The zero range operation is analogous to fallocate with the exception of
converting the range to zeroes. E.g., it attempts to allocate zeroed
blocks over the range specified by the caller. The XFS implementation
kills all delalloc blocks currently over the aligned range, converts the
range to allocated zero blocks (unwritten extents) and handles the
partial pages at the ends of the range by sending writes through the
pagecache.

The current implementation suffers from several problems associated with
inode size. If the aligned range covers an extending I/O, said I/O is
discarded and an inode size update from a previous write never makes it
to disk. Further, if an unaligned zero range extends beyond eof, the
page write induced for the partial end page can itself increase the
inode size, even if the zero range request is not supposed to update
i_size (via KEEP_SIZE, similar to an fallocate beyond EOF).

The latter behavior not only incorrectly increases the inode size, but
can lead to stray delalloc blocks on the inode. Typically, post-eof
preallocation blocks are either truncated on release or inode eviction
or explicitly written to by xfs_zero_eof() on natural file size
extension. If the inode size increases due to zero range, however,
associated blocks leak into the address space having never been
converted or mapped to pagecache pages. A direct I/O to such an
uncovered range cannot convert the extent via writeback and will BUG().
For example:

$ xfs_io -fc "pwrite 0 128k" -c "fzero -k 1m 54321" <file>
...
$ xfs_io -d -c "pread 128k 128k" <file>
<BUG>

If the entire delalloc extent happens to not have page coverage
whatsoever (e.g., delalloc conversion couldn't find a large enough free
space extent), even a full file writeback won't convert what's left of
the extent and we'll assert on inode eviction.

Rework xfs_zero_file_space() to avoid buffered I/O for partial pages.
Use the existing hole punch and prealloc mechanisms as primitives for
zero range. This implementation is not efficient nor ideal as we
writeback dirty data over the range and remove existing extents rather
than convert to unwrittern. The former writeback, however, is currently
the only mechanism available to ensure consistency between pagecache and
extent state. Even a pagecache truncate/delalloc punch prior to hole
punch has lead to inconsistencies due to racing with writeback.

This provides a consistent, correct implementation of zero range that
survives fsstress/fsx testing without assert failures. The
implementation can be optimized from this point forward once the
fundamental issue of pagecache and delalloc extent state consistency is
addressed.
Signed-off-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

The WARN_ON checking whether i_mutex is held in
pagecache_isize_extended() was wrong because some filesystems (e.g.
XFS) use different locks for serialization of truncates / writes. So
just remove the check.
Signed-off-by: NJan Kara <jack@suse.cz>
Reviewed-by: NDave Chinner <dchinner@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

xfs_bulkstat() doesn't check error return from xfs_btree_increment(). In
case of specific fs corruption that could result in xfs_bulkstat()
entering an infinite loop because we would be looping over the same
chunk over and over again. Fix the problem by checking the return value
and terminating the loop properly.

Coverity-id: 1231338
cc: <stable@vger.kernel.org>
Signed-off-by: NJan Kara <jack@suse.cz>
Reviewed-by: NJie Liu <jeff.u.liu@gmail.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

The recent refactoring of the bulkstat code left a small landmine in
the code. If a inobt read fails, then the tree walk is aborted and
returns without releasing the AGI buffer or freeing the cursor. This
can lead to a subsequent bulkstat call hanging trying to grab the
AGI buffer again.

cc: <stable@vger.kernel.org>
Signed-off-by: NDave Chinner <dchinner@redhat.com>
Reviewed-by: NBrian Foster <bfoster@redhat.com>
Reviewed-by: NEric Sandeen <sandeen@redhat.com>
Signed-off-by: NDave Chinner <david@fromorbit.com>

Pull ARM SoC fixes from Olof Johansson:
 "Another week, another small batch of fixes.

  Most of these make zynq, socfpga and sunxi platforms work a bit
  better:

   - due to new requirements for regulators, DWMMC on socfpga broke past
     v3.17
   - SMP spinup fix for socfpga
   - a few DT fixes for zynq
   - another option (FIXED_REGULATOR) for sunxi is needed that used to
     be selected by other options but no longer is.
   - a couple of small DT fixes for at91
   - ...and a couple for i.MX"

* tag 'armsoc-for-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
  ARM: dts: imx28-evk: Let i2c0 run at 100kHz
  ARM: i.MX6: Fix "emi" clock name typo
  ARM: multi_v7_defconfig: enable CONFIG_MMC_DW_ROCKCHIP
  ARM: sunxi_defconfig: enable CONFIG_REGULATOR_FIXED_VOLTAGE
  ARM: dts: socfpga: Add a 3.3V fixed regulator node
  ARM: dts: socfpga: Fix SD card detect
  ARM: dts: socfpga: rename gpio nodes
  ARM: at91/dt: sam9263: fix PLLB frequencies
  power: reset: at91-reset: fix power down register
  MAINTAINERS: add atmel ssc driver maintainer entry
  arm: socfpga: fix fetching cpu1start_addr for SMP
  ARM: zynq: DT: trivial: Fix mc node
  ARM: zynq: DT: Add cadence watchdog node
  ARM: zynq: DT: Add missing reference for memory-controller
  ARM: zynq: DT: Add missing reference for ADC
  ARM: zynq: DT: Add missing address for L2 pl310
  ARM: zynq: DT: Remove 222 MHz OPP
  ARM: zynq: DT: Fix GEM register area size

Pull vfs updates from Al Viro:
 "overlayfs merge + leak fix for d_splice_alias() failure exits"

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  overlayfs: embed middle into overlay_readdir_data
  overlayfs: embed root into overlay_readdir_data
  overlayfs: make ovl_cache_entry->name an array instead of pointer
  overlayfs: don't hold ->i_mutex over opening the real directory
  fix inode leaks on d_splice_alias() failure exits
  fs: limit filesystem stacking depth
  overlay: overlay filesystem documentation
  overlayfs: implement show_options
  overlayfs: add statfs support
  overlay filesystem
  shmem: support RENAME_WHITEOUT
  ext4: support RENAME_WHITEOUT
  vfs: add RENAME_WHITEOUT
  vfs: add whiteout support
  vfs: export check_sticky()
  vfs: introduce clone_private_mount()
  vfs: export __inode_permission() to modules
  vfs: export do_splice_direct() to modules
  vfs: add i_op->dentry_open()

Merge tag 'imx-fixes-3.18' of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux into fixes

Merge "ARM: imx: fixes for 3.18" from Shawn Guo:

The i.MX fixes for 3.18:
 - Revert one patch which increases I2C bus frequency on imx28-evk
 - Fix a typo on imx6q EIM clock name

* tag 'imx-fixes-3.18' of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux:
  ARM: dts: imx28-evk: Let i2c0 run at 100kHz
  ARM: i.MX6: Fix "emi" clock name typo
Signed-off-by: NOlof Johansson <olof@lixom.net>

Commit 78b81f46 ("ARM: dts: imx28-evk: Run I2C0 at 400kHz") caused issues
when doing the following sequence in loop:

- Boot the kernel
- Perform audio playback
- Reboot the system via 'reboot' command

In many times the audio card cannot be probed, which causes playback to fail.

After restoring to the original i2c0 frequency of 100kHz there is no such
problem anymore.

This reverts commit 78b81f46.

Cc: <stable@vger.kernel.org> # 3.16+
Signed-off-by: NFabio Estevam <fabio.estevam@freescale.com>
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

Fix a typo error, the "emi" names refer to the eim clocks.

The change fixes typo in EIM and EIM_SLOW pre-output dividers and
selectors clock names. Notably EIM_SLOW clock itself is named correctly.
Signed-off-by: NSteve Longerbeam <steve_longerbeam@mentor.com>
[vladimir_zapolskiy@mentor.com: ported to v3.17]
Signed-off-by: NVladimir Zapolskiy <vladimir_zapolskiy@mentor.com>
Cc: Sascha Hauer <kernel@pengutronix.de>
Signed-off-by: NShawn Guo <shawn.guo@linaro.org>

same story...
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

no sense having it a pointer - all instances have it pointing to
local variable in the same stack frame
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

just use it to serialize the assignment
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Pull MIPS fixes from Ralf Baechle:
 "This is the first round of fixes and tying up loose ends for MIPS.

   - plenty of fixes for build errors in specific obscure configurations
   - remove redundant code on the Lantiq platform
   - removal of a useless SEAD I2C driver that was causing a build issue
   - fix an earlier TLB exeption handler fix to also work on Octeon.
   - fix ISA level dependencies in FPU emulator's instruction decoding.
   - don't hardcode kernel command line in Octeon software emulator.
   - fix an earlier fix for the Loondson 2 clock setting"

* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus:
  MIPS: SEAD3: Fix I2C device registration.
  MIPS: SEAD3: Nuke PIC32 I2C driver.
  MIPS: ftrace: Fix a microMIPS build problem
  MIPS: MSP71xx: Fix build error
  MIPS: Malta: Do not build the malta-amon.c file if CMP is not enabled
  MIPS: Prevent compiler warning from cop2_{save,restore}
  MIPS: Kconfig: Add missing MIPS_CPS dependencies to PM and cpuidle
  MIPS: idle: Remove leftover __pastwait symbol and its references
  MIPS: Sibyte: Include the swarm subdir to the sb1250 LittleSur builds
  MIPS: ptrace.h: Add a missing include
  MIPS: ath79: Fix compilation error when CONFIG_PCI is disabled
  MIPS: MSP71xx: Remove compilation error when CONFIG_MIPS_MT is present
  MIPS: Octeon: Remove special case for simulator command line.
  MIPS: tlbex: Properly fix HUGE TLB Refill exception handler
  MIPS: loongson2_cpufreq: Fix CPU clock rate setting mismerge
  pci: pci-lantiq: remove duplicate check on resource
  MIPS: Lasat: Add missing CONFIG_PROC_FS dependency to PICVUE_PROC
  MIPS: cp1emu: Fix ISA restrictions for cop1x_op instructions

Pull arm64 fixes from Catalin Marinas:

 - enable 48-bit VA space now that KVM has been fixed, together with a
   couple of fixes for pgd allocation alignment and initial memblock
   current_limit.  There is still a dependency on !ARM_SMMU which needs
   to be updated as it uses the page table manipulation macros of the
   host kernel
 - eBPF fixes following changes/conflicts during the merging window
 - Compat types affecting compat_elf_prpsinfo
 - Compilation error on UP builds
 - ASLR fix when /proc/sys/kernel/randomize_va_space == 0
 - DT definitions for CLCD support on ARMv8 model platform

* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
  arm64: Fix memblock current_limit with 64K pages and 48-bit VA
  arm64: ASLR: Don't randomise text when randomise_va_space == 0
  arm64: vexpress: Add CLCD support to the ARMv8 model platform
  arm64: Fix compilation error on UP builds
  Documentation/arm64/memory.txt: fix typo
  net: bpf: arm64: minor fix of type in jited
  arm64: bpf: add 'load 64-bit immediate' instruction
  arm64: bpf: add 'shift by register' instructions
  net: bpf: arm64: address randomize and write protect JIT code
  arm64: mm: Correct fixmap pagetable types
  arm64: compat: fix compat types affecting struct compat_elf_prpsinfo
  arm64: Align less than PAGE_SIZE pgds naturally
  arm64: Allow 48-bits VA space without ARM_SMMU

Pull two sparc fixes from David Miller:

 1) Fix boots with gcc-4.9 compiled sparc64 kernels.

 2) Add missing __get_user_pages_fast() on sparc64 to fix hangs on
    futexes used in transparent hugepage areas.

    It's really idiotic to have a weak symbolled fallback that just
    returns zero, and causes this kind of bug.  There should be no
    backup implementation and the link should fail if the architecture
    fails to provide __get_user_pages_fast() and supports transparent
    hugepages.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc:
  sparc64: Implement __get_user_pages_fast().
  sparc64: Fix register corruption in top-most kernel stack frame during boot.

Pull kvm fixes from Paolo Bonzini:
 "This is a pretty large update.  I think it is roughly as big as what I
  usually had for the _whole_ rc period.

  There are a few bad bugs where the guest can OOPS or crash the host.
  We have also started looking at attack models for nested
  virtualization; bugs that usually result in the guest ring 0 crashing
  itself become more worrisome if you have nested virtualization,
  because the nested guest might bring down the non-nested guest as
  well.  For current uses of nested virtualization these do not really
  have a security impact, but you never know and bugs are bugs
  nevertheless.

  A lot of these bugs are in 3.17 too, resulting in a large number of
  stable@ Ccs.  I checked that all the patches apply there with no
  conflicts"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  kvm: vfio: fix unregister kvm_device_ops of vfio
  KVM: x86: Wrong assertion on paging_tmpl.h
  kvm: fix excessive pages un-pinning in kvm_iommu_map error path.
  KVM: x86: PREFETCH and HINT_NOP should have SrcMem flag
  KVM: x86: Emulator does not decode clflush well
  KVM: emulate: avoid accessing NULL ctxt->memopp
  KVM: x86: Decoding guest instructions which cross page boundary may fail
  kvm: x86: don't kill guest on unknown exit reason
  kvm: vmx: handle invvpid vm exit gracefully
  KVM: x86: Handle errors when RIP is set during far jumps
  KVM: x86: Emulator fixes for eip canonical checks on near branches
  KVM: x86: Fix wrong masking on relative jump/call
  KVM: x86: Improve thread safety in pit
  KVM: x86: Prevent host from panicking on shared MSR writes.
  KVM: x86: Check non-canonical addresses upon WRMSR

Merge tag 'stable/for-linus-3.18-b-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip

Pull xen bug fixes from David Vrabel:

 - Fix regression in xen_clocksource_read() which caused all Xen guests
   to crash early in boot.
 - Several fixes for super rare race conditions in the p2m.
 - Assorted other minor fixes.

* tag 'stable/for-linus-3.18-b-rc1-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen/pci: Allocate memory for physdev_pci_device_add's optarr
  x86/xen: panic on bad Xen-provided memory map
  x86/xen: Fix incorrect per_cpu accessor in xen_clocksource_read()
  x86/xen: avoid race in p2m handling
  x86/xen: delay construction of mfn_list_list
  x86/xen: avoid writing to freed memory after race in p2m handling
  xen/balloon: Don't continue ballooning when BP_ECANCELED is encountered

Pull sound fixes from Takashi Iwai:
 "Here are a chunk of small fixes since rc1: two PCM core fixes, one is
  a long-standing annoyance about lockdep and another is an ARM64 mmap
  fix.

  The rest are a HD-audio HDMI hotplug notification fix, a fix for
  missing NULL termination in Realtek codec quirks and a few new
  device/codec-specific quirks as usual"

* tag 'sound-3.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hda - Add missing terminating entry to SND_HDA_PIN_QUIRK macro
  ALSA: pcm: Fix false lockdep warnings
  ALSA: hda - Fix inverted LED gpio setup for Lenovo Ideapad
  ALSA: hda - hdmi: Fix missing ELD change event on plug/unplug
  ALSA: usb-audio: Add support for Steinberg UR22 USB interface
  ALSA: ALC283 codec - Avoid pop noise on headphones during suspend/resume
  ALSA: pcm: use the same dma mmap codepath both for arm and arm64

Pull /dev/random updates from Ted Ts'o:
 "This adds a memzero_explicit() call which is guaranteed not to be
  optimized away by GCC.  This is important when we are wiping
  cryptographically sensitive material"

* tag 'random_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/random:
  crypto: memzero_explicit - make sure to clear out sensitive data
  random: add and use memzero_explicit() for clearing data

Pull ACPI and power management updates from Rafael Wysocki:
 "This is material that didn't make it to my 3.18-rc1 pull request for
  various reasons, mostly related to timing and travel (LinuxCon EU /
  LPC) plus a couple of fixes for recent bugs.

  The only really new thing here is the PM QoS class for memory
  bandwidth, but it is simple enough and users of it will be added in
  the next cycle.  One major change in behavior is that platform devices
  enumerated by ACPI will use 32-bit DMA mask by default.  Also included
  is an ACPICA update to a new upstream release, but that's mostly
  cleanups, changes in tools and similar.  The rest is fixes and
  cleanups mostly.

  Specifics:

   - Fix for a recent PCI power management change that overlooked the
     fact that some IRQ chips might not be able to configure PCIe PME
     for system wakeup from Lucas Stach.

   - Fix for a bug introduced in 3.17 where acpi_device_wakeup() is
     called with a wrong ordering of arguments from Zhang Rui.

   - A bunch of intel_pstate driver fixes (all -stable candidates) from
     Dirk Brandewie, Gabriele Mazzotta and Pali Rohár.

   - Fixes for a rather long-standing problem with the OOM killer and
     the freezer that frozen processes killed by the OOM do not actually
     release any memory until they are thawed, so OOM-killing them is
     rather pointless, with a couple of cleanups on top (Michal Hocko,
     Cong Wang, Rafael J Wysocki).

   - ACPICA update to upstream release 20140926, inlcuding mostly
     cleanups reducing differences between the upstream ACPICA and the
     kernel code, tools changes (acpidump, acpiexec) and support for the
     _DDN object (Bob Moore, Lv Zheng).

   - New PM QoS class for memory bandwidth from Tomeu Vizoso.

   - Default 32-bit DMA mask for platform devices enumerated by ACPI
     (this change is mostly needed for some drivers development in
     progress targeted at 3.19) from Heikki Krogerus.

   - ACPI EC driver cleanups, mostly related to debugging, from Lv
     Zheng.

   - cpufreq-dt driver updates from Thomas Petazzoni.

   - powernv cpuidle driver update from Preeti U Murthy"

* tag 'pm+acpi-3.18-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (34 commits)
  intel_pstate: Correct BYT VID values.
  intel_pstate: Fix BYT frequency reporting
  intel_pstate: Don't lose sysfs settings during cpu offline
  cpufreq: intel_pstate: Reflect current no_turbo state correctly
  cpufreq: expose scaling_cur_freq sysfs file for set_policy() drivers
  cpufreq: intel_pstate: Fix setting max_perf_pct in performance policy
  PCI / PM: handle failure to enable wakeup on PCIe PME
  ACPI: invoke acpi_device_wakeup() with correct parameters
  PM / freezer: Clean up code after recent fixes
  PM: convert do_each_thread to for_each_process_thread
  OOM, PM: OOM killed task shouldn't escape PM suspend
  freezer: remove obsolete comments in __thaw_task()
  freezer: Do not freeze tasks killed by OOM killer
  ACPI / platform: provide default DMA mask
  cpuidle: powernv: Populate cpuidle state details by querying the device-tree
  cpufreq: cpufreq-dt: adjust message related to regulators
  cpufreq: cpufreq-dt: extend with platform_data
  cpufreq: allow driver-specific data
  ACPI / EC: Cleanup coding style.
  ACPI / EC: Refine event/query debugging messages.
  ...

Pull thermal management updates from Zhang Rui:
 "Sorry that I missed the merge window as there is a bug found in the
  last minute, and I have to fix it and wait for the code to be tested
  in linux-next tree for a few days.  Now the buggy patch has been
  dropped entirely from my next branch.  Thus I hope those changes can
  still be merged in 3.18-rc2 as most of them are platform thermal
  driver changes.

  Specifics:

   - introduce ACPI INT340X thermal drivers.

     Newer laptops and tablets may have thermal sensors and other
     devices with thermal control capabilities that are exposed for the
     OS to use via the ACPI INT340x device objects.  Several drivers are
     introduced to expose the temperature information and cooling
     ability from these objects to user-space via the normal thermal
     framework.

     From: Lu Aaron, Lan Tianyu, Jacob Pan and Zhang Rui.

   - introduce a new thermal governor, which just uses a hysteresis to
     switch abruptly on/off a cooling device.  This governor can be used
     to control certain fan devices that can not be throttled but just
     switched on or off.  From: Peter Feuerer.

   - introduce support for some new thermal interrupt functions on
     i.MX6SX, in IMX thermal driver.  From: Anson, Huang.

   - introduce tracing support on thermal framework.  From: Punit
     Agrawal.

   - small fixes in OF thermal and thermal step_wise governor"

* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux: (25 commits)
  Thermal: int340x thermal: select ACPI fan driver
  Thermal: int3400_thermal: use acpi_thermal_rel parsing APIs
  Thermal: int340x_thermal: expose acpi thermal relationship tables
  Thermal: introduce int3403 thermal driver
  Thermal: introduce INT3402 thermal driver
  Thermal: move the KELVIN_TO_MILLICELSIUS macro to thermal.h
  ACPI / Fan: support INT3404 thermal device
  ACPI / Fan: add ACPI 4.0 style fan support
  ACPI / fan: convert to platform driver
  ACPI / fan: use acpi_device_xxx_power instead of acpi_bus equivelant
  ACPI / fan: remove no need check for device pointer
  ACPI / fan: remove unused macro
  Thermal: int3400 thermal: register to thermal framework
  Thermal: int3400 thermal: add capability to detect supporting UUIDs
  Thermal: introduce int3400 thermal driver
  ACPI: add ACPI_TYPE_LOCAL_REFERENCE support to acpi_extract_package()
  ACPI: make acpi_create_platform_device() an external API
  thermal: step_wise: fix: Prevent from binary overflow when trend is dropping
  ACPI: introduce ACPI int340x thermal scan handler
  thermal: Added Bang-bang thermal governor
  ...

With 48-bit VA space, the 64K page configuration uses 3 levels instead
of 2 and PUD_SIZE != PMD_SIZE. Since with 64K pages we only cover
PMD_SIZE with the initial swapper_pg_dir populated in head.S, the
memblock current_limit needs to be set accordingly in map_mem() to avoid
allocating unmapped memory. The memblock current_limit is progressively
increased as more blocks are mapped.
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

It is not sufficient to only implement get_user_pages_fast(), you
must also implement the atomic version __get_user_pages_fast()
otherwise you end up using the weak symbol fallback implementation
which simply returns zero.

This is dangerous, because it causes the futex code to loop forever
if transparent hugepages are supported (see get_futex_key()).
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Meelis Roos reported that kernels built with gcc-4.9 do not boot, we
eventually narrowed this down to only impacting machines using
UltraSPARC-III and derivitive cpus.

The crash happens right when the first user process is spawned:

[   54.451346] Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000004
[   54.451346]
[   54.571516] CPU: 1 PID: 1 Comm: init Not tainted 3.16.0-rc2-00211-gd7933ab7 #96
[   54.666431] Call Trace:
[   54.698453]  [0000000000762f8c] panic+0xb0/0x224
[   54.759071]  [000000000045cf68] do_exit+0x948/0x960
[   54.823123]  [000000000042cbc0] fault_in_user_windows+0xe0/0x100
[   54.902036]  [0000000000404ad0] __handle_user_windows+0x0/0x10
[   54.978662] Press Stop-A (L1-A) to return to the boot prom
[   55.050713] ---[ end Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000004

Further investigation showed that compiling only per_cpu_patch() with
an older compiler fixes the boot.

Detailed analysis showed that the function is not being miscompiled by
gcc-4.9, but it is using a different register allocation ordering.

With the gcc-4.9 compiled function, something during the code patching
causes some of the %i* input registers to get corrupted.  Perhaps
we have a TLB miss path into the firmware that is deep enough to
cause a register window spill and subsequent restore when we get
back from the TLB miss trap.

Let's plug this up by doing two things:

1) Stop using the firmware stack for client interface calls into
   the firmware.  Just use the kernel's stack.

2) As soon as we can, call into a new function "start_early_boot()"
   to put a one-register-window buffer between the firmware's
   deepest stack frame and the top-most initial kernel one.
Reported-by: NMeelis Roos <mroos@linux.ee>
Tested-by: NMeelis Roos <mroos@linux.ee>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When user asks to turn off ASLR by writing "0" to
/proc/sys/kernel/randomize_va_space there should not be
any randomization to mmap base, stack, VDSO, libs, text and heap

Currently arm64 violates this behavior by randomising text.
Fix this by defining a constant ELF_ET_DYN_BASE. The randomisation of
mm->mmap_base is done by setup_new_exec -> arch_pick_mmap_layout ->
mmap_base -> mmap_rnd.
Signed-off-by: NArun Chandran <achandran@mvista.com>
Signed-off-by: NCatalin Marinas <catalin.marinas@arm.com>

This isn't a module and shouldn't be one.
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

After commit 80ce1639 (KVM: VFIO: register kvm_device_ops dynamically),
kvm_device_ops of vfio can be registered dynamically. Commit 3c3c29fd
(kvm-vfio: do not use module_init) move the dynamic register invoked by
kvm_init in order to fix broke unloading of the kvm module. However,
kvm_device_ops of vfio is unregistered after rmmod kvm-intel module
which lead to device type collision detection warning after kvm-intel
module reinsmod.

    WARNING: CPU: 1 PID: 10358 at /root/cathy/kvm/arch/x86/kvm/../../../virt/kvm/kvm_main.c:3289 kvm_init+0x234/0x282 [kvm]()
    Modules linked in: kvm_intel(O+) kvm(O) nfsv3 nfs_acl auth_rpcgss oid_registry nfsv4 dns_resolver nfs fscache lockd sunrpc pci_stub bridge stp llc autofs4 8021q cpufreq_ondemand ipv6 joydev microcode pcspkr igb i2c_algo_bit ehci_pci ehci_hcd e1000e i2c_i801 ixgbe ptp pps_core hwmon mdio tpm_tis tpm ipmi_si ipmi_msghandler acpi_cpufreq isci libsas scsi_transport_sas button dm_mirror dm_region_hash dm_log dm_mod [last unloaded: kvm_intel]
    CPU: 1 PID: 10358 Comm: insmod Tainted: G        W  O   3.17.0-rc1 #2
    Hardware name: Intel Corporation S2600CP/S2600CP, BIOS RMLSDP.86I.00.29.D696.1311111329 11/11/2013
     0000000000000cd9 ffff880ff08cfd18 ffffffff814a61d9 0000000000000cd9
     0000000000000000 ffff880ff08cfd58 ffffffff810417b7 ffff880ff08cfd48
     ffffffffa045bcac ffffffffa049c420 0000000000000040 00000000000000ff
    Call Trace:
     [<ffffffff814a61d9>] dump_stack+0x49/0x60
     [<ffffffff810417b7>] warn_slowpath_common+0x7c/0x96
     [<ffffffffa045bcac>] ? kvm_init+0x234/0x282 [kvm]
     [<ffffffff810417e6>] warn_slowpath_null+0x15/0x17
     [<ffffffffa045bcac>] kvm_init+0x234/0x282 [kvm]
     [<ffffffffa016e995>] vmx_init+0x1bf/0x42a [kvm_intel]
     [<ffffffffa016e7d6>] ? vmx_check_processor_compat+0x64/0x64 [kvm_intel]
     [<ffffffff810002ab>] do_one_initcall+0xe3/0x170
     [<ffffffff811168a9>] ? __vunmap+0xad/0xb8
     [<ffffffff8109c58f>] do_init_module+0x2b/0x174
     [<ffffffff8109d414>] load_module+0x43e/0x569
     [<ffffffff8109c6d8>] ? do_init_module+0x174/0x174
     [<ffffffff8109c75a>] ? copy_module_from_user+0x39/0x82
     [<ffffffff8109b7dd>] ? module_sect_show+0x20/0x20
     [<ffffffff8109d65f>] SyS_init_module+0x54/0x81
     [<ffffffff814a9a12>] system_call_fastpath+0x16/0x1b
    ---[ end trace 0626f4a3ddea56f3 ]---

The bug can be reproduced by:

    rmmod kvm_intel.ko
    insmod kvm_intel.ko

without rmmod/insmod kvm.ko
This patch fixes the bug by unregistering kvm_device_ops of vfio when the
kvm-intel module is removed.
Reported-by: NLiu Rongrong <rongrongx.liu@intel.com>
Fixes: 3c3c29fdSigned-off-by: NWanpeng Li <wanpeng.li@linux.intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Even after the recent fix, the assertion on paging_tmpl.h is triggered.
Apparently, the assertion wants to check that the PAE is always set on
long-mode, but does it in incorrect way.  Note that the assertion is not
enabled unless the code is debugged by defining MMU_DEBUG.
Signed-off-by: NNadav Amit <namit@cs.technion.ac.il>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The third parameter of kvm_unpin_pages() when called from
kvm_iommu_map_pages() is wrong, it should be the number of pages to un-pin
and not the page size.

This error was facilitated with an inconsistent API: kvm_pin_pages() takes
a size, but kvn_unpin_pages() takes a number of pages, so fix the problem
by matching the two.

This was introduced by commit 350b8bdd ("kvm: iommu: fix the third parameter
of kvm_iommu_put_pages (CVE-2014-3601)"), which fixes the lack of
un-pinning for pages intended to be un-pinned (i.e. memory leak) but
unfortunately potentially aggravated the number of pages we un-pin that
should have stayed pinned. As far as I understand though, the same
practical mitigations apply.

This issue was found during review of Red Hat 6.6 patches to prepare
Ksplice rebootless updates.

Thanks to Vegard for his time on a late Friday evening to help me in
understanding this code.

Fixes: 350b8bdd ("kvm: iommu: fix the third parameter of... (CVE-2014-3601)")
Cc: stable@vger.kernel.org
Signed-off-by: NQuentin Casasnovas <quentin.casasnovas@oracle.com>
Signed-off-by: NVegard Nossum <vegard.nossum@oracle.com>
Signed-off-by: NJamie Iles <jamie.iles@oracle.com>
Reviewed-by: NSasha Levin <sasha.levin@oracle.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The decode phase of the x86 emulator assumes that every instruction with the
ModRM flag, and which can be used with RIP-relative addressing, has either
SrcMem or DstMem.  This is not the case for several instructions - prefetch,
hint-nop and clflush.

Adding SrcMem|NoAccess for prefetch and hint-nop and SrcMem for clflush.

This fixes CVE-2014-8480.

Fixes: 41061cdb
Cc: stable@vger.kernel.org
Signed-off-by: NNadav Amit <namit@cs.technion.ac.il>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Currently, all group15 instructions are decoded as clflush (e.g., mfence,
xsave).  In addition, the clflush instruction requires no prefix (66/f2/f3)
would exist. If prefix exists it may encode a different instruction (e.g.,
clflushopt).

Creating a group for clflush, and different group for each prefix.

This has been the case forever, but the next patch needs the cflush group
in order to fix a bug introduced in 3.17.

Fixes: 41061cdb
Cc: stable@vger.kernel.org
Signed-off-by: NNadav Amit <namit@cs.technion.ac.il>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

A failure to decode the instruction can cause a NULL pointer access.
This is fixed simply by moving the "done" label as close as possible
to the return.

This fixes CVE-2014-8481.
Reported-by: NAndy Lutomirski <luto@amacapital.net>
Cc: stable@vger.kernel.org
Fixes: 41061cdbSigned-off-by: NPaolo Bonzini <pbonzini@redhat.com>

A platform driver for which nothing ever registers the corresponding
platform device.

Also it was driving the same hardware as sead3-i2c-drv.c so redundant
anyway and couldn't co-exist with that driver because each of them was
using a private spinlock to protect access to the same hardware
resources.

This also fixes a randconfig problem:

arch/mips/mti-sead3/sead3-pic32-i2c-drv.c: In function 'i2c_platform_probe':
arch/mips/mti-sead3/sead3-pic32-i2c-drv.c:345:2: error: implicit declaration of
function 'i2c_add_numbered_adapter' [-Werror=implicit-function-declaration]
  ret = i2c_add_numbered_adapter(&priv->adap);
    ^
arch/mips/mti-sead3/sead3-pic32-i2c-drv.c: In function
'i2c_platform_remove':
arch/mips/mti-sead3/sead3-pic32-i2c-drv.c:361:2: error: implicit declaration
of function 'i2c_del_adapter' [-Werror=implicit-function-declaration]
i2c_del_adapter(&priv->adap);
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Once an instruction crosses a page boundary, the size read from the second page
disregards the common case that part of the operand resides on the first page.
As a result, fetch of long insturctions may fail, and thereby cause the
decoding to fail as well.

Cc: stable@vger.kernel.org
Fixes: 5cfc7e0fSigned-off-by: NNadav Amit <namit@cs.technion.ac.il>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

KVM_EXIT_UNKNOWN is a kvm bug, we don't really know whether it was
triggered by a priveledged application.  Let's not kill the guest: WARN
and inject #UD instead.

Cc: stable@vger.kernel.org
Signed-off-by: NMichael S. Tsirkin <mst@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

On systems with invvpid instruction support (corresponding bit in
IA32_VMX_EPT_VPID_CAP MSR is set) guest invocation of invvpid
causes vm exit, which is currently not handled and results in
propagation of unknown exit to userspace.

Fix this by installing an invvpid vm exit handler.

This is CVE-2014-3646.

Cc: stable@vger.kernel.org
Signed-off-by: NPetr Matousek <pmatouse@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Far jmp/call/ret may fault while loading a new RIP.  Currently KVM does not
handle this case, and may result in failed vm-entry once the assignment is
done.  The tricky part of doing so is that loading the new CS affects the
VMCS/VMCB state, so if we fail during loading the new RIP, we are left in
unconsistent state.  Therefore, this patch saves on 64-bit the old CS
descriptor and restores it if loading RIP failed.

This fixes CVE-2014-3647.

Cc: stable@vger.kernel.org
Signed-off-by: NNadav Amit <namit@cs.technion.ac.il>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Before changing rip (during jmp, call, ret, etc.) the target should be asserted
to be canonical one, as real CPUs do.  During sysret, both target rsp and rip
should be canonical. If any of these values is noncanonical, a #GP exception
should occur.  The exception to this rule are syscall and sysenter instructions
in which the assigned rip is checked during the assignment to the relevant
MSRs.

This patch fixes the emulator to behave as real CPUs do for near branches.
Far branches are handled by the next patch.

This fixes CVE-2014-3647.

Cc: stable@vger.kernel.org
Signed-off-by: NNadav Amit <namit@cs.technion.ac.il>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>