Summary: | i915 classic -- screen fails to update in apps | ||
---|---|---|---|
Product: | Mesa | Reporter: | col |
Component: | Drivers/DRI/i915 | Assignee: | Ian Romanick <idr> |
Status: | RESOLVED FIXED | QA Contact: | |
Severity: | normal | ||
Priority: | medium | CC: | idr, kenneth |
Version: | git | Keywords: | NEEDINFO |
Hardware: | All | ||
OS: | All | ||
Whiteboard: | |||
i915 platform: | i915 features: |
Description
col
2011-04-25 19:28:00 UTC
Since you know approximately when it worked, can you bisect? Ken has also been running gnome3, so he might be able to bisect it too. (In reply to comment #1) > Since you know approximately when it worked, can you bisect? > > Ken has also been running gnome3, so he might be able to bisect it too. I'm working on it. I don't "know" anyone here, so if Ken is watching, maybe he can reproduce the bug on his hardware too? Anyway, knowing nothing much about git bisecting I'm trying with little success. It would seem there are too many changes for my system to go back that far. I'd have to start with a re-installation... really not what I want to do. Anyway, I'll see if this one 3370f9b606217f53297d61b025a06f7bcf62648a builds and works, or if it fixes the bug. Rinse and repeat? I *think* the problem started for me when 7.10.1 hit me, up from 7.9.1, but I rebuilt other libraries, not suspecting a bug in mesa. Now, I can't seem to use my desktop when mesa is so old (without possibly rebuilding other libraries). Maybe somewhere in between I'll get lucky. IDK... seems really like a long-shot as I have to test this in a working desktop. (In reply to comment #2) > (In reply to comment #1) > > Since you know approximately when it worked, can you bisect? > > > > Ken has also been running gnome3, so he might be able to bisect it too. > > I'm working on it. I don't "know" anyone here, so if Ken is watching, maybe he > can reproduce the bug on his hardware too? > Okay... I tried every version released/packaged by gentoo in succession back to 7.9.1. No change. Finally reverted git versions of gnome-shell, mutter, clutter, etc., to packaged releases. Problem went away. Sorry for the noise. I'll see if I can reproduce the problem via upgrades, and probably close this. (In reply to comment #3) > (In reply to comment #2) > > (In reply to comment #1) > > > Since you know approximately when it worked, can you bisect? > > > > > > Ken has also been running gnome3, so he might be able to bisect it too. > > > > I'm working on it. I don't "know" anyone here, so if Ken is watching, maybe he > > can reproduce the bug on his hardware too? > > > > Okay... I tried every version released/packaged by gentoo in succession back to > 7.9.1. No change. Finally reverted git versions of gnome-shell, mutter, > clutter, etc., to packaged releases. Problem went away. > > Sorry for the noise. I'll see if I can reproduce the problem via upgrades, and > probably close this. OK, some progress. I thought the problem was disappeared, but it returned between 7.9.1 and updating to 7.9.2. I suppose I might try to find out exactly where, then. (In reply to comment #4) > (In reply to comment #3) > > (In reply to comment #2) > > > (In reply to comment #1) > > > > Since you know approximately when it worked, can you bisect? [8<] > > > > Sorry for the noise. I'll see if I can reproduce the problem via upgrades, and > > probably close this. > > OK, some progress. I thought the problem was disappeared, but it returned > between 7.9.1 and updating to 7.9.2. I suppose I might try to find out exactly > where, then. Right, I can't be any help here. Seeing redraw problems with 7.10.2, but not the same severity... My guess is there is a bug, just some combinations (of x server, WM) show it more than others. But, I can't reproduce the same bug anymore. Now I get, so far, only some artifacts in scrolling text -- this I can live with for now. Close this? I'd say if your initial problem is gone, close it. Note that these sorts of failure-to-update-with-compositing-wm bugs are generally things that are bugs in the 2D driver (xf86-video-intel) and the server, as opposed to Mesa, so you might look there for where the regression was for your remaining issue. Closing as that seems to be the general consensus. |
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.