Enhance the fix of PR17631

- The fix to PR17631 fixes part of the cases where 'vzeroupper' should
  not be issued before 'call' insn. There're other cases where helper
  calls will be inserted not limited to epilog. These helper calls do
  not follow the standard calling convention and won't clobber any YMM
  registers. (So far, all call conventions will clobber any or part of
  YMM registers.)
  This patch enhances the previous fix to cover more cases 'vzerosupper' should
  not be inserted by checking if that function call won't clobber any YMM
  registers and skipping it if so.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@196261 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 16 insertions, 4 deletions

diff --git a/test/CodeGen/X86/pr17631.ll b/test/CodeGen/X86/pr17631.ll
index a572ff2e3b..98f951f1b1 100644
--- a/test/CodeGen/X86/pr17631.ll
+++ b/test/CodeGen/X86/pr17631.ll
@@ -1,16 +1,16 @@
 ; RUN: llc < %s -mcpu=core-avx-i -mtriple=i386-pc-win32 | FileCheck %s
- 
+
 %struct_type = type { [64 x <8 x float>], <8 x float> }
- 
+
 ; Function Attrs: nounwind readnone
 declare i32 @llvm.x86.avx.movmsk.ps.256(<8 x float>)
- 
+
 ; Function Attrs: nounwind
 define i32 @equal(<8 x i32> %A) {
 allocas:
   %first_alloc  = alloca [64 x <8 x i32>]
   %second_alloc = alloca %struct_type
- 
+
   %A1 = bitcast <8 x i32> %A to <8 x float>
   %A2 = call i32 @llvm.x86.avx.movmsk.ps.256(<8 x float> %A1)
   ret i32 %A2
@@ -20,3 +20,15 @@ allocas:
 ; CHECK-NOT: vzeroupper
 ; CHECK: _chkstk
 ; CHECK: ret
+
+define <8 x float> @foo(<8 x float> %y, i64* %p, double %x) {
+  %i = fptoui double %x to i64
+  store i64 %i, i64* %p
+  %ret = fadd <8 x float> %y, %y
+  ret <8 x float> %ret
+}
+
+; CHECK: foo
+; CHECK-NOT: vzeroupper
+; CHECK: _ftol2
+; CHECK: ret