linux-kernel - Re: [PATCH] drm/dp: Don't attempt AUX transfers when eDP panels are not powered

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: Mon, 19 Feb 2024 15:45:39 +1100
From: Eizan Miyamoto <eizan@...omium.org>
To: Doug Anderson <dianders@...omium.org>
Cc: neil.armstrong@...aro.org, Javier Martinez Canillas <javierm@...hat.com>, 
	Jani Nikula <jani.nikula@...el.com>, Maxime Ripard <mripard@...nel.org>, 
	Thomas Zimmermann <tzimmermann@...e.de>, Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>, 
	Hsin-Yi Wang <hsinyi@...omium.org>, dri-devel@...ts.freedesktop.org, 
	Ankit Nautiyal <ankit.k.nautiyal@...el.com>, Daniel Vetter <daniel@...ll.ch>, 
	David Airlie <airlied@...il.com>, Heiner Kallweit <hkallweit1@...il.com>, 
	Imre Deak <imre.deak@...el.com>, Jessica Zhang <quic_jesszhan@...cinc.com>, 
	Sam Ravnborg <sam@...nborg.org>, Stanislav Lisovskiy <stanislav.lisovskiy@...el.com>, 
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] drm/dp: Don't attempt AUX transfers when eDP panels are
 not powered

Hiya,

On Fri, Feb 16, 2024 at 4:09 AM Doug Anderson <dianders@...omium.org> wrote:
[...snip...]
> Sounds good. I know Eizan (CCed, but also a ChromeOS person) was going
> to poke at it a bit and seemed willing to provide a Tested-by. I'll
> let him chime in.

Yup, I've tested this like so:

1. I started by slightly modifying a recent chromeos-5.15 kernel checked
   out to _right before_ the patch we are discussing to emit some
   tracing info to dmesg at entry/exit of auxdev_read_iter(). I
   installed it on a "tentacool" Corsola device.
2. I then rebooted the device and ran these commands:

   # dmesg -w &
   # while /bin/true; do echo -n "dpms: "; cat
/sys/devices/platform/soc/14000000.syscon/mediatek-drm.7.auto/drm/card0/card0-eDP-1/dpms;
dd if=/dev/drm_dp_aux1  count=1 of=/dev/null; sleep 30; done

And after a few minutes, I saw the following output:

dpms: On
[  435.603257] auxdev_open by pid 6327 inode num 108 dev 256901121
[  435.603369] start auxdev_read_iter by pid 6327
1+0 records in
1+0 records out
[  435.756547] finish auxdev_read_iter by pid 6327 status 512
[  435.756632] auxdev_release by pid 6327 inode num 108 dev 256901121
512 bytes copied, 0.153862 s, 3.3 kB/s
[  455.418637] [drm] mtk_crtc_ddp_hw_fini 459 event 0x0000000000000000
0xffffff80c0277080 0xffffff80c0277080
dpms: Off
[  465.775104] auxdev_open by pid 6399 inode num 108 dev 256901121
[  465.775218] start auxdev_read_iter by pid 6399
dd: error reading '/dev/drm_dp_aux1': Connection timed out
0+0 records in
0+0 records out
0 bytes copied, 16.6631 s, 0.0 kB/s
[  482.437762] finish auxdev_read_iter by pid 6399 status -110
[  482.438200] auxdev_release by pid 6399 inode num 108 dev 256901121

(OK, so what to look for in the above is the ETIMEDOUT returned by
auxdev_read_iter after about 17s when dpms was turned off.)

I then checked out the repo to the patch we are discussing and did the
same thing, and after a few minutes, I saw:

dpms: On
[  441.892692] auxdev_open by pid 6317 inode num 108 dev 256901121
[  441.892786] start auxdev_read_iter by pid 6317
1+0 records in
1+0 records out
512 bytes copied, 0.148004 s, 3.5 kB/s
[  442.040597] finish auxdev_read_iter by pid 6317 status 512
[  442.040652] auxdev_release by pid 6317 inode num 108 dev 256901121
[  455.395549] [drm] mtk_crtc_ddp_hw_fini 459 event 0x0000000000000000
0xffffff80c3993080 0xffffff80c3993080
dpms: Off
dd: error reading '/dev/drm_dp_aux1': Device or resource busy
0+0 records in
0+0 records out
0 bytes copied, 0.000241 s, 0.0 kB/s
[  472.055296] auxdev_open by pid 6439 inode num 108 dev 256901121
[  472.055388] start auxdev_read_iter by pid 6439
[  472.055421] finish auxdev_read_iter by pid 6439 status -16
[  472.055571] auxdev_release by pid 6439 inode num 108 dev 256901121

Tested-by: Eizan Miyamoto <eizan@...omium.org>