Implement return_address() and use it instead of __builtin_return_address(n).

__builtin_return_address(n) is not guaranteed to work for n > 0,
therefore implement a private return_address() function which walks
the stack frames and returns the proper return address.

This way we get also rid of a compile warning which gcc 6.1 emits and
look like all other architectures.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Since commit 3b9d6da6

 ("cpu/hotplug: Fix rollback during error-out
in __cpu_disable()") it is ensured that callbacks of CPU_ONLINE and
CPU_DOWN_PREPARE are processed on the hotplugged CPU. Due to this SMP
function calls are no longer required.

Replace smp_call_function_single() with a direct call of
setup_pmc_cpu(). To keep the calling convention, interrupts are
explicitly disabled around the call.

Cc: linux-s390@vger.kernel.org
Signed-off-by: Anna-Maria Gleixner <anna-maria@linutronix.de>
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Since commit 3b9d6da6

 ("cpu/hotplug: Fix rollback during error-out
in __cpu_disable()") it is ensured that callbacks of CPU_ONLINE and
CPU_DOWN_PREPARE are processed on the hotplugged CPU. Due to this SMP
function calls are no longer required.

Replace smp_call_function_single() with a direct call of
setup_pmc_cpu(). To keep the calling convention, interrupts are
explicitly disabled around the call.

Cc: linux-s390@vger.kernel.org
Signed-off-by: Anna-Maria Gleixner <anna-maria@linutronix.de>
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

The current default processor type is z900. The BPF jit compiler
depends on PACK_STACK && HAVE_MARCH_Z196_FEATURES. To have the
BPF jit code included in compiles with 'make allmodconfig' set
the default processor type to z196.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Fix the initialization of a local variable to remove this warning:

All warnings (new ones prefixed by >>):

   drivers/s390/char/sclp_pci.c: In function 'sclp_pci_report':
>> drivers/s390/char/sclp_pci.c:127:9: warning: missing braces around initializer [-Wmissing-braces]
     struct sclp_req req = {0};
            ^
   drivers/s390/char/sclp_pci.c:127:9: warning: (near initialization for 'req.list') [-Wmissing-braces]

Fixes: 12283a40

 "s390/sclp: add error notification command"
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Analog to git commit 0c8c0f03


"x86/fpu, sched: Dynamically allocate 'struct fpu'"
move the struct fpu to the end of the struct thread_struct,
set CONFIG_ARCH_WANTS_DYNAMIC_TASK_STRUCT and add the
setup_task_size() function to calculate the correct size
fo the task struct.

For the performance_defconfig this increases the size of
struct task_struct from 7424 bytes to 7936 bytes (MACHINE_HAS_VX==1)
or 7552 bytes (MACHINE_HAS_VX==0). The dynamic allocation of the
struct fpu is removed. The slab cache uses an 8KB block for the
task struct in all cases, there is enough room for the struct fpu.
For MACHINE_HAS_VX==1 each task now needs 512 bytes less memory.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

The CPACF instructions are going be used in KVM as well, move the
defines and the inline functions from arch/s390/crypt/crypt_s390.h
to arch/s390/include/asm. Rename the header to cpacf.h and replace
the crypt_s390_xxx names with cpacf_xxx.

While we are at it, cleanup the header as well. The encoding for
the CPACF operations is odd, there is an enum for each of the CPACF
instructions with the hardware function code in the lower 8 bits of
each entry and a software defined number for the CPACF instruction
in the upper 8 bits. Remove the superfluous software number and
replace the enums with simple defines.

The crypt_s390_func_available() function tests for the presence
of a specific CPACF operations. The new name of the function is
cpacf_query and it works slightly different than before. It gets
passed an opcode of an CPACF instruction and a function code for
this instruction. The facility_mask parameter is gone, the opcode
is used to find the correct MSA facility bit to check if the CPACF
instruction itself is available. If it is the query function of the
given instruction is used to test if the requested CPACF operation
is present.

Acked-by: David Hildenbrand <dahi@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Implement the function type specific function measurement block used
in new machines.

Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Implement new DASD IOCTL BIODASDCHECKFMT to check a range of tracks on a
DASD volume for correct formatting. The following characteristics are
checked:
- Block size
- ECKD key length
- ECKD record ID
- Number of records per track

Signed-off-by: Jan Höppner <hoeppner@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Sort the sclp event type defines and use a macro to create the
corresponding event type masks. In addition to that one unused
type/mask pair is removed and another previously unused define
is used now (it was probably unused/unknown because it didn't
follow the EVTYP_X EVTYP_X_MASK convention).

Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Provide an report_error attribute to send an adapter-error
notification associated with a PCI function.

Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Add SCLP event 24 "Adapter-error notification".

Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

sclp commands only used by the PCI code shouldn't be build for
!CONFIG_PCI. Instead of adding more ifdefs to sclp_cmd.c just
move them to a new file.

Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Commit 6c2ff84e979a ("s390: add SMT support") added the nosmt and smt
early kernel parameters for restricting the use of symmetric
multithreading (SMT). They are quite useful for debugging and testing,
so document them.

Signed-off-by: Sascha Silbe <silbe@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

While looking at set_task_state() users I stumbled over the s390 pfault
interrupt code.  Since Heiko provided a great explanation on how it
worked, I figured we ought to preserve this.

Also make a few little tweaks to the code to aid in readability and
explicitly comment the unusual blocking scheme.

Based-on-text-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

With this feature, applications can query if a DASD volume is online
to another operating system instances by checking the online status of
all attached hosts from the storage server.

Reviewed-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Reviewed-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Stefan Haberland <stefan.haberland@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Make sure that s390 appears to be a big endian machine by defining
this config option.

Without this s390 appears to be little endian as seen by e.g. the
recordmount script: "perl ./scripts/recordmcount.pl "s390" "little"
"64""
This has no practical impact within the script since the endian
variable is only evaluated for mips. However there are already a
couple of common code places which evaluate this config option. None
of them is relevant for s390 currently though.

To avoid any issues in the future (and fix the recordmcount oddity)
add the new config option.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

arch_spin_lock_wait_flags() checks if a spinlock is not held before
trying a compare and swap instruction. If the lock is unlocked it
tries the compare and swap instruction, however if a different cpu
grabbed the lock in the meantime the instruction will fail as
expected.

Subsequently the arch_spin_lock_wait_flags() incorrectly tries to
figure out if the cpu that holds the lock is running. However it is
using the wrong cpu number for this (-1) and then will also yield the
current cpu to the wrong cpu.

Fix this by adding a missing continue statement.

Fixes: 470ada6b

 ("s390/spinlock: refactor arch_spin_lock_wait[_flags]")
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

dcssblk_remove_store() holds the dcssblk_devices_sem semaphore while
calling device_unregister(), which in turn tries to acquire the kernfs
kn->dev_map rwsem for the device sysfs subtree. The same rwsem is also
acquired when using the per-device sysfs attributes in the device sub-tree,
and the attribute handlers then also acquire the dcssblk_devices_sem.

This can lead to a deadlock when removing a DCSS while concurrently
reading from / writing to one of its sysfs attributes. The following
lockdep warning hinted towards the issue (CPU0 = dcssblk_remove_store,
CPU1 = dcssblk_shared_store):

[   76.496047]  Possible unsafe locking scenario:

[   76.496054]        CPU0                    CPU1
[   76.496059]        ----                    ----
[   76.496087]   lock(&dcssblk_devices_sem);
[   76.496090]                                lock(s_active#175);
[   76.496106]                                lock(&dcssblk_devices_sem);
[   76.496110]   lock(s_active#175);
[   76.496115]
 *** DEADLOCK ***

Fix this by releasing the dcssblk_devices_sem semaphore, which only
protects internal DCSS data, before calling device_unregister().

Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Fixes this build error on linux-next:

kernel/seccomp.c: In function '__secure_computing_strict':
kernel/seccomp.c:526:3: error: implicit declaration of function
				'get_compat_mode1_syscalls'

The retrieval of compat syscall numbers were moved into inline function
defined in asm-generic header but the asm-generic header is not being
used by s390.

[heiko.carstens@de.ibm.com]: even though the build error will trigger
only in the next merge window it makes sense to include the generic
header file already now.

Fixes: ("seccomp: Get compat syscalls from asm-generic header")
Cc: Matt Redfearn <matt.redfearn@imgtec.com>
Signed-off-by: Sudip Mukherjee <sudip.mukherjee@codethink.co.uk>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Newer machines might use a different (larger) format for function
measurement blocks. To ensure that we comply with the alignment
requirement on these machines and prevent memory corruption (when
firmware writes more data than we expect) add 16 padding bytes
at the end of the fmb.

Cc: stable@vger.kernel.org # v4.1+
Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

When we refuse a non REQ_TYPE_FS request in the build request function
we already hold the queue lock. Thus we must not call blk_end_request_all
but __blk_end_request_all.

Reported-by: Peter Oberparleiter <oberpar@linux.vnet.ibm.com>
Fixes: de9587a2

 ('s390/scm_blk: fix endless loop for requests != REQ_TYPE_FS')
Signed-off-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

Pull drm fixes from Dave Airlie:
 "Nothing too crazy in here: a bunch of AMD fixes/quirks, two msm fixes,
  some rockchip fixes, and a udl warning fix, along with one locking fix
  for displayport that seems to fix some dodgy monitors"

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
  drm/udl: Use unlocked gem unreferencing
  drm/dp: move hw_mutex up the call stack
  drm/amdgpu: Don't move pinned BOs
  drm/radeon: Don't move pinned BOs
  drm/radeon: add a dpm quirk for all R7 370 parts
  drm/radeon: add another R7 370 quirk
  drm/rockchip: dw_hdmi: Don't call platform_set_drvdata()
  drm/rockchip: vop: Fix vop crtc cleanup
  drm/rockchip: dw_hdmi: Call drm_encoder_cleanup() in error path
  drm/rockchip: vop: Disable planes when disabling CRTC
  drm/rockchip: vop: Don't reject empty modesets
  drm/rockchip: cancel pending vblanks on close
  drm/rockchip: vop: fix crtc size in plane check
  drm/radeon: add a dpm quirk for sapphire Dual-X R7 370 2G D5
  drm/amd: Beef up ACP Kconfig menu text
  drm/msm: fix typo in the !COMMON_CLK case
  drm/msm: fix bug after preclose removal

Pull powerpc fixes from Michael Ellerman:
 - Fixup preempt underflow with huge pages from Sebastian Siewior
 - Fix altivec SPR not being saved from Oliver O'Halloran
 - Correct used_vsr comment from Simon Guo

* tag 'powerpc-4.6-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc: Correct used_vsr comment
  powerpc/process: Fix altivec SPR not being saved
  powerpc/mm: Fixup preempt underflow with huge pages

Pull s390 fixes from Martin Schwidefsky:
 - A proper fix for the locking issue in the dasd driver
 - Wire up the new preadv2 nad pwritev2 system calls
 - Add the mark_rodata_ro function and set DEBUG_RODATA=y
 - A few more bug fixes.

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
  s390: wire up preadv2/pwritev2 syscalls
  s390/pci: PCI function group 0 is valid for clp_query_pci_fn
  s390/crypto: provide correct file mode at device register.
  s390/mm: handle PTE-mapped tail pages in fast gup
  s390: add DEBUG_RODATA support
  s390: disable postinit-readonly for now
  s390/dasd: reorder lcu and device lock
  s390/cpum_sf: Fix cpu hotplug notifier transitions
  s390/cpum_cf: Fix missing cpu hotplug notifier transition

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

The PCI function group 0 is a valid function group,
it is wrong to reject it.

Let's accept PCI function group 0.

Signed-off-by: Pierre Morel <pmorel@linux.vnet.ibm.com>
Acked-by: Sebastian Ott <sebott@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

For drm_gem_object_unreference callers are required to hold
dev->struct_mutex, which these paths don't. Enforcing this requirement
has become a bit more strict with

commit ef4c6270


Author: Daniel Vetter <daniel.vetter@ffwll.ch>
Date:   Thu Oct 15 09:36:25 2015 +0200

    drm/gem: Check locking in drm_gem_object_unreference

Cc: stable@vger.kernel.org
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

1) don't let other threads trying to bang on aux channel interrupt the
defer timeout/logic
2) don't let other threads interrupt the i2c over aux logic

Technically, according to people who actually have the DP spec, this
should not be required.  In practice, it makes some troublesome Dell
monitor (and perhaps others) work, so probably a case of "It's compliant
if it works with windows" on the hw vendor's part..

v2: rebased to come before DPCD/AUX logging patch for easier backport
to stable branches.

Reported-by: Dave Wysochanski <dwysocha@redhat.com>
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1274157


Cc: stable@vger.kernel.org
Signed-off-by: Rob Clark <robdclark@gmail.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>

Merge branch 'drm-rockchip-next-fixes-2016-03-28' of https://github.com/markyzq/kernel-drm-rockchip into drm-fixes

bunch of rockchip fixes.

* 'drm-rockchip-next-fixes-2016-03-28' of https://github.com/markyzq/kernel-drm-rockchip:
  drm/rockchip: dw_hdmi: Don't call platform_set_drvdata()
  drm/rockchip: vop: Fix vop crtc cleanup
  drm/rockchip: dw_hdmi: Call drm_encoder_cleanup() in error path
  drm/rockchip: vop: Disable planes when disabling CRTC
  drm/rockchip: vop: Don't reject empty modesets
  drm/rockchip: cancel pending vblanks on close
  drm/rockchip: vop: fix crtc size in plane check

two minor msm fixes.

* 'msm-fixes-4.6-rc1' of git://people.freedesktop.org/~robclark/linux:
  drm/msm: fix typo in the !COMMON_CLK case
  drm/msm: fix bug after preclose removal

Just a few fixes for 4.6 this week:
- Add some SI DPM quirks
- Improve the ACP Kconfig text
- Additional BO pinning checks

* 'drm-next-4.6' of git://people.freedesktop.org/~agd5f/linux:
  drm/amdgpu: Don't move pinned BOs
  drm/radeon: Don't move pinned BOs
  drm/radeon: add a dpm quirk for all R7 370 parts
  drm/radeon: add another R7 370 quirk
  drm/radeon: add a dpm quirk for sapphire Dual-X R7 370 2G D5
  drm/amd: Beef up ACP Kconfig menu text

Pull parisc fixes from Helge Deller:
 "Fix seccomp filter support and SIGSYS signals on compat kernel.

  Both patches are tagged for v4.5 stable kernel"

* 'parisc-4.6-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
  parisc: Fix and enable seccomp filter support
  parisc: Fix SIGSYS signals in compat case

Pull vfs fix from Al Viro.

Automount handling was broken by commit e3c13928 ("namei: massage
lookup_slow() to be usable by lookup_one_len_unlocked()") moving the
test for negative dentry too early.

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  fix the braino in "namei: massage lookup_slow() to be usable by lookup_one_len_unlocked()"

Pull GPIO fixes from Linus Walleij:
 "Here are two GPIO fixes for the v4.6 series, both in drivers:

   - Prevent NULL dereference in the Xgene driver
   - Fix an uninitialized spinlock in the menz127 driver"

* tag 'gpio-v4.6-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio:
  gpio: xgene: Prevent NULL pointer dereference
  gpio: menz127: Drop lock field from struct men_z127_gpio

Pull nvdimm mcsafe_memcpy use from Dan Williams:
 "Now that mcsafe_memcpy() has landed, and the return value was been
  clarified in commit cbf8b5a2 ("x86/mm, x86/mce: Fix return
  type/value for memcpy_mcsafe()"), let's hook up its primary usage in
  the pmem driver.

  The compilation problems from the initial posting have been fixed,
  this has appeared in a -next release with no reported issues, and it
  picked up an ack from Ingo.  There is no pressing need to merge this
  in 4.6- rc2.  However, if we wait until 4.7 the new memcpy_mcsafe()
  capability will ship without a user in 4.6-final"

* 'libnvdimm-for-next' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm:
  x86, pmem: use memcpy_mcsafe() for memcpy_from_pmem()

The seccomp filter support requires careful handling of task registers.  This
includes reloading of the return value (%r28) and proper syscall exit if
secure_computing() returned -1.

Additionally we need to sign-extend the syscall number from signed 32bit to
signed 64bit in do_syscall_trace_enter() since the ptrace interface only allows
storing 32bit values in compat mode.

Signed-off-by: Helge Deller <deller@gmx.de>
Cc: stable@vger.kernel.org # v4.5

Signed-off-by: Helge Deller <deller@gmx.de>
Cc: stable@vger.kernel.org # v4.5

We should try to trigger automount *before* bailing out on negative dentry.

Reported-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Reported-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Reported-by: Arend van Spriel <arend@broadcom.com>
Tested-by: Arend van Spriel <arend@broadcom.com>
Tested-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

Pull nios2 fix from Ley Foon Tan:
 "Replace fdt_translate_address with of_flat_dt_translate_address"

Fixes a build failure.

* tag 'nios2-v4.6-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/lftan/nios2:
  nios2: Replace fdt_translate_address with of_flat_dt_translate_address