[PATCH] D76239: [X86] X86CallFrameOptimization - generalize slow push code path

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon marked an inline comment as done.
RKSimon added inline comments.


================
Comment at: llvm/test/CodeGen/X86/atomic-idempotent.ll:4
+; RUN: llc < %s -mtriple=i686-- -verify-machineinstrs           -mattr=+sse2      | FileCheck %s --check-prefixes=CHECK,X86,X86-GENERIC,X86-SSE2
+; RUN: llc < %s -mtriple=i686-- -verify-machineinstrs -mcpu=slm -mattr=-sse2      | FileCheck %s --check-prefixes=CHECK,X86,X86-GENERIC,X86-SLM
+; RUN: llc < %s -mtriple=i686-- -verify-machineinstrs -mcpu=goldmont -mattr=-sse2 | FileCheck %s --check-prefixes=CHECK,X86,X86-GENERIC,X86-SLM
----------------
craig.topper wrote:
> Why did SLM/GLM/KNL lose sse2, but atom got to keep it?
Disabling sse2 was more of a technicality to force the SLM-like targets to match, but I've set atom to follow this as well to stop the fence insertion


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D76239/new/

https://reviews.llvm.org/D76239