From 94656b1c8c60994a8bcc6c9b03981a78f0bba3a7 Mon Sep 17 00:00:00 2001 From: Chris Lattner Date: Sat, 28 Aug 2010 17:59:08 +0000 Subject: fix the buildvector->insertp[sd] logic to not always create a redundant insertp[sd] $0, which is a noop. Before: _f32: ## @f32 pshufd $1, %xmm1, %xmm2 pshufd $1, %xmm0, %xmm3 addss %xmm2, %xmm3 addss %xmm1, %xmm0 ## kill: XMM0 XMM0 XMM0 insertps $0, %xmm0, %xmm0 insertps $16, %xmm3, %xmm0 ret after: _f32: ## @f32 movdqa %xmm0, %xmm2 addss %xmm1, %xmm2 pshufd $1, %xmm1, %xmm1 pshufd $1, %xmm0, %xmm3 addss %xmm1, %xmm3 movdqa %xmm2, %xmm0 insertps $16, %xmm3, %xmm0 ret The extra movs are due to a random (poor) scheduling decision. llvm-svn: 112379 --- llvm/test/CodeGen/X86/vec_insert-9.ll | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'llvm/test/CodeGen/X86/vec_insert-9.ll') diff --git a/llvm/test/CodeGen/X86/vec_insert-9.ll b/llvm/test/CodeGen/X86/vec_insert-9.ll index 2e829df1f8d..e5a7ccc5ef9 100644 --- a/llvm/test/CodeGen/X86/vec_insert-9.ll +++ b/llvm/test/CodeGen/X86/vec_insert-9.ll @@ -1,5 +1,5 @@ ; RUN: llc < %s -march=x86 -mattr=+sse41 > %t -; RUN: grep pinsrd %t | count 2 +; RUN: grep pinsrd %t | count 1 define <4 x i32> @var_insert2(<4 x i32> %x, i32 %val, i32 %idx) nounwind { entry: -- cgit v1.2.3