linux-kernel - Re: [PATCH] firmware_loader: Use init

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date: Fri, 23 Feb 2024 14:29:19 +0000
From: John Garry <john.g.garry@...cle.com>
To: Luis Chamberlain <mcgrof@...nel.org>
Cc: russ.weight@...ux.dev, gregkh@...uxfoundation.org, rafael@...nel.org,
        linux-kernel@...r.kernel.org, masahiroy@...nel.org
Subject: Re: [PATCH] firmware_loader: Use init_utsname()->release

>> ---
>> Note: As mentioned by Masahiro in [0], when CONFIG_MODVERSIONS=y it
>> could be possible for a driver to be built as a module with a different
>> kernel baseline and so use a different UTS_RELEASE from the baseline. So
>> now using init_utsname()->release could lead to a change in behaviour
>> in this driver. However, considering the nature of this driver and how it
>> would not make sense to build as module against a different tree, this
> 
> would not make sense to build it as an external module against against a
> different tree, so this change should not have any effect to users.
> 

ok

>> change should be ok.
>>
>> [0] https://urldefense.com/v3/__https://lore.kernel.org/lkml/CAK7LNAQ_r5yUjNpOppLkDBQ12sDxBYQTvRZGn1ng8D1POfZr_A@mail.gmail.com/__;!!ACWV5N9M2RV99hQ!KxnFzsitgB8e0kDndt3nYwqw0FcAPKzJjuRl9BzwLQ9dbAtqK_SZOkhHw9ssT2PobYVkh8UU7WryWKwGXg$
> 
> Thanks for doing this, could you send a v2 with the "Note:" removed but
> instead include that blurb as part of the commit log as it is important
> information to retain.

ok

> 
> Could you also test the selftests to ensure nothing is broken ?
> 
> ./tools/testing/selftests/firmware/fw_run_tests.sh

I am running this now, but it does not seem stable on a v6.8-rc5 
baseline. I am seeing hangs. Are there any known issues?

This one passed, it seems to me:

https://pastebin.com/ZySPmH9h

But then on another run I see a hang at:

...
Testing with the file missing...
Batched request_firmware() nofile try #1: OK
Batched request_firmware() nofile try #2: OK
Batched request_firmware() nofile try #3: OK
Batched request_firmware() nofile try #4: OK
Batched request_firmware() nofile try #5: OK
Batched request_firmware_into_buf() nofile try #1: OK
Batched request_firmware_into_buf() nofile try #2: OK
Batched request_firmware_into_buf() nofile try #3: OK
Batched request_firmware_into_buf() nofile try #4: OK
Batched request_firmware_into_buf() nofile try #5: OK
Batched request_firmware_direct() nofile try #1: OK
Batched request_firmware_direct() nofile try #2: OK
Batched request_firmware_direct() nofile try #3: OK
Batched request_firmware_direct() nofile try #4: OK
Batched request_firmware_direct() nofile try #5: OK
Batched request_firmware_nowait(uevent=true) nofile try #1: OK
Batched request_firmware_nowait(uevent=true) nofile try #2: OK
Batched request_firmware_nowait(uevent=true) nofile try #3: OK
Batched request_firmware_nowait(uevent=true) nofile try #4: OK
Batched request_firmware_nowait(uevent=true) nofile try #5: OK
Batched request_firmware_nowait(uevent=false) nofile try #1: OK
Batched request_firmware_nowait(uevent=false) nofile try #2: OK
Batched request_firmware_nowait(uevent=false) nofile try #3: OK
Batched request_firmware_nowait(uevent=false) nofile try #4:

And kernel log:

[  385.105004] test_firmware: #3: failed to async load firmware
[  385.181221] test_firmware: #0: failed to async load firmware
[  385.252231] test_firmware: #1: failed to async load firmware
[  385.323307] test_firmware: #2: failed to async load firmware
[  385.397444] test_firmware: #3: failed to async load firmware
[  644.462125] INFO: task fw_filesystem.s:5466 blocked for more than 120 
seconds.
[  644.548803]       Not tainted 6.8.0-rc5-g7e58e46976f2 #32
[  644.613581] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" 
disables this message.
[  765.294106] INFO: task fw_filesystem.s:5466 blocked for more than 241 
seconds.
[  765.380805]       Not tainted 6.8.0-rc5-g7e58e46976f2 #32
[  765.445610] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" 
disables this message.
[  886.126154] INFO: task fw_filesystem.s:5466 blocked for more than 362 
seconds.
[  886.212883]       Not tainted 6.8.0-rc5-g7e58e46976f2 #32
[  886.277695] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" 
disables this message.

FWIW, with my patch we get same output for "passing" run upto another hang.

Thanks,
John

>