[PATCH] D136285: Bad optimization with alloca and intrinsic function stackrestore

[Arthur Eubanks via Phabricator via llvm-commits]

aeubanks added a comment.

seems like the stackrestore shouldn't be marked as a tail call?



================
Comment at: llvm/test/Transforms/MemCpyOpt/stackrestore.ll:3
 ; RUN: opt -S -memcpyopt < %s -verify-memoryssa | FileCheck %s
+; RUN: opt -S -passes=memcpyopt -verify-memoryssa < %s | FileCheck %s --check-prefix=CHECK-TEST2
 
----------------
can you use update_test_checks.py instead of adding a new RUN line?


================
Comment at: llvm/test/Transforms/MemCpyOpt/stackrestore.ll:88
+; Function Attrs: nobuiltin norecurse
+define dso_local void @test() {
+; CHECK-TEST2-LABEL: @test
----------------
test could use some reduction, e.g. no `dso_local`, some of instructions seem unnecessary for a repro


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D136285/new/

https://reviews.llvm.org/D136285