[PATCH] D113739: [X86][ABI] Change the alignment of f80 in 32-bit calling conversion to meet with different data layout

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added a comment.

Should we be testing more target triples? I'm not clear how much variety there is on 32-bit targets.



================
Comment at: llvm/test/CodeGen/X86/inline-asm-fpstack.ll:29
 ; Asm consumes stack, nothing should be popped.
-define void @test3(x86_fp80 %X) {
+define void @test3(x86_fp80 %X) nounwind {
 ; CHECK-LABEL: test3:
----------------
precommit the nounwind change so its obvious where the stack manip is coming from


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113739/new/

https://reviews.llvm.org/D113739