Re: [Intel-gfx] [PATCH 3/3] drm/i915: mark GEM objects as dirty when written by the CPU

On Tue, Dec 08, 2015 at 06:24:40PM +, Dave Gordon wrote: > On 08/12/15 17:03, Chris Wilson wrote: > >On Tue, Dec 08, 2015 at 04:51:18PM +, Dave Gordon wrote: > >>This patch covers a couple more places where a GEM object is (or may be) > >>modified by means of CPU writes, and should therefor

Re: [Intel-gfx] [PATCH 3/3] drm/i915: mark GEM objects as dirty when written by the CPU

On 08/12/15 17:03, Chris Wilson wrote: On Tue, Dec 08, 2015 at 04:51:18PM +, Dave Gordon wrote: This patch covers a couple more places where a GEM object is (or may be) modified by means of CPU writes, and should therefore be marked dirty to ensure that the changes are not lost in the evenof

Re: [Intel-gfx] [PATCH 3/3] drm/i915: mark GEM objects as dirty when written by the CPU

On Tue, Dec 08, 2015 at 04:51:18PM +, Dave Gordon wrote: > This patch covers a couple more places where a GEM object is (or may be) > modified by means of CPU writes, and should therefore be marked dirty to > ensure that the changes are not lost in the evenof of the object is > evicted under me

[Intel-gfx] [PATCH 3/3] drm/i915: mark GEM objects as dirty when written by the CPU

This patch covers a couple more places where a GEM object is (or may be) modified by means of CPU writes, and should therefore be marked dirty to ensure that the changes are not lost in the evenof of the object is evicted under memory pressure. It may be possible to optimise these paths later, by