "online_movable" > /sys/devices/system/memory/memory43/state
  :/# cat /proc/zoneinfo
  Node 1, zone  Movable
          spanned  98304
          present  65536
          managed  65536
  :/# echo 0 > /sys/devices/system/memory/memory43/online
  :/# cat /proc/zoneinfo
  Node 1, zone  Movable
          spanned  32768
          present  32768
          managed  32768
  :/# echo 0 > /sys/devices/system/memory/memory41/online
  :/# cat /proc/zoneinfo
  Node 1, zone  Movable
          spanned  0
          present  0
          managed  0

This patch (of 10):

With an altmap, the memmap falling into the reserved altmap space are not
initialized and, therefore, contain a garbage NID and a garbage zone.
Make sure to read the NID/zone from a memmap that was initialized.

This fixes a kernel crash that is observed when destroying a namespace:

  kernel BUG at include/linux/mm.h:1107!
  cpu 0x1: Vector: 700 (Program Check) at [c000000274087890]
      pc: c0000000004b9728: memunmap_pages+0x238/0x340
      lr: c0000000004b9724: memunmap_pages+0x234/0x340
  ...
      pid   = 3669, comm = ndctl
  kernel BUG at include/linux/mm.h:1107!
    devm_action_release+0x30/0x50
    release_nodes+0x268/0x2d0
    device_release_driver_internal+0x174/0x240
    unbind_store+0x13c/0x190
    drv_attr_store+0x44/0x60
    sysfs_kf_write+0x70/0xa0
    kernfs_fop_write+0x1ac/0x290
    __vfs_write+0x3c/0x70
    vfs_write+0xe4/0x200
    ksys_write+0x7c/0x140
    system_call+0x5c/0x68

The "page_zone(pfn_to_page(pfn)" was introduced by 69324b8f ("mm,
devm_memremap_pages: add MEMORY_DEVICE_PRIVATE support"), however, I
think we will never have driver reserved memory with
MEMORY_DEVICE_PRIVATE (no altmap AFAIKS).

[david@redhat.com: minimze code changes, rephrase description]
Link: http://lkml.kernel.org/r/20191006085646.5768-2-david@redhat.com
Fixes: 2c2a5af6

 ("mm, memory_hotplug: add nid parameter to arch_remove_memory")
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: David Hildenbrand <david@redhat.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Logan Gunthorpe <logang@deltatee.com>
Cc: Ira Weiny <ira.weiny@intel.com>
Cc: Damian Tometzki <damian.tometzki@gmail.com>
Cc: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Anshuman Khandual <anshuman.khandual@arm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Christophe Leroy <christophe.leroy@c-s.fr>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Halil Pasic <pasic@linux.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jun Yao <yaojun8558363@gmail.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: Pankaj Gupta <pagupta@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Pavel Tatashin <pasha.tatashin@soleen.com>
Cc: Pavel Tatashin <pavel.tatashin@microsoft.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qian Cai <cai@lca.pw>
Cc: Rich Felker <dalias@libc.org>
Cc: Robin Murphy <robin.murphy@arm.com>
Cc: Steve Capper <steve.capper@arm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Wei Yang <richard.weiyang@gmail.com>
Cc: Wei Yang <richardw.yang@linux.intel.com>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Yu Zhao <yuzhao@google.com>
Cc: <stable@vger.kernel.org>	[5.0+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

We might use the nid of memmaps that were never initialized.  For
example, if the memmap was poisoned, we will crash the kernel in
pfn_to_nid() right now.  Let's use the calculated boundaries of the
separate zones instead.  This now also avoids having to iterate over a
whole bunch of subsections again, after shrinking one zone.

Before commit d0dc12e8 ("mm/memory_hotplug: optimize memory
hotplug"), the memmap was initialized to 0 and the node was set to the
right value.  After that commit, the node might be garbage.

We'll have to fix shrink_zone_span() next.

Link: http://lkml.kernel.org/r/20191006085646.5768-4-david@redhat.com
Fixes: f1dd2cd1 ("mm, memory_hotplug: do not associate hotadded memory to zones until online")	[d0dc12e8

]
Signed-off-by: David Hildenbrand <david@redhat.com>
Reported-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Cc: Oscar Salvador <osalvador@suse.de>
Cc: David Hildenbrand <david@redhat.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Pavel Tatashin <pasha.tatashin@soleen.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Wei Yang <richardw.yang@linux.intel.com>
Cc: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Anshuman Khandual <anshuman.khandual@arm.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: Christophe Leroy <christophe.leroy@c-s.fr>
Cc: Damian Tometzki <damian.tometzki@gmail.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Halil Pasic <pasic@linux.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Ira Weiny <ira.weiny@intel.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Jun Yao <yaojun8558363@gmail.com>
Cc: Logan Gunthorpe <logang@deltatee.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Masahiro Yamada <yamada.masahiro@socionext.com>
Cc: "Matthew Wilcox (Oracle)" <willy@infradead.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Pankaj Gupta <pagupta@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Pavel Tatashin <pavel.tatashin@microsoft.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Qian Cai <cai@lca.pw>
Cc: Rich Felker <dalias@libc.org>
Cc: Robin Murphy <robin.murphy@arm.com>
Cc: Steve Capper <steve.capper@arm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Vasily Gorbik <gor@linux.ibm.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Wei Yang <richard.weiyang@gmail.com>
Cc: Will Deacon <will@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Yu Zhao <yuzhao@google.com>
Cc: <stable@vger.kernel.org>	[4.13+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Uninitialized memmaps contain garbage and in the worst case trigger
kernel BUGs, especially with CONFIG_PAGE_POISONING.  They should not get
touched.

For example, when not onlining a memory block that is spanned by a zone
and reading /proc/pagetypeinfo with CONFIG_DEBUG_VM_PGFLAGS and
CONFIG_PAGE_POISONING, we can trigger a kernel BUG:

  :/# echo 1 > /sys/devices/system/memory/memory40/online
  :/# echo 1 > /sys/devices/system/memory/memory42/online
  :/# cat /proc/pagetypeinfo > test.file
   page:fffff2c585200000 is uninitialized and poisoned
   raw: ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff
   raw: ffffffffffffffff ffffffffffffffff ffffffffffffffff ffffffffffffffff
   page dumped because: VM_BUG_ON_PAGE(PagePoisoned(p))
   There is not page extension available.
   ------------[ cut here ]------------
   kernel BUG at include/linux/mm.h:1107!
   invalid opcode: 0000 [#1] SMP NOPTI

Please note th...

When CONFIG_PRINTK_CALLER is set, struct printk_log contains an
additional member caller_id.  This affects the offset of the log text.
Account for this by using the type information from gdb to determine all
the offsets instead of using hardcoded values.

This fixes following error:

  (gdb) lx-dmesg
  Python Exception <class 'ValueError'> embedded null character:
  Error occurred in Python command: embedded null character

The read_u* utility functions now take an offset argument to make them
easier to use.

Link: http://lkml.kernel.org/r/20191011142500.2339-1-joel.colledge@linbit.com


Signed-off-by: Joel Colledge <joel.colledge@linbit.com>
Reviewed-by: Jan Kiszka <jan.kiszka@siemens.com>
Cc: Kieran Bingham <kbingham@kernel.org>
Cc: Leonard Crestez <leonard.crestez@nxp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

We should check for pfn_to_online_page() to not access uninitialized
memmaps.  Reshuffle the code so we don't have to duplicate the error
message.

Link: http://lkml.kernel.org/r/20191009142435.3975-3-david@redhat.com


Signed-off-by: David Hildenbrand <david@redhat.com>
Fixes: f1dd2cd1 ("mm, memory_hotplug: do not associate hotadded memory to zones until online")	[visible after d0dc12e8

]
Acked-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: <stable@vger.kernel.org>	[4.13+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

There are three places where we access uninitialized memmaps, namely:
- /proc/kpagecount
- /proc/kpageflags
- /proc/kpagecgroup

We have initialized memmaps either when the section is online or when the
page was initialized to the ZONE_DEVICE.  Uninitialized memmaps contain
garbage and in the worst case trigger kernel BUGs, especially with
CONFIG_PAGE_POISONING.

For example, not onlining a DIMM during boot and calling /proc/kpagecount
with CONFIG_PAGE_POISONING:

  :/# cat /proc/kpagecount > tmp.test
  BUG: unable to handle page fault for address: fffffffffffffffe
  #PF: supervisor read access in kernel mode
  #PF: error_code(0x0000) - not-present page
  PGD 114616067 P4D 114616067 PUD 114618067 PMD 0
  Oops: 0000 [#1] SMP NOPTI
  CPU: 0 PID: 469 Comm: cat Not tainted 5.4.0-rc1-next-20191004+ #11
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.4
  RIP: 0010:kpagecount_read+0xce/0x1e0
  Code: e8 09 83 e0 3f 48 0f a3 02 73 2d 4c 89 e7 48 c1 e7 06 48 03 3d ab 51 01 01 74 1d 48 8b 57 08 480
  RSP: 0018:ffffa14e409b7e78 EFLAGS: 00010202
  RAX: fffffffffffffffe RBX: 0000000000020000 RCX: 0000000000000000
  RDX: 0000000000000001 RSI: 00007f76b5595000 RDI: fffff35645000000
  RBP: 00007f76b5595000 R08: 0000000000000001 R09: 0000000000000000
  R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000140000
  R13: 0000000000020000 R14: 00007f76b5595000 R15: ffffa14e409b7f08
  FS:  00007f76b577d580(0000) GS:ffff8f41bd400000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: fffffffffffffffe CR3: 0000000078960000 CR4: 00000000000006f0
  Call Trace:
   proc_reg_read+0x3c/0x60
   vfs_read+0xc5/0x180
   ksys_read+0x68/0xe0
   do_syscall_64+0x5c/0xa0
   entry_SYSCALL_64_after_hwframe+0x49/0xbe

For now, let's drop support for ZONE_DEVICE from the three pseudo files
in order to fix this.  To distinguish offline memory (with garbage
memmap) from ZONE_DEVICE memory with properly initialized memmaps, we
would have to check get_dev_pagemap() and pfn_zone_device_reserved()
right now.  The usage of both (especially, special casing devmem) is
frowned upon and needs to be reworked.

The fundamental issue we have is:

	if (pfn_to_online_page(pfn)) {
		/* memmap initialized */
	} else if (pfn_valid(pfn)) {
		/*
		 * ???
		 * a) offline memory. memmap garbage.
		 * b) devmem: memmap initialized to ZONE_DEVICE.
		 * c) devmem: reserved for driver. memmap garbage.
		 * (d) devmem: memmap currently initializing - garbage)
		 */
	}

We'll leave the pfn_zone_device_reserved() check in stable_page_flags()
in place as that function is also used from memory failure.  We now no
longer dump information about pages that are not in use anymore -
offline.

Link: http://lkml.kernel.org/r/20191009142435.3975-2-david@redhat.com
Fixes: f1dd2cd1 ("mm, memory_hotplug: do not associate hotadded memory to zones until online")	[visible after d0dc12e8

]
Signed-off-by: David Hildenbrand <david@redhat.com>
Reported-by: Qian Cai <cai@lca.pw>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Toshiki Fukasawa <t-fukasawa@vx.jp.nec.com>
Cc: Pankaj gupta <pagupta@redhat.com>
Cc: Mike Rapoport <rppt@linux.vnet.ibm.com>
Cc: Anthony Yznaga <anthony.yznaga@oracle.com>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>
Cc: <stable@vger.kernel.org>	[4.13+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Uninitialized memmaps contain garbage and in the worst case trigger kernel
BUGs, especially with CONFIG_PAGE_POISONING.  They should not get touched.

Right now, when trying to soft-offline a PFN that resides on a memory
block that was never onlined, one gets a misleading error with
CONFIG_PAGE_POISONING:

  :/# echo 5637144576 > /sys/devices/system/memory/soft_offline_page
  [   23.097167] soft offline: 0x150000 page already poisoned

But the actual result depends on the garbage in the memmap.

soft_offline_page() can only work with online pages, it returns -EIO in
case of ZONE_DEVICE.  Make sure to only forward pages that are online
(iow, managed by the buddy) and, therefore, have an initialized memmap.

Add a check against pfn_to_online_page() and similarly return -EIO.

Link: http://lkml.kernel.org/r/20191010141200.8985-1-david@redhat.com
Fixes: f1dd2cd1 ("mm, memory_hotplug: do not associate hotadded memory to zones until online")	[visible after d0dc12e8

]
Signed-off-by: David Hildenbrand <david@redhat.com>
Acked-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>
Cc: <stable@vger.kernel.org>	[4.13+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

This was always meant to be a temporary thing, just for testing and to
see if it actually ever triggered.

The only thing that reported it was syzbot doing disk image fuzzing, and
then that warning is expected.  So let's just remove it before -rc4,
because the extra sanity testing should probably go to -stable, but we
don't want the warning to do so.

Reported-by:  <syzbot+3031f712c7ad5dd4d926@syzkaller.appspotmail.com>
Fixes: 8a23eb80

 ("Make filldir[64]() verify the directory entry filename is valid")
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Pull ceph fixes from Ilya Dryomov:
 "A future-proofing decoding fix from Jeff intended for stable and a
  patch for a mostly benign race from Dongsheng"

* tag 'ceph-for-5.4-rc4' of git://github.com/ceph/ceph-client:
  rbd: cancel lock_dwork if the wait is interrupted
  ceph: just skip unrecognized info in ceph_reply_info_extra

Merge tag 'for-5.4/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm

Pull device mapper fixes from Mike Snitzer:

 - Fix DM snapshot deadlock that can occur due to COW throttling
   preventing locks from being released.

 - Fix DM cache's GFP_NOWAIT allocation failure error paths by switching
   to GFP_NOIO.

 - Make __hash_find() static in the DM clone target.

* tag 'for-5.4/dm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm cache: fix bugs when a GFP_NOWAIT allocation fails
  dm snapshot: rework COW throttling to fix deadlock
  dm snapshot: introduce account_start_copy() and account_end_copy()
  dm clone: Make __hash_find static

Pull iommu fixes from Joerg Roedel:

 - Fixes for page-table issues on Mali GPUs

 - Missing free in an error path for ARM-SMMU

 - PASID decoding in the AMD IOMMU Event log code

 - Another update for the locking fixes in the AMD IOMMU driver

 - Reduce the calls to platform_get_irq() in the IPMMU-VMSA and Rockchip
   IOMMUs to get rid of the warning message added to this function
   recently

* tag 'iommu-fixes-v5.4-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
  iommu/amd: Check PM_LEVEL_SIZE() condition in locked section
  iommu/amd: Fix incorrect PASID decoding from event log
  iommu/ipmmu-vmsa: Only call platform_get_irq() when interrupt is mandatory
  iommu/rockchip: Don't use platform_get_irq to implicitly count irqs
  iommu/io-pgtable-arm: Support all Mali configurations
  iommu/io-pgtable-arm: Correct Mali attributes
  iommu/arm-smmu: Free context bitmap in the err path of arm_smmu_init_domain_context

Merge tag 'copy-struct-from-user-v5.4-rc4' of gitolite.kernel.org:pub/scm/linux/kernel/git/brauner/linux

Pull usercopy test fixlets from Christian Brauner:
 "This contains two improvements for the copy_struct_from_user() tests:

   - a coding style change to get rid of the ugly "if ((ret |= test()))"
     pointed out when pulling the original patchset.

   - avoid a soft lockups when running the usercopy tests on machines
     with large page sizes by scanning only a 1024 byte region"

* tag 'copy-struct-from-user-v5.4-rc4' of gitolite.kernel.org:pub/scm/linux/kernel/git/brauner/linux:
  usercopy: Avoid soft lockups in test_check_nonzero_user()
  lib: test_user_copy: style cleanup

Pull MMC fixes from Ulf Hansson:
 "MMC host:
   - sdhci-iproc: Prevent some spurious interrupts
   - renesas_sdhi/sh_mmcif: Avoid false warnings about IRQs not found

  MEMSTICK host:
   - jmb38x_ms: Fix an error handling path at ->probe()"

* tag 'mmc-v5.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc:
  memstick: jmb38x_ms: Fix an error handling path in 'jmb38x_ms_probe()'
  mmc: sdhci-iproc: fix spurious interrupts on Multiblock reads with bcm2711
  mmc: sh_mmcif: Use platform_get_irq_optional() for optional interrupt
  mmc: renesas_sdhi: Do not use platform_get_irq() to count interrupts

Pull sound fixes from Takashi Iwai:
 "Just a few small fixes for the usual suspect, HD- and USB-audio:
  enablement of runtime PM for Nvidia due to the recent PCI changes, a
  fix for potential hangs with recent HD-audio platforms, and the rest
  device-specific quirks"

* tag 'sound-5.4-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hda - Force runtime PM on Nvidia HDMI codecs
  ALSA: hda/realtek - Enable headset mic on Asus MJ401TA
  ALSA: usb-audio: Disable quirks for BOSS Katana amplifiers
  ALSA: hdac: clear link output stream mapping
  ALSA: hda/realtek: Reduce the Headphone static noise on XPS 9350/9360

Pull ACPI fixes from Rafael Wysocki:
 "Fix possible use-after-free in the ACPI CPPC support code (John Garry)
  and prevent the ACPI HMAT parsing code from using possibly incorrect
  data coming from the platform firmware (Daniel Black)"

* tag 'acpi-5.4-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI: CPPC: Set pcc_data[pcc_ss_id] to NULL in acpi_cppc_processor_exit()
  ACPI: HMAT: ACPI_HMAT_MEMORY_PD_VALID is deprecated since ACPI-6.3

Pull power management fixes from Rafael Wysocki:
 "These include a fix for a recent regression in the ACPI CPU
performance scaling code, a PCI device power management fix,
a system shutdown fix related to cpufreq, a removal of an ACPI
suspend-to-idle blacklist entry and a build warning fix.

Specifics:

   - Fix possible NULL pointer dereference in the ACPI processor scaling
     initialization code introduced by a recent cpufreq update (Rafael
     Wysocki).

   - Fix possible deadlock due to suspending cpufreq too late during
     system shutdown (Rafael Wysocki).

   - Make the PCI device system resume code path be more consistent with
     its PM-runtime counterpart to fix an issue with missing delay on
     transitions from D3cold to D0 during system resume from
     suspend-to-idle on some systems (Rafael Wysocki).

   - Drop Dell XPS13 9360 from the LPS0 Idle _DSM blacklist to make it
     use suspend-to-idle by default (Mario Limonciello).

   - Fix build warning in the core system suspend support code (Ben
     Dooks)"

* tag 'pm-5.4-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI: processor: Avoid NULL pointer dereferences at init time
  PCI: PM: Fix pci_power_up()
  PM: sleep: include <linux/pm_runtime.h> for pm_wq
  cpufreq: Avoid cpufreq_suspend() deadlock on system shutdown
  ACPI: PM: Drop Dell XPS13 9360 from LPS0 Idle _DSM blacklist

Pull scsi fixes from Martin Petersen:
 "These two commits were in a separate postmerge branch due to a
  dependency on changes merged for 5.4 in the block tree.

  They fix two issues in the intersection of the request cleanup changes
  from block (b7e9e1fb) and the request batching changes
  (8930a6c2) that were made to SCSI during the 5.4 cycle"

* tag 'mkp-scsi-postmerge' of git://git.kernel.org/pub/scm/linux/kernel/git/mkp/scsi:
  scsi: core: fix dh and multipathing for SCSI hosts without request batching
  scsi: core: fix missing .cleanup_rq for SCSI hosts without request batching

The increase_address_space() function has to check the PM_LEVEL_SIZE()
condition again under the domain->lock to avoid a false trigger of the
WARN_ON_ONCE() and to avoid that the address space is increase more
often than necessary.

Reported-by: Qian Cai <cai@lca.pw>
Fixes: 754265bc

 ("iommu/amd: Fix race in increase_address_space()")
Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>

* acpi-tables:
  ACPI: HMAT: ACPI_HMAT_MEMORY_PD_VALID is deprecated since ACPI-6.3

When enabling KASAN and DEBUG_TEST_DRIVER_REMOVE, I find this KASAN
warning:

[   20.872057] BUG: KASAN: use-after-free in pcc_data_alloc+0x40/0xb8
[   20.878226] Read of size 4 at addr ffff00236cdeb684 by task swapper/0/1
[   20.884826]
[   20.886309] CPU: 19 PID: 1 Comm: swapper/0 Not tainted 5.4.0-rc1-00009-ge7f7df3db5bf-dirty #289
[   20.894994] Hardware name: Huawei D06 /D06, BIOS Hisilicon D06 UEFI RC0 - V1.16.01 03/15/2019
[   20.903505] Call trace:
[   20.905942]  dump_backtrace+0x0/0x200
[   20.909593]  show_stack+0x14/0x20
[   20.912899]  dump_stack+0xd4/0x130
[   20.916291]  print_address_description.isra.9+0x6c/0x3b8
[   20.921592]  __kasan_report+0x12c/0x23c
[   20.925417]  kasan_report+0xc/0x18
[   20.928808]  __asan_load4+0x94/0xb8
[   20.932286]  pcc_data_alloc+0x40/0xb8
[   20.935938]  acpi_cppc_processor_probe+0x4e8/0xb08
[   20.940717]  __acpi_processor_start+0x48/0xb0
[   20.945062]  acpi_processor_start+0x40/0x60
[   20.949235]  really_probe+0x118/0x548
[   20.952887]  driver_probe_device+0x7c/0x148
[   20.957059]  device_driver_attach+0x94/0xa0
[   20.961231]  __driver_attach+0xa4/0x110
[   20.965055]  bus_for_each_dev+0xe8/0x158
[   20.968966]  driver_attach+0x30/0x40
[   20.972531]  bus_add_driver+0x234/0x2f0
[   20.976356]  driver_register+0xbc/0x1d0
[   20.980182]  acpi_processor_driver_init+0x40/0xe4
[   20.984875]  do_one_initcall+0xb4/0x254
[   20.988700]  kernel_init_freeable+0x24c/0x2f8
[   20.993047]  kernel_init+0x10/0x118
[   20.996524]  ret_from_fork+0x10/0x18
[   21.000087]
[   21.001567] Allocated by task 1:
[   21.004785]  save_stack+0x28/0xc8
[   21.008089]  __kasan_kmalloc.isra.9+0xbc/0xd8
[   21.012435]  kasan_kmalloc+0xc/0x18
[   21.015913]  pcc_data_alloc+0x94/0xb8
[   21.019564]  acpi_cppc_processor_probe+0x4e8/0xb08
[   21.024343]  __acpi_processor_start+0x48/0xb0
[   21.028689]  acpi_processor_start+0x40/0x60
[   21.032860]  really_probe+0x118/0x548
[   21.036512]  driver_probe_device+0x7c/0x148
[   21.040684]  device_driver_attach+0x94/0xa0
[   21.044855]  __driver_attach+0xa4/0x110
[   21.048680]  bus_for_each_dev+0xe8/0x158
[   21.052591]  driver_attach+0x30/0x40
[   21.056155]  bus_add_driver+0x234/0x2f0
[   21.059980]  driver_register+0xbc/0x1d0
[   21.063805]  acpi_processor_driver_init+0x40/0xe4
[   21.068497]  do_one_initcall+0xb4/0x254
[   21.072322]  kernel_init_freeable+0x24c/0x2f8
[   21.076667]  kernel_init+0x10/0x118
[   21.080144]  ret_from_fork+0x10/0x18
[   21.083707]
[   21.085186] Freed by task 1:
[   21.088056]  save_stack+0x28/0xc8
[   21.091360]  __kasan_slab_free+0x118/0x180
[   21.095445]  kasan_slab_free+0x10/0x18
[   21.099183]  kfree+0x80/0x268
[   21.102139]  acpi_cppc_processor_exit+0x1a8/0x1b8
[   21.106832]  acpi_processor_stop+0x70/0x80
[   21.110917]  really_probe+0x174/0x548
[   21.114568]  driver_probe_device+0x7c/0x148
[   21.118740]  device_driver_attach+0x94/0xa0
[   21.122912]  __driver_attach+0xa4/0x110
[   21.126736]  bus_for_each_dev+0xe8/0x158
[   21.130648]  driver_attach+0x30/0x40
[   21.134212]  bus_add_driver+0x234/0x2f0
[   21.0x10/0x18
[   21.161764]
[   21.163244] The buggy address belongs to the object at ffff00236cdeb600
[   21.163244]  which belongs to the cache kmalloc-256 of size 256
[   21.175750] The buggy address is located 132 bytes inside of
[   21.175750]  256-byte region [ffff00236cdeb600, ffff00236cdeb700)
[   21.187473] The buggy address belongs to the page:
[   21.192254] page:fffffe008d937a00 refcount:1 mapcount:0 mapping:ffff002370c0fa00 index:0x0 compound_mapcount: 0
[   21.202331] flags: 0x1ffff00000010200(slab|head)
[   21.206940] raw: 1ffff00000010200 dead000000000100 dead000000000122 ffff002370c0fa00
[   21.214671] raw: 0000000000000000 00000000802a002a 00000001ffffffff 0000000000000000
[   21.222400] page dumped because: kasan: bad access detected
[   21.227959]
[   21.229438] Memory state around the buggy address:
[   21.234218]  ffff00236cdeb580: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[   21.241427]  ffff00236cdeb600: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
[   21.248637] >ffff00236cdeb680: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
[   21.255845]                    ^
[   21.259062]  ffff00236cdeb700: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[   21.266272]  ffff00236cdeb780: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
[   21.273480] ==================================================================

It seems that global pcc_data[pcc_ss_id] can be freed in
acpi_cppc_processor_exit(), but we may later reference this value, so
NULLify it when freed.

Also remove the useless setting of data "pcc_channel_acquired", which
we're about to free.

Fixes: 85b1407b

 ("ACPI / CPPC: Make CPPC ACPI driver aware of PCC subspace IDs")
Signed-off-by: John Garry <john.garry@huawei.com>
Cc: 4.15+ <stable@vger.kernel.org> # 4.15+
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>

* pm-cpufreq:
  ACPI: processor: Avoid NULL pointer dereferences at init time
  cpufreq: Avoid cpufreq_suspend() deadlock on system shutdown

* pm-sleep:
  PM: sleep: include <linux/pm_runtime.h> for pm_wq
  ACPI: PM: Drop Dell XPS13 9360 from LPS0 Idle _DSM blacklist

Pull arm64 fixes from Will Deacon:
 "The main thing here is a long-awaited workaround for a CPU erratum on
  ThunderX2 which we have developed in conjunction with engineers from
  Cavium/Marvell.

  At the moment, the workaround is unconditionally enabled for affected
  CPUs at runtime but we may add a command-line option to disable it in
  future if performance numbers show up indicating a significant cost
  for real workloads.

  Summary:

   - Work around Cavium/Marvell ThunderX2 erratum #219

   - Fix regression in mlock() ABI caused by sign-extension of TTBR1 addresses

   - More fixes to the spurious kernel fault detection logic

   - Fix pathological preemption race when enabling some CPU features at boot

   - Drop broken kcore macros in favour of generic implementations

   - Fix userspace view of ID_AA64ZFR0_EL1 when SVE is disabled

   - Avoid NULL dereference on allocation failure during hibernation"

* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
  arm64: tags: Preserve tags for addresses translated via TTBR1
  arm64: mm: fix inverted PAR_EL1.F check
  arm64: sysreg: fix incorrect definition of SYS_PAR_EL1_F
  arm64: entry.S: Do not preempt from IRQ before all cpufeatures are enabled
  arm64: hibernate: check pgd table allocation
  arm64: cpufeature: Treat ID_AA64ZFR0_EL1 as RAZ when SVE is not enabled
  arm64: Fix kcore macros after 52-bit virtual addressing fallout
  arm64: Allow CAVIUM_TX2_ERRATUM_219 to be selected
  arm64: Avoid Cavium TX2 erratum 219 when switching TTBR
  arm64: Enable workaround for Cavium TX2 erratum 219 when running SMT
  arm64: KVM: Trap VM ops when ARM64_WORKAROUND_CAVIUM_TX2_219_TVM is set

Pull Xtensa fixes from Max Filippov:

 - fix {get,put}_user() for 64bit values

 - fix warning about static EXPORT_SYMBOL from modpost

 - fix PCI IO ports mapping for the virt board

 - fix pasto in change_bit for exclusive access option

* tag 'xtensa-20191017' of git://github.com/jcmvbkbc/linux-xtensa:
  xtensa: fix change_bit in exclusive access option
  xtensa: virt: fix PCI IO ports mapping
  xtensa: drop EXPORT_SYMBOL for outs*/ins*
  xtensa: fix type conversion in __get_user_[no]check
  xtensa: clean up assembly arguments in uaccess macros
  xtensa: fix {get,put}_user() for 64bit values

Pull xfs fix from Darrick Wong:
 "The single fix converts the seconds field in the recently added XFS
  bulkstat structure to a signed 64-bit quantity.

  The structure layout doesn't change and so far there are no users of
  the ioctl to break because we only publish xfs ioctl interfaces
  through the XFS userspace development libraries, and we're still
  working on a 5.3 release"

* tag 'xfs-5.4-fixes-4' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: change the seconds fields in xfs_bulkstat to signed

Pull drm fixes from Dave Airlie:
 "This is this weeks fixes for drm.

  The dma-resv one is probably the more important one a fair few people
  have reported it, besides that it's a couple of panfrost, a few i915
  and a few amdgpu fixes.

  One radeon patch to fix some ppc64 related issues caused an x86
  regression so is getting reverted for now.

  Summary:

  dma-resv:
   - shared fences for lima/panfrost

  ttm:
   - prefault regression fix
   - lifetime fix

  panfrost:
   - stopped job timeout fix
   - missing register values

  amdgpu:
   - smu7 powerplay fix
   - bail earlier for cik/si detection
   - navi SDMA fix

  radeon:
   - revert a ppc64 shutdown fix that broke x86

  i915:
   - VBT information handling fix
   - Circular locking fix
   - preemption vs resubmission virtual requests fix"

* tag 'drm-fixes-2019-10-18' of git://anongit.freedesktop.org/drm/drm:
  drm/i915: Fixup preempt-to-busy vs resubmission of a virtual request
  drm/i915/userptr: Never allow userptr into the mappable GGTT
  drm/i915: Favor last VBT child device with conflicting AUX ch/DDC pin
  drm/i915/execlists: Refactor -EIO markup of hung requests
  drm/panfrost: Handle resetting on timeout better
  drm/panfrost: Add missing GPU feature registers
  drm/ttm: fix handling in ttm_bo_add_mem_to_lru
  drm/ttm: Restore ttm prefaulting
  drm/ttm: fix busy reference in ttm_mem_evict_first
  drm/amdgpu/sdma5: fix mask value of POLL_REGMEM packet for pipe sync
  drm/amdgpu: Bail earlier when amdgpu.cik_/si_support is not set to 1
  Revert "drm/radeon: Fix EEH during kexec"
  drm/msm/dsi: Implement reset correctly
  dma-buf/resv: fix exclusive fence get
  drm/edid: Add 6 bpc quirk for SDC panel in Lenovo G50
  drm/tiny: Kconfig: Remove always-y THERMAL dep. from TINYDRM_REPAPER
  drm/amdgpu/powerplay: fix typo in mvdd table setup

Workaround for Cavium/Marvell ThunderX2 erratum #219.

* errata/tx2-219:
  arm64: Allow CAVIUM_TX2_ERRATUM_219 to be selected
  arm64: Avoid Cavium TX2 erratum 219 when switching TTBR
  arm64: Enable workaround for Cavium TX2 erratum 219 when running SMT
  arm64: KVM: Trap VM ops when ARM64_WORKAROUND_CAVIUM_TX2_219_TVM is set

-dma-resv: Change shared_count to post-increment to fix lima crash (Qiang)
-ttm: A couple fixes related to lifetime and restore prefault behavior
 (Christian & Thomas)
-panfrost: Fill in missing feature reg values and fix stoppedjob timeouts
 (Steven)

Cc: Qiang Yu <yuq825@gmail.com>
Cc: Thomas Hellstrom <thellstrom@vmware.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Steven Price <steven.price@arm.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Sean Paul <sean@poorly.run>
Link: https://patchwork.freedesktop.org/patch/msgid/20191017203419.GA142909@art_vandelay

drm-fixes-5.4-2019-10-16:

amdgpu:
- Powerplay fix for SMU7 parts
- Bail earlier when cik/si support is not set to 1
- Fix an SDMA issue on navi

radeon:
- revert a PPC fix which broken x86

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexdeucher@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191017022443.3853-1-alexander.deucher@amd.com

Merge tag 'drm-intel-fixes-2019-10-17' of git://anongit.freedesktop.org/drm/drm-intel

 into drm-fixes

- Display fix on handling VBT information.
- Important circular locking fix
- Fix for preemption vs resubmission on virtual requests
  - and a prep patch to make this last one to apply cleanly

Signed-off-by: Dave Airlie <airlied@redhat.com>

From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191017135444.GA12255@intel.com

Pull input fixes from Dmitry Torokhov:
 "The main change is that we are reverting blanket enablement of SMBus
  mode for devices with Elan touchpads that report BIOS release date as
  2018+ because there are older boxes with updated BIOSes that still do
  not work well in SMbus mode.

  We will have to establish whitelist for SMBus mode it looks like"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
  Revert "Input: elantech - enable SMBus on new (2018+) systems"
  Input: synaptics-rmi4 - avoid processing unknown IRQs
  Input: soc_button_array - partial revert of support for newer surface devices
  Input: goodix - add support for 9-bytes reports
  Input: da9063 - fix capability and drop KEY_SLEEP

While it is useful for new drivers to use devm_platform_ioremap_resource,
this script is currently used to spam maintainers, often updating very
old drivers.  The net benefit is the removal of 2 lines of code in the
driver but the review load for the maintainers is huge.  As of now, more
that 560 patches have been sent, some of them obviously broken, as in:

 https://lore.kernel.org/lkml/9bbcce19c777583815c92ce3c2ff2586@www.loen.fr/



Remove the script to reduce the spam.

Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Acked-by: Julia Lawall <Julia.Lawall@lip6.fr>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Przemysław Kopa reports that since commit b516ea58 ("PCI: Enable
NVIDIA HDA controllers"), the discrete GPU Nvidia GeForce GT 540M on his
2011 Samsung laptop refuses to runtime suspend, resulting in a power
regression and excessive heat.

Rivera Valdez witnesses the same issue with a GeForce GT 525M (GF108M)
of the same era, as does another Arch Linux user named "R0AR" with a
more recent GeForce GTX 1050 Ti (GP107M).

The commit exposes the discrete GPU's HDA controller and all four codecs
on the controller do not set the CLKSTOP and EPSS bits in the Supported
Power States Response.  They also do not set the PS-ClkStopOk bit in the
Get Power State Response.  hda_codec_runtime_suspend() therefore does
not call snd_hdac_codec_link_down(), which prevents each codec and the
PCI device from runtime suspending.

The same issue is present on some AMD discrete GPUs and we addressed it
by forcing runtime PM despite the bits not being set, see commit
57cb54e5 ("ALSA: hda - Force to link down at runtime suspend on
ATI/AMD HDMI").

Do the same for Nvidia HDMI codecs.

Fixes: b516ea58 ("PCI: Enable NVIDIA HDA controllers")
Link: https://bbs.archlinux.org/viewtopic.php?pid=1865512
Link: https://bugs.freedesktop.org/show_bug.cgi?id=75985#c81


Reported-by: Przemysław Kopa <prymoo@gmail.com>
Reported-by: Rivera Valdez <riveravaldez@ysinembargo.com>
Signed-off-by: Lukas Wunner <lukas@wunner.de>
Cc: Daniel Drake <dan@reactivated.net>
Cc: stable@vger.kernel.org # v5.3+
Link: https://lore.kernel.org/r/3086bc75135c1e3567c5bc4f3cc4ff5cbf7a56c2.1571324194.git.lukas@wunner.de


Signed-off-by: Takashi Iwai <tiwai@suse.de>

Pull x86 platform driver fixes from Andy Shevchenko:

 - Users of Intel P-Unit IPC driver might be surprised by harmless
   warning. Thus, switch to API which doesn't issue a warning at all.

 - I²C multi-instantiate driver continues to add slave devices even when
   IRQ resource is not found. For devices in the market IRQ resource is
   mandatory, so, fail the ->probe() of the parent driver to avoid
   slaves being probed.

 - Avoid compiler warning due to unused variable in Classmate laptop
   driver.

* tag 'platform-drivers-x86-v5.4-3' of git://git.infradead.org/linux-platform-drivers-x86:
  platform/x86: i2c-multi-instantiate: Fail the probe if no IRQ provided
  platform/x86: intel_punit_ipc: Avoid error message when retrieving IRQ
  platform/x86: classmate-laptop: remove unused variable

GFP_NOWAIT allocation can fail anytime - it doesn't wait for memory being
available and it fails if the mempool is exhausted and there is not enough
memory.

If we go down this path:
  map_bio -> mg_start -> alloc_migration -> mempool_alloc(GFP_NOWAIT)
we can see that map_bio() doesn't check the return value of mg_start(),
and the bio is leaked.

If we go down this path:
  map_bio -> mg_start -> mg_lock_writes -> alloc_prison_cell ->
  dm_bio_prison_alloc_cell_v2 -> mempool_alloc(GFP_NOWAIT) ->
  mg_lock_writes -> mg_complete
the bio is ended with an error - it is unacceptable because it could
cause filesystem corruption if the machine ran out of memory
temporarily.

Change GFP_NOWAIT to GFP_NOIO, so that the mempool code will properly
wait until memory becomes available. mempool_alloc with GFP_NOIO can't
fail, so remove the code paths that deal with allocation failure.

Cc: stable@vger.kernel.org
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>

Pull GPIO fixes from Linus Walleij:
 "The fixes pertain to a problem with initializing the Intel GPIO
  irqchips when adding gpiochips.

  Andy fixed it up elegantly by adding a hardware initialization
  callback to the struct gpio_irq_chip so let's use this. Tested and
  verified on the target hardware"

* tag 'gpio-v5.4-3' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio:
  gpio: lynxpoint: set default handler to be handle_bad_irq()
  gpio: merrifield: Move hardware initialization to callback
  gpio: lynxpoint: Move hardware initialization to callback
  gpio: intel-mid: Move hardware initialization to callback
  gpiolib: Initialize the hardware with a callback
  gpio: merrifield: Restore use of irq_base

On Asus MJ401TA (with Realtek ALC256), the headset mic is connected to
pin 0x19, with default configuration value 0x411111f0 (indicating no
physical connection).

Enable this by quirking the pin. Mic jack detection was also tested and
found to be working.

This enables use of the headset mic on this product.

Signed-off-by: Daniel Drake <drake@endlessm.com>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20191017081501.17135-1-drake@endlessm.com


Signed-off-by: Takashi Iwai <tiwai@suse.de>

BOSS Katana amplifiers cannot be used for recording or playback if quirks
are applied

BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=195223


Signed-off-by: Szabolcs Szőke <szszoke.code@gmail.com>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20191011171937.8013-1-szszoke.code@gmail.com


Signed-off-by: Takashi Iwai <tiwai@suse.de>

As preempt-to-busy leaves the request on the HW as the resubmission is
processed, that request may complete in the background and even cause a
second virtual request to enter queue. This second virtual request
breaks our "single request in the virtual pipeline" assumptions.
Furthermore, as the virtual request may be completed and retired, we
lose the reference the virtual engine assumes is held. Normally, just
removing the request from the scheduler queue removes it from the
engine, but the virtual engine keeps track of its singleton request via
its ve->request. This pointer needs protecting with a reference.

v2: Drop unnecessary motion of rq->engine = owner

Fixes: 22b7a426

 ("drm/i915/execlists: Preempt-to-busy")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190923152844.8914-1-chris@chris-wilson.co.uk
(cherry picked from commit b647c7df

)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

Daniel Vetter uncovered a nasty cycle in using the mmu-notifiers to
invalidate userptr objects which also happen to be pulled into GGTT
mmaps. That is when we unbind the userptr object (on mmu invalidation),
we revoke all CPU mmaps, which may then recurse into mmu invalidation.

We looked for ways of breaking the cycle, but the revocation on
invalidation is required and cannot be avoided. The only solution we
could see was to not allow such GGTT bindings of userptr objects in the
first place. In practice, no one really wants to use a GGTT mmapping of
a CPU pointer...

Just before Daniel's explosive lockdep patches land in v5.4-rc1, we got
a genuine blip from CI:

<4>[  246.793958] ======================================================
<4>[  246.793972] WARNING: possible circular locking dependency detected
<4>[  246.793989] 5.3.0-gbd6c56f50d15-drmtip_372+ #1 Tainted: G     U
<4>[  246.794003] ------------------------------------------------------
<4>[  246.794017] kswapd0/145 is trying to acquire lock:
<4>[  246.794030] 000000003f565be6 (&dev->struct_mutex/1){+.+.}, at: userptr_mn_invalidate_range_start+0x18f/0x220 [i915]
<4>[  246.794250]
                  but task is already holding lock:
<4>[  246.794263] 000000001799cef9 (&anon_vma->rwsem){++++}, at: page_lock_anon_vma_read+0xe6/0x2a0
<4>[  246.794291]
                  which lock already depends on the new lock.

<4>[  246.794307]
                  the existing dependency chain (in reverse order) is:
<4>[  246.794322]
                  -> #3 (&anon_vma->rwsem){++++}:
<4>[  246.794344]        down_write+0x33/0x70
<4>[  246.794357]        __vma_adjust+0x3d9/0x7b0
<4>[  246.794370]        __split_vma+0x16a/0x180
<4>[  246.794385]        mprotect_fixup+0x2a5/0x320
<4>[  246.794399]        do_mprotect_pkey+0x208/0x2e0
<4>[  246.794413]        __x64_sys_mprotect+0x16/0x20
<4>[  246.794429]        do_syscall_64+0x55/0x1c0
<4>[  246.794443]        entry_SYSCALL_64_after_hwframe+0x49/0xbe
<4>[  246.794456]
                  -> #2 (&mapping->i_mmap_rwsem){++++}:
<4>[  246.794478]        down_write+0x33/0x70
<4>[  246.794493]        unmap_mapping_pages+0x48/0x130
<4>[  246.794519]        i915_vma_revoke_mmap+0x81/0x1b0 [i915]
<4>[  246.794519]        i915_vma_unbind+0x11d/0x4a0 [i915]
<4>[  246.794519]        i915_vma_destroy+0x31/0x300 [i915]
<4>[  246.794519]        __i915_gem_free_objects+0xb8/0x4b0 [i915]
<4>[  246.794519]        drm_file_free.part.0+0x1e6/0x290
<4>[  246.794519]        drm_release+0xa6/0xe0
<4>[  246.794519]        __fput+0xc2/0x250
<4>[  246.794519]        task_work_run+0x82/0xb0
<4>[  246.794519]        do_exit+0x35b/0xdb0
<4>[  246.794519]        do_group_exit+0x34/0xb0
<4>[  246.794519]        __x64_sys_exit_group+0xf/0x10
<4>[  246.794519]        do_syscall_64+0x55/0x1c0
<4>[  246.794519]        entry_SYSCALL_64_after_hwframe+0x49/0xbe
<4>[  246.794519]
                  -> #1 (&vm->mutex){+.+.}:
<4>[  246.794519]        i915_gem_shrinker_taints_mutex+0x6d/0xe0 [i915]
<4>[  246.794519]        i915_address_space_init+0x9f/0x160 [i915]
<4>[  246.794519]        i915_ggtt_init_hw+0x55/0x170 [i915]
<4>[  246.794519]        i915_driver_probe+0xc9f/0x1620 [i915]
<4>[  246.794519]        i915_pci_probe+0x43/0x1b0 [i915]
<4>[  246.794519]        pci_device_probe+0x9e/0x120
<4>[  246.794519]        really_probe+0xea/0x3d0
<4>[  246.794519]        driver_probe_device+0x10b/0x120
<4>[  246.794519]        device_driver_attach+0x4a/0x50
<4>[  246.794519]        __driver_attach+0x97/0x130
<4>[  246.794519]        bus_for_each_dev+0x74/0xc0
<4>[  246.794519]        bus_add_driver+0x13f/0x210
<4>[  246.794519]        driver_register+0x56/0xe0
<4>[  246.794519]        do_one_initcall+0x58/0x300
<4>[  246.794519]        do_init_module+0x56/0x1f6
<4>[  246.794519]        load_module+0x25bd/0x2a40
<4>[  246.794519]        __se_sys_finit_module+0xd3/0xf0
<4>[  246.794519]        do_syscall_64+0x55/0x1c0
<4>[  246.794519]        entry_SYSCALL_64_after_hwframe+0x49/0xbe
<4>[  246.794519]
                  -> #0 (&dev->struct_mutex/1){+.+.}:
<4>[  246.794519]        __lock_acquire+0x15d8/0x1e90
<4>[  246.794519]        lock_acquire+0xa6/0x1c0
<4>[  246.794519]        __mutex_lock+0x9d/0x9b0
<4>[  246.794519]        userptr_mn_invalidate_range_start+0x18f/0x220 [i915]
<4>[  246.794519]        __mmu_notifier_invalidate_range_start+0x85/0x110
<4>[  246.794519]        try_to_unmap_one+0x76b/0x860
<4>[  246.794519]        rmap_walk_anon+0x104/0x280
<4>[  246.794519]        try_to_unmap+0xc0/0xf0
<4>[  246.794519]        shrink_page_list+0x561/0xc10
<4>[  246.794519]        shrink_inactive_list+0x220/0x440
<4>[  246.794519]        shrink_node_memcg+0x36e/0x740
<4>[  246.794519]        shrink_node+0xcb/0x490
<4>[  246.794519]        balance_pgdat+0x241/0x580
<4>[  246.794519]        kswapd+0x16c/0x530
<4>[  246.794519]        kthread+0x119/0x130
<4>[  246.794519]        ret_from_fork+0x24/0x50
<4>[  246.794519]
                  other info that might help us debug this:

<4>[  246.794519] Chain exists of:
                    &dev->struct_mutex/1 --> &mapping->i_mmap_rwsem --> &anon_vma->rwsem

<4>[  246.794519]  Possible unsafe locking scenario:

<4>[  246.794519]        CPU0                    CPU1
<4>[  246.794519]        ----                    ----
<4>[  246.794519]   lock(&anon_vma->rwsem);
<4>[  246.794519]                                lock(&mapping->i_mmap_rwsem);
<4>[  246.794519]                                lock(&anon_vma->rwsem);
<4>[  246.794519]   lock(&dev->struct_mutex/1);
<4>[  246.794519]
                   *** DEADLOCK ***

v2: Say no to mmap_ioctl

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111744
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111870


Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: stable@vger.kernel.org
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190928082546.3473-1-chris@chris-wilson.co.uk
(cherry picked from commit a4311745

)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

The first come first served apporoach to handling the VBT
child device AUX ch conflicts has backfired. We have machines
in the wild where the VBT specifies both port A eDP and
port E DP (in that order) with port E being the real one.

So let's try to flip the preference around and let the last
child device win once again.

Cc: stable@vger.kernel.org
Cc: Jani Nikula <jani.nikula@intel.com>
Tested-by: Masami Ichikawa <masami256@gmail.com>
Tested-by: Torsten <freedesktop201910@liggy.de>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=111966
Fixes: 36a0f920

 ("drm/i915/bios: make child device order the priority order")
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191011202030.8829-1-ville.syrjala@linux.intel.com


Acked-by: Jani Nikula <jani.nikula@intel.com>
(cherry picked from commit 41e35ffb

)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>