On Mon, Jan 04, 2016 at 09:12:11PM +0100, Pavel Machek wrote: > Hi! > > > > I then ran a git bissect between v4.0 and v4.1 from Linus's tree and > > > found the "guilty" commit was > > > > > > commit 317b4e903636305cfe702ab3e5b3d68547a69e72 > > > Author: Ben Widawsky <benjamin.widawsky@xxxxxxxxx> > > > Date: Mon Mar 16 16:00:55 2015 +0000 > > > > > > drm/i915: Extract context switch skip and add pd load logic > > > > Damnit, paste fail. > > > > I meant to paste : > > > > commit 678d96fbb3b5995a2fdff2bca5e1ab4a40b7e968 > > Author: Ben Widawsky <benjamin.widawsky@xxxxxxxxx> > > Date: Mon Mar 16 16:00:56 2015 +0000 > > > > drm/i915: Track GEN6 page table usage > > > > (as indicated in the title and in the git bisect log) > > Can you verify that reverting this patch (on top of 4.4?) fixes it? > > If so, is it time to revert it? > > Thanks, > Pavel It's highly unlikely you'll be able to revert this on top of 4.4. Unfortunately, this is the cornerstone of a whole lot of infrastructure needed to support the per process page tables. I'm just the measly original author of the patch. Lots of others tried to right my wrongs, and they're better suited to debug this. If there hasn't been much traction, it's probably because of the holidays. I assume suspend to disk is also failing? -- Ben Widawsky, Intel Open Source Technology Center _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
