Re: (subset) [PATCH 0/3] drm/bridge: imx: A few fixes

On Wed, Jan 28, 2026 at 04:26:22PM +0100, Luca Ceresoli wrote:

Hello Liu,

Hello Luca,

On Wed Jan 28, 2026 at 2:50 AM CET, Liu Ying wrote:

quoted

Hi Luca,

On Tue, Jan 27, 2026 at 12:35:47PM +0100, Luca Ceresoli wrote:

quoted

On Fri, 23 Jan 2026 17:22:14 +0800, Liu Ying wrote:

quoted

This patch series contains 3 bug fixes for i.MX8QXP DRM bridges.
Each patch fixes a NULL pointer dereference issue and stands alone.

Applied, thanks!

[1/3] drm/bridge: imx8qxp-ldb: Fix NULL pointer dereference in imx8qxp_ldb_bridge_destroy()
      commit: 852c68bf42965ee38b465d2d6f7b965eb0b5dc1d
[2/3] drm/bridge: imx8qxp-pxl2dpi: Fix NULL pointer dereference in imx8qxp_pxl2dpi_bridge_destroy()
      commit: db7e7ea838c916ee4cdf26bee126fd36f58295dc

Any reason why patch 3/3 is not applied?

Yes, the reason is there's a patch conflict and I still have to find out
how to handle it.

Patch 3 by the rules [0] should go to drm-misc-fixes because the bug (added
by [1]) is already on Linus' master tree (since v6.17), but it does not
apply there due to a conflict with an unrelated change [2] currently on
drm-misc-next.

Note that [1] and [2] are the two commits mentioned in your 'Fixes: 'tags.

I had a chat with Maxime on #dir-devel about how to handle this situation
but still haven't got a final answer.

I guess the correct way is:

 1. apply patch 3 on drm-misc-next, where it applies cleanly
 2. send a modified patch for drm-misc-fixes, to fix ASAP the master branch

The conflict between the modified patch and commit [2] still needs to
be addressed in the future, right?  Not sure if there is any proper
process to address that.

I think it's acceptable to wait for a while and apply 3/3 after commit [2]
shows up in the master branch, because i.MX8QXP pixel combiner is not
added to device tree yet, that is, no one uses it for now.

Would you send the patch as in item 2, so it can be applied to
drm-misc-fixes and have master fixed ASAP? Note iit should not mention [2]
in the Fixes: tag because [2] is not yet on master. I can send it of course
but you can test on hardware so that would be the best option (and I'm a
bit overloaded at the moment).

[0] https://drm.pages.freedesktop.org/maintainer-tools/committer/committer-drm-misc.html
[1] commit 99764593528f ("drm/bridge: imx8qxp-pixel-combiner: convert to devm_drm_bridge_alloc() API")
[2] commit ae754f049ce1 ("drm/bridge: imx8qxp-pixel-combiner: get/put the next bridge")

Luca

--
Luca Ceresoli, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com/

-- 
Regards,
Liu Ying