On Tue, 2013-07-23 at 20:13 +1000, iiNET wrote: > Hi Kernel Maintainers > > I have been tracking related bugs 704987 and 703715, I have suggested a > fix that has been under test for the last week and seems to fix the > reported issue, the following is the patch from the stable tree that > reverts Debian patch, > features/all/Input-add-Synaptics-USB-device-driver.patch.
I don't know how you think these are connected... > Testing has been editing, creating and reading various Libre Office > files without any symptoms reported, all with various touch pad > interactions, suspend, resume. > > > commit e24fb4d67f53530038a9711d0c1f65937490bb8c > Author: Ben Hutchings <b...@decadent.org.uk> > Date: Tue Jun 25 04:15:27 2013 +0100 > > Revert "drm/i915: GFX_MODE Flush TLB Invalidate Mode must be '1' > for scanli > > This reverts commit 393143615d9f2f581d87387268dc11b95adc339c, which > was commit f05bb0c7b624252a5e768287e340e8e45df96e42 upstream. > > This has been found to cause GPU hangs when backported to 3.2, though > not in mainline. [...] We are actually using the DRM code from 3.4, not 3.2. But this patch is applied and should not be. Thanks for finding this. If anyone else would like to test and confirm that this fixes the hangs they are seeing, please use the attached patch, following the instructions at <http://kernel-handbook.alioth.debian.org/ch-common-tasks.html#s-common-official>. Ben. -- Ben Hutchings Once a job is fouled up, anything done to improve it makes it worse.
reverted: --- b/drivers/gpu/drm/i915/intel_ringbuffer.c +++ a/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -407,11 +407,6 @@ if (INTEL_INFO(dev)->gen >= 6) I915_WRITE(MI_MODE, GFX_MODE_ENABLE(ASYNC_FLIP_PERF_DISABLE)); - /* Required for the hardware to program scanline values for waiting */ - if (INTEL_INFO(dev)->gen == 6) - I915_WRITE(GFX_MODE, - GFX_MODE_ENABLE(GFX_TLB_INVALIDATE_ALWAYS)); - if (IS_GEN7(dev)) I915_WRITE(GFX_MODE_GEN7, GFX_MODE_DISABLE(GFX_TLB_INVALIDATE_ALWAYS) |
signature.asc
Description: This is a digitally signed message part