[Mesa-dev] [PATCH] i965: Avoid blocking on the GPU for setting the HiZ op vertex data.

Chad Versace chad.versace at linux.intel.com
Mon Feb 27 16:23:21 PST 2012


On 02/27/2012 01:04 PM, Eric Anholt wrote:
> We need to allocate a new buffer every batch to avoid blocking on the
> last HiZ op completing.  There are two easy ways to do this:
> brw_state_batch() and intel_upload_data().  brw_state_batch() is
> simpler and avoids another buffer allocation.
> 
> Improves Unigine Tropics performance 0.376416% +/- 0.148722% (n=7).
> ---
>  src/mesa/drivers/dri/i965/brw_context.h |   15 ----------
>  src/mesa/drivers/dri/i965/brw_vtbl.c    |    1 -
>  src/mesa/drivers/dri/i965/gen6_hiz.c    |   44 ++++++------------------------
>  src/mesa/drivers/dri/i965/gen7_hiz.c    |    9 ------
>  4 files changed, 9 insertions(+), 60 deletions(-)

Looks good to me. This makes the code a little simpler too.
Reviewed-by: Chad Versace <chad.versace at linux.intel.com>


More information about the mesa-dev mailing list