aco: improve readfirstlane after uniform ssbo loads on GFX7
pipeline-db changes for GFX7: 80310 shaders in 40472 tests Totals: SGPRS: 3655900 -> 3643916 (-0.33 %) VGPRS: 2678324 -> 2686324 (0.30 %) Spilled SGPRs: 1730 -> 1634 (-5.55 %) Spilled VGPRs: 14 -> 21 (50.00 %) Scratch size: 15540 -> 15536 (-0.03 %) dwords per thread Code Size: 136106120 -> 135457616 (-0.48 %) bytes LDS: 1259 -> 1259 (0.00 %) blocks Max Waves: 601014 -> 600206 (-0.13 %) Totals from affected shaders: SGPRS: 307832 -> 295848 (-3.89 %) VGPRS: 267864 -> 275864 (2.99 %) Spilled SGPRs: 770 -> 674 (-12.47 %) Spilled VGPRs: 14 -> 21 (50.00 %) Scratch size: 16 -> 12 (-25.00 %) dwords per thread Code Size: 22007488 -> 21358984 (-2.95 %) bytes LDS: 65 -> 65 (0.00 %) blocks Max Waves: 28668 -> 27860 (-2.82 %) Reviewed-by: Rhys Perry <pendingchaos02@gmail.com>
This commit is contained in:
parent
0837471463
commit
df3e674fb3
|
@ -3447,10 +3447,11 @@ void load_buffer(isel_context *ctx, unsigned num_components, Temp dst,
|
|||
Temp vec = bld.tmp(RegType::vgpr, dst.size());
|
||||
instr->definitions[0] = Definition(vec);
|
||||
bld.insert(std::move(instr));
|
||||
bld.pseudo(aco_opcode::p_as_uniform, Definition(dst), vec);
|
||||
expand_vector(ctx, vec, dst, num_components, (1 << num_components) - 1);
|
||||
} else {
|
||||
instr->definitions[0] = Definition(dst);
|
||||
bld.insert(std::move(instr));
|
||||
emit_split_vector(ctx, dst, num_components);
|
||||
}
|
||||
} else {
|
||||
switch (num_bytes) {
|
||||
|
@ -3506,9 +3507,8 @@ void load_buffer(isel_context *ctx, unsigned num_components, Temp dst,
|
|||
} else {
|
||||
bld.insert(std::move(load));
|
||||
}
|
||||
|
||||
}
|
||||
emit_split_vector(ctx, dst, num_components);
|
||||
}
|
||||
}
|
||||
|
||||
void visit_load_ubo(isel_context *ctx, nir_intrinsic_instr *instr)
|
||||
|
|
Loading…
Reference in New Issue