| Summary: | [UXA] rotation before gnome startup causes desktop corruption | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | xorg | Reporter: | zhao jian <jian.j.zhao> | ||||||||||
| Component: | Driver/intel | Assignee: | Jesse Barnes <jbarnes> | ||||||||||
| Status: | VERIFIED FIXED | QA Contact: | Xorg Project Team <xorg-team> | ||||||||||
| Severity: | major | ||||||||||||
| Priority: | high | ||||||||||||
| Version: | unspecified | ||||||||||||
| Hardware: | Other | ||||||||||||
| OS: | Linux (All) | ||||||||||||
| Whiteboard: | |||||||||||||
| i915 platform: | i915 features: | ||||||||||||
| Attachments: |
|
||||||||||||
Created attachment 23251 [details]
xorg.0.log
Created attachment 23252 [details]
desktop in uxa with tiling on
Jian, is rotation a required step to reproduce this bug? If so, this may be related to bug#20265. (In reply to comment #3) > Jian, is rotation a required step to reproduce this bug? If so, this may be > related to bug#20265. Yes, it's a required step. But Jesse's patch to #20265 have no effect to this bug. Does this happen if rotating after gnome-session (i.e. move step 4 before step 2)? If so, I'll consider this as major bug. (In reply to comment #5) > Does this happen if rotating after gnome-session (i.e. move step 4 before step > 2)? If so, I'll consider this as major bug. If I start gnome-session directly, and rotating in gnome, it will have a disordered display. But when rotate it back to normal, it displays well. How about using KMS? With KMS, it will crash X with the newest code. As #20943. This should be fixed now? With the newest code, both in UMS and KMS, it works well if I rotate after I started gnome-session, but it will still has the blurred screen if I rotate it and then start gnome-session. Libdrm: (master)1faab66cfd1a854925da6ff7109aa614292dea90 Mesa: (mesa_7_4_branch)de197cf991416f0cd65ad2e2d2ca9aa599b52075 Xserver: (server-1.6-branch)60c161545af80eb78eb790a05bde79409dfdf16e Xf86_video_intel: (2.7)3e5586cace98f73a9f8403a6446d380899ecbce9 Kernel: (drm-intel-2.6.29)71d7aec4bbf923eaf43563c03459726a746cd332 Yeah I see this too; but only after rotating then rotating back to normal and starting the session, both with and without tiling. Even happens with EXA... Happens on 965 too. Ah no I see it with EXA but only if my pixmap management patch is applied, which points to the prepare/finish access hooks... Seems like some sort of reuse or stale pointer issue. I can reliably reproduce this but only if no other windows pop up before I start the GNOME session. I had a failure dialog pop up last time, and when I started gnome-session after that it worked fine. This commit may affect weird tiling issues: commit 87332a7cc16af82aa47e07fbf90da3635b071dbf Author: Keith Packard <keithp@keithp.com> Date: Mon May 11 13:52:00 2009 -0700 Use drm_intel_bo_disable_reuse API to flag scanout and cursor buffers (only kms though, probably) I had the same problem with 2.7.1 when i just updated to karmic (ubuntu pre-alpha). I found this bug, updated to git and now rotation seems to work fine again. Im using a gma945 and exa (defaults to it). Jian, retest please. (In reply to comment #18) > Jian, retest please. It still has such problem with the newest code. increasing priority, assuming users setting rotation in xorg.conf will be impacted. I see this even with tiling disabled, can you confirm? (In reply to comment #21) > I see this even with tiling disabled, can you confirm? Yes, with either server-1.6-branch and mesa_7_5_branch or all in master branch, with tiling disabled it still existed. Created attachment 27343 [details] [review] clear bo of scratch pixmap to handle reuse Can you confirm that this fixes things for you in both KMS and UMS mode? It works for me, but I'm afraid it's papering over a bigger problem with scratch pixmaps. Keith? Fix pushed, please verify. commit 882d992e7375d090f6ef031129e027412a51edf1 Author: Jesse Barnes <jbarnes@virtuousgeek.org> Date: Thu Jul 2 13:04:51 2009 -0700 Clear the bo on the rotate scratch pixmap err had to rebase before pushing: commit c1755599db1d9a20954b84ccc07afd892bb6ac9e Author: Jesse Barnes <jbarnes@virtuousgeek.org> Date: Thu Jul 2 13:04:51 2009 -0700 Clear the bo on the rotate scratch pixmap (In reply to comment #25) > err had to rebase before pushing: > commit c1755599db1d9a20954b84ccc07afd892bb6ac9e > Author: Jesse Barnes <jbarnes@virtuousgeek.org> > Date: Thu Jul 2 13:04:51 2009 -0700 > Clear the bo on the rotate scratch pixmap Yes. It now works well. verified. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.
Created attachment 23250 [details] xorg.conf System Environment: ---------------------- Platform: 945gm Arch: i386 OSD: Fedora release 9 (Sulphur) Libdrm: (master)4a0d19ef4f210cea9e60c5acc355df03723ef808 Mesa: (mesa_7_4_branch)e2092bb23c956ba9ab940935f803ef843db81af2 Xserver: (server-1.6-branch)4557b3f6c4273cd83b701beaf7a150c806fed298 Xf86_video_intel: (master)81c652e9a666a7459bcc5217c8a5ec518b6e00da Kernel: (for-airlied)66647dc60d16fae9f6963fd98b6d9baa1a8dac69 Bug Description: --------------------- In UXA mode, set the tiling on, if you use xrandr to rotate the desktop, it will hsa a messed output and when rotate back to normal its output is OK. Then log in to gnome, it will be almost blackscreen only can see the xterm with the title blurred. You drag it it appears no change.(see the attached photo) In UXA mode disabling the tiling, the desktop will be OK. And in EXA mode the desktop will be OK both with tiling on and off. Reproduce Steps: --------------------- 1. set tiling on 1. xinit& 2. xrandr --output LVDS --rotate left 3. xrandr --output LVDS --rotate normal 4. gnome-session&