On Wed, 05 Sep 2018 at 10:37:39 +0100 Eric Engestrom wrote:
Hey Karsten, thanks for the patch! I don't know if it's correct or not, but I'll comment on a few other things.
The title of the commit should be a description of what your commit does; in this case you should probably keep the title that git gave you when you did `git revert`, as well as the first line:
Revert "drm/atomic: Handling the case when setting old crtc for plane"
This reverts commit fc2a69f3903dfd97cd47f593e642b47918c949df.
After that comes your explanation of why this commit should be reverted, which you've done here:
On Wednesday, 2018-09-05 00:26:27 +0200, Karsten Hohmeier wrote:
Original commit causes "plane A assertion failure" on lid close/lid open with older HP Compaq 6720s laptops (Intel Mobile GME965/GLE960).
The "Fixes" line that Chris gave you should got at the end of the commit message, after the explanation.
Full bug report at "https://bugs.freedesktop.org/show_bug.cgi?id=107827".
The format for bug report references is (on its own line): Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107827
drivers/gpu/drm/drm_atomic.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index 895741e9cd7d..a59ded403657 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -1433,9 +1433,7 @@ drm_atomic_set_crtc_for_plane(struct drm_plane_state *plane_state,
It looks like thunderbird mangled the patch when sending the email; it's recommended to use `git send-email` instead, as it does the right thing: https://git-scm.com/docs/git-send-email
{ struct drm_plane *plane = plane_state->plane; struct drm_crtc_state *crtc_state;
- /* Nothing to do for same crtc*/
- if (plane_state->crtc == crtc)
return 0;
- if (plane_state->crtc) { crtc_state = drm_atomic_get_crtc_state(plane_state->state, plane_state->crtc);
--
Hi Karsten, Thanks for the patch. However, your own comments at the below link are contradictory
https://bugs.freedesktop.org/show_bug.cgi?id=107827 Here, you mention that the commit 82bcb0e7a45d was still in place and screen corruption is gone and logs are clean.
Karsten Hohmeier 2018-09-05 17:11:14 UTC I cherry-picked commit 05c72e77ccda ("drm/i915: Nuke the LVDS lid notifier") on top of 4.18.5 as you suggested. Yes, it works. The screen corruption is gone and the logs are clean. The 82bcb0e7a45d ("drm/atomic: Handling the case when setting old crtc for plane") change was still in place.
From above comments it looks like this patch was not the reason for screen corruption.
Please clarify and let me know if I am missing something here. Thanks -Satendra