Re: [Intel-gfx] [PATCH 2/2] drm/i915/bxt: work around HW context corruption due to coherency problem

2015-09-17 Thread Imre Deak
On ti, 2015-09-15 at 21:30 +0300, Imre Deak wrote: > The execlist context object is mapped with a CPU/GPU coherent mapping > everywhere, but on BXT A stepping due to a HW issue the coherency is not > guaranteed. To work around this flush the CPU cache after any change > from the CPU to the context

Re: [Intel-gfx] [PATCH 2/2] drm/i915/bxt: work around HW context corruption due to coherency problem

2015-09-16 Thread Imre Deak
On ke, 2015-09-16 at 09:17 +0100, Chris Wilson wrote: > On Tue, Sep 15, 2015 at 09:30:20PM +0300, Imre Deak wrote: > > The execlist context object is mapped with a CPU/GPU coherent mapping > > everywhere, but on BXT A stepping due to a HW issue the coherency is not > > guaranteed. To work around th

Re: [Intel-gfx] [PATCH 2/2] drm/i915/bxt: work around HW context corruption due to coherency problem

2015-09-16 Thread Chris Wilson
On Tue, Sep 15, 2015 at 09:30:20PM +0300, Imre Deak wrote: > The execlist context object is mapped with a CPU/GPU coherent mapping > everywhere, but on BXT A stepping due to a HW issue the coherency is not > guaranteed. To work around this flush the CPU cache after any change > from the CPU to the

[Intel-gfx] [PATCH 2/2] drm/i915/bxt: work around HW context corruption due to coherency problem

2015-09-15 Thread Imre Deak
The execlist context object is mapped with a CPU/GPU coherent mapping everywhere, but on BXT A stepping due to a HW issue the coherency is not guaranteed. To work around this flush the CPU cache after any change from the CPU to the context object. Note that this also includes any changes done by th