AlexIndustrial/mesa
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

aco: use p_parallelcopy for uniform reduction with zero source

Browse Source 
I think v_mov_b32 was only used because a sub-dword p_parallelcopy
couldn't take constants on some gfx levels. That shouldn't be the case
anymore.

Signed-off-by: Rhys Perry <pendingchaos02@gmail.com>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/16595>
This commit is contained in:
Rhys Perry
2022-05-13 13:39:01 +01:00
committed by Marge Bot
parent 3e73fd40a6
commit 8384189b6c
1 changed files with 1 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/amd/compiler/aco_instruction_selection.cpp
+1 -3
View File
@@ -7725,10 +7725,8 @@ emit_addition_uniform_reduce(isel_context* ctx, nir_op op, Definition dst, nir_s
bld.pseudo(aco_opcode::p_extract_vector, dst, count, Operand::zero());
else if (nir_src_as_uint(src) == 1)
bld.copy(dst, count);
else if (nir_src_as_uint(src) == 0 && dst.bytes() <= 2)
bld.vop1(aco_opcode::v_mov_b32, dst, Operand::zero()); /* RA will use SDWA if possible */
else if (nir_src_as_uint(src) == 0)
bld.copy(dst, Operand::zero());
bld.copy(dst, Operand::zero(dst.bytes()));
else if (count.type() == RegType::vgpr)
bld.v_mul_imm(dst, count, nir_src_as_uint(src));
else