Re: [Intel-gfx] [PATCH 4/4] i915: fix remap_io_sg to verify the pgprot
Christoph Hellwig writes: > As an ad-hoc experiment: can you replace the call to remap_pfn_range > with remap_pfn_range_notrack (and export it if you build i915 modular) > in remap_io_sg and see if that makes any difference? That worked, thanks -- no artifacts seen. ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [Intel-gfx] [PATCH 4/4] i915: fix remap_io_sg to verify the pgprot
I have another problem with this patch since it landed in mainline. On my m3-6Y30 skylake HD Graphics 515 (rev 07), it causes visual artifacts that look like bunch of one pixel high horizontal streaks, seen most often in firefox while scrolling or in menu controls. Reverting this patch on top of current mainline fixes the problem. ___ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 2/2] drm/i915: Skip object locking around a no-op set-domain ioctl
Hi! > This patch causes lockups in firefox. They appear like non-fatal hangs > of the webpage contents, "fixable" with alt-tab or a background system > load. I have verified that reverting the commit 754a254427 on top of > current Linus tree fixes the problem. This is still broken in v5.2-rc3. I have also verified that the particular commit, if backported to v5.1 release, breaks firefox there too in the same way. (for reference:) commit 754a25442705c4f90e0d05f1a7bd303ffe700ca9 Author: Chris Wilson Date: Thu Mar 21 16:19:08 2019 + drm/i915: Skip object locking around a no-op set-domain ioctl
Re: [PATCH 2/2] drm/i915: Skip object locking around a no-op set-domain ioctl
Hi! This patch causes lockups in firefox. They appear like non-fatal hangs of the webpage contents, "fixable" with alt-tab or a background system load. I have verified that reverting the commit 754a254427 on top of current Linus tree fixes the problem.