[llvm] [AMDGPU][GISel] Add RegBankLegalize support for G_STRICT_{FADD|FSUB|FMUL} (PR #169406)

[Petar Avramovic via llvm-commits]

================
@@ -1,9 +1,17 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx900 < %s | FileCheck -check-prefixes=GCN,GFX9 %s
-; RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=fiji < %s | FileCheck -check-prefixes=GCN,GFX8 %s
-; RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1010 < %s | FileCheck -check-prefixes=GFX10PLUS,GFX10 %s
-; RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1100 -mattr=+real-true16 -amdgpu-enable-delay-alu=0 < %s | FileCheck -check-prefixes=GFX10PLUS,GFX11,GFX11-TRUE16 %s
-; RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1100 -mattr=-real-true16 -amdgpu-enable-delay-alu=0 < %s | FileCheck -check-prefixes=GFX10PLUS,GFX11,GFX11-FAKE16 %s
----------------
petar-avramovic wrote:

What are you trying to test with all these new RUN lines? true16/no-true16/sdag behavior i not in scope of this patch. 
Minimal set of targets to test this patch is for example gfx11(no salu float), gfx12(has salu float) and gfx942(v_pk_Op_f32)


https://github.com/llvm/llvm-project/pull/169406