openwrt/staging/blogic.git
4 years agoMerge tag 'io_uring-5.5-20191220' of git://git.kernel.dk/linux-block
Linus Torvalds [Fri, 20 Dec 2019 21:30:49 +0000 (13:30 -0800)]
Merge tag 'io_uring-5.5-20191220' of git://git.kernel.dk/linux-block

Pull io_uring fixes from Jens Axboe:
 "Here's a set of fixes that should go into 5.5-rc3 for io_uring.

  This is bigger than I'd like it to be, mainly because we're fixing the
  case where an application reuses sqe data right after issue. This
  really must work, or it's confusing. With 5.5 we're flagging us as
  submit stable for the actual data, this must also be the case for
  SQEs.

  Honestly, I'd really like to add another series on top of this, since
  it cleans it up considerable and prevents any SQE reuse by design. I
  posted that here:

    https://lore.kernel.org/io-uring/20191220174742.7449-1-axboe@kernel.dk/T/#u

  and may still send it your way early next week once it's been looked
  at and had some more soak time (does pass all regression tests). With
  that series, we've unified the prep+issue handling, and only the prep
  phase even has access to the SQE.

  Anyway, outside of that, fixes in here for a few other issues that
  have been hit in testing or production"

* tag 'io_uring-5.5-20191220' of git://git.kernel.dk/linux-block:
  io_uring: io_wq_submit_work() should not touch req->rw
  io_uring: don't wait when under-submitting
  io_uring: warn about unhandled opcode
  io_uring: read opcode and user_data from SQE exactly once
  io_uring: make IORING_OP_TIMEOUT_REMOVE deferrable
  io_uring: make IORING_OP_CANCEL_ASYNC deferrable
  io_uring: make IORING_POLL_ADD and IORING_POLL_REMOVE deferrable
  io_uring: make HARDLINK imply LINK
  io_uring: any deferred command must have stable sqe data
  io_uring: remove 'sqe' parameter to the OP helpers that take it
  io_uring: fix pre-prepped issue with force_nonblock == true
  io-wq: re-add io_wq_current_is_worker()
  io_uring: fix sporadic -EFAULT from IORING_OP_RECVMSG
  io_uring: fix stale comment and a few typos

4 years agoMerge tag 'iommu-fixes-v5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 20 Dec 2019 18:42:25 +0000 (10:42 -0800)]
Merge tag 'iommu-fixes-v5.5-rc2' of git://git./linux/kernel/git/joro/iommu

Pull iommu fixes from Joerg Roedel:

 - Fix kmemleak warning in IOVA code

 - Fix compile warnings on ARM32/64 in dma-iommu code due to dma_mask
   type mismatches

 - Make ISA reserved regions relaxable, so that VFIO can assign devices
   which have such regions defined

 - Fix mapping errors resulting in IO page-faults in the VT-d driver

 - Make sure direct mappings for a domain are created after the default
   domain is updated

 - Map ISA reserved regions in the VT-d driver with correct permissions

 - Remove unneeded check for PSI capability in the IOTLB flush code of
   the VT-d driver

 - Lockdep fix iommu_dma_prepare_msi()

* tag 'iommu-fixes-v5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
  iommu/dma: Relax locking in iommu_dma_prepare_msi()
  iommu/vt-d: Remove incorrect PSI capability check
  iommu/vt-d: Allocate reserved region for ISA with correct permission
  iommu: set group default domain before creating direct mappings
  iommu/vt-d: Fix dmar pte read access not set error
  iommu/vt-d: Set ISA bridge reserved region as relaxable
  iommu/dma: Rationalise types for DMA masks
  iommu/iova: Init the struct iova to fix the possible memleak

4 years agoMerge tag 'platform-drivers-x86-v5.5-2' of git://git.infradead.org/linux-platform...
Linus Torvalds [Fri, 20 Dec 2019 18:38:21 +0000 (10:38 -0800)]
Merge tag 'platform-drivers-x86-v5.5-2' of git://git.infradead.org/linux-platform-drivers-x86

Pull x86 platform driver fixes from Andy Shevchenko:
 "Bucket of fixes for PDx86. Note, that there is no ABI breakage in
  Mellanox driver because it has been introduced in v5.5-rc1, so we can
  change it.

  Summary:

   - Add support of APUv4 and fix an assignment of simswap GPIO

   - Add Siemens CONNECT X300 to DMI table to avoid stuck during boot

   - Correct arguments of WMI call on HP Envy x360 15-cp0xxx model

   - Fix the mlx-bootctl sysfs attributes to be device related"

* tag 'platform-drivers-x86-v5.5-2' of git://git.infradead.org/linux-platform-drivers-x86:
  platform/x86: pcengines-apuv2: Spelling fixes in the driver
  platform/x86: pcengines-apuv2: detect apuv4 board
  platform/x86: pcengines-apuv2: fix simswap GPIO assignment
  platform/x86: pmc_atom: Add Siemens CONNECT X300 to critclk_systems DMI table
  platform/x86: hp-wmi: Make buffer for HPWMI_FEATURE2_QUERY 128 bytes
  platform/mellanox: fix the mlx-bootctl sysfs

4 years agoMerge tag 'mmc-v5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc
Linus Torvalds [Fri, 20 Dec 2019 18:36:49 +0000 (10:36 -0800)]
Merge tag 'mmc-v5.5-rc2' of git://git./linux/kernel/git/ulfh/mmc

Pull MMC host fixes from Ulf Hansson:

 - mtk-sd: Fix tuning for MT8173 HS200/HS400 mode

 - sdhci: Revert a fix for incorrect switch to HS mode

 - sdhci-msm: Fixup accesses to the DDR_CONFIG register

 - sdhci-of-esdhc: Revert a bad fix for erratum A-009204

 - sdhci-of-esdhc: Re-implement fix for erratum A-009204

 - sdhci-of-esdhc: Fixup P2020 errata handling

 - sdhci-pci: Disable broken CMDQ on Intel GLK based Lenovo systems

* tag 'mmc-v5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc:
  mmc: sdhci-of-esdhc: re-implement erratum A-009204 workaround
  mmc: sdhci: Add a quirk for broken command queuing
  mmc: sdhci: Workaround broken command queuing on Intel GLK
  mmc: sdhci-of-esdhc: fix P2020 errata handling
  mmc: sdhci: Update the tuning failed messages to pr_debug level
  mmc: sdhci-of-esdhc: Revert "mmc: sdhci-of-esdhc: add erratum A-009204 support"
  mmc: mediatek: fix CMD_TA to 2 for MT8173 HS200/HS400 mode
  mmc: sdhci-msm: Correct the offset and value for DDR_CONFIG register
  Revert "mmc: sdhci: Fix incorrect switch to HS mode"

4 years agoMerge tag 'char-misc-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Fri, 20 Dec 2019 18:11:30 +0000 (10:11 -0800)]
Merge tag 'char-misc-5.5-rc3' of git://git./linux/kernel/git/gregkh/char-misc

Pull char/misc driver fixes from Greg KH:
 "Here are some small char and other driver fixes for 5.5-rc3.

  The most noticable one is a much-reported fix for a random driver
  issue that came up from 5.5-rc1 compat_ioctl cleanups. The others are
  a chunk of habanalab driver fixes and intel_th driver fixes and new
  device ids.

  All have been in linux-next with no reported issues"

* tag 'char-misc-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  random: don't forget compat_ioctl on urandom
  intel_th: msu: Fix window switching without windows
  intel_th: Fix freeing IRQs
  intel_th: pci: Add Elkhart Lake SOC support
  intel_th: pci: Add Comet Lake PCH-V support
  habanalabs: remove variable 'val' set but not used
  habanalabs: rate limit error msg on waiting for CS

4 years agoMerge tag 'staging-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Fri, 20 Dec 2019 18:09:21 +0000 (10:09 -0800)]
Merge tag 'staging-5.5-rc3' of git://git./linux/kernel/git/gregkh/staging

Pull staging driver fixes from Greg KH:
 "Here are some small staging driver fixes for a number of reported
  issues.

  The majority here are some fixes for the wfx driver, but also in here
  is a comedi driver fix found during some code review, and an axis-fifo
  build dependancy issue to resolve some reported testing problems.

  All of these have been in linux-next with no reported issues"

* tag 'staging-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
  staging: wfx: fix wrong error message
  staging: wfx: fix hif_set_mfp() with big endian hosts
  staging: wfx: detect race condition in WEP authentication
  staging: wfx: ensure that retry policy always fallbacks to MCS0 / 1Mbps
  staging: wfx: fix rate control handling
  staging: wfx: firmware does not support more than 32 total retries
  staging: wfx: use boolean appropriately
  staging: wfx: fix counter overflow
  staging: wfx: fix case of lack of tx_retry_policies
  staging: wfx: fix the cache of rate policies on interface reset
  staging: axis-fifo: add unspecified HAS_IOMEM dependency
  staging: comedi: gsc_hpdi: check dma_alloc_coherent() return value

4 years agoMerge tag 'tty-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
Linus Torvalds [Fri, 20 Dec 2019 17:55:28 +0000 (09:55 -0800)]
Merge tag 'tty-5.5-rc3' of git://git./linux/kernel/git/gregkh/tty

Pull tty/serial fixes from Greg KH:
 "Here are some small tty and serial driver fixes for 5.5-rc3.

  Only four small patches here:

   - atmel serial driver fix

   - msm_serial driver fix

   - sprd serial driver fix

   - tty core port fix

  The last tty core fix should resolve a long-standing bug with a race
  at port creation time that some people would see, and Sudip finally
  tracked down.

  All of these have been in linux-next with no reported issues"

* tag 'tty-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  tty/serial: atmel: fix out of range clock divider handling
  tty: link tty and port before configuring it as console
  serial: sprd: Add clearing break interrupt operation
  tty: serial: msm_serial: Fix lockup for sysrq and oops

4 years agoMerge tag 'usb-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Linus Torvalds [Fri, 20 Dec 2019 17:53:24 +0000 (09:53 -0800)]
Merge tag 'usb-5.5-rc3' of git://git./linux/kernel/git/gregkh/usb

Pull USB fixes from Greg KH:
 "Here are some small USB fixes for some reported issues.

  Included in here are:

   - xhci build warning fix

   - ehci disconnect warning fix

   - usbip lockup fix and error cleanup fix

   - typec build fix

  All of these have been in linux-next with no reported issues"

* tag 'usb-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
  usb: xhci: Fix build warning seen with CONFIG_PM=n
  usbip: Fix error path of vhci_recv_ret_submit()
  usbip: Fix receive error in vhci-hcd when using scatter-gather
  USB: EHCI: Do not return -EPIPE when hub is disconnected
  usb: typec: fusb302: Fix an undefined reference to 'extcon_get_state'

4 years agoMerge tag 'pinctrl-v5.5-3' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw...
Linus Torvalds [Fri, 20 Dec 2019 17:49:05 +0000 (09:49 -0800)]
Merge tag 'pinctrl-v5.5-3' of git://git./linux/kernel/git/linusw/linux-pinctrl

Pull pin control fixes from Linus Walleij:
 "Sorry that this fixes pull request took a while. Too much christmas
  business going on.

  This contains a few really important Intel fixes and some odd fixes:

   - A host of fixes for the Intel baytrail and cherryview: properly
     serialize all register accesses and add the irqchip with the
     gpiochip as we need to, fix some pin lists and initialize the
     hardware in the right order.

   - Fix the Aspeed G6 LPC configuration.

   - Handle a possible NULL pointer exception in the core.

   - Fix the Kconfig dependencies for the Equilibrium driver"

* tag 'pinctrl-v5.5-3' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:
  pinctrl: ingenic: Fixup PIN_CONFIG_OUTPUT config
  pinctrl: Modify Kconfig to fix linker error
  pinctrl: pinmux: fix a possible null pointer in pinmux_can_be_used_for_gpio
  pinctrl: aspeed-g6: Fix LPC/eSPI mux configuration
  pinctrl: cherryview: Pass irqchip when adding gpiochip
  pinctrl: cherryview: Add GPIO <-> pin mapping ranges via callback
  pinctrl: cherryview: Split out irq hw-init into a separate helper function
  pinctrl: baytrail: Pass irqchip when adding gpiochip
  pinctrl: baytrail: Add GPIO <-> pin mapping ranges via callback
  pinctrl: baytrail: Update North Community pin list
  pinctrl: baytrail: Really serialize all register accesses

4 years agoplatform/x86: pcengines-apuv2: Spelling fixes in the driver
Andy Shevchenko [Wed, 18 Dec 2019 12:48:07 +0000 (14:48 +0200)]
platform/x86: pcengines-apuv2: Spelling fixes in the driver

Mainly does:
 - capitalize gpio and bios to GPIO and BIOS
 - capitalize beginning of comments
 - add periods in multi-line comments

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
4 years agoplatform/x86: pcengines-apuv2: detect apuv4 board
Enrico Weigelt, metux IT consult [Thu, 12 Dec 2019 13:27:56 +0000 (14:27 +0100)]
platform/x86: pcengines-apuv2: detect apuv4 board

GPIO stuff on APUv4 seems to be the same as on APUv2, so we just
need to match on DMI data.

Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
4 years agoplatform/x86: pcengines-apuv2: fix simswap GPIO assignment
Enrico Weigelt, metux IT consult [Thu, 12 Dec 2019 11:23:48 +0000 (12:23 +0100)]
platform/x86: pcengines-apuv2: fix simswap GPIO assignment

The mapping entry has to hold the GPIO line index instead of
controller's register number.

Fixes: 5037d4ddda31 ("platform/x86: pcengines-apuv2: wire up simswitch gpio as led")
Signed-off-by: Enrico Weigelt, metux IT consult <info@metux.net>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
4 years agoplatform/x86: pmc_atom: Add Siemens CONNECT X300 to critclk_systems DMI table
Michael Haener [Fri, 29 Nov 2019 09:16:49 +0000 (10:16 +0100)]
platform/x86: pmc_atom: Add Siemens CONNECT X300 to critclk_systems DMI table

The CONNECT X300 uses the PMC clock for on-board components and gets
stuck during boot if the clock is disabled. Therefore, add this
device to the critical systems list.
Tested on CONNECT X300.

Fixes: 648e921888ad ("clk: x86: Stop marking clocks as CLK_IS_CRITICAL")
Signed-off-by: Michael Haener <michael.haener@siemens.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
4 years agoplatform/x86: hp-wmi: Make buffer for HPWMI_FEATURE2_QUERY 128 bytes
Hans de Goede [Tue, 17 Dec 2019 19:06:04 +0000 (20:06 +0100)]
platform/x86: hp-wmi: Make buffer for HPWMI_FEATURE2_QUERY 128 bytes

At least on the HP Envy x360 15-cp0xxx model the WMI interface
for HPWMI_FEATURE2_QUERY requires an outsize of at least 128 bytes,
otherwise it fails with an error code 5 (HPWMI_RET_INVALID_PARAMETERS):

Dec 06 00:59:38 kernel: hp_wmi: query 0xd returned error 0x5

We do not care about the contents of the buffer, we just want to know
if the HPWMI_FEATURE2_QUERY command is supported.

This commits bumps the buffer size, fixing the error.

Fixes: 8a1513b4932 ("hp-wmi: limit hotkey enable")
Cc: stable@vger.kernel.org
BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1520703
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
4 years agoplatform/mellanox: fix the mlx-bootctl sysfs
Liming Sun [Wed, 18 Dec 2019 18:35:27 +0000 (13:35 -0500)]
platform/mellanox: fix the mlx-bootctl sysfs

This is a follow-up commit for the sysfs attributes to change
from DRIVER_ATTR to DEVICE_ATTR according to some initial comments.
In such case, it's better to point the sysfs path to the device
itself instead of the driver. The ABI document is also updated.

Fixes: 79e29cb8fbc5 ("platform/mellanox: Add bootctl driver for Mellanox BlueField Soc")
Signed-off-by: Liming Sun <lsun@mellanox.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
4 years agoMerge branch 'akpm' (patches from Andrew)
Linus Torvalds [Thu, 19 Dec 2019 16:13:04 +0000 (08:13 -0800)]
Merge branch 'akpm' (patches from Andrew)

Merge fixes from Andrew Morton:
 "6 fixes"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  lib/Kconfig.debug: fix some messed up configurations
  mm: vmscan: protect shrinker idr replace with CONFIG_MEMCG
  kasan: don't assume percpu shadow allocations will succeed
  kasan: use apply_to_existing_page_range() for releasing vmalloc shadow
  mm/memory.c: add apply_to_existing_page_range() helper
  kasan: fix crashes on access to memory mapped by vm_map_ram()

4 years agoMerge tag 'pm-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Linus Torvalds [Thu, 19 Dec 2019 16:09:43 +0000 (08:09 -0800)]
Merge tag 'pm-5.5-rc3' of git://git./linux/kernel/git/rafael/linux-pm

Pull power management fix from Rafael Wysocki:
 "Fix a problem related to CPU offline/online and cpufreq governors that
  in some system configurations may lead to a system-wide deadlock
  during CPU online"

* tag 'pm-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  cpufreq: Avoid leaving stale IRQ work items during CPU offline

4 years agoMerge branch 'pm-cpufreq'
Rafael J. Wysocki [Thu, 19 Dec 2019 15:10:52 +0000 (16:10 +0100)]
Merge branch 'pm-cpufreq'

* pm-cpufreq:
  cpufreq: Avoid leaving stale IRQ work items during CPU offline

4 years agommc: sdhci-of-esdhc: re-implement erratum A-009204 workaround
Yangbo Lu [Thu, 19 Dec 2019 03:23:35 +0000 (11:23 +0800)]
mmc: sdhci-of-esdhc: re-implement erratum A-009204 workaround

The erratum A-009204 workaround patch was reverted because of
incorrect implementation.

8b6dc6b mmc: sdhci-of-esdhc: Revert "mmc: sdhci-of-esdhc: add
        erratum A-009204 support"

This patch is to re-implement the workaround (add a 5 ms delay
before setting SYSCTL[RSTD] to make sure all the DMA transfers
are finished).

Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Link: https://lore.kernel.org/r/20191219032335.26528-1-yangbo.lu@nxp.com
Fixes: 5dd195522562 ("mmc: sdhci-of-esdhc: add erratum A-009204 support")
Cc: stable@vger.kernel.org
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
4 years agoMerge tag 'tpmdd-next-20191219' of git://git.infradead.org/users/jjs/linux-tpmdd
Linus Torvalds [Thu, 19 Dec 2019 01:17:36 +0000 (17:17 -0800)]
Merge tag 'tpmdd-next-20191219' of git://git.infradead.org/users/jjs/linux-tpmdd

Pull tpm fixes from Jarkko Sakkinen:
 "Bunch of fixes for rc3"

* tag 'tpmdd-next-20191219' of git://git.infradead.org/users/jjs/linux-tpmdd:
  tpm/tpm_ftpm_tee: add shutdown call back
  tpm: selftest: cleanup after unseal with wrong auth/policy test
  tpm: selftest: add test covering async mode
  tpm: fix invalid locking in NONBLOCKING mode
  security: keys: trusted: fix lost handle flush
  tpm_tis: reserve chip for duration of tpm_tis_core_init
  KEYS: asymmetric: return ENOMEM if akcipher_request_alloc() fails
  KEYS: remove CONFIG_KEYS_COMPAT

4 years agotpm/tpm_ftpm_tee: add shutdown call back
Pavel Tatashin [Wed, 4 Dec 2019 15:40:38 +0000 (10:40 -0500)]
tpm/tpm_ftpm_tee: add shutdown call back

Add shutdown call back to close existing session with fTPM TA
to support kexec scenario.

Add parentheses to function names in comments as specified in kdoc.

Signed-off-by: Thirupathaiah Annapureddy <thiruan@microsoft.com>
Signed-off-by: Pavel Tatashin <pasha.tatashin@soleen.com>
Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Tested-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
4 years agoio_uring: io_wq_submit_work() should not touch req->rw
Jens Axboe [Wed, 18 Dec 2019 19:19:41 +0000 (12:19 -0700)]
io_uring: io_wq_submit_work() should not touch req->rw

I've been chasing a weird and obscure crash that was userspace stack
corruption, and finally narrowed it down to a bit flip that made a
stack address invalid. io_wq_submit_work() unconditionally flips
the req->rw.ki_flags IOCB_NOWAIT bit, but since it's a generic work
handler, this isn't valid. Normal read/write operations own that
part of the request, on other types it could be something else.

Move the IOCB_NOWAIT clear to the read/write handlers where it belongs.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
4 years agousb: xhci: Fix build warning seen with CONFIG_PM=n
Guenter Roeck [Wed, 18 Dec 2019 01:19:11 +0000 (17:19 -0800)]
usb: xhci: Fix build warning seen with CONFIG_PM=n

The following build warning is seen if CONFIG_PM is disabled.

drivers/usb/host/xhci-pci.c:498:13: warning:
unused function 'xhci_pci_shutdown'

Fixes: f2c710f7dca8 ("usb: xhci: only set D3hot for pci device")
Cc: Henry Lin <henryl@nvidia.com>
Cc: stable@vger.kernel.org # all stable releases with f2c710f7dca8
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Acked-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Link: https://lore.kernel.org/r/20191218011911.6907-1-linux@roeck-us.net
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agoio_uring: don't wait when under-submitting
Pavel Begunkov [Wed, 18 Dec 2019 16:53:45 +0000 (19:53 +0300)]
io_uring: don't wait when under-submitting

There is no reliable way to submit and wait in a single syscall, as
io_submit_sqes() may under-consume sqes (in case of an early error).
Then it will wait for not-yet-submitted requests, deadlocking the user
in most cases.

Don't wait/poll if can't submit all sqes

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
4 years agoMerge tag 'sound-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai...
Linus Torvalds [Wed, 18 Dec 2019 16:54:15 +0000 (08:54 -0800)]
Merge tag 'sound-5.5-rc3' of git://git./linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "A slightly high amount at this time, but all good and small fixes:

   - A PCM core fix that initializes the buffer properly for avoiding
     information leaks; it is a long-standing minor problem, but good to
     fix better now

   - A few ASoC core fixes for the init / cleanup ordering issues that
     surfaced after the recent refactoring

   - Lots of SOF and topology-related fixes went in, as usual as such
     hot topics

   - Several ASoC codec and platform-specific small fixes: wm89xx,
     realtek, and max98090, AMD, Intel-SST

   - A fix for the previous incomplete regression of HD-audio, now
     hitting Nvidia HDMI

   - A few HD-audio CA0132 codec fixes"

* tag 'sound-5.5-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (27 commits)
  ALSA: hda - Downgrade error message for single-cmd fallback
  ASoC: wm8962: fix lambda value
  ALSA: hda: Fix regression by strip mask fix
  ALSA: hda/ca0132 - Fix work handling in delayed HP detection
  ALSA: hda/ca0132 - Avoid endless loop
  ALSA: hda/ca0132 - Keep power on during processing DSP response
  ALSA: pcm: Avoid possible info leaks from PCM stream buffers
  ASoC: Intel: common: work-around incorrect ACPI HID for CML boards
  ASoC: SOF: Intel: split cht and byt debug window sizes
  ASoC: SOF: loader: fix snd_sof_fw_parse_ext_data
  ASoC: SOF: loader: snd_sof_fw_parse_ext_data log warning on unknown header
  ASoC: simple-card: Don't create separate link when platform is present
  ASoC: topology: Check return value for soc_tplg_pcm_create()
  ASoC: topology: Check return value for snd_soc_add_dai_link()
  ASoC: core: only flush inited work during free
  ASoC: Intel: bytcr_rt5640: Update quirk for Teclast X89
  ASoC: core: Init pcm runtime work early to avoid warnings
  ASoC: Intel: sst: Add missing include <linux/io.h>
  ASoC: max98090: fix possible race conditions
  ASoC: max98090: exit workaround earlier if PLL is locked
  ...

4 years agoiommu/dma: Relax locking in iommu_dma_prepare_msi()
Robin Murphy [Mon, 9 Dec 2019 19:47:25 +0000 (19:47 +0000)]
iommu/dma: Relax locking in iommu_dma_prepare_msi()

Since commit ece6e6f0218b ("iommu/dma-iommu: Split iommu_dma_map_msi_msg()
in two parts"), iommu_dma_prepare_msi() should no longer have to worry
about preempting itself, nor being called in atomic context at all. Thus
we can downgrade the IRQ-safe locking to a simple mutex to avoid angering
the new might_sleep() check in iommu_map().

Reported-by: Qian Cai <cai@lca.pw>
Tested-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
4 years agoiommu/vt-d: Remove incorrect PSI capability check
Lu Baolu [Wed, 20 Nov 2019 06:10:16 +0000 (14:10 +0800)]
iommu/vt-d: Remove incorrect PSI capability check

The PSI (Page Selective Invalidation) bit in the capability register
is only valid for second-level translation. Intel IOMMU supporting
scalable mode must support page/address selective IOTLB invalidation
for first-level translation. Remove the PSI capability check in SVA
cache invalidation code.

Fixes: 8744daf4b0699 ("iommu/vt-d: Remove global page flush support")
Cc: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
4 years agostaging: wfx: fix wrong error message
Jérôme Pouiller [Tue, 17 Dec 2019 16:14:40 +0000 (16:14 +0000)]
staging: wfx: fix wrong error message

The driver checks that the number of retries made by the device is
coherent with the rate policy. However, this check make sense only if
the device has returned RETRY_EXCEEDED.

Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20191217161318.31402-11-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: wfx: fix hif_set_mfp() with big endian hosts
Jérôme Pouiller [Tue, 17 Dec 2019 16:14:38 +0000 (16:14 +0000)]
staging: wfx: fix hif_set_mfp() with big endian hosts

struct hif_mib_protected_mgmt_policy is an array of u8. There is no
reason to swap its bytes.

Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20191217161318.31402-10-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: wfx: detect race condition in WEP authentication
Jérôme Pouiller [Tue, 17 Dec 2019 16:14:37 +0000 (16:14 +0000)]
staging: wfx: detect race condition in WEP authentication

Current code has a special case to handle association with WEP. Before
to rework the tx data handling, let's try to detect any possible misuse
of this code.

Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20191217161318.31402-9-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: wfx: ensure that retry policy always fallbacks to MCS0 / 1Mbps
Jérôme Pouiller [Tue, 17 Dec 2019 16:14:36 +0000 (16:14 +0000)]
staging: wfx: ensure that retry policy always fallbacks to MCS0 / 1Mbps

When not using HT mode, minstrel always includes 1Mbps as fallback rate.
But, when using HT mode, this fallback is not included. Yet, it seems
that it could save some frames. So, this patch add it unconditionally.

Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20191217161318.31402-8-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: wfx: fix rate control handling
Jérôme Pouiller [Tue, 17 Dec 2019 16:14:34 +0000 (16:14 +0000)]
staging: wfx: fix rate control handling

A tx_retry_policy (the equivalent of a list of ieee80211_tx_rate in
hardware API) is not able to include a rate multiple time. So currently,
the driver merges the identical rates from the policy provided by
minstrel (and it try to do the best choice it can in the associated
flags) before to sent it to firmware.

Until now, when rates are merged, field "count" is set to
max(count1, count2). But, it means that the sum of retries for all rates
could be far less than initial number of retries. So, this patch changes
the value of field "count" to count1 + count2. Thus, sum of all retries
for all rates stay the same.

Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20191217161318.31402-7-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: wfx: firmware does not support more than 32 total retries
Jérôme Pouiller [Tue, 17 Dec 2019 16:14:33 +0000 (16:14 +0000)]
staging: wfx: firmware does not support more than 32 total retries

The sum of all retries for a Tx frame cannot be superior to 32.

There are 4 rates at most. So this patch limits number of retries per
rate to 8.

Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20191217161318.31402-6-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: wfx: use boolean appropriately
Jérôme Pouiller [Tue, 17 Dec 2019 16:14:32 +0000 (16:14 +0000)]
staging: wfx: use boolean appropriately

The field 'uploaded' is used as a boolean, so call it a boolean.

Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20191217161318.31402-5-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: wfx: fix counter overflow
Jérôme Pouiller [Tue, 17 Dec 2019 16:14:30 +0000 (16:14 +0000)]
staging: wfx: fix counter overflow

Some weird behaviors were observed when connection is really good and
packets are small. It appears that sometime, number of packets in queues
can exceed 255 and generate an overflow in field usage_count.

Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20191217161318.31402-4-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: wfx: fix case of lack of tx_retry_policies
Jérôme Pouiller [Tue, 17 Dec 2019 16:14:29 +0000 (16:14 +0000)]
staging: wfx: fix case of lack of tx_retry_policies

In some rare cases, driver may not have any available tx_retry_policies.
In this case, the driver asks to mac80211 to stop sending data. However,
it seems that a race is possible and a few frames can be sent to the
driver. In this case, driver can't wait for free tx_retry_policies since
wfx_tx() must be atomic. So, this patch fix this case by sending these
frames with the special policy number 15.

The firmware normally use policy 15 to send internal frames (PS-poll,
beacons, etc...). So, it is not a so bad fallback.

Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20191217161318.31402-3-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: wfx: fix the cache of rate policies on interface reset
Jérôme Pouiller [Tue, 17 Dec 2019 16:14:27 +0000 (16:14 +0000)]
staging: wfx: fix the cache of rate policies on interface reset

Device and driver maintain a cache of rate policies (aka.
tx_retry_policy in hardware API).

When hif_reset() is sent to hardware, device resets its cache of rate
policies. In order to keep driver in sync, it is necessary to do the
same on driver.

Note, when driver tries to use a rate policy that has not been defined
on device, data is sent at 1Mbps. So, this patch should fix abnormal
throughput observed sometime after a reset of the interface.

Signed-off-by: Jérôme Pouiller <jerome.pouiller@silabs.com>
Link: https://lore.kernel.org/r/20191217161318.31402-2-Jerome.Pouiller@silabs.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agommc: sdhci: Add a quirk for broken command queuing
Adrian Hunter [Tue, 17 Dec 2019 09:53:49 +0000 (11:53 +0200)]
mmc: sdhci: Add a quirk for broken command queuing

Command queuing has been reported broken on some systems based on Intel
GLK. A separate patch disables command queuing in some cases.

This patch adds a quirk for broken command queuing, which enables users
with problems to disable command queuing using sdhci module parameters for
quirks.

Fixes: 8ee82bda230f ("mmc: sdhci-pci: Add CQHCI support for Intel GLK")
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20191217095349.14592-2-adrian.hunter@intel.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
4 years agommc: sdhci: Workaround broken command queuing on Intel GLK
Adrian Hunter [Tue, 17 Dec 2019 09:53:48 +0000 (11:53 +0200)]
mmc: sdhci: Workaround broken command queuing on Intel GLK

Command queuing has been reported broken on some Lenovo systems based on
Intel GLK. This is likely a BIOS issue, so disable command queuing for
Intel GLK if the BIOS vendor string is "LENOVO".

Fixes: 8ee82bda230f ("mmc: sdhci-pci: Add CQHCI support for Intel GLK")
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20191217095349.14592-1-adrian.hunter@intel.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
4 years agommc: sdhci-of-esdhc: fix P2020 errata handling
Yangbo Lu [Mon, 16 Dec 2019 03:18:42 +0000 (11:18 +0800)]
mmc: sdhci-of-esdhc: fix P2020 errata handling

Two previous patches introduced below quirks for P2020 platforms.
- SDHCI_QUIRK_RESET_AFTER_REQUEST
- SDHCI_QUIRK_BROKEN_TIMEOUT_VAL

The patches made a mistake to add them in quirks2 of sdhci_host
structure, while they were defined for quirks.
host->quirks2 |= SDHCI_QUIRK_RESET_AFTER_REQUEST;
host->quirks2 |= SDHCI_QUIRK_BROKEN_TIMEOUT_VAL;

This patch is to fix them.
host->quirks |= SDHCI_QUIRK_RESET_AFTER_REQUEST;
host->quirks |= SDHCI_QUIRK_BROKEN_TIMEOUT_VAL;

Fixes: 05cb6b2a66fa ("mmc: sdhci-of-esdhc: add erratum eSDHC-A001 and A-008358 support")
Fixes: a46e42712596 ("mmc: sdhci-of-esdhc: add erratum eSDHC5 support")
Signed-off-by: Yangbo Lu <yangbo.lu@nxp.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20191216031842.40068-1-yangbo.lu@nxp.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
4 years agolib/Kconfig.debug: fix some messed up configurations
Changbin Du [Wed, 18 Dec 2019 04:51:56 +0000 (20:51 -0800)]
lib/Kconfig.debug: fix some messed up configurations

Some configuration items are messed up during conflict resolving.  For
example, STRICT_DEVMEM should not in testing menu, but kunit should.
This patch fixes all of them.

[akpm@linux-foundation.org: coding style fixes]
Link: http://lkml.kernel.org/r/20191209155653.7509-1-changbin.du@gmail.com
Signed-off-by: Changbin Du <changbin.du@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
4 years agomm: vmscan: protect shrinker idr replace with CONFIG_MEMCG
Yang Shi [Wed, 18 Dec 2019 04:51:52 +0000 (20:51 -0800)]
mm: vmscan: protect shrinker idr replace with CONFIG_MEMCG

Since commit 0a432dcbeb32 ("mm: shrinker: make shrinker not depend on
memcg kmem"), shrinkers' idr is protected by CONFIG_MEMCG instead of
CONFIG_MEMCG_KMEM, so it makes no sense to protect shrinker idr replace
with CONFIG_MEMCG_KMEM.

And in the CONFIG_MEMCG && CONFIG_SLOB case, shrinker_idr contains only
shrinker, and it is deferred_split_shrinker.  But it is never actually
called, since idr_replace() is never compiled due to the wrong #ifdef.
The deferred_split_shrinker all the time is staying in half-registered
state, and it's never called for subordinate mem cgroups.

Link: http://lkml.kernel.org/r/1575486978-45249-1-git-send-email-yang.shi@linux.alibaba.com
Fixes: 0a432dcbeb32 ("mm: shrinker: make shrinker not depend on memcg kmem")
Signed-off-by: Yang Shi <yang.shi@linux.alibaba.com>
Reviewed-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Roman Gushchin <guro@fb.com>
Cc: <stable@vger.kernel.org> [5.4+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
4 years agokasan: don't assume percpu shadow allocations will succeed
Daniel Axtens [Wed, 18 Dec 2019 04:51:49 +0000 (20:51 -0800)]
kasan: don't assume percpu shadow allocations will succeed

syzkaller and the fault injector showed that I was wrong to assume that
we could ignore percpu shadow allocation failures.

Handle failures properly.  Merge all the allocated areas back into the
free list and release the shadow, then clean up and return NULL.  The
shadow is released unconditionally, which relies upon the fact that the
release function is able to tolerate pages not being present.

Also clean up shadows in the recovery path - currently they are not
released, which leaks a bit of memory.

Link: http://lkml.kernel.org/r/20191205140407.1874-3-dja@axtens.net
Fixes: 3c5c3cfb9ef4 ("kasan: support backing vmalloc space with real shadow memory")
Signed-off-by: Daniel Axtens <dja@axtens.net>
Reported-by: syzbot+82e323920b78d54aaed5@syzkaller.appspotmail.com
Reported-by: syzbot+59b7daa4315e07a994f1@syzkaller.appspotmail.com
Reviewed-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Qian Cai <cai@lca.pw>
Cc: Uladzislau Rezki (Sony) <urezki@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
4 years agokasan: use apply_to_existing_page_range() for releasing vmalloc shadow
Daniel Axtens [Wed, 18 Dec 2019 04:51:46 +0000 (20:51 -0800)]
kasan: use apply_to_existing_page_range() for releasing vmalloc shadow

kasan_release_vmalloc uses apply_to_page_range to release vmalloc
shadow.  Unfortunately, apply_to_page_range can allocate memory to fill
in page table entries, which is not what we want.

Also, kasan_release_vmalloc is called under free_vmap_area_lock, so if
apply_to_page_range does allocate memory, we get a sleep in atomic bug:

BUG: sleeping function called from invalid context at mm/page_alloc.c:4681
in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 15087, name:

Call Trace:
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x199/0x216 lib/dump_stack.c:118
 ___might_sleep.cold.97+0x1f5/0x238 kernel/sched/core.c:6800
 __might_sleep+0x95/0x190 kernel/sched/core.c:6753
 prepare_alloc_pages mm/page_alloc.c:4681 [inline]
 __alloc_pages_nodemask+0x3cd/0x890 mm/page_alloc.c:4730
 alloc_pages_current+0x10c/0x210 mm/mempolicy.c:2211
 alloc_pages include/linux/gfp.h:532 [inline]
 __get_free_pages+0xc/0x40 mm/page_alloc.c:4786
 __pte_alloc_one_kernel include/asm-generic/pgalloc.h:21 [inline]
 pte_alloc_one_kernel include/asm-generic/pgalloc.h:33 [inline]
 __pte_alloc_kernel+0x1d/0x200 mm/memory.c:459
 apply_to_pte_range mm/memory.c:2031 [inline]
 apply_to_pmd_range mm/memory.c:2068 [inline]
 apply_to_pud_range mm/memory.c:2088 [inline]
 apply_to_p4d_range mm/memory.c:2108 [inline]
 apply_to_page_range+0x77d/0xa00 mm/memory.c:2133
 kasan_release_vmalloc+0xa7/0xc0 mm/kasan/common.c:970
 __purge_vmap_area_lazy+0xcbb/0x1f30 mm/vmalloc.c:1313
 try_purge_vmap_area_lazy mm/vmalloc.c:1332 [inline]
 free_vmap_area_noflush+0x2ca/0x390 mm/vmalloc.c:1368
 free_unmap_vmap_area mm/vmalloc.c:1381 [inline]
 remove_vm_area+0x1cc/0x230 mm/vmalloc.c:2209
 vm_remove_mappings mm/vmalloc.c:2236 [inline]
 __vunmap+0x223/0xa20 mm/vmalloc.c:2299
 __vfree+0x3f/0xd0 mm/vmalloc.c:2356
 __vmalloc_area_node mm/vmalloc.c:2507 [inline]
 __vmalloc_node_range+0x5d5/0x810 mm/vmalloc.c:2547
 __vmalloc_node mm/vmalloc.c:2607 [inline]
 __vmalloc_node_flags mm/vmalloc.c:2621 [inline]
 vzalloc+0x6f/0x80 mm/vmalloc.c:2666
 alloc_one_pg_vec_page net/packet/af_packet.c:4233 [inline]
 alloc_pg_vec net/packet/af_packet.c:4258 [inline]
 packet_set_ring+0xbc0/0x1b50 net/packet/af_packet.c:4342
 packet_setsockopt+0xed7/0x2d90 net/packet/af_packet.c:3695
 __sys_setsockopt+0x29b/0x4d0 net/socket.c:2117
 __do_sys_setsockopt net/socket.c:2133 [inline]
 __se_sys_setsockopt net/socket.c:2130 [inline]
 __x64_sys_setsockopt+0xbe/0x150 net/socket.c:2130
 do_syscall_64+0xfa/0x780 arch/x86/entry/common.c:294
 entry_SYSCALL_64_after_hwframe+0x49/0xbe

Switch to using the apply_to_existing_page_range() helper instead, which
won't allocate memory.

[akpm@linux-foundation.org: s/apply_to_existing_pages/apply_to_existing_page_range/]
Link: http://lkml.kernel.org/r/20191205140407.1874-2-dja@axtens.net
Fixes: 3c5c3cfb9ef4 ("kasan: support backing vmalloc space with real shadow memory")
Signed-off-by: Daniel Axtens <dja@axtens.net>
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Reviewed-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Qian Cai <cai@lca.pw>
Cc: Uladzislau Rezki (Sony) <urezki@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
4 years agomm/memory.c: add apply_to_existing_page_range() helper
Daniel Axtens [Wed, 18 Dec 2019 04:51:41 +0000 (20:51 -0800)]
mm/memory.c: add apply_to_existing_page_range() helper

apply_to_page_range() takes an address range, and if any parts of it are
not covered by the existing page table hierarchy, it allocates memory to
fill them in.

In some use cases, this is not what we want - we want to be able to
operate exclusively on PTEs that are already in the tables.

Add apply_to_existing_page_range() for this.  Adjust the walker
functions for apply_to_page_range to take 'create', which switches them
between the old and new modes.

This will be used in KASAN vmalloc.

[akpm@linux-foundation.org: reduce code duplication]
[akpm@linux-foundation.org: s/apply_to_existing_pages/apply_to_existing_page_range/]
[akpm@linux-foundation.org: initialize __apply_to_page_range::err]
Link: http://lkml.kernel.org/r/20191205140407.1874-1-dja@axtens.net
Signed-off-by: Daniel Axtens <dja@axtens.net>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Uladzislau Rezki (Sony) <urezki@gmail.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Daniel Axtens <dja@axtens.net>
Cc: Qian Cai <cai@lca.pw>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
4 years agokasan: fix crashes on access to memory mapped by vm_map_ram()
Andrey Ryabinin [Wed, 18 Dec 2019 04:51:38 +0000 (20:51 -0800)]
kasan: fix crashes on access to memory mapped by vm_map_ram()

With CONFIG_KASAN_VMALLOC=y any use of memory obtained via vm_map_ram()
will crash because there is no shadow backing that memory.

Instead of sprinkling additional kasan_populate_vmalloc() calls all over
the vmalloc code, move it into alloc_vmap_area(). This will fix
vm_map_ram() and simplify the code a bit.

[aryabinin@virtuozzo.com: v2]
Link: http://lkml.kernel.org/r/20191205095942.1761-1-aryabinin@virtuozzo.comLink:
Fixes: 3c5c3cfb9ef4 ("kasan: support backing vmalloc space with real shadow memory")
Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Reviewed-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Cc: Daniel Axtens <dja@axtens.net>
Cc: Alexander Potapenko <glider@google.com>
Cc: Daniel Axtens <dja@axtens.net>
Cc: Qian Cai <cai@lca.pw>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
4 years agoio_uring: warn about unhandled opcode
Jens Axboe [Wed, 18 Dec 2019 02:45:06 +0000 (19:45 -0700)]
io_uring: warn about unhandled opcode

Now that we have all the opcodes handled in terms of command prep and
SQE reuse, add a printk_once() to warn about any potentially new and
unhandled ones.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
4 years agoio_uring: read opcode and user_data from SQE exactly once
Jens Axboe [Wed, 18 Dec 2019 02:53:05 +0000 (19:53 -0700)]
io_uring: read opcode and user_data from SQE exactly once

If we defer a request, we can't be reading the opcode again. Ensure that
the user_data and opcode fields are stable. For the user_data we already
have a place for it, for the opcode we can fill a one byte hold and store
that as well. For both of them, assign them when we originally read the
SQE in io_get_sqring(). Any code that uses sqe->opcode or sqe->user_data
is switched to req->opcode and req->user_data.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
4 years agoio_uring: make IORING_OP_TIMEOUT_REMOVE deferrable
Jens Axboe [Wed, 18 Dec 2019 01:50:29 +0000 (18:50 -0700)]
io_uring: make IORING_OP_TIMEOUT_REMOVE deferrable

If we defer this command as part of a link, we have to make sure that
the SQE data has been read upfront. Integrate the timeout remove op into
the prep handling to make it safe for SQE reuse.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
4 years agoio_uring: make IORING_OP_CANCEL_ASYNC deferrable
Jens Axboe [Wed, 18 Dec 2019 01:45:56 +0000 (18:45 -0700)]
io_uring: make IORING_OP_CANCEL_ASYNC deferrable

If we defer this command as part of a link, we have to make sure that
the SQE data has been read upfront. Integrate the async cancel op into
the prep handling to make it safe for SQE reuse.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
4 years agoio_uring: make IORING_POLL_ADD and IORING_POLL_REMOVE deferrable
Jens Axboe [Wed, 18 Dec 2019 01:40:57 +0000 (18:40 -0700)]
io_uring: make IORING_POLL_ADD and IORING_POLL_REMOVE deferrable

If we defer these commands as part of a link, we have to make sure that
the SQE data has been read upfront. Integrate the poll add/remove into
the prep handling to make it safe for SQE reuse.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
4 years agoio_uring: make HARDLINK imply LINK
Pavel Begunkov [Tue, 17 Dec 2019 17:57:05 +0000 (20:57 +0300)]
io_uring: make HARDLINK imply LINK

The rules are as follows, if IOSQE_IO_HARDLINK is specified, then it's a
link and there is no need to set IOSQE_IO_LINK separately, though it
could be there. Add proper check and ensure that IOSQE_IO_HARDLINK
implies IOSQE_IO_LINK.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
4 years agoio_uring: any deferred command must have stable sqe data
Jens Axboe [Mon, 16 Dec 2019 18:55:28 +0000 (11:55 -0700)]
io_uring: any deferred command must have stable sqe data

We're currently not retaining sqe data for accept, fsync, and
sync_file_range. None of these commands need data outside of what
is directly provided, hence it can't go stale when the request is
deferred. However, it can get reused, if an application reuses
SQE entries.

Ensure that we retain the information we need and only read the sqe
contents once, off the submission path. Most of this is just moving
code into a prep and finish function.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
4 years agoio_uring: remove 'sqe' parameter to the OP helpers that take it
Jens Axboe [Tue, 10 Dec 2019 21:38:45 +0000 (14:38 -0700)]
io_uring: remove 'sqe' parameter to the OP helpers that take it

We pass in req->sqe for all of them, no need to pass it in as the
request is always passed in. This is a necessary prep patch to be
able to cleanup/fix the request prep path.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
4 years agoio_uring: fix pre-prepped issue with force_nonblock == true
Jens Axboe [Mon, 16 Dec 2019 05:13:43 +0000 (22:13 -0700)]
io_uring: fix pre-prepped issue with force_nonblock == true

Some of these code paths assume that any force_nonblock == true issue
is not prepped, but that's not true if we did prep as part of link setup
earlier. Check if we already have an async context allocate before
setting up a new one.

Cleanup the async context setup in general, we have a lot of duplicated
code there.

Fixes: 03b1230ca12a ("io_uring: ensure async punted sendmsg/recvmsg requests copy data")
Fixes: f67676d160c6 ("io_uring: ensure async punted read/write requests copy iovec")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
4 years agoio-wq: re-add io_wq_current_is_worker()
Jens Axboe [Tue, 17 Dec 2019 21:13:37 +0000 (14:13 -0700)]
io-wq: re-add io_wq_current_is_worker()

This reverts commit 8cdda87a4414, we now have several use csaes for this
helper. Reinstate it.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
4 years agoMerge tag 'for-5.5-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave...
Linus Torvalds [Tue, 17 Dec 2019 21:27:02 +0000 (13:27 -0800)]
Merge tag 'for-5.5-rc2-tag' of git://git./linux/kernel/git/kdave/linux

Pull btrfs fixes from David Sterba:
 "A mix of regression fixes and regular fixes for stable trees:

   - fix swapped error messages for qgroup enable/rescan

   - fixes for NO_HOLES feature with clone range

   - fix deadlock between iget/srcu lock/synchronize srcu while freeing
     an inode

   - fix double lock on subvolume cross-rename

   - tree log fixes
      * fix missing data checksums after replaying a log tree
      * also teach tree-checker about this problem
      * skip log replay on orphaned roots

   - fix maximum devices constraints for RAID1C -3 and -4

   - send: don't print warning on read-only mount regarding orphan
     cleanup

   - error handling fixes"

* tag 'for-5.5-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: send: remove WARN_ON for readonly mount
  btrfs: do not leak reloc root if we fail to read the fs root
  btrfs: skip log replay on orphaned roots
  btrfs: handle ENOENT in btrfs_uuid_tree_iterate
  btrfs: abort transaction after failed inode updates in create_subvol
  Btrfs: fix hole extent items with a zero size after range cloning
  Btrfs: fix removal logic of the tree mod log that leads to use-after-free issues
  Btrfs: make tree checker detect checksum items with overlapping ranges
  Btrfs: fix missing data checksums after replaying a log tree
  btrfs: return error pointer from alloc_test_extent_buffer
  btrfs: fix devs_max constraints for raid1c3 and raid1c4
  btrfs: tree-checker: Fix error format string for size_t
  btrfs: don't double lock the subvol_sem for rename exchange
  btrfs: handle error in btrfs_cache_block_group
  btrfs: do not call synchronize_srcu() in inode_tree_del
  Btrfs: fix cloning range with a hole when using the NO_HOLES feature
  btrfs: Fix error messages in qgroup_rescan_init

4 years agoearly init: fix error handling when opening /dev/console
Linus Torvalds [Tue, 17 Dec 2019 21:10:11 +0000 (13:10 -0800)]
early init: fix error handling when opening /dev/console

The comment says "this should never fail", but it definitely can fail
when you have odd initial boot filesystems, or kernel configurations.

So get the error handling right: filp_open() returns an error pointer.

Reported-by: Jesse Barnes <jsbarnes@google.com>
Reported-by: youling 257 <youling257@gmail.com>
Fixes: 8243186f0cc7 ("fs: remove ksys_dup()")
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
4 years agoMerge tag 'regulator-fix-v5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Tue, 17 Dec 2019 21:08:41 +0000 (13:08 -0800)]
Merge tag 'regulator-fix-v5.5-rc2' of git://git./linux/kernel/git/broonie/regulator

Pull regulator fixes from Mark Brown:
 "A small set of fixes for mostly minor issues here, the only real code
  ones are Wen Yang's fixes for error handling in the core and Christian
  Marussi's list_voltage() change which is a fix for disruptively bad
  performance for regulators with continuous voltage control (which are
  rare)"

* tag 'regulator-fix-v5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:
  regulator: rn5t618: fix module aliases
  regulator: max77650: add of_match table
  regulator: core: avoid unneeded .list_voltage calls
  regulator: s5m8767: Fix a warning message
  regulator: core: fix regulator_register() error paths to properly release rdev
  regulator: fix use after free issue

4 years agoMerge tag 'spi-fix-v5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi
Linus Torvalds [Tue, 17 Dec 2019 21:06:31 +0000 (13:06 -0800)]
Merge tag 'spi-fix-v5.5-rc2' of git://git./linux/kernel/git/broonie/spi

Pull spi fixes from Mark Brown:
 "A relatively large set of fixes here, the biggest part of it is for
  fallout from the GPIO descriptor rework that affected several of the
  devices with usable native chip select support. There's also some new
  PCI IDs for Intel Jasper Lake devices.

  The conversion to platform_get_irq() in the fsl driver is an
  incremental fix for build errors introduced on SPARC by the earlier
  fix for error handling in probe in that driver"

* tag 'spi-fix-v5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
  spi: fsl: use platform_get_irq() instead of of_irq_to_resource()
  spi: nxp-fspi: Ensure width is respected in spi-mem operations
  spi: spi-ti-qspi: Fix a bug when accessing non default CS
  spi: fsl: don't map irq during probe
  spi: spi-cavium-thunderx: Add missing pci_release_regions()
  spi: sprd: Fix the incorrect SPI register
  gpiolib: of: Make of_gpio_spi_cs_get_count static
  spi: fsl: Handle the single hardwired chipselect case
  gpio: Handle counting of Freescale chipselects
  spi: fsl: Fix GPIO descriptor support
  spi: dw: Correct handling of native chipselect
  spi: cadence: Correct handling of native chipselect
  spi: pxa2xx: Add support for Intel Jasper Lake

4 years agoMerge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Tue, 17 Dec 2019 19:17:03 +0000 (11:17 -0800)]
Merge branch 'x86-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull x86 fix from Ingo Molnar:
 "Fix kexec booting with certain EFI memory map layouts"

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/efi: Update e820 with reserved EFI boot services data to fix kexec breakage

4 years agoMerge branch 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Tue, 17 Dec 2019 19:11:08 +0000 (11:11 -0800)]
Merge branch 'timers-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull timer fixes from Ingo Molnar:
 "Add HPET quirks for the Intel 'Coffee Lake H' and 'Ice Lake' platforms"

* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/intel: Disable HPET on Intel Ice Lake platforms
  x86/intel: Disable HPET on Intel Coffee Lake H platforms

4 years agoMerge branch 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Tue, 17 Dec 2019 19:09:05 +0000 (11:09 -0800)]
Merge branch 'sched-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull scheduler fix from Ingo Molnar:
 "Fix the guest-nice cpustat values in /proc"

* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/cputime, proc/stat: Fix incorrect guest nice cpustat value

4 years agoMerge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Tue, 17 Dec 2019 19:03:57 +0000 (11:03 -0800)]
Merge branch 'perf-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull perf tooling fixes from Ingo  Molnar:
 "These are all perf tooling changes: most of them are fixes.

  Note that the large CPU count related fixes go beyond regression
  fixes, but the IPI-flood symptoms are severe enough that I think
  justifies their inclusion"

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (38 commits)
  perf vendor events s390: Remove name from L1D_RO_EXCL_WRITES description
  perf vendor events s390: Fix counter long description for DTLB1_GPAGE_WRITES
  libtraceevent: Allow custom libdir path
  perf header: Fix false warning when there are no duplicate cache entries
  perf metricgroup: Fix printing event names of metric group with multiple events
  perf/x86/pmu-events: Fix Kernel_Utilization metric
  perf top: Do not bail out when perf_env__read_cpuid() returns ENOSYS
  perf arch: Make the default get_cpuid() return compatible error
  tools headers kvm: Sync linux/kvm.h with the kernel sources
  tools headers UAPI: Update tools's copy of drm.h headers
  tools headers UAPI: Sync drm/i915_drm.h with the kernel sources
  perf inject: Fix processing of ID index for injected instruction tracing
  perf report: Bail out --mem-mode if mem info is not available
  perf report: Make -F more strict like -s
  perf report/top TUI: Replace pr_err() with ui__error()
  libtraceevent: Copy pkg-config file to output folder when using O=
  libtraceevent: Fix lib installation with O=
  perf kvm: Clarify the 'perf kvm' -i and -o command line options
  tools arch x86: Sync asm/cpufeatures.h with the kernel sources
  perf beauty: Add CLEAR_SIGHAND support for clone's flags arg
  ...

4 years agoMerge branch 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Tue, 17 Dec 2019 19:00:46 +0000 (11:00 -0800)]
Merge branch 'locking-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull locking fixes from Ingo Molnar:
 "Tone down mutex debugging complaints, and annotate/fix spinlock
  debugging data accesses for KCSAN"

* 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  Revert "locking/mutex: Complain upon mutex API misuse in IRQ contexts"
  locking/spinlock/debug: Fix various data races

4 years agoMerge branch 'efi-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Tue, 17 Dec 2019 18:39:55 +0000 (10:39 -0800)]
Merge branch 'efi-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull EFI fixes from Ingo Molnar:
 "Protect presistent EFI memory reservations from kexec, fix EFIFB early
  console, EFI stub graphics output fixes and other misc fixes."

* 'efi-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  efi: Don't attempt to map RCI2 config table if it doesn't exist
  efi/earlycon: Remap entire framebuffer after page initialization
  efi: Fix efi_loaded_image_t::unload type
  efi/gop: Fix memory leak in __gop_query32/64()
  efi/gop: Return EFI_SUCCESS if a usable GOP was found
  efi/gop: Return EFI_NOT_FOUND if there are no usable GOPs
  efi/memreserve: Register reservations as 'reserved' in /proc/iomem

4 years agorandom: don't forget compat_ioctl on urandom
Jason A. Donenfeld [Tue, 17 Dec 2019 17:24:55 +0000 (18:24 +0100)]
random: don't forget compat_ioctl on urandom

Recently, there's been some compat ioctl cleanup, in which large
hardcoded lists were replaced with compat_ptr_ioctl. One of these
changes involved removing the random.c hardcoded list entries and adding
a compat ioctl function pointer to the random.c fops. In the process,
urandom was forgotten about, so this commit fixes that oversight.

Fixes: 507e4e2b430b ("compat_ioctl: remove /dev/random commands")
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Link: https://lore.kernel.org/r/20191217172455.186395-1-Jason@zx2c4.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agousbip: Fix error path of vhci_recv_ret_submit()
Suwan Kim [Fri, 13 Dec 2019 02:30:55 +0000 (11:30 +0900)]
usbip: Fix error path of vhci_recv_ret_submit()

If a transaction error happens in vhci_recv_ret_submit(), event
handler closes connection and changes port status to kick hub_event.
Then hub tries to flush the endpoint URBs, but that causes infinite
loop between usb_hub_flush_endpoint() and vhci_urb_dequeue() because
"vhci_priv" in vhci_urb_dequeue() was already released by
vhci_recv_ret_submit() before a transmission error occurred. Thus,
vhci_urb_dequeue() terminates early and usb_hub_flush_endpoint()
continuously calls vhci_urb_dequeue().

The root cause of this issue is that vhci_recv_ret_submit()
terminates early without giving back URB when transaction error
occurs in vhci_recv_ret_submit(). That causes the error URB to still
be linked at endpoint list without “vhci_priv".

So, in the case of transaction error in vhci_recv_ret_submit(),
unlink URB from the endpoint, insert proper error code in
urb->status and give back URB.

Reported-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Tested-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Signed-off-by: Suwan Kim <suwan.kim027@gmail.com>
Cc: stable <stable@vger.kernel.org>
Acked-by: Shuah Khan <skhan@linuxfoundation.org>
Link: https://lore.kernel.org/r/20191213023055.19933-3-suwan.kim027@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agousbip: Fix receive error in vhci-hcd when using scatter-gather
Suwan Kim [Fri, 13 Dec 2019 02:30:54 +0000 (11:30 +0900)]
usbip: Fix receive error in vhci-hcd when using scatter-gather

When vhci uses SG and receives data whose size is smaller than SG
buffer size, it tries to receive more data even if it acutally
receives all the data from the server. If then, it erroneously adds
error event and triggers connection shutdown.

vhci-hcd should check if it received all the data even if there are
more SG entries left. So, check if it receivces all the data from
the server in for_each_sg() loop.

Fixes: ea44d190764b ("usbip: Implement SG support to vhci-hcd and stub driver")
Reported-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Tested-by: Marek Marczykowski-Górecki <marmarek@invisiblethingslab.com>
Signed-off-by: Suwan Kim <suwan.kim027@gmail.com>
Acked-by: Shuah Khan <skhan@linuxfoundation.org>
Cc: stable <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20191213023055.19933-2-suwan.kim027@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agoUSB: EHCI: Do not return -EPIPE when hub is disconnected
Erkka Talvitie [Wed, 11 Dec 2019 08:08:39 +0000 (10:08 +0200)]
USB: EHCI: Do not return -EPIPE when hub is disconnected

When disconnecting a USB hub that has some child device(s) connected to it
(such as a USB mouse), then the stack tries to clear halt and
reset device(s) which are _already_ physically disconnected.

The issue has been reproduced with:

CPU: IMX6D5EYM10AD or MCIMX6D5EYM10AE.
SW: U-Boot 2019.07 and kernel 4.19.40.

CPU: HP Proliant Microserver Gen8.
SW: Linux version 4.2.3-300.fc23.x86_64

In this situation there will be error bit for MMF active yet the
CERR equals EHCI_TUNE_CERR + halt. Existing implementation
interprets this as a stall [1] (chapter 8.4.5).

The possible conditions when the MMF will be active + halt
can be found from [2] (Table 4-13).

Fix for the issue is to check whether MMF is active and PID Code is
IN before checking for the stall. If these conditions are true then
it is not a stall.

What happens after the fix is that when disconnecting a hub with
attached device(s) the situation is not interpret as a stall.

[1] [https://www.usb.org/document-library/usb-20-specification, usb_20.pdf]
[2] [https://www.intel.com/content/dam/www/public/us/en/documents/
     technical-specifications/ehci-specification-for-usb.pdf]

Signed-off-by: Erkka Talvitie <erkka.talvitie@vincit.fi>
Reviewed-by: Alan Stern <stern@rowland.harvard.edu>
Cc: stable <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/ef70941d5f349767f19c0ed26b0dd9eed8ad81bb.1576050523.git.erkka.talvitie@vincit.fi
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agousb: typec: fusb302: Fix an undefined reference to 'extcon_get_state'
zhong jiang [Fri, 13 Dec 2019 12:16:18 +0000 (20:16 +0800)]
usb: typec: fusb302: Fix an undefined reference to 'extcon_get_state'

Fixes the following compile error:

drivers/usb/typec/tcpm/fusb302.o: In function `tcpm_get_current_limit':
fusb302.c:(.text+0x3ee): undefined reference to `extcon_get_state'
fusb302.c:(.text+0x422): undefined reference to `extcon_get_state'
fusb302.c:(.text+0x450): undefined reference to `extcon_get_state'
fusb302.c:(.text+0x48c): undefined reference to `extcon_get_state'
drivers/usb/typec/tcpm/fusb302.o: In function `fusb302_probe':
fusb302.c:(.text+0x980): undefined reference to `extcon_get_extcon_dev'
make: *** [vmlinux] Error 1

It is because EXTCON is build as a module, but FUSB302 is not.

Suggested-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Signed-off-by: zhong jiang <zhongjiang@huawei.com>
Acked-by: Heikki Krogerus <heikki.krogerus@linux.intel.com>
Reviewed-by: Guenter Roeck <linux@roeck-us.net>
Link: https://lore.kernel.org/r/1576239378-50795-1-git-send-email-zhongjiang@huawei.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agointel_th: msu: Fix window switching without windows
Alexander Shishkin [Tue, 17 Dec 2019 11:55:27 +0000 (13:55 +0200)]
intel_th: msu: Fix window switching without windows

Commit 6cac7866c2741 ("intel_th: msu: Add a sysfs attribute to trigger
window switch") adds a NULL pointer dereference in the case when there are
no windows allocated:

> BUG: kernel NULL pointer dereference, address: 0000000000000000
> #PF: supervisor read access in kernel mode
> #PF: error_code(0x0000) - not-present page
> PGD 0 P4D 0
> Oops: 0000 1 SMP
> CPU: 5 PID: 1110 Comm: bash Not tainted 5.5.0-rc1+ #1
> RIP: 0010:msc_win_switch+0xa/0x80 [intel_th_msu]
> Call Trace:
> ? win_switch_store+0x9b/0xc0 [intel_th_msu]
> dev_attr_store+0x17/0x30
> sysfs_kf_write+0x3e/0x50
> kernfs_fop_write+0xda/0x1b0
> __vfs_write+0x1b/0x40
> vfs_write+0xb9/0x1a0
> ksys_write+0x67/0xe0
> __x64_sys_write+0x1a/0x20
> do_syscall_64+0x57/0x1d0
> entry_SYSCALL_64_after_hwframe+0x44/0xa9

Fix that by disallowing window switching with multiwindow buffers without
windows.

Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Fixes: 6cac7866c274 ("intel_th: msu: Add a sysfs attribute to trigger window switch")
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Reported-by: Ammy Yi <ammy.yi@intel.com>
Tested-by: Ammy Yi <ammy.yi@intel.com>
Cc: stable@vger.kernel.org # v5.2+
Link: https://lore.kernel.org/r/20191217115527.74383-5-alexander.shishkin@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agointel_th: Fix freeing IRQs
Alexander Shishkin [Tue, 17 Dec 2019 11:55:26 +0000 (13:55 +0200)]
intel_th: Fix freeing IRQs

Commit aac8da65174a ("intel_th: msu: Start handling IRQs") implicitly
relies on the use of devm_request_irq() to subsequently free the irqs on
device removal, but in case of the pci_free_irq_vectors() API, the
handlers need to be freed before it is called. Therefore, at the moment
the driver's remove path trips a BUG_ON(irq_has_action()):

> kernel BUG at drivers/pci/msi.c:375!
> invalid opcode: 0000 1 SMP
> CPU: 2 PID: 818 Comm: rmmod Not tainted 5.5.0-rc1+ #1
> RIP: 0010:free_msi_irqs+0x67/0x1c0
> pci_disable_msi+0x116/0x150
> pci_free_irq_vectors+0x1b/0x20
> intel_th_pci_remove+0x22/0x30 [intel_th_pci]
> pci_device_remove+0x3e/0xb0
> device_release_driver_internal+0xf0/0x1c0
> driver_detach+0x4c/0x8f
> bus_remove_driver+0x5c/0xd0
> driver_unregister+0x31/0x50
> pci_unregister_driver+0x40/0x90
> intel_th_pci_driver_exit+0x10/0xad6 [intel_th_pci]
> __x64_sys_delete_module+0x147/0x290
> ? exit_to_usermode_loop+0xd7/0x120
> do_syscall_64+0x57/0x1b0
> entry_SYSCALL_64_after_hwframe+0x44/0xa9

Fix this by explicitly freeing irqs before freeing the vectors. We keep
using the devm_* variants because they are still useful in early error
paths.

Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Fixes: aac8da65174a ("intel_th: msu: Start handling IRQs")
Reported-by: Ammy Yi <ammy.yi@intel.com>
Tested-by: Ammy Yi <ammy.yi@intel.com>
Cc: stable@vger.kernel.org # v5.2+
Link: https://lore.kernel.org/r/20191217115527.74383-4-alexander.shishkin@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agointel_th: pci: Add Elkhart Lake SOC support
Alexander Shishkin [Tue, 17 Dec 2019 11:55:25 +0000 (13:55 +0200)]
intel_th: pci: Add Elkhart Lake SOC support

This adds support for Intel Trace Hub in Elkhart Lake.

Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20191217115527.74383-3-alexander.shishkin@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agointel_th: pci: Add Comet Lake PCH-V support
Alexander Shishkin [Tue, 17 Dec 2019 11:55:24 +0000 (13:55 +0200)]
intel_th: pci: Add Comet Lake PCH-V support

This adds Intel(R) Trace Hub PCI ID for Comet Lake PCH-V.

Signed-off-by: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20191217115527.74383-2-alexander.shishkin@linux.intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agoMerge tag 'asoc-fix-v5.5-rc2' of https://git.kernel.org/pub/scm/linux/kernel/git...
Takashi Iwai [Tue, 17 Dec 2019 13:18:32 +0000 (14:18 +0100)]
Merge tag 'asoc-fix-v5.5-rc2' of https://git./linux/kernel/git/broonie/sound into for-linus

ASoC: Fixes for v5.5

A collection of fixes since the merge window, mostly driver specific but
there's a few in the core that clean up fallout from the refactorings
done in the last cycle.

4 years agotty/serial: atmel: fix out of range clock divider handling
David Engraf [Mon, 16 Dec 2019 08:54:03 +0000 (09:54 +0100)]
tty/serial: atmel: fix out of range clock divider handling

Use MCK_DIV8 when the clock divider is > 65535. Unfortunately the mode
register was already written thus the clock selection is ignored.

Fix by doing the baud rate calulation before setting the mode.

Fixes: 5bf5635ac170 ("tty/serial: atmel: add fractional baud rate support")
Signed-off-by: David Engraf <david.engraf@sysgo.com>
Acked-by: Ludovic Desroches <ludovic.desroches@microchip.com>
Acked-by: Richard Genoud <richard.genoud@gmail.com>
Cc: stable <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20191216085403.17050-1-david.engraf@sysgo.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agotty: link tty and port before configuring it as console
Sudip Mukherjee [Thu, 12 Dec 2019 13:16:02 +0000 (13:16 +0000)]
tty: link tty and port before configuring it as console

There seems to be a race condition in tty drivers and I could see on
many boot cycles a NULL pointer dereference as tty_init_dev() tries to
do 'tty->port->itty = tty' even though tty->port is NULL.
'tty->port' will be set by the driver and if the driver has not yet done
it before we open the tty device we can get to this situation. By adding
some extra debug prints, I noticed that:

6.650130: uart_add_one_port
6.663849: register_console
6.664846: tty_open
6.674391: tty_init_dev
6.675456: tty_port_link_device

uart_add_one_port() registers the console, as soon as it registers, the
userspace tries to use it and that leads to tty_open() but
uart_add_one_port() has not yet done tty_port_link_device() and so
tty->port is not yet configured when control reaches tty_init_dev().

Further look into the code and tty_port_link_device() is done by
uart_add_one_port(). After registering the console uart_add_one_port()
will call tty_port_register_device_attr_serdev() and
tty_port_link_device() is called from this.

Call add tty_port_link_device() before uart_configure_port() is done and
add a check in tty_port_link_device() so that it only links the port if
it has not been done yet.

Suggested-by: Jiri Slaby <jslaby@suse.com>
Signed-off-by: Sudip Mukherjee <sudipm.mukherjee@gmail.com>
Cc: stable <stable@vger.kernel.org>
Link: https://lore.kernel.org/r/20191212131602.29504-1-sudipm.mukherjee@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agoMerge tag 'perf-urgent-for-mingo-5.5-20191216' of git://git.kernel.org/pub/scm/linux...
Ingo Molnar [Tue, 17 Dec 2019 11:28:02 +0000 (12:28 +0100)]
Merge tag 'perf-urgent-for-mingo-5.5-20191216' of git://git./linux/kernel/git/acme/linux into perf/urgent

Pull perf/urgent fixes:

perf top:

 Arnaldo Carvalho de Melo:

 - Do not bail out when perf_env__read_cpuid() returns ENOSYS, which
   has been reported happening on aarch64.

perf metricgroup:

  Kajol Jain:

  - Fix printing event names of metric group with multiple events

vendor events:

x86:

  Ravi Bangoria:

  - Fix Kernel_Utilization metric.

s390:

  Ed Maste:

  - Fix counter long description for DTLB1_GPAGE_WRITES and L1D_RO_EXCL_WRITES.

perf header:

  Michael Petlan:

  - Fix false warning when there are no duplicate cache entries

libtraceevent:

  Sudip Mukherjee:

  - Allow custom libdir path

API headers:

  Arnaldo Carvalho de Melo:

  - Sync linux/kvm.h with the kernel sources.

Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
4 years agoiommu/vt-d: Allocate reserved region for ISA with correct permission
Jerry Snitselaar [Fri, 13 Dec 2019 05:36:42 +0000 (22:36 -0700)]
iommu/vt-d: Allocate reserved region for ISA with correct permission

Currently the reserved region for ISA is allocated with no
permissions. If a dma domain is being used, mapping this region will
fail. Set the permissions to DMA_PTE_READ|DMA_PTE_WRITE.

Cc: Joerg Roedel <jroedel@suse.de>
Cc: Lu Baolu <baolu.lu@linux.intel.com>
Cc: iommu@lists.linux-foundation.org
Cc: stable@vger.kernel.org # v5.3+
Fixes: d850c2ee5fe2 ("iommu/vt-d: Expose ISA direct mapping region via iommu_get_resv_regions")
Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
Acked-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
4 years agoiommu: set group default domain before creating direct mappings
Jerry Snitselaar [Tue, 10 Dec 2019 18:56:06 +0000 (11:56 -0700)]
iommu: set group default domain before creating direct mappings

iommu_group_create_direct_mappings uses group->default_domain, but
right after it is called, request_default_domain_for_dev calls
iommu_domain_free for the default domain, and sets the group default
domain to a different domain. Move the
iommu_group_create_direct_mappings call to after the group default
domain is set, so the direct mappings get associated with that domain.

Cc: Joerg Roedel <jroedel@suse.de>
Cc: Lu Baolu <baolu.lu@linux.intel.com>
Cc: iommu@lists.linux-foundation.org
Cc: stable@vger.kernel.org
Fixes: 7423e01741dd ("iommu: Add API to request DMA domain for device")
Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
4 years agoiommu/vt-d: Fix dmar pte read access not set error
Lu Baolu [Wed, 11 Dec 2019 01:40:15 +0000 (09:40 +0800)]
iommu/vt-d: Fix dmar pte read access not set error

If the default DMA domain of a group doesn't fit a device, it
will still sit in the group but use a private identity domain.
When map/unmap/iova_to_phys come through iommu API, the driver
should still serve them, otherwise, other devices in the same
group will be impacted. Since identity domain has been mapped
with the whole available memory space and RMRRs, we don't need
to worry about the impact on it.

Link: https://www.spinics.net/lists/iommu/msg40416.html
Cc: Jerry Snitselaar <jsnitsel@redhat.com>
Reported-by: Jerry Snitselaar <jsnitsel@redhat.com>
Fixes: 942067f1b6b97 ("iommu/vt-d: Identify default domains replaced with private")
Cc: stable@vger.kernel.org # v5.3+
Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com>
Tested-by: Jerry Snitselaar <jsnitsel@redhat.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
4 years agoiommu/vt-d: Set ISA bridge reserved region as relaxable
Alex Williamson [Wed, 11 Dec 2019 20:28:29 +0000 (13:28 -0700)]
iommu/vt-d: Set ISA bridge reserved region as relaxable

Commit d850c2ee5fe2 ("iommu/vt-d: Expose ISA direct mapping region via
iommu_get_resv_regions") created a direct-mapped reserved memory region
in order to replace the static identity mapping of the ISA address
space, where the latter was then removed in commit df4f3c603aeb
("iommu/vt-d: Remove static identity map code").  According to the
history of this code and the Kconfig option surrounding it, this direct
mapping exists for the benefit of legacy ISA drivers that are not
compatible with the DMA API.

In conjuntion with commit 9b77e5c79840 ("vfio/type1: check dma map
request is within a valid iova range") this change introduced a
regression where the vfio IOMMU backend enforces reserved memory regions
per IOMMU group, preventing userspace from creating IOMMU mappings
conflicting with prescribed reserved regions.  A necessary prerequisite
for the vfio change was the introduction of "relaxable" direct mappings
introduced by commit adfd37382090 ("iommu: Introduce
IOMMU_RESV_DIRECT_RELAXABLE reserved memory regions").  These relaxable
direct mappings provide the same identity mapping support in the default
domain, but also indicate that the reservation is software imposed and
may be relaxed under some conditions, such as device assignment.

Convert the ISA bridge direct-mapped reserved region to relaxable to
reflect that the restriction is self imposed and need not be enforced
by drivers such as vfio.

Fixes: 1c5c59fbad20 ("iommu/vt-d: Differentiate relaxable and non relaxable RMRRs")
Cc: stable@vger.kernel.org # v5.3+
Link: https://lore.kernel.org/linux-iommu/20191211082304.2d4fab45@x1.home
Reported-by: cprt <cprt@protonmail.com>
Tested-by: cprt <cprt@protonmail.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Acked-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Jerry Snitselaar <jsnitsel@redhat.com>
Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
4 years agotpm: selftest: cleanup after unseal with wrong auth/policy test
Tadeusz Struk [Thu, 12 Dec 2019 17:48:59 +0000 (09:48 -0800)]
tpm: selftest: cleanup after unseal with wrong auth/policy test

Unseal with wrong auth or wrong policy test affects DA lockout
and eventually causes the tests to fail with:
"ProtocolError: TPM_RC_LOCKOUT: rc=0x00000921"
when the tests run multiple times.
Send tpm clear command after the test to reset the DA counters.

Signed-off-by: Tadeusz Struk <tadeusz.struk@intel.com>
Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
4 years agotpm: selftest: add test covering async mode
Tadeusz Struk [Thu, 12 Dec 2019 17:48:53 +0000 (09:48 -0800)]
tpm: selftest: add test covering async mode

Add a test that sends a tpm cmd in an async mode.
Currently there is a gap in test coverage with regards
to this functionality.

Signed-off-by: Tadeusz Struk <tadeusz.struk@intel.com>
Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
4 years agotpm: fix invalid locking in NONBLOCKING mode
Tadeusz Struk [Thu, 12 Dec 2019 17:48:47 +0000 (09:48 -0800)]
tpm: fix invalid locking in NONBLOCKING mode

When an application sends TPM commands in NONBLOCKING mode
the driver holds chip->tpm_mutex returning from write(),
which triggers: "WARNING: lock held when returning to user space".
To fix this issue the driver needs to release the mutex before
returning and acquire it again in tpm_dev_async_work() before
sending the command.

Cc: stable@vger.kernel.org
Fixes: 9e1b74a63f776 (tpm: add support for nonblocking operation)
Reported-by: Jeffrin Jose T <jeffrin@rajagiritech.edu.in>
Tested-by: Jeffrin Jose T <jeffrin@rajagiritech.edu.in>
Signed-off-by: Tadeusz Struk <tadeusz.struk@intel.com>
Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
4 years agoiommu/dma: Rationalise types for DMA masks
Robin Murphy [Wed, 11 Dec 2019 18:33:26 +0000 (18:33 +0000)]
iommu/dma: Rationalise types for DMA masks

Since iommu_dma_alloc_iova() combines incoming masks with the u64 bus
limit, it makes more sense to pass them around in their native u64
rather than converting to dma_addr_t early. Do that, and resolve the
remaining type discrepancy against the domain geometry with a cheeky
cast to keep things simple.

Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Tested-by: Nathan Chancellor <natechancellor@gmail.com> # build
Reviewed-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
4 years agoiommu/iova: Init the struct iova to fix the possible memleak
Xiaotao Yin [Tue, 10 Dec 2019 04:27:04 +0000 (12:27 +0800)]
iommu/iova: Init the struct iova to fix the possible memleak

During ethernet(Marvell octeontx2) set ring buffer test:
ethtool -G eth1 rx <rx ring size> tx <tx ring size>
following kmemleak will happen sometimes:

unreferenced object 0xffff000b85421340 (size 64):
  comm "ethtool", pid 867, jiffies 4295323539 (age 550.500s)
  hex dump (first 64 bytes):
    80 13 42 85 0b 00 ff ff ff ff ff ff ff ff ff ff  ..B.............
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    ff ff ff ff ff ff ff ff 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<000000001b204ddf>] kmem_cache_alloc+0x1b0/0x350
    [<00000000d9ef2e50>] alloc_iova+0x3c/0x168
    [<00000000ea30f99d>] alloc_iova_fast+0x7c/0x2d8
    [<00000000b8bb2f1f>] iommu_dma_alloc_iova.isra.0+0x12c/0x138
    [<000000002f1a43b5>] __iommu_dma_map+0x8c/0xf8
    [<00000000ecde7899>] iommu_dma_map_page+0x98/0xf8
    [<0000000082004e59>] otx2_alloc_rbuf+0xf4/0x158
    [<000000002b107f6b>] otx2_rq_aura_pool_init+0x110/0x270
    [<00000000c3d563c7>] otx2_open+0x15c/0x734
    [<00000000a2f5f3a8>] otx2_dev_open+0x3c/0x68
    [<00000000456a98b5>] otx2_set_ringparam+0x1ac/0x1d4
    [<00000000f2fbb819>] dev_ethtool+0xb84/0x2028
    [<0000000069b67c5a>] dev_ioctl+0x248/0x3a0
    [<00000000af38663a>] sock_ioctl+0x280/0x638
    [<000000002582384c>] do_vfs_ioctl+0x8b0/0xa80
    [<000000004e1a2c02>] ksys_ioctl+0x84/0xb8

The reason:
When alloc_iova_mem() without initial with Zero, sometimes fpn_lo will
equal to IOVA_ANCHOR by chance, so when return with -ENOMEM(iova32_full)
from __alloc_and_insert_iova_range(), the new_iova will not be freed in
free_iova_mem().

Fixes: bb68b2fbfbd6 ("iommu/iova: Add rbtree anchor node")
Signed-off-by: Xiaotao Yin <xiaotao.yin@windriver.com>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
4 years agosecurity: keys: trusted: fix lost handle flush
James Bottomley [Thu, 12 Dec 2019 17:58:35 +0000 (12:58 -0500)]
security: keys: trusted: fix lost handle flush

The original code, before it was moved into security/keys/trusted-keys
had a flush after the blob unseal.  Without that flush, the volatile
handles increase in the TPM until it becomes unusable and the system
either has to be rebooted or the TPM volatile area manually flushed.
Fix by adding back the lost flush, which we now have to export because
of the relocation of the trusted key code may cause the consumer to be
modular.

Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Fixes: 2e19e10131a0 ("KEYS: trusted: Move TPM2 trusted keys code")
Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com>
Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
4 years agostaging: axis-fifo: add unspecified HAS_IOMEM dependency
Brendan Higgins [Wed, 11 Dec 2019 19:27:41 +0000 (11:27 -0800)]
staging: axis-fifo: add unspecified HAS_IOMEM dependency

Currently CONFIG_XIL_AXIS_FIFO=y implicitly depends on
CONFIG_HAS_IOMEM=y; consequently, on architectures without IOMEM we get
the following build error:

ld: drivers/staging/axis-fifo/axis-fifo.o: in function `axis_fifo_probe':
drivers/staging/axis-fifo/axis-fifo.c:809: undefined reference to `devm_ioremap_resource'

Fix the build error by adding the unspecified dependency.

Reported-by: Brendan Higgins <brendanhiggins@google.com>
Signed-off-by: Brendan Higgins <brendanhiggins@google.com>
Link: https://lore.kernel.org/r/20191211192742.95699-7-brendanhiggins@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agostaging: comedi: gsc_hpdi: check dma_alloc_coherent() return value
Ian Abbott [Mon, 16 Dec 2019 11:08:23 +0000 (11:08 +0000)]
staging: comedi: gsc_hpdi: check dma_alloc_coherent() return value

The "auto-attach" handler function `gsc_hpdi_auto_attach()` calls
`dma_alloc_coherent()` in a loop to allocate some DMA data buffers, and
also calls it to allocate a buffer for a DMA descriptor chain.  However,
it does not check the return value of any of these calls.  Change
`gsc_hpdi_auto_attach()` to return `-ENOMEM` if any of these
`dma_alloc_coherent()` calls fail.  This will result in the comedi core
calling the "detach" handler `gsc_hpdi_detach()` as part of the
clean-up, which will call `gsc_hpdi_free_dma()` to free any allocated
DMA coherent memory buffers.

Cc: <stable@vger.kernel.org> #4.6+
Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
Link: https://lore.kernel.org/r/20191216110823.216237-1-abbotti@mev.co.uk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
4 years agotpm_tis: reserve chip for duration of tpm_tis_core_init
Jerry Snitselaar [Wed, 11 Dec 2019 23:54:55 +0000 (16:54 -0700)]
tpm_tis: reserve chip for duration of tpm_tis_core_init

Instead of repeatedly calling tpm_chip_start/tpm_chip_stop when
issuing commands to the tpm during initialization, just reserve the
chip after wait_startup, and release it when we are ready to call
tpm_chip_register.

Cc: Christian Bundy <christianbundy@fraction.io>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Peter Huewe <peterhuewe@gmx.de>
Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Stefan Berger <stefanb@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org
Cc: linux-integrity@vger.kernel.org
Fixes: a3fbfae82b4c ("tpm: take TPM chip power gating out of tpm_transmit()")
Fixes: 5b359c7c4372 ("tpm_tis_core: Turn on the TPM before probing IRQ's")
Suggested-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Reviewed-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Signed-off-by: Jerry Snitselaar <jsnitsel@redhat.com>
Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
4 years agoMerge tag 'armsoc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc
Linus Torvalds [Tue, 17 Dec 2019 00:43:07 +0000 (16:43 -0800)]
Merge tag 'armsoc-fixes' of git://git./linux/kernel/git/soc/soc

Pull ARM SoC fixes from Olof Johansson:
 "I didn't get a batch in this weekend, so here's what we queued up last
  week and today.

   - A couple of defconfigs add back debugfs -- it used to be implicitly
     enabled through CONFIG_TRACING, but 0e4a459f56c32d3e ("tracing:
     Remove unnecessary DEBUG_FS dependency") removed that.

   - The rest are mostly minor fixlets of the usual kind; some DT
     tweaks, a headerfile refactor that needs a build fix now, etc"

* tag 'armsoc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (30 commits)
  ARM: bcm: Add missing sentinel to bcm2711_compat[]
  ARM: shmobile: defconfig: Restore debugfs support
  bus: ti-sysc: Fix missing reset delay handling
  ARM: imx: Fix boot crash if ocotp is not found
  ARM: imx_v6_v7_defconfig: Explicitly restore CONFIG_DEBUG_FS
  ARM: dts: imx6ul-evk: Fix peripheral regulator
  arm64: dts: ls1028a: fix reboot node
  ARM: mmp: include the correct cputype.h
  ARM: dts: am437x-gp/epos-evm: fix panel compatible
  arm64: dts: ls1028a: fix typo in TMU calibration data
  ARM: imx: Correct ocotp id for serial number support of i.MX6ULL/ULZ SoCs
  ARM: dts: bcm283x: Fix critical trip point
  ARM: omap2plus_defconfig: Add back DEBUG_FS
  ARM: omap2plus_defconfig: enable NET_SWITCHDEV
  ARM: dts: am335x-sancloud-bbe: fix phy mode
  bus: ti-sysc: Fix missing force mstandby quirk handling
  reset: Do not register resource data for missing resets
  reset: Fix {of,devm}_reset_control_array_get kerneldoc return types
  reset: brcmstb: Remove resource checks
  dt-bindings: reset: Fix brcmstb-reset example
  ...

4 years agoMerge tag 'arm-soc/for-5.5/soc-fixes' of https://github.com/Broadcom/stblinux into...
Olof Johansson [Mon, 16 Dec 2019 19:33:29 +0000 (11:33 -0800)]
Merge tag 'arm-soc/for-5.5/soc-fixes' of https://github.com/Broadcom/stblinux into arm/fixes

This pull request contains Broadcom ARM-based SoCs machine fixes for
5.5-rc1, please pull the following:

- H. Nikolaus adds a missing sentinel entry to the BCM2711 machine
  descriptor compatible array which would make multiplatform kernels fail
  to boot

* tag 'arm-soc/for-5.5/soc-fixes' of https://github.com/Broadcom/stblinux:
  ARM: bcm: Add missing sentinel to bcm2711_compat[]

Link: https://lore.kernel.org/r/20191216035701.15534-1-f.fainelli@gmail.com
Signed-off-by: Olof Johansson <olof@lixom.net>
4 years agoMerge tag 'samsung-fixes-5.5' of https://git.kernel.org/pub/scm/linux/kernel/git...
Olof Johansson [Mon, 16 Dec 2019 19:33:13 +0000 (11:33 -0800)]
Merge tag 'samsung-fixes-5.5' of https://git./linux/kernel/git/krzk/linux into arm/fixes

Samsung fixes for v5.5

1. Restore debugfs support in exynos_defconfig (as now it is not
   selected as dependency of tracing).  Debugfs is required by systemd
   and several tests.
2. Maintainers updates.

* tag 'samsung-fixes-5.5' of https://git.kernel.org/pub/scm/linux/kernel/git/krzk/linux:
  ARM: exynos_defconfig: Restore debugfs support
  MAINTAINERS: Include Samsung SoC serial driver in Samsung SoC entry
  MAINTAINERS: Update Lukasz Luba's email address

Link: https://lore.kernel.org/r/20191215121316.32091-1-krzk@kernel.org
Signed-off-by: Olof Johansson <olof@lixom.net>
4 years agoMerge tag 'renesas-fixes-for-v5.5-tag1' of git://git.kernel.org/pub/scm/linux/kernel...
Olof Johansson [Mon, 16 Dec 2019 19:33:04 +0000 (11:33 -0800)]
Merge tag 'renesas-fixes-for-v5.5-tag1' of git://git./linux/kernel/git/geert/renesas-devel into arm/fixes

Renesas fixes for v5.5

  - Restore debugfs support

* tag 'renesas-fixes-for-v5.5-tag1' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/renesas-devel:
  ARM: shmobile: defconfig: Restore debugfs support

Link: https://lore.kernel.org/r/20191213213719.18122-1-geert+renesas@glider.be
Signed-off-by: Olof Johansson <olof@lixom.net>
4 years agoMerge tag 'linux-kselftest-5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Mon, 16 Dec 2019 18:06:04 +0000 (10:06 -0800)]
Merge tag 'linux-kselftest-5.5-rc2' of git://git./linux/kernel/git/shuah/linux-kselftest

Pull kselftest fixes from Shuah Khan:

 - ftrace and safesetid test fixes from Masami Hiramatsu

 - Kunit fixes from Brendan Higgins, Iurii Zaikin, and Heidi Fahim

 - Kselftest framework fixes from SeongJae Park and Michael Ellerman

* tag 'linux-kselftest-5.5-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:
  kselftest: Support old perl versions
  kselftest/runner: Print new line in print of timeout log
  selftests: Fix dangling documentation references to kselftest_module.sh
  Documentation: kunit: add documentation for kunit_tool
  Documentation: kunit: fix typos and gramatical errors
  kunit: testing kunit: Bug fix in test_run_timeout function
  fs/ext4/inode-test: Fix inode test on 32 bit platforms.
  selftests: safesetid: Fix Makefile to set correct test program
  selftests: safesetid: Check the return value of setuid/setgid
  selftests: safesetid: Move link library to LDLIBS
  selftests/ftrace: Fix multiple kprobe testcase
  selftests/ftrace: Do not to use absolute debugfs path
  selftests/ftrace: Fix ftrace test cases to check unsupported
  selftests/ftrace: Fix to check the existence of set_ftrace_filter

4 years agoiommu: fix KASAN use-after-free in iommu_insert_resv_region
Eric Auger [Tue, 26 Nov 2019 17:54:13 +0000 (18:54 +0100)]
iommu: fix KASAN use-after-free in iommu_insert_resv_region

In case the new region gets merged into another one, the nr list node is
freed.  Checking its type while completing the merge algorithm leads to
a use-after-free.  Use new->type instead.

Fixes: 4dbd258ff63e ("iommu: Revisit iommu_insert_resv_region() implementation")
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reported-by: Qian Cai <cai@lca.pw>
Reviewed-by: Jerry Snitselaar <jsnitsel@redhat.com>
Cc: Stable <stable@vger.kernel.org> #v5.3+
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
4 years agoFix root mounting with no mount options
Linus Torvalds [Mon, 16 Dec 2019 03:50:23 +0000 (19:50 -0800)]
Fix root mounting with no mount options

The "trivial conversion" in commit cccaa5e33525 ("init: use do_mount()
instead of ksys_mount()") was totally broken, since it didn't handle the
case of a NULL mount data pointer.  And while I had "tested" it (and
presumably Dominik had too) that bug was hidden by me having options.

Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Cc: Arnd Bergmann <arnd@arndb.de>
Reported-by: Ondřej Jirman <megi@xff.cz>
Reported-by: Guenter Roeck <linux@roeck-us.net>
Reported-by: Naresh Kamboju <naresh.kamboju@linaro.org>
Reported-and-tested-by: Borislav Petkov <bp@suse.de>
Tested-by: Chris Clayton <chris2553@googlemail.com>
Tested-by: Eric Biggers <ebiggers@kernel.org>
Tested-by: Geert Uytterhoeven <geert@linux-m68k.org>
Tested-by: Guido Günther <agx@sigxcpu.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
4 years agoperf vendor events s390: Remove name from L1D_RO_EXCL_WRITES description
Ed Maste [Thu, 12 Dec 2019 14:53:46 +0000 (14:53 +0000)]
perf vendor events s390: Remove name from L1D_RO_EXCL_WRITES description

In 7fcfa9a2d9 an unintended prefix "Counter:18 Name:" was removed from
the description for L1D_RO_EXCL_WRITES, but the extra name remained in
the description.  Remove it too.

Fixes: 7fcfa9a2d9a7 ("perf list: Fix s390 counter long description for L1D_RO_EXCL_WRITES")
Signed-off-by: Ed Maste <emaste@freebsd.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Greentime Hu <green.hu@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Nick Hu <nickhu@andestech.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Richter <tmricht@linux.ibm.com>
Cc: Vincent Chen <deanbo422@gmail.com>
Link: http://lore.kernel.org/lkml/20191212145346.5026-1-emaste@freefall.freebsd.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>