| Bug ID | 100673 |
|---|---|
| Summary | Tonga agd5f drm-next-4.12-wip xorg segfault on startx |
| Product | DRI |
| Version | DRI git |
| Hardware | Other |
| OS | All |
| Status | NEW |
| Severity | normal |
| Priority | medium |
| Component | DRM/AMDgpu |
| Assignee | dri-devel@lists.freedesktop.org |
| Reporter | adf.lists@gmail.com |
Created attachment 130829=
[details]
xorg log with segfault
With current agd5f drm-next-4.12-wip I get a segfault on startx (xserver
1.19.3, git ddx,mesa,llvm. Log attached.
On the previous incarnation of the same kernel I can startx, but there are a
lot of warnings in dmesg every time the screen is turned on - ie. startx or
coming out of dpms. dmesg attached.
Don't know when I'll get time to bisect, so just getting these reported now=
.
Created attachment 130830 [details]
dmesg with lots of warnings on startx (older kernel)
(In reply to Andy Furniss from comment #1) > Created attachment 130830 [details] > dmesg with lots of warnings on startx (older kernel) This was fixed here: https://cgit.fre= edesktop.org/~agd5f/linux/commit/?h=3Ddrm-next-4.12-wip&id=3D53552d5df6= 694f68f276b2e86f8162d0ef7cad93
Reverting the following commit fixes this branch for me: eb8eb02 drm: Drop modeset_lock_all from the getproperty ioctl Let me check if I can make a quick fix for it.
Should be fixed by: commit 8cb68c83ab99a474ae847602f0c514d0fe17cc82 Author: Daniel Vetter <dan= iel.vetter@ffwll.ch> Date: Mon Apr 10 13:54:45 2017 +0200 drm: Fix get_property logic fumble Yet again I've proven that I can't negate conditions :( Testcase: igt/kms_properties/get_property-sanity Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Sean Paul <seanpaul@chromium.org> Fixes: eb8eb02ed850 ("drm: Drop modeset_lock_all from the getprope= rty ioctl")
(In reply to Alex Deucher from comment #4) > Should be fixed by: >=20 > commit 8cb68c83ab99a474ae847602f0c514d0fe17cc82 > Author: Daniel Vetter <daniel.vetter@ffwll.ch> > Date: Mon Apr 10 13:54:45 2017 +0200 >=20 > drm: Fix get_property logic fumble >=20 > Yet again I've proven that I can't negate conditions :( >=20 > Testcase: igt/kms_properties/get_property-sanity > Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> > Reviewed-by: Sean Paul <seanpaul@chromium.org> > Fixes: eb8eb02ed850 ("drm: Drop modeset_lock_all from the get= property > ioctl") Ok, but it doesn't seem to be anywhere in your tree.
(In reply to Andy Furniss from comment #5) > (In reply to Alex Deucher from comment #4) > > Should be fixed by: > >=20 > > commit 8cb68c83ab99a474ae847602f0c514d0fe17cc82 > > Author: Daniel Vetter <daniel.vetter@ffwll.ch> > > Date: Mon Apr 10 13:54:45 2017 +0200 > >=20 > > drm: Fix get_property logic fumble > >=20 > > Yet again I've proven that I can't negate conditions :( > >=20 > > Testcase: igt/kms_properties/get_property-sanity > > Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> > > Reviewed-by: Sean Paul <seanpaul@chromium.org> > > Fixes: eb8eb02ed850 ("drm: Drop modeset_lock_all from th= e getproperty > > ioctl") >=20 > Ok, but it doesn't seem to be anywhere in your tree. It hasn't been pulled into drm-next yet.
| What | Removed | Added |
|---|---|---|
| Resolution | --- | FIXED |
| Status | NEW | RESOLVED |