linux-2.6-microblaze.git
3 years agonvmem: sprd: Fix return value of sprd_efuse_probe()
Tiezhu Yang [Wed, 22 Jul 2020 10:06:52 +0000 (11:06 +0100)]
nvmem: sprd: Fix return value of sprd_efuse_probe()

When call function devm_platform_ioremap_resource(), we should use IS_ERR()
to check the return value and return PTR_ERR() if failed.

Fixes: 096030e7f449 ("nvmem: sprd: Add Spreadtrum SoCs eFuse support")
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20200722100705.7772-2-srinivas.kandagatla@linaro.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agodrivers: android: Fix the SPDX comment style
Mrinal Pandey [Fri, 24 Jul 2020 13:14:49 +0000 (18:44 +0530)]
drivers: android: Fix the SPDX comment style

C source files should have `//` as SPDX comment and not `/**/`. Fix this
by running checkpatch on the file.

Signed-off-by: Mrinal Pandey <mrinalmni@gmail.com>
Link: https://lore.kernel.org/r/20200724131449.zvjutbemg3vqhrzh@mrinalpandey
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agodrivers: android: Fix a variable declaration coding style issue
Mrinal Pandey [Fri, 24 Jul 2020 13:14:33 +0000 (18:44 +0530)]
drivers: android: Fix a variable declaration coding style issue

Add a blank line after variable declarations as suggested by checkpatch.

Signed-off-by: Mrinal Pandey <mrinalmni@gmail.com>
Link: https://lore.kernel.org/r/20200724131433.stf3ycooogawyzb3@mrinalpandey
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agodrivers: android: Remove braces for a single statement if-else block
Mrinal Pandey [Fri, 24 Jul 2020 13:14:03 +0000 (18:44 +0530)]
drivers: android: Remove braces for a single statement if-else block

Remove braces for both if and else block as suggested by checkpatch.

Signed-off-by: Mrinal Pandey <mrinalmni@gmail.com>
Link: https://lore.kernel.org/r/20200724131403.dahfhdwa3wirzkxj@mrinalpandey
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agodrivers: android: Remove the use of else after return
Mrinal Pandey [Fri, 24 Jul 2020 13:13:48 +0000 (18:43 +0530)]
drivers: android: Remove the use of else after return

Remove the unnecessary else branch after return statement as suggested by
checkpatch.

Signed-off-by: Mrinal Pandey <mrinalmni@gmail.com>
Link: https://lore.kernel.org/r/20200724131348.haz4ocxcferdcsgn@mrinalpandey
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agodrivers: android: Fix a variable declaration coding style issue
Mrinal Pandey [Fri, 24 Jul 2020 13:12:54 +0000 (18:42 +0530)]
drivers: android: Fix a variable declaration coding style issue

Add a blank line after variable declarations as suggested by checkpatch.

Signed-off-by: Mrinal Pandey <mrinalmni@gmail.com>
Link: https://lore.kernel.org/r/20200724131254.qxbvderrws36dzzq@mrinalpandey
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agobinder: Prevent context manager from incrementing ref 0
Jann Horn [Mon, 27 Jul 2020 12:04:24 +0000 (14:04 +0200)]
binder: Prevent context manager from incrementing ref 0

Binder is designed such that a binder_proc never has references to
itself. If this rule is violated, memory corruption can occur when a
process sends a transaction to itself; see e.g.
<https://syzkaller.appspot.com/bug?extid=09e05aba06723a94d43d>.

There is a remaining edgecase through which such a transaction-to-self
can still occur from the context of a task with BINDER_SET_CONTEXT_MGR
access:

 - task A opens /dev/binder twice, creating binder_proc instances P1
   and P2
 - P1 becomes context manager
 - P2 calls ACQUIRE on the magic handle 0, allocating index 0 in its
   handle table
 - P1 dies (by closing the /dev/binder fd and waiting a bit)
 - P2 becomes context manager
 - P2 calls ACQUIRE on the magic handle 0, allocating index 1 in its
   handle table
   [this triggers a warning: "binder: 1974:1974 tried to acquire
   reference to desc 0, got 1 instead"]
 - task B opens /dev/binder once, creating binder_proc instance P3
 - P3 calls P2 (via magic handle 0) with (void*)1 as argument (two-way
   transaction)
 - P2 receives the handle and uses it to call P3 (two-way transaction)
 - P3 calls P2 (via magic handle 0) (two-way transaction)
 - P2 calls P2 (via handle 1) (two-way transaction)

And then, if P2 does *NOT* accept the incoming transaction work, but
instead closes the binder fd, we get a crash.

Solve it by preventing the context manager from using ACQUIRE on ref 0.
There shouldn't be any legitimate reason for the context manager to do
that.

Additionally, print a warning if someone manages to find another way to
trigger a transaction-to-self bug in the future.

Cc: stable@vger.kernel.org
Fixes: 457b9a6f09f0 ("Staging: android: add binder driver")
Acked-by: Todd Kjos <tkjos@google.com>
Signed-off-by: Jann Horn <jannh@google.com>
Reviewed-by: Martijn Coenen <maco@android.com>
Link: https://lore.kernel.org/r/20200727120424.1627555-1-jannh@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agogreybus: Use fallthrough pseudo-keyword
Gustavo A. R. Silva [Mon, 27 Jul 2020 18:32:58 +0000 (13:32 -0500)]
greybus: Use fallthrough pseudo-keyword

Replace the existing /* fall through */ comments and its variants with
the new pseudo-keyword macro fallthrough[1].

[1] https://www.kernel.org/doc/html/v5.7/process/deprecated.html?highlight=fallthrough#implicit-switch-case-fall-through

Signed-off-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Link: https://lore.kernel.org/r/20200727183258.GA28571@embeddedor
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agomei: add device kind to sysfs
Alexander Usyskin [Tue, 28 Jul 2020 19:22:42 +0000 (22:22 +0300)]
mei: add device kind to sysfs

Some of the mei device heads are not generic and have
a specific purpose, we need to announce it to the user space
so it is possible to detect the correct device node via
matching attributes.

Generic heads are marked as 'mei' while special purpose heads
have their own names. Currently we are adding 'itouch' string
for Intel IPTS 1.0, 2.0 devices.

This is done via new sysfs attribute 'kind'.

Signed-off-by: Alexander Usyskin <alexander.usyskin@intel.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Link: https://lore.kernel.org/r/20200728192242.3117779-1-tomas.winkler@intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agoMAINTAINERS: Fix maintainer entry for mei driver
Tomas Winkler [Wed, 29 Jul 2020 11:05:40 +0000 (14:05 +0300)]
MAINTAINERS: Fix maintainer entry for mei driver

mei driver has sub modules, those are not
listed via scripts/get_maintainer.pl when using asterisk:
drivers/misc/mei/*
The correct notation is:
drivers/misc/mei/

Cc: Joe Perches <joe@perches.com>
Signed-off-by: Tomas Winkler <tomas.winkler@intel.com>
Reviewed-by: Gustavo A. R. Silva <gustavoars@kernel.org>
Link: https://lore.kernel.org/r/20200729110540.3205585-1-tomas.winkler@intel.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agohabanalabs: goya_ctx_init() can be static
kernel test robot [Wed, 29 Jul 2020 00:03:13 +0000 (08:03 +0800)]
habanalabs: goya_ctx_init() can be static

Signed-off-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/r/20200729000313.GA14680@e442e3f624c4
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agohabanalabs: fix up absolute include instructions
Greg Kroah-Hartman [Tue, 28 Jul 2020 17:18:51 +0000 (19:18 +0200)]
habanalabs: fix up absolute include instructions

There's no need to try to be cute with the include file locations in the
Makefile, so just specify exactly where the files are.

Bonus is this fixes the problem of building with O= as well as trying to
just build the subdirectory alone.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Oded Gabbay <oded.gabbay@gmail.com>
Cc: Omer Shpigelman <oshpigelman@habana.ai>
Cc: Tomer Tayar <ttayar@habana.ai>
Cc: Moti Haimovski <mhaimovski@habana.ai>
Cc: Ofir Bitton <obitton@habana.ai>
Cc: Ben Segal <bpsegal20@gmail.com>
Cc: Christine Gharzuzi <cgharzuzi@habana.ai>
Cc: Pawel Piskorski <ppiskorski@habana.ai>
Link: https://lore.kernel.org/r/20200728171851.55842-1-gregkh@linuxfoundation.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agoMerge 5.8-rc7 into char-misc-next
Greg Kroah-Hartman [Mon, 27 Jul 2020 09:49:37 +0000 (11:49 +0200)]
Merge 5.8-rc7 into char-misc-next

This should resolve the merge/build issues reported when trying to
create linux-next.

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agoLinux 5.8-rc7
Linus Torvalds [Sun, 26 Jul 2020 21:14:06 +0000 (14:14 -0700)]
Linux 5.8-rc7

3 years agoMerge tag 'kbuild-fixes-v5.8-3' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 26 Jul 2020 20:46:57 +0000 (13:46 -0700)]
Merge tag 'kbuild-fixes-v5.8-3' of git://git./linux/kernel/git/masahiroy/linux-kbuild into master

Pull Kbuild fixes from Masahiro Yamada:

 - do not use non-portable strsep() in a host program

 - fix single target builds for external modules

 - change Clang's --prefix option to make it work for the latest Clang

* tag 'kbuild-fixes-v5.8-3' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
  Makefile: Fix GCC_TOOLCHAIN_DIR prefix for Clang cross compilation
  kbuild: fix single target builds for external modules
  modpost: remove use of non-standard strsep() in HOSTCC code

3 years agoMerge branch 'parisc-5.8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller...
Linus Torvalds [Sun, 26 Jul 2020 19:14:46 +0000 (12:14 -0700)]
Merge branch 'parisc-5.8-2' of git://git./linux/kernel/git/deller/parisc-linux into master

Pull parisc fixes from Helge Deller:
 "Two fixes:

   - Add the cmpxchg() function for pointers to u8 values. This fixes a
     kernel linking error when building the tusb1210 driver (from Liam
     Beguin).

   - Add a define for atomic64_set_release() to fix CPU soft lockups
     which happen because of missing unlocks while processing bit
     operations (from John David Anglin)"

* 'parisc-5.8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
  parisc: Add atomic64_set_release() define to avoid CPU soft lockups
  parisc: add support for cmpxchg on u8 pointers

3 years agoMerge tag 'char-misc-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Sun, 26 Jul 2020 16:33:25 +0000 (09:33 -0700)]
Merge tag 'char-misc-5.8-rc7' of git://git./linux/kernel/git/gregkh/char-misc into master

Pull char/misc driver fixes from Greg KH:
 "Here are a few small driver fixes for 5.8-rc7

  They include:

   - habanalabs fixes

   - tiny fpga driver fixes

   - /dev/mem fixup from previous changes

   - interconnect driver fixes

   - binder fix

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

* tag 'char-misc-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
  interconnect: msm8916: Fix buswidth of pcnoc_s nodes
  interconnect: Do not skip aggregation for disabled paths
  /dev/mem: Add missing memory barriers for devmem_inode
  binder: Don't use mmput() from shrinker function.
  habanalabs: prevent possible out-of-bounds array access
  fpga: dfl: fix bug in port reset handshake
  fpga: dfl: pci: reduce the scope of variable 'ret'
  habanalabs: set 4s timeout for message to device CPU
  habanalabs: set clock gating per engine
  habanalabs: block WREG_BULK packet on PDMA

3 years agoMerge tag 'driver-core-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 26 Jul 2020 16:29:22 +0000 (09:29 -0700)]
Merge tag 'driver-core-5.8-rc7' of git://git./linux/kernel/git/gregkh/driver-core into master

Pull driver core fix from Greg KH:
 "A single driver core fix for 5.8-rc7. It resolves a problem found in
  the previous fix for this code made in 5.8-rc6. Hopefully this is all
  now cleared up, as this seems to be the last of the reported issues in
  this area, and was tested on the problem hardware.

  This patch has been in linux-next with no reported problems"

* tag 'driver-core-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core:
  device property: Avoid NULL pointer dereference in device_get_next_child_node()

3 years agoMerge tag 'staging-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh...
Linus Torvalds [Sun, 26 Jul 2020 16:14:59 +0000 (09:14 -0700)]
Merge tag 'staging-5.8-rc7' of git://git./linux/kernel/git/gregkh/staging into master

Pull staging driver fixes from Greg KH:
 "Five small staging driver fixes for 5.8-rc7 to resolve some reported
  problems:

   - four comedi driver fixes for problems found with them

   - a syzbot-found fix for the wlang-ng driver that resolves a much
     reported problem.

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

* tag 'staging-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
  staging: wlan-ng: properly check endpoint types
  staging: comedi: addi_apci_1564: check INSN_CONFIG_DIGITAL_TRIG shift
  staging: comedi: addi_apci_1500: check INSN_CONFIG_DIGITAL_TRIG shift
  staging: comedi: addi_apci_1032: check INSN_CONFIG_DIGITAL_TRIG shift
  staging: comedi: ni_6527: fix INSN_CONFIG_DIGITAL_TRIG support

3 years agoMerge tag 'tty-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty...
Linus Torvalds [Sun, 26 Jul 2020 16:09:43 +0000 (09:09 -0700)]
Merge tag 'tty-5.8-rc7' of git://git./linux/kernel/git/gregkh/tty into master

Pull tty/serial/fbcon fixes from Greg KH:
 "Here are some small tty and serial and fbcon fixes for 5.8-rc7 to
  resolve some reported issues.

  The fbcon fix is in here as it was simpler to take it this way (and it
  was acked by the maintainer) as it was related to the vt console fix
  as well, both of which resolve syzbot-found issues in the console
  handling code.

  The other serial driver fixes are for small issues reported in the -rc
  releases.

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

* tag 'tty-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
  serial: exar: Fix GPIO configuration for Sealevel cards based on XR17V35X
  fbdev: Detect integer underflow at "struct fbcon_ops"->clear_margins.
  serial: 8250_mtk: Fix high-speed baud rates clamping
  serial: 8250: fix null-ptr-deref in serial8250_start_tx()
  serial: tegra: drop bogus NULL tty-port checks
  serial: tegra: fix CREAD handling for PIO
  tty: xilinx_uartps: Really fix id assignment
  vt: Reject zero-sized screen buffer size.

3 years agoMerge tag 'usb-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb...
Linus Torvalds [Sun, 26 Jul 2020 16:02:29 +0000 (09:02 -0700)]
Merge tag 'usb-5.8-rc7' of git://git./linux/kernel/git/gregkh/usb into master

Pull USB fixes from Greg KH:
 "Three small USB XHCI driver fixes for 5.8-rc7.

  They all resolve some minor issues that have been reported on some
  different platforms.

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

* tag 'usb-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb:
  usb: tegra: Fix allocation for the FPCI context
  usb: xhci: Fix ASM2142/ASM3142 DMA addressing
  usb: xhci-mtk: fix the failure of bandwidth allocation

3 years agoMerge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi...
Linus Torvalds [Sun, 26 Jul 2020 15:59:15 +0000 (08:59 -0700)]
Merge tag 'scsi-fixes' of git://git./linux/kernel/git/jejb/scsi into master

Pull SCSI fix from James Bottomley:
 "Small core patch to fix a corner case bug: we forgot to run the queues
  to handle starvation in the error exit from the scsi_queue_rq routine,
  which can lead to hangs on error conditions"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: core: Run queue in case of I/O resource contention failure

3 years agoMerge tag 'riscv-for-linus-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sat, 25 Jul 2020 21:42:11 +0000 (14:42 -0700)]
Merge tag 'riscv-for-linus-5.8-rc7' of git://git./linux/kernel/git/riscv/linux into master

Pull RISC-V fixes from Palmer Dabbelt:
 "A few more fixes this week:

   - A fix to avoid using SBI calls during kasan initialization, as the
     SBI calls themselves have not been probed yet.

   - Three fixes related to systems with multiple memory regions"

* tag 'riscv-for-linus-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
  riscv: Parse all memory blocks to remove unusable memory
  RISC-V: Do not rely on initrd_start/end computed during early dt parsing
  RISC-V: Set maximum number of mapped pages correctly
  riscv: kasan: use local_tlb_flush_all() to avoid uninitialized __sbi_rfence

3 years agoMerge tag 'x86-urgent-2020-07-25' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 25 Jul 2020 21:25:47 +0000 (14:25 -0700)]
Merge tag 'x86-urgent-2020-07-25' of git://git./linux/kernel/git/tip/tip into master

Pull x86 fixes from Ingo Molnar:
 "Misc fixes:

   - Fix a section end page alignment assumption that was causing
     crashes

   - Fix ORC unwinding on freshly forked tasks which haven't executed
     yet and which have empty user task stacks

   - Fix the debug.exception-trace=1 sysctl dumping of user stacks,
     which was broken by recent maccess changes"

* tag 'x86-urgent-2020-07-25' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/dumpstack: Dump user space code correctly again
  x86/stacktrace: Fix reliable check for empty user task stacks
  x86/unwind/orc: Fix ORC for newly forked tasks
  x86, vmlinux.lds: Page-align end of ..page_aligned sections

3 years agoMerge tag 'perf-urgent-2020-07-25' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 25 Jul 2020 20:55:38 +0000 (13:55 -0700)]
Merge tag 'perf-urgent-2020-07-25' of git://git./linux/kernel/git/tip/tip into master

Pull uprobe fix from Ingo Molnar:
 "Fix an interaction/regression between uprobes based shared library
  tracing & GDB"

* tag 'perf-urgent-2020-07-25' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  uprobes: Change handle_swbp() to send SIGTRAP with si_code=SI_KERNEL, to fix GDB regression

3 years agoMerge tag 'timers-urgent-2020-07-25' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sat, 25 Jul 2020 20:27:12 +0000 (13:27 -0700)]
Merge tag 'timers-urgent-2020-07-25' of git://git./linux/kernel/git/tip/tip into master

Pull timer fix from Ingo Molnar:
 "Fix a suspend/resume regression (crash) on TI AM3/AM4 SoC's"

* tag 'timers-urgent-2020-07-25' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  clocksource/drivers/timer-ti-dm: Fix suspend and resume for am3 and am4

3 years agoMerge tag 'sched-urgent-2020-07-25' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sat, 25 Jul 2020 20:24:40 +0000 (13:24 -0700)]
Merge tag 'sched-urgent-2020-07-25' of git://git./linux/kernel/git/tip/tip into master

Pull scheduler fixes from Ingo Molnar:
 "Fix a race introduced by the recent loadavg race fix, plus add a debug
  check for a hard to debug case of bogus wakeup function flags"

* tag 'sched-urgent-2020-07-25' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched: Warn if garbage is passed to default_wake_function()
  sched: Fix race against ptrace_freeze_trace()

3 years agoMerge tag 'efi-urgent-2020-07-25' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 25 Jul 2020 20:18:42 +0000 (13:18 -0700)]
Merge tag 'efi-urgent-2020-07-25' of git://git./linux/kernel/git/tip/tip into master

Pull EFI fixes from Ingo Molnar:
 "Various EFI fixes:

   - Fix the layering violation in the use of the EFI runtime services
     availability mask in users of the 'efivars' abstraction

   - Revert build fix for GCC v4.8 which is no longer supported

   - Clean up some x86 EFI stub details, some of which are borderline
     bugs that copy around garbage into padding fields - let's fix these
     out of caution.

   - Fix build issues while working on RISC-V support

   - Avoid --whole-archive when linking the stub on arm64"

* tag 'efi-urgent-2020-07-25' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  efi: Revert "efi/x86: Fix build with gcc 4"
  efi/efivars: Expose RT service availability via efivars abstraction
  efi/libstub: Move the function prototypes to header file
  efi/libstub: Fix gcc error around __umoddi3 for 32 bit builds
  efi/libstub/arm64: link stub lib.a conditionally
  efi/x86: Only copy upto the end of setup_header
  efi/x86: Remove unused variables

3 years agoMerge tag '5.8-rc6-cifs-fix' of git://git.samba.org/sfrench/cifs-2.6 into master
Linus Torvalds [Sat, 25 Jul 2020 19:53:46 +0000 (12:53 -0700)]
Merge tag '5.8-rc6-cifs-fix' of git://git.samba.org/sfrench/cifs-2.6 into master

Pull cifs fix from Steve French:
 "A fix for a recently discovered regression in rename to older servers
  caused by a recent patch"

* tag '5.8-rc6-cifs-fix' of git://git.samba.org/sfrench/cifs-2.6:
  Revert "cifs: Fix the target file was deleted when rename failed."

3 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net into master
Linus Torvalds [Sat, 25 Jul 2020 18:50:59 +0000 (11:50 -0700)]
Merge git://git./linux/kernel/git/netdev/net into master

Pull networking fixes from David Miller:

 1) Fix RCU locaking in iwlwifi, from Johannes Berg.

 2) mt76 can access uninitialized NAPI struct, from Felix Fietkau.

 3) Fix race in updating pause settings in bnxt_en, from Vasundhara
    Volam.

 4) Propagate error return properly during unbind failures in ax88172a,
    from George Kennedy.

 5) Fix memleak in adf7242_probe, from Liu Jian.

 6) smc_drv_probe() can leak, from Wang Hai.

 7) Don't muck with the carrier state if register_netdevice() fails in
    the bonding driver, from Taehee Yoo.

 8) Fix memleak in dpaa_eth_probe, from Liu Jian.

 9) Need to check skb_put_padto() return value in hsr_fill_tag(), from
    Murali Karicheri.

10) Don't lose ionic RSS hash settings across FW update, from Shannon
    Nelson.

11) Fix clobbered SKB control block in act_ct, from Wen Xu.

12) Missing newlink in "tx_timeout" sysfs output, from Xiongfeng Wang.

13) IS_UDPLITE cleanup a long time ago, incorrectly handled
    transformations involving UDPLITE_RECV_CC. From Miaohe Lin.

14) Unbalanced locking in netdevsim, from Taehee Yoo.

15) Suppress false-positive error messages in qed driver, from Alexander
    Lobakin.

16) Out of bounds read in ax25_connect and ax25_sendmsg, from Peilin Ye.

17) Missing SKB release in cxgb4's uld_send(), from Navid Emamdoost.

18) Uninitialized value in geneve_changelink(), from Cong Wang.

19) Fix deadlock in xen-netfront, from Andera Righi.

19) flush_backlog() frees skbs with IRQs disabled, so should use
    dev_kfree_skb_irq() instead of kfree_skb(). From Subash Abhinov
    Kasiviswanathan.

* git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (111 commits)
  drivers/net/wan: lapb: Corrected the usage of skb_cow
  dev: Defer free of skbs in flush_backlog
  qrtr: orphan socket in qrtr_release()
  xen-netfront: fix potential deadlock in xennet_remove()
  flow_offload: Move rhashtable inclusion to the source file
  geneve: fix an uninitialized value in geneve_changelink()
  bonding: check return value of register_netdevice() in bond_newlink()
  tcp: allow at most one TLP probe per flight
  AX.25: Prevent integer overflows in connect and sendmsg
  cxgb4: add missing release on skb in uld_send()
  net: atlantic: fix PTP on AQC10X
  AX.25: Prevent out-of-bounds read in ax25_sendmsg()
  sctp: shrink stream outq when fails to do addstream reconf
  sctp: shrink stream outq only when new outcnt < old outcnt
  AX.25: Fix out-of-bounds read in ax25_connect()
  enetc: Remove the mdio bus on PF probe bailout
  net: ethernet: ti: add NETIF_F_HW_TC hw feature flag for taprio offload
  net: ethernet: ave: Fix error returns in ave_init
  drivers/net/wan/x25_asy: Fix to make it work
  ipvs: fix the connection sync failed in some cases
  ...

3 years agoriscv: Parse all memory blocks to remove unusable memory
Atish Patra [Wed, 15 Jul 2020 23:30:09 +0000 (16:30 -0700)]
riscv: Parse all memory blocks to remove unusable memory

Currently, maximum physical memory allowed is equal to -PAGE_OFFSET.
That's why we remove any memory blocks spanning beyond that size. However,
it is done only for memblock containing linux kernel which will not work
if there are multiple memblocks.

Process all memory blocks to figure out how much memory needs to be removed
and remove at the end instead of updating the memblock list in place.

Signed-off-by: Atish Patra <atish.patra@wdc.com>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
3 years agoRISC-V: Do not rely on initrd_start/end computed during early dt parsing
Atish Patra [Wed, 15 Jul 2020 23:30:08 +0000 (16:30 -0700)]
RISC-V: Do not rely on initrd_start/end computed during early dt parsing

Currently, initrd_start/end are computed during early_init_dt_scan
but used during arch_setup. We will get the following panic if initrd is used
and CONFIG_DEBUG_VIRTUAL is turned on.

[    0.000000] ------------[ cut here ]------------
[    0.000000] kernel BUG at arch/riscv/mm/physaddr.c:33!
[    0.000000] Kernel BUG [#1]
[    0.000000] Modules linked in:
[    0.000000] CPU: 0 PID: 0 Comm: swapper Not tainted 5.8.0-rc4-00015-ged0b226fed02 #886
[    0.000000] epc: ffffffe0002058d2 ra : ffffffe0000053f0 sp : ffffffe001001f40
[    0.000000]  gp : ffffffe00106e250 tp : ffffffe001009d40 t0 : ffffffe00107ee28
[    0.000000]  t1 : 0000000000000000 t2 : ffffffe000a2e880 s0 : ffffffe001001f50
[    0.000000]  s1 : ffffffe0001383e8 a0 : ffffffe00c087e00 a1 : 0000000080200000
[    0.000000]  a2 : 00000000010bf000 a3 : ffffffe00106f3c8 a4 : ffffffe0010bf000
[    0.000000]  a5 : ffffffe000000000 a6 : 0000000000000006 a7 : 0000000000000001
[    0.000000]  s2 : ffffffe00106f068 s3 : ffffffe00106f070 s4 : 0000000080200000
[    0.000000]  s5 : 0000000082200000 s6 : 0000000000000000 s7 : 0000000000000000
[    0.000000]  s8 : 0000000080011010 s9 : 0000000080012700 s10: 0000000000000000
[    0.000000]  s11: 0000000000000000 t3 : 000000000001fe30 t4 : 000000000001fe30
[    0.000000]  t5 : 0000000000000000 t6 : ffffffe00107c471
[    0.000000] status: 0000000000000100 badaddr: 0000000000000000 cause: 0000000000000003
[    0.000000] random: get_random_bytes called from print_oops_end_marker+0x22/0x46 with crng_init=0

To avoid the error, initrd_start/end can be computed from phys_initrd_start/size
in setup itself. It also improves the initrd placement by aligning the start
and size with the page size.

Fixes: 76d2a0493a17 ("RISC-V: Init and Halt Code")
Signed-off-by: Atish Patra <atish.patra@wdc.com>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
3 years agodrivers/net/wan: lapb: Corrected the usage of skb_cow
Xie He [Fri, 24 Jul 2020 16:33:47 +0000 (09:33 -0700)]
drivers/net/wan: lapb: Corrected the usage of skb_cow

This patch fixed 2 issues with the usage of skb_cow in LAPB drivers
"lapbether" and "hdlc_x25":

1) After skb_cow fails, kfree_skb should be called to drop a reference
to the skb. But in both drivers, kfree_skb is not called.

2) skb_cow should be called before skb_push so that is can ensure the
safety of skb_push. But in "lapbether", it is incorrectly called after
skb_push.

More details about these 2 issues:

1) The behavior of calling kfree_skb on failure is also the behavior of
netif_rx, which is called by this function with "return netif_rx(skb);".
So this function should follow this behavior, too.

2) In "lapbether", skb_cow is called after skb_push. This results in 2
logical issues:
   a) skb_push is not protected by skb_cow;
   b) An extra headroom of 1 byte is ensured after skb_push. This extra
      headroom has no use in this function. It also has no use in the
      upper-layer function that this function passes the skb to
      (x25_lapb_receive_frame in net/x25/x25_dev.c).
So logically skb_cow should instead be called before skb_push.

Cc: Eric Dumazet <edumazet@google.com>
Cc: Martin Schiller <ms@dev.tdt.de>
Signed-off-by: Xie He <xie.he.0141@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agodev: Defer free of skbs in flush_backlog
Subash Abhinov Kasiviswanathan [Thu, 23 Jul 2020 17:31:48 +0000 (11:31 -0600)]
dev: Defer free of skbs in flush_backlog

IRQs are disabled when freeing skbs in input queue.
Use the IRQ safe variant to free skbs here.

Fixes: 145dd5f9c88f ("net: flush the softnet backlog in process context")
Signed-off-by: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoRISC-V: Set maximum number of mapped pages correctly
Atish Patra [Wed, 15 Jul 2020 23:30:07 +0000 (16:30 -0700)]
RISC-V: Set maximum number of mapped pages correctly

Currently, maximum number of mapper pages are set to the pfn calculated
from the memblock size of the memblock containing kernel. This will work
until that memblock spans the entire memory. However, it will be set to
a wrong value if there are multiple memblocks defined in kernel
(e.g. with efi runtime services).

Set the the maximum value to the pfn calculated from dram size.

Signed-off-by: Atish Patra <atish.patra@wdc.com>
Signed-off-by: Palmer Dabbelt <palmerdabbelt@google.com>
3 years agoMerge tag 'pci-v5.8-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas...
Linus Torvalds [Sat, 25 Jul 2020 01:30:24 +0000 (18:30 -0700)]
Merge tag 'pci-v5.8-fixes-2' of git://git./linux/kernel/git/helgaas/pci into master

Pull PCI fixes from Bjorn Helgaas:

 - Reject invalid IRQ 0 command line argument for virtio_mmio because
   IRQ 0 now generates warnings (Bjorn Helgaas)

 - Revert "PCI/PM: Assume ports without DLL Link Active train links in
   100 ms", which broke nouveau (Bjorn Helgaas)

* tag 'pci-v5.8-fixes-2' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
  Revert "PCI/PM: Assume ports without DLL Link Active train links in 100 ms"
  virtio-mmio: Reject invalid IRQ 0 command line argument

3 years agoqrtr: orphan socket in qrtr_release()
Cong Wang [Fri, 24 Jul 2020 16:45:51 +0000 (09:45 -0700)]
qrtr: orphan socket in qrtr_release()

We have to detach sock from socket in qrtr_release(),
otherwise skb->sk may still reference to this socket
when the skb is released in tun->queue, particularly
sk->sk_wq still points to &sock->wq, which leads to
a UAF.

Reported-and-tested-by: syzbot+6720d64f31c081c2f708@syzkaller.appspotmail.com
Fixes: 28fb4e59a47d ("net: qrtr: Expose tunneling endpoint to user space")
Cc: Bjorn Andersson <bjorn.andersson@linaro.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoMerge tag 'wireless-drivers-2020-07-24' of git://git.kernel.org/pub/scm/linux/kernel...
David S. Miller [Sat, 25 Jul 2020 00:26:09 +0000 (17:26 -0700)]
Merge tag 'wireless-drivers-2020-07-24' of git://git./linux/kernel/git/kvalo/wireless-drivers

Kalle Valo says:

====================
wireless-drivers fixes for v5.8

Second set of fixes for v5.8, and hopefully also the last. Three
important regressions fixed.

ath9k

* fix a regression which broke support for all ath9k usb devices

ath10k

* fix a regression which broke support for all QCA4019 AHB devices

iwlwifi

* fix a regression which broke support for some Killer Wireless-AC 1550 cards
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoxen-netfront: fix potential deadlock in xennet_remove()
Andrea Righi [Fri, 24 Jul 2020 08:59:10 +0000 (10:59 +0200)]
xen-netfront: fix potential deadlock in xennet_remove()

There's a potential race in xennet_remove(); this is what the driver is
doing upon unregistering a network device:

  1. state = read bus state
  2. if state is not "Closed":
  3.    request to set state to "Closing"
  4.    wait for state to be set to "Closing"
  5.    request to set state to "Closed"
  6.    wait for state to be set to "Closed"

If the state changes to "Closed" immediately after step 1 we are stuck
forever in step 4, because the state will never go back from "Closed" to
"Closing".

Make sure to check also for state == "Closed" in step 4 to prevent the
deadlock.

Also add a 5 sec timeout any time we wait for the bus state to change,
to avoid getting stuck forever in wait_event().

Signed-off-by: Andrea Righi <andrea.righi@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoMerge tag 'nfsd-5.8-2' of git://linux-nfs.org/~bfields/linux into master
Linus Torvalds [Fri, 24 Jul 2020 23:27:54 +0000 (16:27 -0700)]
Merge tag 'nfsd-5.8-2' of git://linux-nfs.org/~bfields/linux into master

Pull nfsd fix from Bruce Fields:
 "Just one fix for a NULL dereference if someone happens to read
  /proc/fs/nfsd/client/../state at the wrong moment"

* tag 'nfsd-5.8-2' of git://linux-nfs.org/~bfields/linux:
  nfsd4: fix NULL dereference in nfsd/clients display code

3 years agoflow_offload: Move rhashtable inclusion to the source file
Herbert Xu [Fri, 24 Jul 2020 00:50:22 +0000 (10:50 +1000)]
flow_offload: Move rhashtable inclusion to the source file

I noticed that touching linux/rhashtable.h causes lib/vsprintf.c to
be rebuilt.  This dependency came through a bogus inclusion in the
file net/flow_offload.h.  This patch moves it to the right place.

This patch also removes a lingering rhashtable inclusion in cls_api
created by the same commit.

Fixes: 4e481908c51b ("flow_offload: move tc indirect block to...")
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
3 years agoMerge branch 'akpm' into master (patches from Andrew)
Linus Torvalds [Fri, 24 Jul 2020 21:24:35 +0000 (14:24 -0700)]
Merge branch 'akpm' into master (patches from Andrew)

Merge misc fixes from Andrew Morton:
 "Subsystems affected by this patch series: mm/pagemap, mm/shmem,
  mm/hotfixes, mm/memcg, mm/hugetlb, mailmap, squashfs, scripts,
  io-mapping, MAINTAINERS, and gdb"

* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
  scripts/gdb: fix lx-symbols 'gdb.error' while loading modules
  MAINTAINERS: add KCOV section
  io-mapping: indicate mapping failure
  scripts/decode_stacktrace: strip basepath from all paths
  squashfs: fix length field overlap check in metadata reading
  mailmap: add entry for Mike Rapoport
  khugepaged: fix null-pointer dereference due to race
  mm/hugetlb: avoid hardcoding while checking if cma is enabled
  mm: memcg/slab: fix memory leak at non-root kmem_cache destroy
  mm/memcg: fix refcount error while moving and swapping
  mm/memcontrol: fix OOPS inside mem_cgroup_get_nr_swap_pages()
  mm: initialize return of vm_insert_pages
  vfs/xattr: mm/shmem: kernfs: release simple xattr entry in a right way
  mm/mmap.c: close race between munmap() and expand_upwards()/downwards()

3 years agoMerge branch 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs into...
Linus Torvalds [Fri, 24 Jul 2020 21:19:00 +0000 (14:19 -0700)]
Merge branch 'fixes' of git://git./linux/kernel/git/viro/vfs into master

Pull xtensa csum regression fix from Al Viro:
 "Max Filippov caught a breakage introduced in xtensa this cycle
  by the csum_and_copy_..._user() series.

  Cut'n'paste from the wrong source - the check that belongs
  in csum_and_copy_to_user() ended up both there and in
  csum_and_copy_from_user()"

* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
  xtensa: fix access check in csum_and_copy_from_user

3 years agoMerge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux...
Linus Torvalds [Fri, 24 Jul 2020 21:16:12 +0000 (14:16 -0700)]
Merge tag 'arm64-fixes' of git://git./linux/kernel/git/arm64/linux into master

Pull arm64 fix from Will Deacon:
 "Fix compat vDSO build flags for recent versions of clang to tell it
  where to find the assembler"

* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
  arm64: vdso32: Fix '--prefix=' value for newer versions of clang

3 years agoMerge tag 'for-5.8-rc6-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave...
Linus Torvalds [Fri, 24 Jul 2020 21:11:43 +0000 (14:11 -0700)]
Merge tag 'for-5.8-rc6-tag' of git://git./linux/kernel/git/kdave/linux into master

Pull btrfs fixes from David Sterba:
 "A few resouce leak fixes from recent patches, all are stable material.

  The problems have been observed during testing or have a reproducer"

* tag 'for-5.8-rc6-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: fix mount failure caused by race with umount
  btrfs: fix page leaks after failure to lock page for delalloc
  btrfs: qgroup: fix data leak caused by race between writeback and truncate
  btrfs: fix double free on ulist after backref resolution failure

3 years agoMerge tag 'zonefs-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal...
Linus Torvalds [Fri, 24 Jul 2020 21:09:19 +0000 (14:09 -0700)]
Merge tag 'zonefs-5.8-rc7' of git://git./linux/kernel/git/dlemoal/zonefs into master

Pull zonefs fixes from Damien Le Moal:
 "Two fixes, the first one to remove compilation warnings and the second
  to avoid potentially inefficient allocation of BIOs for direct writes
  into sequential zones"

* tag 'zonefs-5.8-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/zonefs:
  zonefs: count pages after truncating the iterator
  zonefs: Fix compilation warning

3 years agoMerge tag 'io_uring-5.8-2020-07-24' of git://git.kernel.dk/linux-block into master
Linus Torvalds [Fri, 24 Jul 2020 21:02:41 +0000 (14:02 -0700)]
Merge tag 'io_uring-5.8-2020-07-24' of git://git.kernel.dk/linux-block into master

Pull io_uring fixes from Jens Axboe:

 - Fix discrepancy in how sqe->flags are treated for a few requests,
   this makes it consistent (Daniele)

 - Ensure that poll driven retry works with double waitqueue poll users

 - Fix a missing io_req_init_async() (Pavel)

* tag 'io_uring-5.8-2020-07-24' of git://git.kernel.dk/linux-block:
  io_uring: missed req_init_async() for IOSQE_ASYNC
  io_uring: always allow drain/link/hardlink/async sqe flags
  io_uring: ensure double poll additions work with both request types

3 years agoMerge tag 'iommu-fix-v5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/joro...
Linus Torvalds [Fri, 24 Jul 2020 20:58:05 +0000 (13:58 -0700)]
Merge tag 'iommu-fix-v5.8-rc6' of git://git./linux/kernel/git/joro/iommu into master

Pull iommu fix from Joerg Roedel:
 "Fix a NULL-ptr dereference in the QCOM IOMMU driver"

* tag 'iommu-fix-v5.8-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
  iommu/qcom: Use domain rather than dev as tlb cookie

3 years agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma...
Linus Torvalds [Fri, 24 Jul 2020 20:48:57 +0000 (13:48 -0700)]
Merge tag 'for-linus' of git://git./linux/kernel/git/rdma/rdma into master

Pull rdma fixes from Jason Gunthorpe:
 "One merge window regression, some corruption bugs in HNS and a few
  more syzkaller fixes:

   - Two long standing syzkaller races

   - Fix incorrect HW configuration in HNS

   - Restore accidentally dropped locking in IB CM

   - Fix ODP prefetch bug added in the big rework several versions ago"

* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
  RDMA/mlx5: Prevent prefetch from racing with implicit destruction
  RDMA/cm: Protect access to remote_sidr_table
  RDMA/core: Fix race in rdma_alloc_commit_uobject()
  RDMA/hns: Fix wrong PBL offset when VA is not aligned to PAGE_SIZE
  RDMA/hns: Fix wrong assignment of lp_pktn_ini in QPC
  RDMA/mlx5: Use xa_lock_irq when access to SRQ table

3 years agoMerge tag 'for-5.8/dm-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/devic...
Linus Torvalds [Fri, 24 Jul 2020 20:44:14 +0000 (13:44 -0700)]
Merge tag 'for-5.8/dm-fixes-3' of git://git./linux/kernel/git/device-mapper/linux-dm into master

Pull device mapper fix from Mike Snitzer:
 "A stable fix for DM integrity target's integrity recalculation that
  gets skipped when resuming a device. This is a fix for a previous
  stable@ fix"

* tag 'for-5.8/dm-fixes-3' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm integrity: fix integrity recalculation that is improperly skipped

3 years agoMerge branch 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa...
Linus Torvalds [Fri, 24 Jul 2020 20:41:13 +0000 (13:41 -0700)]
Merge branch 'i2c/for-current' of git://git./linux/kernel/git/wsa/linux into master

Pull i2c fixes from Wolfram Sang:
 "Again some driver bugfixes and some documentation fixes"

* 'i2c/for-current' of git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux:
  i2c: i2c-qcom-geni: Fix DMA transfer race
  i2c: rcar: always clear ICSAR to avoid side effects
  MAINTAINERS: i2c: at91: handover maintenance to Codrin Ciubotariu
  i2c: drop duplicated word in the header file
  i2c: cadence: Clear HOLD bit at correct time in Rx path
  Revert "i2c: cadence: Fix the hold bit setting"

3 years agoMerge tag 'mmc-v5.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc...
Linus Torvalds [Fri, 24 Jul 2020 20:37:38 +0000 (13:37 -0700)]
Merge tag 'mmc-v5.8-rc5' of git://git./linux/kernel/git/ulfh/mmc into master

Pull MMC fix from Ulf Hansson:
 "Fix clock divider calculation in the ASPEED SDHCI controller"

* tag 'mmc-v5.8-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc:
  mmc: sdhci-of-aspeed: Fix clock divider calculation

3 years agoMerge tag 'drm-fixes-2020-07-24' of git://anongit.freedesktop.org/drm/drm into master
Linus Torvalds [Fri, 24 Jul 2020 20:35:55 +0000 (13:35 -0700)]
Merge tag 'drm-fixes-2020-07-24' of git://anongit.freedesktop.org/drm/drm into master

Pull drm fixes from Dave Airlie:
 "Quiet fixes, I may have a single regression fix follow up to this for
  nouveau, but it might be next week, Ben was testing it a bit more .

  Otherwise two amdgpu fixes, one lima and one sun4i:

  amdgpu:
    - Fix crash when overclocking VegaM
    - Fix possible crash when editing dpm levels

  sun4i:
    - Fix inverted HPD result; fixes an earlier fix

  lima:
    - fix timeout during reset"

* tag 'drm-fixes-2020-07-24' of git://anongit.freedesktop.org/drm/drm:
  drm/amdgpu: Fix NULL dereference in dpm sysfs handlers
  drm/amd/powerplay: fix a crash when overclocking Vega M
  drm/lima: fix wait pp reset timeout
  drm: sun4i: hdmi: Fix inverted HPD result

3 years agoscripts/gdb: fix lx-symbols 'gdb.error' while loading modules
Stefano Garzarella [Fri, 24 Jul 2020 04:15:52 +0000 (21:15 -0700)]
scripts/gdb: fix lx-symbols 'gdb.error' while loading modules

Commit ed66f991bb19 ("module: Refactor section attr into bin attribute")
removed the 'name' field from 'struct module_sect_attr' triggering the
following error when invoking lx-symbols:

  (gdb) lx-symbols
  loading vmlinux
  scanning for modules in linux/build
  loading @0xffffffffc014f000: linux/build/drivers/net/tun.ko
  Python Exception <class 'gdb.error'> There is no member named name.:
  Error occurred in Python: There is no member named name.

This patch fixes the issue taking the module name from the 'struct
attribute'.

Fixes: ed66f991bb19 ("module: Refactor section attr into bin attribute")
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Jan Kiszka <jan.kiszka@siemens.com>
Reviewed-by: Kieran Bingham <kbingham@kernel.org>
Link: http://lkml.kernel.org/r/20200722102239.313231-1-sgarzare@redhat.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agoMAINTAINERS: add KCOV section
Andrey Konovalov [Fri, 24 Jul 2020 04:15:49 +0000 (21:15 -0700)]
MAINTAINERS: add KCOV section

To link KCOV to the kasan-dev@ mailing list.

Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Dmitry Vyukov <dvyukov@google.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Marco Elver <elver@google.com>
Link: http://lkml.kernel.org/r/5fa344db7ac4af2213049e5656c0f43d6ecaa379.1595331682.git.andreyknvl@google.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agoio-mapping: indicate mapping failure
Michael J. Ruhl [Fri, 24 Jul 2020 04:15:46 +0000 (21:15 -0700)]
io-mapping: indicate mapping failure

The !ATOMIC_IOMAP version of io_maping_init_wc will always return
success, even when the ioremap fails.

Since the ATOMIC_IOMAP version returns NULL when the init fails, and
callers check for a NULL return on error this is unexpected.

During a device probe, where the ioremap failed, a crash can look like
this:

    BUG: unable to handle page fault for address: 0000000000210000
     #PF: supervisor write access in kernel mode
     #PF: error_code(0x0002) - not-present page
     Oops: 0002 [#1] PREEMPT SMP
     CPU: 0 PID: 177 Comm:
     RIP: 0010:fill_page_dma [i915]
       gen8_ppgtt_create [i915]
       i915_ppgtt_create [i915]
       intel_gt_init [i915]
       i915_gem_init [i915]
       i915_driver_probe [i915]
       pci_device_probe
       really_probe
       driver_probe_device

The remap failure occurred much earlier in the probe.  If it had been
propagated, the driver would have exited with an error.

Return NULL on ioremap failure.

[akpm@linux-foundation.org: detect ioremap_wc() errors earlier]

Fixes: cafaf14a5d8f ("io-mapping: Always create a struct to hold metadata about the io-mapping")
Signed-off-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Mike Rapoport <rppt@linux.ibm.com>
Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20200721171936.81563-1-michael.j.ruhl@intel.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agoscripts/decode_stacktrace: strip basepath from all paths
Pi-Hsun Shih [Fri, 24 Jul 2020 04:15:43 +0000 (21:15 -0700)]
scripts/decode_stacktrace: strip basepath from all paths

Currently the basepath is removed only from the beginning of the string.
When the symbol is inlined and there's multiple line outputs of
addr2line, only the first line would have basepath removed.

Change to remove the basepath prefix from all lines.

Fixes: 31013836a71e ("scripts/decode_stacktrace: match basepath using shell prefix operator, not regex")
Co-developed-by: Shik Chen <shik@chromium.org>
Signed-off-by: Pi-Hsun Shih <pihsun@chromium.org>
Signed-off-by: Shik Chen <shik@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Cc: Sasha Levin <sashal@kernel.org>
Cc: Nicolas Boichat <drinkcat@chromium.org>
Cc: Jiri Slaby <jslaby@suse.cz>
Link: http://lkml.kernel.org/r/20200720082709.252805-1-pihsun@chromium.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agosquashfs: fix length field overlap check in metadata reading
Phillip Lougher [Fri, 24 Jul 2020 04:15:40 +0000 (21:15 -0700)]
squashfs: fix length field overlap check in metadata reading

This is a regression introduced by the "migrate from ll_rw_block usage
to BIO" patch.

Squashfs packs structures on byte boundaries, and due to that the length
field (of the metadata block) may not be fully in the current block.
The new code rewrote and introduced a faulty check for that edge case.

Fixes: 93e72b3c612adcaca1 ("squashfs: migrate from ll_rw_block usage to BIO")
Reported-by: Bernd Amend <bernd.amend@gmail.com>
Signed-off-by: Phillip Lougher <phillip@squashfs.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Adrien Schildknecht <adrien+dev@schischi.me>
Cc: Guenter Roeck <groeck@chromium.org>
Cc: Daniel Rosenberg <drosen@google.com>
Link: http://lkml.kernel.org/r/20200717195536.16069-1-phillip@squashfs.org.uk
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agomailmap: add entry for Mike Rapoport
Mike Rapoport [Fri, 24 Jul 2020 04:15:37 +0000 (21:15 -0700)]
mailmap: add entry for Mike Rapoport

Add an entry to correct my email addresses.

Signed-off-by: Mike Rapoport <rppt@linux.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20200708095414.12275-1-rppt@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agokhugepaged: fix null-pointer dereference due to race
Kirill A. Shutemov [Fri, 24 Jul 2020 04:15:34 +0000 (21:15 -0700)]
khugepaged: fix null-pointer dereference due to race

khugepaged has to drop mmap lock several times while collapsing a page.
The situation can change while the lock is dropped and we need to
re-validate that the VMA is still in place and the PMD is still subject
for collapse.

But we miss one corner case: while collapsing an anonymous pages the VMA
could be replaced with file VMA.  If the file VMA doesn't have any
private pages we get NULL pointer dereference:

general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
anon_vma_lock_write include/linux/rmap.h:120 [inline]
collapse_huge_page mm/khugepaged.c:1110 [inline]
khugepaged_scan_pmd mm/khugepaged.c:1349 [inline]
khugepaged_scan_mm_slot mm/khugepaged.c:2110 [inline]
khugepaged_do_scan mm/khugepaged.c:2193 [inline]
khugepaged+0x3bba/0x5a10 mm/khugepaged.c:2238

The fix is to make sure that the VMA is anonymous in
hugepage_vma_revalidate().  The helper is only used for collapsing
anonymous pages.

Fixes: 99cb0dbd47a1 ("mm,thp: add read-only THP support for (non-shmem) FS")
Reported-by: syzbot+ed318e8b790ca72c5ad0@syzkaller.appspotmail.com
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Acked-by: Yang Shi <yang.shi@linux.alibaba.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20200722121439.44328-1-kirill.shutemov@linux.intel.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agomm/hugetlb: avoid hardcoding while checking if cma is enabled
Barry Song [Fri, 24 Jul 2020 04:15:30 +0000 (21:15 -0700)]
mm/hugetlb: avoid hardcoding while checking if cma is enabled

hugetlb_cma[0] can be NULL due to various reasons, for example, node0
has no memory.  so NULL hugetlb_cma[0] doesn't necessarily mean cma is
not enabled.  gigantic pages might have been reserved on other nodes.
This patch fixes possible double reservation and CMA leak.

[akpm@linux-foundation.org: fix CONFIG_CMA=n warning]
[sfr@canb.auug.org.au: better checks before using hugetlb_cma]
Link: http://lkml.kernel.org/r/20200721205716.6dbaa56b@canb.auug.org.au
Fixes: cf11e85fc08c ("mm: hugetlb: optionally allocate gigantic hugepages using cma")
Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Acked-by: Roman Gushchin <guro@fb.com>
Cc: Jonathan Cameron <jonathan.cameron@huawei.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20200710005726.36068-1-song.bao.hua@hisilicon.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agomm: memcg/slab: fix memory leak at non-root kmem_cache destroy
Muchun Song [Fri, 24 Jul 2020 04:15:27 +0000 (21:15 -0700)]
mm: memcg/slab: fix memory leak at non-root kmem_cache destroy

If the kmem_cache refcount is greater than one, we should not mark the
root kmem_cache as dying.  If we mark the root kmem_cache dying
incorrectly, the non-root kmem_cache can never be destroyed.  It
resulted in memory leak when memcg was destroyed.  We can use the
following steps to reproduce.

  1) Use kmem_cache_create() to create a new kmem_cache named A.
  2) Coincidentally, the kmem_cache A is an alias for kmem_cache B,
     so the refcount of B is just increased.
  3) Use kmem_cache_destroy() to destroy the kmem_cache A, just
     decrease the B's refcount but mark the B as dying.
  4) Create a new memory cgroup and alloc memory from the kmem_cache
     B. It leads to create a non-root kmem_cache for allocating memory.
  5) When destroy the memory cgroup created in the step 4), the
     non-root kmem_cache can never be destroyed.

If we repeat steps 4) and 5), this will cause a lot of memory leak.  So
only when refcount reach zero, we mark the root kmem_cache as dying.

Fixes: 92ee383f6daa ("mm: fix race between kmem_cache destroy, create and deactivate")
Signed-off-by: Muchun Song <songmuchun@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Shakeel Butt <shakeelb@google.com>
Acked-by: Roman Gushchin <guro@fb.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/20200716165103.83462-1-songmuchun@bytedance.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agomm/memcg: fix refcount error while moving and swapping
Hugh Dickins [Fri, 24 Jul 2020 04:15:24 +0000 (21:15 -0700)]
mm/memcg: fix refcount error while moving and swapping

It was hard to keep a test running, moving tasks between memcgs with
move_charge_at_immigrate, while swapping: mem_cgroup_id_get_many()'s
refcount is discovered to be 0 (supposedly impossible), so it is then
forced to REFCOUNT_SATURATED, and after thousands of warnings in quick
succession, the test is at last put out of misery by being OOM killed.

This is because of the way moved_swap accounting was saved up until the
task move gets completed in __mem_cgroup_clear_mc(), deferred from when
mem_cgroup_move_swap_account() actually exchanged old and new ids.
Concurrent activity can free up swap quicker than the task is scanned,
bringing id refcount down 0 (which should only be possible when
offlining).

Just skip that optimization: do that part of the accounting immediately.

Fixes: 615d66c37c75 ("mm: memcontrol: fix memcg id ref counter on swap charge move")
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Alex Shi <alex.shi@linux.alibaba.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/alpine.LSU.2.11.2007071431050.4726@eggly.anvils
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agomm/memcontrol: fix OOPS inside mem_cgroup_get_nr_swap_pages()
Bhupesh Sharma [Fri, 24 Jul 2020 04:15:21 +0000 (21:15 -0700)]
mm/memcontrol: fix OOPS inside mem_cgroup_get_nr_swap_pages()

Prabhakar reported an OOPS inside mem_cgroup_get_nr_swap_pages()
function in a corner case seen on some arm64 boards when kdump kernel
runs with "cgroup_disable=memory" passed to the kdump kernel via
bootargs.

The root-cause behind the same is that currently mem_cgroup_swap_init()
function is implemented as a subsys_initcall() call instead of a
core_initcall(), this means 'cgroup_memory_noswap' still remains set to
the default value (false) even when memcg is disabled via
"cgroup_disable=memory" boot parameter.

This may result in premature OOPS inside mem_cgroup_get_nr_swap_pages()
function in corner cases:

  Unable to handle kernel NULL pointer dereference at virtual address 0000000000000188
  Mem abort info:
    ESR = 0x96000006
    EC = 0x25: DABT (current EL), IL = 32 bits
    SET = 0, FnV = 0
    EA = 0, S1PTW = 0
  Data abort info:
    ISV = 0, ISS = 0x00000006
    CM = 0, WnR = 0
  [0000000000000188] user address but active_mm is swapper
  Internal error: Oops: 96000006 [#1] SMP
  Modules linked in:
  <..snip..>
  Call trace:
    mem_cgroup_get_nr_swap_pages+0x9c/0xf4
    shrink_lruvec+0x404/0x4f8
    shrink_node+0x1a8/0x688
    do_try_to_free_pages+0xe8/0x448
    try_to_free_pages+0x110/0x230
    __alloc_pages_slowpath.constprop.106+0x2b8/0xb48
    __alloc_pages_nodemask+0x2ac/0x2f8
    alloc_page_interleave+0x20/0x90
    alloc_pages_current+0xdc/0xf8
    atomic_pool_expand+0x60/0x210
    __dma_atomic_pool_init+0x50/0xa4
    dma_atomic_pool_init+0xac/0x158
    do_one_initcall+0x50/0x218
    kernel_init_freeable+0x22c/0x2d0
    kernel_init+0x18/0x110
    ret_from_fork+0x10/0x18
  Code: aa1403e3 91106000 97f82a27 14000011 (f940c663)
  ---[ end trace 9795948475817de4 ]---
  Kernel panic - not syncing: Fatal exception
  Rebooting in 10 seconds..

Fixes: eccb52e78809 ("mm: memcontrol: prepare swap controller setup for integration")
Reported-by: Prabhakar Kushwaha <pkushwaha@marvell.com>
Signed-off-by: Bhupesh Sharma <bhsharma@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: James Morse <james.morse@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Link: http://lkml.kernel.org/r/1593641660-13254-2-git-send-email-bhsharma@redhat.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agomm: initialize return of vm_insert_pages
Tom Rix [Fri, 24 Jul 2020 04:15:18 +0000 (21:15 -0700)]
mm: initialize return of vm_insert_pages

clang static analysis reports a garbage return

  In file included from mm/memory.c:84:
  mm/memory.c:1612:2: warning: Undefined or garbage value returned to caller [core.uninitialized.UndefReturn]
          return err;
          ^~~~~~~~~~

The setting of err depends on a loop executing.  So initialize err.

Signed-off-by: Tom Rix <trix@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20200703155354.29132-1-trix@redhat.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agovfs/xattr: mm/shmem: kernfs: release simple xattr entry in a right way
Chengguang Xu [Fri, 24 Jul 2020 04:15:14 +0000 (21:15 -0700)]
vfs/xattr: mm/shmem: kernfs: release simple xattr entry in a right way

After commit fdc85222d58e ("kernfs: kvmalloc xattr value instead of
kmalloc"), simple xattr entry is allocated with kvmalloc() instead of
kmalloc(), so we should release it with kvfree() instead of kfree().

Fixes: fdc85222d58e ("kernfs: kvmalloc xattr value instead of kmalloc")
Signed-off-by: Chengguang Xu <cgxu519@mykernel.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Daniel Xu <dxu@dxuuu.xyz>
Cc: Chris Down <chris@chrisdown.name>
Cc: Andreas Dilger <adilger@dilger.ca>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: <stable@vger.kernel.org> [5.7]
Link: http://lkml.kernel.org/r/20200704051608.15043-1-cgxu519@mykernel.net
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agomm/mmap.c: close race between munmap() and expand_upwards()/downwards()
Kirill A. Shutemov [Fri, 24 Jul 2020 04:15:11 +0000 (21:15 -0700)]
mm/mmap.c: close race between munmap() and expand_upwards()/downwards()

VMA with VM_GROWSDOWN or VM_GROWSUP flag set can change their size under
mmap_read_lock().  It can lead to race with __do_munmap():

Thread A Thread B
__do_munmap()
  detach_vmas_to_be_unmapped()
  mmap_write_downgrade()
expand_downwards()
  vma->vm_start = address;
  // The VMA now overlaps with
  // VMAs detached by the Thread A
// page fault populates expanded part
// of the VMA
  unmap_region()
    // Zaps pagetables partly
    // populated by Thread B

Similar race exists for expand_upwards().

The fix is to avoid downgrading mmap_lock in __do_munmap() if detached
VMAs are next to VM_GROWSDOWN or VM_GROWSUP VMA.

[akpm@linux-foundation.org: s/mmap_sem/mmap_lock/ in comment]

Fixes: dd2283f2605e ("mm: mmap: zap pages with read mmap_sem in munmap")
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Yang Shi <yang.shi@linux.alibaba.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: <stable@vger.kernel.org> [4.20+]
Link: http://lkml.kernel.org/r/20200709105309.42495-1-kirill.shutemov@linux.intel.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
3 years agoMerge tag 'misc-habanalabs-next-2020-07-24' of git://people.freedesktop.org/~gabbayo...
Greg Kroah-Hartman [Fri, 24 Jul 2020 18:22:25 +0000 (20:22 +0200)]
Merge tag 'misc-habanalabs-next-2020-07-24' of git://people.freedesktop.org/~gabbayo/linux into char-misc-next

Oded writes:

This tag contains the following changes for kernel 5.9-rc1:

- Remove rate limiters from GAUDI configuration (no longer needed).
- Set maximum amount of in-flight CS per ASIC type and increase
  the maximum amount for GAUDI.
- Refactor signal/wait command submissions code
- Calculate trace frequency from PLLs to show accurate profiling data
- Rephrase error messages to make them more clear to the common user
- Add statistics of dropped CS (counter per possible reason for drop)
- Get ECC information from firmware
- Remove support for partial SoC reset in Gaudi
- Halt device CPU only when reset is certain to happen. Sometimes we abort
  the reset procedure and in that case we can't leave device CPU in halt
  mode.
- set each CQ to its own work queue to prevent a race between completions
  on different CQs.
- Use queue pi/ci in order to determine queue occupancy. This is done to
  make the code reusable between current and future ASICs.
- Add more validations for user inputs.
- Refactor PCIe controller configuration to make the code reusable between
  current and future ASICs.
- Update firmware interface headers to latest version
- Move all common code to a dedicated common sub-folder

* tag 'misc-habanalabs-next-2020-07-24' of git://people.freedesktop.org/~gabbayo/linux: (28 commits)
  habanalabs: Fix memory leak in error flow of context initialization
  habanalabs: use no flags on MMU cache invalidation
  habanalabs: enable device before hw_init()
  habanalabs: create internal CB pool
  habanalabs: update hl_boot_if.h from firmware
  habanalabs: create common folder
  habanalabs: check for DMA errors when clearing memory
  habanalabs: verify queue can contain all cs jobs
  habanalabs: Assign each CQ with its own work queue
  habanalabs: halt device CPU only upon certain reset
  habanalabs: remove unused hash
  habanalabs: use queue pi/ci in order to determine queue occupancy
  habanalabs: configure maximum queues per asic
  habanalabs: remove soft-reset support from GAUDI
  habanalabs: PCIe iATU refactoring
  habanalabs: Extract ECC information from FW
  habanalabs: Add dropped cs statistics info struct
  habanalabs: extract cpu boot status lookup
  habanalabs: rephrase error messages
  habanalabs: Increase queues depth
  ...

3 years agohabanalabs: Fix memory leak in error flow of context initialization
Tomer Tayar [Thu, 23 Jul 2020 06:17:57 +0000 (09:17 +0300)]
habanalabs: Fix memory leak in error flow of context initialization

Add a missing free of the cs_pending array in the error flow of context
initialization.

Fixes: c16d45f42b64 ("habanalabs: Use pending CS amount per ASIC")

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: use no flags on MMU cache invalidation
Tomer Tayar [Sun, 19 Jul 2020 08:00:03 +0000 (11:00 +0300)]
habanalabs: use no flags on MMU cache invalidation

gaudi_mmu_invalidate_cache() doesn't use the flags parameter, and thus
it can be set to 0 when the function is called in the gaudi only files.

Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: enable device before hw_init()
Oded Gabbay [Wed, 15 Jul 2020 18:59:32 +0000 (21:59 +0300)]
habanalabs: enable device before hw_init()

Device is now enabled before the hw_init() because part of the
initialization requires communication with the device firmware to get
information that is required for the initialization itself

Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Tomer Tayar <ttayar@habana.ai>
3 years agohabanalabs: create internal CB pool
Ofir Bitton [Mon, 13 Jul 2020 10:36:55 +0000 (13:36 +0300)]
habanalabs: create internal CB pool

Create a device MMU-mapped internal command buffer pool, in order to allow
the driver to allocate CBs for the signal/wait operations
that are fetched by the queues when they are configured with the user's
address space ID.

We must pre-map this internal pool due to performance issues.

This pool is needed for future ASIC support and it is currently unused in
GOYA and GAUDI.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: update hl_boot_if.h from firmware
Oded Gabbay [Mon, 13 Jul 2020 10:11:33 +0000 (13:11 +0300)]
habanalabs: update hl_boot_if.h from firmware

Update the boot interface file from the latest version from firmware.
Defines for secure boot were added.

Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Omer Shpigelman <oshpigelman@habana.ai>
3 years agohabanalabs: create common folder
Oded Gabbay [Mon, 13 Jul 2020 09:21:04 +0000 (12:21 +0300)]
habanalabs: create common folder

For internal needs of our CI we need to move all the common code into a
common folder instead of putting them in the root folder of the driver.

Same applies to the common header files under include/

Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Omer Shpigelman <oshpigelman@habana.ai>
3 years agohabanalabs: check for DMA errors when clearing memory
Moti Haimovski [Wed, 24 Jun 2020 16:40:57 +0000 (19:40 +0300)]
habanalabs: check for DMA errors when clearing memory

In GAUDI we use QMAN0 DMA for clearing the MMU memory region
at initialization. if this operation fails it places the DMA in an error
state and then when trying to initialize QMAN0 we fail and erroneously
assume its the QMAN that failed.

This commit adds a check and clear of such DMA errors at initialization so
we will have a better understanding of what went wrong.

Signed-off-by: Moti Haimovski <mhaimovski@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: verify queue can contain all cs jobs
Ofir Bitton [Wed, 8 Jul 2020 07:16:27 +0000 (10:16 +0300)]
habanalabs: verify queue can contain all cs jobs

In order for the user to be aware of wrong inputs, we must return
error in case the amount of jobs per cs exceeds the corresponding
queue size.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: Assign each CQ with its own work queue
Ofir Bitton [Sun, 5 Jul 2020 10:35:51 +0000 (13:35 +0300)]
habanalabs: Assign each CQ with its own work queue

We identified a possible race during job completion when working
with a single multi-threaded work queue. In order to overcome this
race we suggest using a single threaded work queue per completion
queue, hence we guarantee jobs completion in order.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: halt device CPU only upon certain reset
Oded Gabbay [Sun, 5 Jul 2020 12:48:34 +0000 (15:48 +0300)]
habanalabs: halt device CPU only upon certain reset

Currently the driver halts the device CPU in the halt engines function,
which halts all the engines of the ASIC. The problem is that if later on we
stop the reset process (due to inability to clean memory mappings in time),
the CPU will remain in halt mode. This creates many issues, such as
thermal/power control and FLR handling.

Therefore, move the halting of the device CPU to the very end of the reset
process, just before writing to the registers to initiate the reset. In
addition, the driver now needs to send a message to the device F/W to
disable it from sending interrupts to the host machine because during halt
engines function the driver disables the MSI/MSI-X interrupts.

Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Tomer Tayar <ttayar@habana.ai>
3 years agohabanalabs: remove unused hash
Omer Shpigelman [Tue, 7 Jul 2020 21:29:54 +0000 (00:29 +0300)]
habanalabs: remove unused hash

Remove an old hash that is not in use anymore.

Signed-off-by: Omer Shpigelman <oshpigelman@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: use queue pi/ci in order to determine queue occupancy
Ofir Bitton [Wed, 24 Jun 2020 11:49:43 +0000 (14:49 +0300)]
habanalabs: use queue pi/ci in order to determine queue occupancy

Instead of using the free slots amount on the compute CQ to determine
whether we can submit work to queues, use the queues pi/ci.

This is needed in future ASICs where we don't have CQ per queue.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: configure maximum queues per asic
Ofir Bitton [Tue, 23 Jun 2020 11:50:39 +0000 (14:50 +0300)]
habanalabs: configure maximum queues per asic

Currently the amount of maximum queues is statically configured.
Using a static value is causing redundunt cycles when traversing
all queues and consumes more memory than actually needed.
In this patch we configure each asic with the exact number of
queues needed.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: remove soft-reset support from GAUDI
Oded Gabbay [Fri, 3 Jul 2020 17:58:23 +0000 (20:58 +0300)]
habanalabs: remove soft-reset support from GAUDI

Soft-reset isn't supported in GAUDI. Remove the code that performs it and
print error in case the user wants to do it via sysfs.

Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Tomer Tayar <ttayar@habana.ai>
3 years agohabanalabs: PCIe iATU refactoring
Ofir Bitton [Mon, 15 Jun 2020 14:45:12 +0000 (17:45 +0300)]
habanalabs: PCIe iATU refactoring

Divide iATU initialization into inbound/outbound methods.
We must separate it in order to enable different match mode
per PCIe region.
In addition, added support for PCI address match mode.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: Extract ECC information from FW
Oded Gabbay [Sun, 17 May 2020 05:20:35 +0000 (08:20 +0300)]
habanalabs: Extract ECC information from FW

ECC (Error Correcting Code) interrupts are going to be handled
by the FW. Hence, we define an interface in which the driver can
obtain the relevant ECC information.
This information is needed for monitoring and can also lead
to a hard reset if ECC error is not correctable.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: Add dropped cs statistics info struct
Ofir Bitton [Thu, 18 Jun 2020 06:51:16 +0000 (09:51 +0300)]
habanalabs: Add dropped cs statistics info struct

Add command submission statistics structure which can be obtained
through the info ioctl. Each drop counter describes the reason for
which the command submission was dropped.
This information is needed for the user to be aware of the specific
reason for which the submitted work was dropped. The user can then
utilize the driver more efficiently.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: extract cpu boot status lookup
Christine Gharzuzi [Tue, 23 Jun 2020 16:21:22 +0000 (19:21 +0300)]
habanalabs: extract cpu boot status lookup

Extract detection of the cpu boot status to a function
to allow code reuse

Signed-off-by: Christine Gharzuzi <cgharzuzi@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: rephrase error messages
Oded Gabbay [Mon, 22 Jun 2020 06:52:22 +0000 (09:52 +0300)]
habanalabs: rephrase error messages

rephrase some error/warning/notice messages to make them more accessible to
ordinary users.

There is no need to print context ASID as the driver currently doesn't
support multiple contexts.

Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Tomer Tayar <ttayar@habana.ai>
3 years agohabanalabs: Increase queues depth
Ofir Bitton [Mon, 15 Jun 2020 09:09:50 +0000 (12:09 +0300)]
habanalabs: Increase queues depth

After recent concurrent cs amount increase, we must also
increase queues depth since much more concurrent work can be done.
All external queue depths were increased to 4096 as gaudi's
internal queue depths were also increased to 1024.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: rephrase error message
Omer Shpigelman [Sat, 4 Jul 2020 19:51:16 +0000 (22:51 +0300)]
habanalabs: rephrase error message

Rephrase F/W error message to make it more understandable to ordinary
users.

Signed-off-by: Omer Shpigelman <oshpigelman@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: calculate trace frequency from PLL
Adam Aharon [Tue, 26 May 2020 08:04:30 +0000 (11:04 +0300)]
habanalabs: calculate trace frequency from PLL

The profiler needs to know the PLL values for correctly showing the
profiling data. Because our firmware can use different PLL configurations,
we need to read the PLL values from the ASIC to pass them to the profiler.

Signed-off-by: Adam Aharon <aaharon@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: align armcp_packet structure to 8 bytes
Oded Gabbay [Tue, 9 Jun 2020 16:58:44 +0000 (19:58 +0300)]
habanalabs: align armcp_packet structure to 8 bytes

Once there is a 64-bit field in a structure, GCC compiler for ARM aligns
the structure to 8 bytes. In order to avoid confusion when these
structures are being passed between CPUs from different architectures, we
explicitly align the structure to 8 bytes.

Reviewed-by: Omer Shpigelman <oshpigelman@habana.ai>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agouapi/habanalabs: fix some comments
Oded Gabbay [Tue, 9 Jun 2020 13:14:55 +0000 (16:14 +0300)]
uapi/habanalabs: fix some comments

MAP/UNMAP are done also for device memory.

Reviewed-by: Omer Shpigelman <oshpigelman@habana.ai>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: Use mask instead of shift in sync stream registers
Ofir Bitton [Mon, 1 Jun 2020 07:38:46 +0000 (10:38 +0300)]
habanalabs: Use mask instead of shift in sync stream registers

Use proper bitfield masks instead of shifting values when configuring
packets sent to device.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: sync stream generic functionality
Ofir Bitton [Thu, 14 May 2020 15:25:47 +0000 (18:25 +0300)]
habanalabs: sync stream generic functionality

Currently sync stream is limited only for external queues. We want to
remove this constraint by adding a new queue property dedicated for sync
stream. In addition we move the initialization and reset methods to the
common code since we can re-use them with slight changes.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: Use pending CS amount per ASIC
Ofir Bitton [Tue, 2 Jun 2020 09:28:27 +0000 (12:28 +0300)]
habanalabs: Use pending CS amount per ASIC

Training schemes requires much more concurrent command submissions than
inference does. In addition, training command submissions can be completed
in a non serialized manner. Hence, we add support in which each ASIC will
be able to configure the amount of concurrent pending command submissions,
rather than use a predefined amount. This change will enhance performance
by allowing the user to add more concurrent work without waiting for the
previous work to be completed.

Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agohabanalabs: remove rate limiters from GAUDI
Oded Gabbay [Mon, 15 Jun 2020 16:25:57 +0000 (19:25 +0300)]
habanalabs: remove rate limiters from GAUDI

We no longer need to initialize the rate limiters in GAUDI A1.

Reviewed-by: Omer Shpigelman <oshpigelman@habana.ai>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
3 years agoMerge tag 'icc-5.9-rc1' of https://git.linaro.org/people/georgi.djakov/linux into...
Greg Kroah-Hartman [Fri, 24 Jul 2020 15:24:09 +0000 (17:24 +0200)]
Merge tag 'icc-5.9-rc1' of https://git.linaro.org/people/georgi.djakov/linux into char-misc-next

Georgi writes:

interconnect changes for 5.9

Here are the interconnect changes for the 5.9-rc1 merge window
consisting mostly of changes that give the core more flexibility
in order to support some new provider drivers.

Core changes:
- Export of_icc_get_from_provider()
- Relax requirement in of_icc_get_from_provider()
- Allow inter-provider pairs to be configured
- Mark all dummy functions as static inline

Signed-off-by: Georgi Djakov <georgi.djakov@linaro.org>
* tag 'icc-5.9-rc1' of https://git.linaro.org/people/georgi.djakov/linux:
  interconnect: Mark all dummy functions as static inline
  interconnect: Allow inter-provider pairs to be configured
  interconnect: Relax requirement in of_icc_get_from_provider()
  interconnect: Export of_icc_get_from_provider()

3 years agodyndbg: export ddebug_exec_queries
Jim Cromie [Sun, 19 Jul 2020 23:10:58 +0000 (17:10 -0600)]
dyndbg: export ddebug_exec_queries

Export ddebug_exec_queries() for use by modules.

This will allow module authors to control all their *pr_debug*s
dynamically.  And since ddebug_exec_queries() is what implements
"echo $query >control", it gives the same per-callsite control.

Virtues of this:
- simplicity. just an export.
- full control over any/all subsets of callsites.
- same "query/command-string" in code and console
- full callsite selectivity with module file line format

Format in particular deserves special attention; it is where
low-hanging fruit will be found.

Consider: drivers/gpu/drm/amd/display/include/logger_types.h:

  #define DC_LOG_SURFACE(...)          pr_debug("[SURFACE]:"__VA_ARGS__)
  #define DC_LOG_HW_LINK_TRAINING(...) pr_debug("[HW_LINK_TRAINING]:"__VA_ARGS__)
  .. 9 more ..

Thats 11 string prefixes, used in 804 places in drivers/gpu/**
Clearly this is a systematized classification of those callsites.
And one I'd expect to see repeated often.

Using ddebug_exec_queries(), authors can select on those prefixes
as a unitary set, equivalent to:

    echo "module=MODULE_NAME format=^[SURFACE]: +p" >control

Trivially, those sets can be subsected with the other query terms too,
say file=foo, should the author see fit.

Perhaps as important, users can modify the set of enabled callsites,
presumably to aid debugging by enabling helpful debug callsites, and
disabling those that just clutter the info.

Authors could even alter [fmlt] flags, though I dont see a good reason
why they would.  Perhaps harnessed by bug-logging automation to get
fuller, or more minimal bug-reports.

DRM

drm has both drm.debug, which defines 32 categories of drm_printk
logging, and entirely separate uses of pr_debug, which are dynamic on
this i915 laptop, running mainline.  So I can observe and report on
both.

The i915 driver has 118 dyndbg callsites, with following
"classifications" defined in drivers/gpu/drm/i915/gvt/**

$ grep 915 /proc/dynamic_debug/control | cut -d= -f2 | cut -d: -f1,2 | sort -u
_ "gvt: cmd
_ "gvt: core
_ "gvt: dpy
_ "gvt: el
_ "gvt: irq
_ "gvt: mm
_ "gvt: mmio
_ "gvt: render
_ "gvt: sched
_ "%s for root hub!\012"
_ "Vendor defined info completion code %u\012"

This classification is entirely out-of-band for control by drm.debug,
and is only available to root user at the console.  But module authors
can activate them with ddebug_exec_queries(sprintf("format=^%s +p")),
and then decide how to expose the groups to the user for max utility.

drm.debug

drm.debug has 32 bit-flags, and matching enum drm_debug_category
values to classify the ~2943 DRM_DEBUG*() callsites in drivers/gpu

The drm.debug callback could invoke ddebug_exec_queries() with 32
different hardcoded query strings, needing only (bit) ? " +p" : " -p"
added.

I briefly enabled drm.debug=0xff on my i915 laptop, which yielded
these unique prefixes: (dmesg | cut -c17- | cut -d\] -f1 | sort -u)

[drm:drm_atomic_check_only [drm
[drm:drm_atomic_get_crtc_state [drm
[drm:drm_atomic_get_plane_state [drm
[drm:drm_atomic_nonblocking_commit [drm
[drm:drm_atomic_set_fb_for_plane [drm
[drm:drm_atomic_state_default_clear [drm
[drm:__drm_atomic_state_free [drm
[drm:drm_atomic_state_init [drm
[drm:drm_crtc_vblank_helper_get_vblank_timestamp_internal [drm
[drm:drm_handle_vblank [drm
[drm:drm_ioctl [drm
[drm:drm_mode_addfb2 [drm
[drm:drm_mode_object_get [drm
[drm:drm_mode_object_put.part.0 [drm
[drm:drm_update_vblank_count [drm
[drm:drm_vblank_enable [drm
[drm:drm_vblank_restore [drm
[drm:vblank_disable_fn [drm
i915 0000:00:02.0: [drm:gen9_set_dc_state [i915
i915 0000:00:02.0: [drm:intel_atomic_get_global_obj_state [i915
i915 0000:00:02.0: [drm:__intel_display_power_get_domain.part.0 [i915
i915 0000:00:02.0: [drm:__intel_display_power_put_domain [i915
i915 0000:00:02.0: [drm:intel_plane_atomic_calc_changes [i915
i915 0000:00:02.0: [drm:skl_enable_dc6 [i915

Several good format=^prefixes are apparent there, and some misses.

 ^[drm:drm_atomic_ # misses: [drm:__drm_atomic_state_free [drm
 ^[drm:drm_ioctl
 ^[drm:drm_mode
 ^[drm:drm_vblank_ # misses: [drm:drm_update_vblank_count & [drm:vblank_disable_fn

Its not a perfect 1:1 single format-match per class, but the misses
above can be covered with 1 & 2 additional queries, which can be
concatenated together with ";" separators and submitted with 1 call.

Benefits:

For drm, adapting DRM_DEBUG to use dynamic-debug inside could
replicate (and thereby obsolete) lots of bit-checking in current
DRM_DEBUG callsites, at least with JUMP_LABEL optimized code.
ddebug_exec_queries() and a handful of fixed query-strings can select
and thereby control the already classified callsites.

With the classes mapped to queries, the enum type and parameter can be
eliminated (folded away with macro magic), at least for DYNAMIC_DEBUG
& JUMP_LABEL builds.

Is it safe ?

ddebug_exec_queries() is currently exposed to user space in
several limited ways;

1 it is called from module-load callback, where it implements the
  $modname.dyndbg=+p "fake" parameter provided to all modules.

2 it handles query input via >control directly

IOW, it is "fully" exposed to local root user; exposing the same
functionality to other kernel modules is no additional risk.

The other standard issue to check is locking:

dyndbg has a single mutex, taken by ddebug_change to handle >control,
and by ddebug_proc_(start|stop) to span `cat control`.  Queries
submitted via export will typically have module specified, which
dramatically cuts the scan by ddebug_change vs "module=* +p".
ISTM this proposed export presents no locking problems.

TLDR;

It would be interesting to see how drm.dyndbg=$QUERY and
drm.debug=$HEXY would interact; it might be order dependent, as
if given as modprobe args or in /etc/modprobe.d/

Acked-by: <jbaron@akamai.com>
Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
Link: https://lore.kernel.org/r/20200719231058.1586423-19-jim.cromie@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agodyndbg: shorten our logging prefix, drop __func__
Jim Cromie [Sun, 19 Jul 2020 23:10:57 +0000 (17:10 -0600)]
dyndbg: shorten our logging prefix, drop __func__

For log-message output, reduce column space consumed by current
pr_fmt by dropping __func__ and shortening "dynamic_debug" to
"dyndbg".  This improves readability on narrow consoles, and better
matches other kernel boot info messages.

Acked-by: <jbaron@akamai.com>
Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
Link: https://lore.kernel.org/r/20200719231058.1586423-18-jim.cromie@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
3 years agodyndbg: allow anchored match on format query term
Jim Cromie [Sun, 19 Jul 2020 23:10:56 +0000 (17:10 -0600)]
dyndbg: allow anchored match on format query term

This should work:

  echo module=amd* format=^[IF_TRACE]: +p  >/proc/dynamic_debug/control

consider drivers/gpu/drm/amd/display/include/logger_types.h:
It has 11 defines like:

  #define DC_LOG_IF_TRACE(...) pr_debug("[IF_TRACE]:"__VA_ARGS__)

These defines are used 804 times at recent count; they are a good use
case to evaluate existing format-message based classifications of
*pr_debug*.  Those macros prefix the supplied format with a fixed
string, I'd expect most existing message classification schemes to do
something similar.

Hence we want to be able to anchor our match to the beginning of the
format string, allowing easy construction of clear and precise
queries, leveraging the existing classification scheme to enable and
disable those callsites.

Note that unlike other search terms, formats are implicitly floating
substring matches, without the need for explicit wildcards.

This makes no attempt at wider regex features, just the one we need.

TLDR: Using the anchor also means the []s are less helpful for
disamiguating the prefix from a random in-message occurrence, allowing
shorter prefixes.

Acked-by: <jbaron@akamai.com>
Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
Link: https://lore.kernel.org/r/20200719231058.1586423-17-jim.cromie@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>