mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Renjun Wang <renjunw0@foxmail.com>
To: Louis Chauvet <louis.chauvet@bootlin.com>,
	hamohammed.sa@gmail.com,  simona@ffwll.ch, melissa.srw@gmail.com,
	maarten.lankhorst@linux.intel.com,  mripard@kernel.org,
	tzimmermann@suse.de, airlied@gmail.com
Cc: dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] drm/vkms: fix gamma LUT size check
Date: Thu, 12 Feb 2026 23:04:28 +0800	[thread overview]
Message-ID: <tencent_2AAD56FDA429585DB3A835F1C98CE2566707@qq.com> (raw)
In-Reply-To: <5ee4e72c-0862-40b3-b47b-70ca5d8f89c1@bootlin.com>

Hi Louis,

Actually for a 64-bit machine, the value of sizeof(struct
drm_color_lut) and sizeof(struct drm_color_lut*) is equal. But for
32-bit machine, it is not equal. The definition of struct 
drm_color_lut shown as follows:

struct drm_color_lut {
	__u16 red;
	__u16 green;
	__u16 blue;
	__u16 reserved;
}

There is one test case in igt-gpu-tools for gamma, and the 32-bit
machine test results shown as below:

before patch:

# IGT_FORCE_DRIVER=vkms ./kms_color  --run-subtest gamma
[241590.953610] Console: switching to colour dummy device 80x25
[241590.955637] [IGT] kms_color: executing
IGT-Version: 2.3-NO-GIT (riscv32) (Linux: 6.18.7 riscv32)
Using IGT_SRANDOM=1770864024 for randomisation
Opened device: /dev/dri/card0
[241590.966858] [IGT] kms_color: starting subtest gamma
Starting subtest: gamma
[241590.968030] [IGT] kms_color: starting dynamic subtest pipe-A-
Virtual-1
Starting dynamic subtest: pipe-A-Virtual-1
(kms_color:235) igt_kms-CRITICAL: Test assertion failure function
igt_pipe_commit, file ../lib/igt_kms.c:4212:
(kms_color:235) igt_kms-CRITICAL: Failed assertion: ret == 0
(kms_color:235) igt_kms-CRITICAL: Last errno: 22, Invalid argument
(kms_color:235) igt_kms-CRITICAL: error: -22 != 0
Stack trace: not implemented
Dynamic subtest pipe-A-Virtual-1 failed.

after patch:

# uname -a
Linux buildroot 6.18.7 #2 SMP Mon Feb  9 15:22:27 CST 2026 riscv32
GNU/Linux
# pwd
/usr/libexec/igt-gpu-tools
# lsmod|grep vkms
vkms                   53248  0
drm_shmem_helper       20480  2 vkms
drm_client_lib         12288  1 vkms
drm_kms_helper        122880  3 vkms,drm_shmem_helper,drm_client_lib
drm                   458752  5
vkms,drm_shmem_helper,drm_client_lib,drm_kms_helper
# IGT_FORCE_DRIVER=vkms ./kms_color  --run-subtest gamma
[  111.981381] Console: switching to colour dummy device 80x25
[  111.983078] [IGT] kms_color: executing
IGT-Version: 2.3-NO-GIT (riscv32) (Linux: 6.18.7 riscv32)
Using IGT_SRANDOM=1770864535 for randomisation
Opened device: /dev/dri/card0
[  111.992250] [IGT] kms_color: starting subtest gamma
Starting subtest: gamma
[  111.993166] [IGT] kms_color: starting dynamic subtest pipe-A-
Virtual-1
Starting dynamic subtest: pipe-A-Virtual-1
Dynamic subtest pipe-A-Virtual-1: SUCCESS (7.498s)
[  119.492455] [IGT] kms_color: finished subtest pipe-A-Virtual-1,
SUCCESS
Subtest gamma: SUCCESS (7.501s)
[  119.495286] [IGT] kms_color: finished subtest gamma, SUCCESS
[  119.497233] [IGT] kms_color: exiting, ret=0
[  119.517787] Console: switching to colour frame buffer device 128x48

Best Regards,
Renjun Wang


On Fri, 2026-02-06 at 14:53 +0100, Louis Chauvet wrote:
> 
> 
> On 2/4/26 15:15, Renjun Wang wrote:
> > vkms_atomic_check() computed the gamma LUT entry count using
> > sizeof(struct drm_color_lut *), which uses pointer size and
> > can incorrectly reject or accept LUT sizes. Use
> > drm_color_lut_size() instead to validate against VKMS_LUT_SIZE.
> > 
> > Signed-off-by: Renjun Wang <renjunw0@foxmail.com>
> 
> Reviewed-by: Louis Chauvet <louis.chauvet@bootlin.com>
> 
> Like for the YUV patch, were you able to reproduce it using a tool?
> 
> > ---
> >   drivers/gpu/drm/vkms/vkms_drv.c | 4 ++--
> >   1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/vkms/vkms_drv.c
> > b/drivers/gpu/drm/vkms/vkms_drv.c
> > index dd1402f43773..a09589949f48 100644
> > --- a/drivers/gpu/drm/vkms/vkms_drv.c
> > +++ b/drivers/gpu/drm/vkms/vkms_drv.c
> > @@ -17,6 +17,7 @@
> >   #include <drm/drm_gem.h>
> >   #include <drm/drm_atomic.h>
> >   #include <drm/drm_atomic_helper.h>
> > +#include <drm/drm_color_mgmt.h>
> >   #include <drm/drm_drv.h>
> >   #include <drm/drm_fbdev_shmem.h>
> >   #include <drm/drm_file.h>
> > @@ -111,8 +112,7 @@ static int vkms_atomic_check(struct drm_device
> > *dev, struct drm_atomic_state *st
> >   		if (!new_crtc_state->gamma_lut || !new_crtc_state-
> > >color_mgmt_changed)
> >   			continue;
> >   
> > -		if (new_crtc_state->gamma_lut->length /
> > sizeof(struct drm_color_lut *)
> > -		    > VKMS_LUT_SIZE)
> > +		if (drm_color_lut_size(new_crtc_state->gamma_lut)
> > > VKMS_LUT_SIZE)
> >   			return -EINVAL;
> >   	}
> >   


  reply	other threads:[~2026-02-12 15:04 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-02-04 14:15 Renjun Wang
2026-02-06 13:53 ` Louis Chauvet
2026-02-12 15:04   ` Renjun Wang [this message]
2026-08-20  8:31 [PATCH 1/2] drm/vkms: Cache composer line buffers in vkms_output oushixiong1025
2026-08-20  8:31 ` [PATCH] drm/vkms: Fix gamma_lut size check oushixiong1025
2026-09-09  2:10 oushixiong1025

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=tencent_2AAD56FDA429585DB3A835F1C98CE2566707@qq.com \
    --to=renjunw0@foxmail.com \
    --cc=airlied@gmail.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=hamohammed.sa@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=louis.chauvet@bootlin.com \
    --cc=maarten.lankhorst@linux.intel.com \
    --cc=melissa.srw@gmail.com \
    --cc=mripard@kernel.org \
    --cc=simona@ffwll.ch \
    --cc=tzimmermann@suse.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®