[llvm] r241445 - [X86][SSE] Added missing stack folding test for SQRTSD and SQRTSS instructions.
Simon Pilgrim
llvm-dev at redking.me.uk
Mon Jul 6 07:15:02 PDT 2015
Author: rksimon
Date: Mon Jul 6 09:15:02 2015
New Revision: 241445
URL: http://llvm.org/viewvc/llvm-project?rev=241445&view=rev
Log:
[X86][SSE] Added missing stack folding test for SQRTSD and SQRTSS instructions.
Modified:
llvm/trunk/test/CodeGen/X86/stack-folding-fp-sse42.ll
Modified: llvm/trunk/test/CodeGen/X86/stack-folding-fp-sse42.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/stack-folding-fp-sse42.ll?rev=241445&r1=241444&r2=241445&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/X86/stack-folding-fp-sse42.ll (original)
+++ llvm/trunk/test/CodeGen/X86/stack-folding-fp-sse42.ll Mon Jul 6 09:15:02 2015
@@ -938,13 +938,25 @@ define <4 x float> @stack_fold_sqrtps(<4
}
declare <4 x float> @llvm.x86.sse.sqrt.ps(<4 x float>) nounwind readnone
-; TODO stack_fold_sqrtsd
+define double @stack_fold_sqrtsd(double %a0) optsize {
+ ;CHECK-LABEL: stack_fold_sqrtsd
+ ;CHECK: sqrtsd {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 8-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call double @llvm.sqrt.f64(double %a0)
+ ret double %2
+}
declare double @llvm.sqrt.f64(double) nounwind readnone
; TODO stack_fold_sqrtsd_int
declare <2 x double> @llvm.x86.sse2.sqrt.sd(<2 x double>) nounwind readnone
-; TODO stack_fold_sqrtss
+define float @stack_fold_sqrtss(float %a0) optsize {
+ ;CHECK-LABEL: stack_fold_sqrtss
+ ;CHECK: sqrtss {{-?[0-9]*}}(%rsp), {{%xmm[0-9][0-9]*}} {{.*#+}} 4-byte Folded Reload
+ %1 = tail call <2 x i64> asm sideeffect "nop", "=x,~{xmm1},~{xmm2},~{xmm3},~{xmm4},~{xmm5},~{xmm6},~{xmm7},~{xmm8},~{xmm9},~{xmm10},~{xmm11},~{xmm12},~{xmm13},~{xmm14},~{xmm15},~{flags}"()
+ %2 = call float @llvm.sqrt.f32(float %a0)
+ ret float %2
+}
declare float @llvm.sqrt.f32(float) nounwind readnone
; TODO stack_fold_sqrtss_int
More information about the llvm-commits
mailing list