brintos

brintos / llvm-project-archived public Read only

0
0
Text · 706 B · 6c8dd25 Raw
17 lines · plain
1; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse4.1 | FileCheck %s2; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s3 4; When commuting the operands of a SSE blend, make sure that the resulting blend5; mask can be encoded as a imm8.6; Before, when commuting the operands to the shuffle in function @test, the backend7; produced the following assembly:8;   pblendw $4294967103, %xmm1, %xmm09 10define <4 x i32> @test(<4 x i32> %a, <4 x i32> %b) {11; CHECK: pblendw $63, %xmm1, %xmm012  %shuffle = shufflevector <4 x i32> %a, <4 x i32> %b, <4 x i32> <i32 4, i32 5, i32 6, i32 3>13  ; add forces execution domain14  %sum = add <4 x i32> %shuffle, %shuffle15  ret <4 x i32> %sum16}17