diff options
| author | Sagar Ghuge <[email protected]> | 2019-02-27 14:02:54 -0800 |
|---|---|---|
| committer | Sagar Ghuge <[email protected]> | 2019-03-04 15:50:25 -0800 |
| commit | 1d8994a63b546a5b2dc4feb5bd98a84ee853d6af (patch) | |
| tree | 6ff1fb8b5cca7261fb887e3bc9dfe773121e2352 /src/mesa/program | |
| parent | e551040c602d392019e68f54d9a3a310d2a937a3 (diff) | |
glsl: [u/i]mulExtended optimization for GLSL
Optimize mulExtended to use 32x32->64 multiplication.
Drivers which are not based on NIR, they can set the
MUL64_TO_MUL_AND_MUL_HIGH lowering flag in order to have same old
behavior.
v2: Add missing condition check (Jason Ekstrand)
Signed-off-by: Sagar Ghuge <[email protected]>
Suggested-by: Matt Turner <Matt Turner <[email protected]>
Suggested-by: Jason Ekstrand <[email protected]>
Reviewed-by: Jason Ekstrand <[email protected]>
Diffstat (limited to 'src/mesa/program')
| -rw-r--r-- | src/mesa/program/ir_to_mesa.cpp | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/src/mesa/program/ir_to_mesa.cpp b/src/mesa/program/ir_to_mesa.cpp index e65a6743353..ed194eb13a3 100644 --- a/src/mesa/program/ir_to_mesa.cpp +++ b/src/mesa/program/ir_to_mesa.cpp @@ -3053,6 +3053,7 @@ _mesa_ir_link_shader(struct gl_context *ctx, struct gl_shader_program *prog) do_mat_op_to_vec(ir); lower_instructions(ir, (MOD_TO_FLOOR | DIV_TO_MUL_RCP | EXP_TO_EXP2 | LOG_TO_LOG2 | INT_DIV_TO_MUL_RCP + | MUL64_TO_MUL_AND_MUL_HIGH | ((options->EmitNoPow) ? POW_TO_EXP2 : 0))); progress = do_common_optimization(ir, true, true, |