[Mesa-dev] [PATCH 0/5] r600g: Fix common state for cayman compute

[Tom Stellard]

On Thu, Oct 25, 2012 at 11:55:50AM -0400, alexdeucher at gmail.com wrote:
> From: Alex Deucher <alexander.deucher at amd.com>
> 
> This patch set cleans up and unifies the common state and CS init
> set up between evergreen and cayman and gfx and compute.  No piglit
> regressions on the gfx side here.  This also allows compute to run
> on cayman without causing a CS error in the kernel when VM is enabled
> due to emitting some evergreen-only registers.  However, the GPU
> hangs on the cayman compute command buffer.
> 
> Alex Deucher (5):
>   r600g/compute: always CONTEXT_CONTROL packet at start of CS
>   r600g: rework evergreen_init_common_regs()
>   r600g: there are 16 const buffer size regs for each shader stage
>   r600g: emit some additional regs on cayman
>   r600g: split cayman common state out into a shared function
> 
>  src/gallium/drivers/r600/evergreen_compute.c |   17 ++-
>  src/gallium/drivers/r600/evergreen_state.c   |  177 ++++++++++++++++++--------
>  src/gallium/drivers/r600/evergreend.h        |    1 +
>  src/gallium/drivers/r600/r600_pipe.h         |    4 +
>  4 files changed, 146 insertions(+), 53 deletions(-)
>

For the series:

Reviewed-by: Tom Stellard <thomas.stellard at amd.com>

I've tested these patches on evergreen and found no regressions.

> -- 
> 1.7.7.5
> 
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/mesa-dev