| e253175c9c8034bfc13ecc3e16ab72007bde744a |
|
17-Jul-2012 |
James Benton <jbenton@vmware.com> |
llvmpipe: Fix bug with blend factor in complementary optimisations.
Fixes fdo 52168.
Reviewed-by: José Fonseca <jfonseca@vmware.com>
/external/mesa3d/src/gallium/drivers/llvmpipe/lp_bld_blend.c
|
| 8684ffc141ed024ab1d42fcba31d8517288fac40 |
|
03-Jul-2012 |
James Benton <jbenton@vmware.com> |
llvmpipe: Unified common code between AoS and SoA blending.
Added a new file lp_bld_blend.c for the common code.
Merged and added some simple optimisations.
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
/external/mesa3d/src/gallium/drivers/llvmpipe/lp_bld_blend.c
|
| 2529ed5616b1b152766a3355444260b88184cd6e |
|
09-Aug-2009 |
José Fonseca <jfonseca@vmware.com> |
llvmpipe: SoA blending.
Throughput seems to be 4x higher.
/external/mesa3d/src/gallium/drivers/llvmpipe/lp_bld_blend.c
|
| 9493260fdcef4a8238b9d9a9dc3e753dd89810fe |
|
08-Aug-2009 |
José Fonseca <jfonseca@vmware.com> |
llvmpipe: Centralize lp_build_context initialization.
/external/mesa3d/src/gallium/drivers/llvmpipe/lp_bld_blend.c
|
| 1dd7bb17c7331f9ecd0bc830b61ada235a56fe6d |
|
04-Aug-2009 |
José Fonseca <jfonseca@vmware.com> |
llvmpipe: Optimize blend swizzles by using bitmasks instead of shuffles for ubytes.
/external/mesa3d/src/gallium/drivers/llvmpipe/lp_bld_blend.c
|
| 64611e086dbefa2003773ab541c0381b5713e18d |
|
03-Aug-2009 |
José Fonseca <jfonseca@vmware.com> |
llvmpipe: Separate constant building.
/external/mesa3d/src/gallium/drivers/llvmpipe/lp_bld_blend.c
|
| c87fab0008453567b45dd5e5eb7dd5d026990071 |
|
03-Aug-2009 |
José Fonseca <jfonseca@vmware.com> |
llvmpipe: Move type support functions into a separate file.
/external/mesa3d/src/gallium/drivers/llvmpipe/lp_bld_blend.c
|
| 5778970f15af253ce9d279516287602f28153548 |
|
03-Aug-2009 |
José Fonseca <jfonseca@vmware.com> |
llvmpipe: Minor comments / debug mnemonics.
/external/mesa3d/src/gallium/drivers/llvmpipe/lp_bld_blend.c
|
| 9e922444403d704f9ed0cf50e510fba53c736b57 |
|
03-Aug-2009 |
José Fonseca <jfonseca@vmware.com> |
llvmpipe: Improve generated 8bit blending code.
/external/mesa3d/src/gallium/drivers/llvmpipe/lp_bld_blend.c
|
| a6622e6c544d3530a463d6a274a15bfae58f7ccc |
|
02-Aug-2009 |
José Fonseca <jfonseca@vmware.com> |
llvmpipe: Specialize arithmetic operations.
/external/mesa3d/src/gallium/drivers/llvmpipe/lp_bld_blend.c
|
| 272dadbe4ebeaeb4f942c0f3c2fd140285b0457c |
|
02-Aug-2009 |
José Fonseca <jfonseca@vmware.com> |
llvmpipe: Introduce a custom typing system.
Straightforward representation of floating-point/fixed-point/integer,
normalized/scaled, signed/unsigned SIMD vector types.
/external/mesa3d/src/gallium/drivers/llvmpipe/lp_bld_blend.c
|
| 7d043162c5d9150947d9341cfa22192bd4c70fde |
|
01-Aug-2009 |
José Fonseca <jfonseca@vmware.com> |
llvmpipe: Blending.
The code
/external/mesa3d/src/gallium/drivers/llvmpipe/lp_bld_blend.c
|