On Sat, Nov 5, 2011 at 1:10 AM, Vadim Girlin wrote:
> On Fri, 2011-11-04 at 22:53 +0100, Marek Olšák wrote:
>> FYI, this commit completely breaks RV670. glxgears is completely
>> black, other apps are either black or rendered incorrectly (mostly
>> one-colored, the clear color I guess).
>>
>
> Tha
On Fri, 2011-11-04 at 22:53 +0100, Marek Olšák wrote:
> FYI, this commit completely breaks RV670. glxgears is completely
> black, other apps are either black or rendered incorrectly (mostly
> one-colored, the clear color I guess).
>
Thanks. Probably it's due to missing spi_vs_out_id array initial
On Fri, Nov 4, 2011 at 5:53 PM, Marek Olšák wrote:
> FYI, this commit completely breaks RV670. glxgears is completely
> black, other apps are either black or rendered incorrectly (mostly
> one-colored, the clear color I guess).
I knew I should have tested 6xx before I pushed. Reverted for now.
FYI, this commit completely breaks RV670. glxgears is completely
black, other apps are either black or rendered incorrectly (mostly
one-colored, the clear color I guess).
Marek
On Fri, Nov 4, 2011 at 6:24 PM, Vadim Girlin wrote:
> There is no need to duplicate semantic mapping which is done in h
On Fri, Nov 4, 2011 at 1:24 PM, Vadim Girlin wrote:
> There is no need to duplicate semantic mapping which is done in hw, so get
> rid of r600_find_vs_semantic_index.
>
> TGSI name/sid pair is mapped to the 8-bit semantic index for SPI.
Pushed. Thanks! Note that the hw can use semantic ids for
There is no need to duplicate semantic mapping which is done in hw, so get
rid of r600_find_vs_semantic_index.
TGSI name/sid pair is mapped to the 8-bit semantic index for SPI.
Signed-off-by: Vadim Girlin
---
This patch and the next one were tested on the evergreen,
there are some changes in th
