[GIT PULL] ACPI fixes for v5.16-rc3

[GIT PULL] ACPI fixes for v5.16-rc3

[Rafael J. Wysocki]

Hi Linus,

Please pull from the tag

 git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git \
 acpi-5.16-rc3

with top-most commit 2e13e5aeda156f747919c7111723b9302836fb38

 Merge branch 'acpi-properties'

on top of commit 136057256686de39cc3a07c2e39ef6bc43003ff6

 Linux 5.16-rc2

to receive ACPI fixes for 5.16-rc3.

These fix a NULL pointer dereference in the CPPC library code and
a locking issue related to printing the names of ACPI device nodes
in the device properties framework.

Specifics:

 - Fix NULL pointer dereference in the CPPC library code occurring
   on hybrid systems without CPPC support (Rafael Wysocki).

 - Avoid attempts to acquire a semaphore with interrupts off when
   printing the names of ACPI device nodes and clean up code on
   top of that fix (Sakari Ailus).

Thanks!


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

Rafael J. Wysocki (1):
      ACPI: CPPC: Add NULL pointer check to cppc_get_perf()

Sakari Ailus (2):
      ACPI: Get acpi_device's parent from the parent field
      ACPI: Make acpi_node_get_parent() local

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

 drivers/acpi/cppc_acpi.c |  9 ++++++++-
 drivers/acpi/property.c  | 14 +++++---------
 include/linux/acpi.h     |  7 -------
 3 files changed, 13 insertions(+), 17 deletions(-)

Re: [GIT PULL] ACPI fixes for v5.16-rc3

[pr-tracker-bot]

The pull request you sent on Fri, 26 Nov 2021 20:38:05 +0100:

> git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git acpi-5.16-rc3

has been merged into torvalds/linux.git:
https://git.kernel.org/torvalds/c/5367cf1c3ad02f7f14d79733814302a96cc97b96

Thank you!

-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/prtracker.html

Re: [GIT PULL] ACPI fixes for v5.16-rc3

[Linus Torvalds]

On Fri, Nov 26, 2021 at 11:38 AM Rafael J. Wysocki <rafael@kernel.org> wrote:
>
>  git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git \
>  acpi-5.16-rc3

Hmm. This may be unrelated, but I don't think I've seen this before..
On my laptop, I get

  WARNING: CPU: 4 PID: 95940 at kernel/workqueue.c:1441 __queue_work+0x2f9/0x3b0
  Workqueue: kec_query acpi_ec_event_processor
  ..
  Call Trace:
   <TASK>
   queue_work_on+0x20/0x30
   advance_transaction+0x1a1/0x500
   acpi_ec_transaction+0x15c/0x410
   acpi_ec_space_handler+0xd2/0x270
   acpi_ev_address_space_dispatch+0x216/0x2a3
   ? acpi_ec_resume+0x20/0x20
   acpi_ex_access_region+0x1dc/0x255
   ? acpi_os_wait_semaphore+0x48/0x70
   acpi_ex_field_datum_io+0xfd/0x178
   acpi_ex_read_data_from_field+0x12e/0x171
   acpi_ex_resolve_node_to_value+0x1fe/0x281
   acpi_ds_evaluate_name_path+0x75/0xe9
   acpi_ds_exec_end_op+0x8f/0x411
   acpi_ps_parse_loop+0x495/0x5bc
   acpi_ps_parse_aml+0x94/0x2c2
   acpi_ps_execute_method+0x15e/0x193
   acpi_ns_evaluate+0x1c6/0x25d
   acpi_evaluate_object+0x12e/0x226
   acpi_ec_event_processor+0x63/0x90
   process_one_work+0x217/0x3c0

and it seems to be happening at resume time.

This was when running a51e3ac43ddb, so not the very latest git tree,
but recent.

Maybe I've missed a report of this?

               Linus

Re: [GIT PULL] ACPI fixes for v5.16-rc3

[Rafael J. Wysocki]

On Sat, Dec 4, 2021 at 6:04 PM Linus Torvalds
<torvalds@linux-foundation.org> wrote:
>
> On Fri, Nov 26, 2021 at 11:38 AM Rafael J. Wysocki <rafael@kernel.org> wrote:
> >
> >  git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git \
> >  acpi-5.16-rc3
>
> Hmm. This may be unrelated, but I don't think I've seen this before..

No, this is not related.  It shows up because of the drain_workqueue()
in __acpi_ec_flush_work().

> On my laptop, I get
>
>   WARNING: CPU: 4 PID: 95940 at kernel/workqueue.c:1441 __queue_work+0x2f9/0x3b0
>   Workqueue: kec_query acpi_ec_event_processor
>   ..
>   Call Trace:
>    <TASK>
>    queue_work_on+0x20/0x30

So this tries to add work to ec_wq while it is draining AFAICS.

>    advance_transaction+0x1a1/0x500
>    acpi_ec_transaction+0x15c/0x410
>    acpi_ec_space_handler+0xd2/0x270
>    acpi_ev_address_space_dispatch+0x216/0x2a3
>    ? acpi_ec_resume+0x20/0x20
>    acpi_ex_access_region+0x1dc/0x255
>    ? acpi_os_wait_semaphore+0x48/0x70
>    acpi_ex_field_datum_io+0xfd/0x178
>    acpi_ex_read_data_from_field+0x12e/0x171
>    acpi_ex_resolve_node_to_value+0x1fe/0x281
>    acpi_ds_evaluate_name_path+0x75/0xe9
>    acpi_ds_exec_end_op+0x8f/0x411
>    acpi_ps_parse_loop+0x495/0x5bc
>    acpi_ps_parse_aml+0x94/0x2c2
>    acpi_ps_execute_method+0x15e/0x193
>    acpi_ns_evaluate+0x1c6/0x25d
>    acpi_evaluate_object+0x12e/0x226
>    acpi_ec_event_processor+0x63/0x90
>    process_one_work+0x217/0x3c0
>
> and it seems to be happening at resume time.
>
> This was when running a51e3ac43ddb, so not the very latest git tree,
> but recent.
>
> Maybe I've missed a report of this?

It's not been reported or at least I can't recall that, but it is not
a new issue.

There is a commit to address this in my linux-next branch:

https://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git/commit/?h=linux-next&id=4a9af6cac050dce2e895ec3205c4615383ad9112

but it is interesting that you can reproduce it.  I thought it would
be rather theoretical.

Re: [GIT PULL] ACPI fixes for v5.16-rc3

[Linus Torvalds]

On Sat, Dec 4, 2021 at 12:11 PM Rafael J. Wysocki <rafael@kernel.org> wrote:
>
> but it is interesting that you can reproduce it.  I thought it would
> be rather theoretical.

Well, "reproduce it" may be putting it a bit strong...

It happened this morning, and after that suspend/resume was broken
until a reboot. But I've only ever seen it that one time.

              Linus