On 12.09.2012 18:09, Jonathan Nieder wrote: > Hi Timo, > > Timo Aaltonen wrote: > >> commit d698c826bb93663e4b349a78fbd30443654d37cd upstream. oops, script-fail.. that's the local branch sha, real one is further down (0f91128d88bbb8b0a8e7bb93df2c40680871d45a). >> Please pull this for v3.2.x, should be the last missing (kernel) piece >> for https://bugs.freedesktop.org/show_bug.cgi?id=48838 > [...] >> It should be more reliable to wait for the pending >> operations to complete first, even though we still have a window for >> userspace to submit a broken command buffer during the modeset. > > That sounds hazy --- is it tested? Well, these are usually hard to verify fixed. The commit is mentioned on fdo bugs 45413 and 48838, should fix some GPU hangs. > Presumably this is needed for 3.4.y and 3.5.y as well, right? Do you > know when the bad behavior was introduced? Does 3.0.y need the fix as > well? It's from 3.5-rc1, don't know when it was introduced. Maybe it would make sense for 3.4 at least. -- t
