[llvm] [AMDGPU][GlobalISel] Improve combining on v_ldexp (PR #190236)

[Jay Foad via llvm-commits]

================
@@ -2,34 +2,20 @@
 ; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1200 < %s | FileCheck -check-prefixes=GFX12,GFX12-SDAG %s
 ; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx1200 < %s | FileCheck -check-prefixes=GFX12,GFX12-GISEL %s
 
-; TODO: GlobalISel should avoid generating v_ldexp_f32.
----------------
jayfoad wrote:

@petar-avramovic I think the motiviation is to fix this TODO. We should not generate any VALU instructions for an operation that can be done entirely with scalar (or pseudo-scalar) float.

https://github.com/llvm/llvm-project/pull/190236