ShuriZma
/
xenia
mirror of https://github.com/xenia-project/xenia.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

[GPU] Update v_mad_legacy_f32 comment

This commit is contained in:
Triang3l 2020-12-10 12:39:02 +03:00
parent bc0c2040e2
commit 8bcfcf6452
1 changed files with 5 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
src/xenia/gpu/ucode.h
View File

@ -816,10 +816,11 @@ static_assert_size(TextureFetchInstruction, 12);
// move of the third operand in case of zero multiplicands, because the term
// may be -0, while the result should be +0 in this case.
// http://developer.amd.com/wordpress/media/2013/10/R5xx_Acceleration_v1.5.pdf
// Multiply-add also appears to be not fused (the SM3 behavior instruction on
// GCN is called v_mad_legacy_f32, not v_fma_legacy_f32) - shader translators
// should not use instructions that may be interpreted by the host GPU as
// fused multiply-add.
// Multiply-add also appears to be not fused; the SM3 behavior instruction on
// GCN is called v_mad_legacy_f32, not v_fma_legacy_f32 (in 2012-2020, before
// RDNA 2, which removed v_mad_f32 as well) - shader translators should not
// use instructions that may be interpreted by the host GPU as fused
// multiply-add.
enum class AluScalarOpcode : uint32_t {
// Floating-Point Add