[PATCH] D108994: [AMDGPU] Enable ds_min/ds_max on more subtargets

Tue Aug 31 08:43:37 PDT 2021

foad added inline comments.

================
Comment at: llvm/test/CodeGen/AMDGPU/lds-atomic-add.ll:1
+; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,VI %s
+; RUN: llc -march=amdgcn -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -enable-var-scope -check-prefixes=GCN,GFX9 %s
----------------
Joe_Nash wrote:
> foad wrote:
> > This file looks like a subset of the existing `test/CodeGen/AMDGPU/lds_atomic_f32.ll`?
> I split the test because the lds_atomic_fadd is only on gfx8 and gfx9, but the min/max are available on all architectures
Oh, I hadn't even noticed that your patch deletes lds_atomic_f32.ll. OK then.

The new files should probably be called lds-atomic-fadd.ll and lds-atomic-fmin-fmax.ll.

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D108994/new/

https://reviews.llvm.org/D108994