Re: [Intel-gfx] [PATCH 0/5] Fixup pwrite/pread with gtt mmapped user addresses

2011-10-20 Thread Keith Packard
On Sat, 17 Sep 2011 20:55:44 +0200, Daniel Vetter daniel.vet...@ffwll.ch wrote: io-mapping: ensure io_mapping_map_atomic _is_ atomic drm/i915: drop KM_USER0 argument to k(un)map_atomic These have been merged. drm/i915: fall through pwrite_gtt_slow to the shmem slow path drm/i915:

[Intel-gfx] [PATCH 0/5] Fixup pwrite/pread with gtt mmapped user addresses

2011-09-17 Thread Daniel Vetter
Hi all, Original bugreport: https://bugs.freedesktop.org/show_bug.cgi?id=38115 This bug can e.g. be hit with glBufferData(dst, glBufferMap(src)) Testcase is gem_mmap_gtt. Neglecting the trivial cleanup patch 2, patch 1 converts the possible deadlock in an errornous -EFAULT. I've already

Re: [Intel-gfx] [PATCH 0/5] Fixup pwrite/pread with gtt mmapped user addresses

2011-09-17 Thread Ben Widawsky
On Sat, 17 Sep 2011 20:55:44 +0200 Daniel Vetter daniel.vet...@ffwll.ch wrote: Hi all, Original bugreport: https://bugs.freedesktop.org/show_bug.cgi?id=38115 This bug can e.g. be hit with glBufferData(dst, glBufferMap(src)) Testcase is gem_mmap_gtt. Neglecting the trivial cleanup