X-Git-Url: http://plrg.eecs.uci.edu/git/?p=oota-llvm.git;a=blobdiff_plain;f=test%2FCodeGen%2FARM%2Fvrev.ll;h=a20d4b6baf293172cadf0c6cfe0536079ba7e464;hp=0f0ea2b93220502a73cd73aea4c27aad7675593e;hb=b1337aba53a9112aa5f45a54e7fb6f5102071afa;hpb=2a8eb722c7bb0fac2fe09a876f3471dcb25f465e diff --git a/test/CodeGen/ARM/vrev.ll b/test/CodeGen/ARM/vrev.ll index 0f0ea2b9322..a20d4b6baf2 100644 --- a/test/CodeGen/ARM/vrev.ll +++ b/test/CodeGen/ARM/vrev.ll @@ -1,113 +1,113 @@ -; RUN: llc < %s -march=arm -mattr=+neon | FileCheck %s +; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o - | FileCheck %s define <8 x i8> @test_vrev64D8(<8 x i8>* %A) nounwind { -;CHECK: test_vrev64D8: +;CHECK-LABEL: test_vrev64D8: ;CHECK: vrev64.8 - %tmp1 = load <8 x i8>* %A + %tmp1 = load <8 x i8>, <8 x i8>* %A %tmp2 = shufflevector <8 x i8> %tmp1, <8 x i8> undef, <8 x i32> ret <8 x i8> %tmp2 } define <4 x i16> @test_vrev64D16(<4 x i16>* %A) nounwind { -;CHECK: test_vrev64D16: +;CHECK-LABEL: test_vrev64D16: ;CHECK: vrev64.16 - %tmp1 = load <4 x i16>* %A + %tmp1 = load <4 x i16>, <4 x i16>* %A %tmp2 = shufflevector <4 x i16> %tmp1, <4 x i16> undef, <4 x i32> ret <4 x i16> %tmp2 } define <2 x i32> @test_vrev64D32(<2 x i32>* %A) nounwind { -;CHECK: test_vrev64D32: +;CHECK-LABEL: test_vrev64D32: ;CHECK: vrev64.32 - %tmp1 = load <2 x i32>* %A + %tmp1 = load <2 x i32>, <2 x i32>* %A %tmp2 = shufflevector <2 x i32> %tmp1, <2 x i32> undef, <2 x i32> ret <2 x i32> %tmp2 } define <2 x float> @test_vrev64Df(<2 x float>* %A) nounwind { -;CHECK: test_vrev64Df: +;CHECK-LABEL: test_vrev64Df: ;CHECK: vrev64.32 - %tmp1 = load <2 x float>* %A + %tmp1 = load <2 x float>, <2 x float>* %A %tmp2 = shufflevector <2 x float> %tmp1, <2 x float> undef, <2 x i32> ret <2 x float> %tmp2 } define <16 x i8> @test_vrev64Q8(<16 x i8>* %A) nounwind { -;CHECK: test_vrev64Q8: +;CHECK-LABEL: test_vrev64Q8: ;CHECK: vrev64.8 - %tmp1 = load <16 x i8>* %A + %tmp1 = load <16 x i8>, <16 x i8>* %A %tmp2 = shufflevector <16 x i8> %tmp1, <16 x i8> undef, <16 x i32> ret <16 x i8> %tmp2 } define <8 x i16> @test_vrev64Q16(<8 x i16>* %A) nounwind { -;CHECK: test_vrev64Q16: +;CHECK-LABEL: test_vrev64Q16: ;CHECK: vrev64.16 - %tmp1 = load <8 x i16>* %A + %tmp1 = load <8 x i16>, <8 x i16>* %A %tmp2 = shufflevector <8 x i16> %tmp1, <8 x i16> undef, <8 x i32> ret <8 x i16> %tmp2 } define <4 x i32> @test_vrev64Q32(<4 x i32>* %A) nounwind { -;CHECK: test_vrev64Q32: +;CHECK-LABEL: test_vrev64Q32: ;CHECK: vrev64.32 - %tmp1 = load <4 x i32>* %A + %tmp1 = load <4 x i32>, <4 x i32>* %A %tmp2 = shufflevector <4 x i32> %tmp1, <4 x i32> undef, <4 x i32> ret <4 x i32> %tmp2 } define <4 x float> @test_vrev64Qf(<4 x float>* %A) nounwind { -;CHECK: test_vrev64Qf: +;CHECK-LABEL: test_vrev64Qf: ;CHECK: vrev64.32 - %tmp1 = load <4 x float>* %A + %tmp1 = load <4 x float>, <4 x float>* %A %tmp2 = shufflevector <4 x float> %tmp1, <4 x float> undef, <4 x i32> ret <4 x float> %tmp2 } define <8 x i8> @test_vrev32D8(<8 x i8>* %A) nounwind { -;CHECK: test_vrev32D8: +;CHECK-LABEL: test_vrev32D8: ;CHECK: vrev32.8 - %tmp1 = load <8 x i8>* %A + %tmp1 = load <8 x i8>, <8 x i8>* %A %tmp2 = shufflevector <8 x i8> %tmp1, <8 x i8> undef, <8 x i32> ret <8 x i8> %tmp2 } define <4 x i16> @test_vrev32D16(<4 x i16>* %A) nounwind { -;CHECK: test_vrev32D16: +;CHECK-LABEL: test_vrev32D16: ;CHECK: vrev32.16 - %tmp1 = load <4 x i16>* %A + %tmp1 = load <4 x i16>, <4 x i16>* %A %tmp2 = shufflevector <4 x i16> %tmp1, <4 x i16> undef, <4 x i32> ret <4 x i16> %tmp2 } define <16 x i8> @test_vrev32Q8(<16 x i8>* %A) nounwind { -;CHECK: test_vrev32Q8: +;CHECK-LABEL: test_vrev32Q8: ;CHECK: vrev32.8 - %tmp1 = load <16 x i8>* %A + %tmp1 = load <16 x i8>, <16 x i8>* %A %tmp2 = shufflevector <16 x i8> %tmp1, <16 x i8> undef, <16 x i32> ret <16 x i8> %tmp2 } define <8 x i16> @test_vrev32Q16(<8 x i16>* %A) nounwind { -;CHECK: test_vrev32Q16: +;CHECK-LABEL: test_vrev32Q16: ;CHECK: vrev32.16 - %tmp1 = load <8 x i16>* %A + %tmp1 = load <8 x i16>, <8 x i16>* %A %tmp2 = shufflevector <8 x i16> %tmp1, <8 x i16> undef, <8 x i32> ret <8 x i16> %tmp2 } define <8 x i8> @test_vrev16D8(<8 x i8>* %A) nounwind { -;CHECK: test_vrev16D8: +;CHECK-LABEL: test_vrev16D8: ;CHECK: vrev16.8 - %tmp1 = load <8 x i8>* %A + %tmp1 = load <8 x i8>, <8 x i8>* %A %tmp2 = shufflevector <8 x i8> %tmp1, <8 x i8> undef, <8 x i32> ret <8 x i8> %tmp2 } define <16 x i8> @test_vrev16Q8(<16 x i8>* %A) nounwind { -;CHECK: test_vrev16Q8: +;CHECK-LABEL: test_vrev16Q8: ;CHECK: vrev16.8 - %tmp1 = load <16 x i8>* %A + %tmp1 = load <16 x i8>, <16 x i8>* %A %tmp2 = shufflevector <16 x i8> %tmp1, <16 x i8> undef, <16 x i32> ret <16 x i8> %tmp2 } @@ -115,17 +115,17 @@ define <16 x i8> @test_vrev16Q8(<16 x i8>* %A) nounwind { ; Undef shuffle indices should not prevent matching to VREV: define <8 x i8> @test_vrev64D8_undef(<8 x i8>* %A) nounwind { -;CHECK: test_vrev64D8_undef: +;CHECK-LABEL: test_vrev64D8_undef: ;CHECK: vrev64.8 - %tmp1 = load <8 x i8>* %A + %tmp1 = load <8 x i8>, <8 x i8>* %A %tmp2 = shufflevector <8 x i8> %tmp1, <8 x i8> undef, <8 x i32> ret <8 x i8> %tmp2 } define <8 x i16> @test_vrev32Q16_undef(<8 x i16>* %A) nounwind { -;CHECK: test_vrev32Q16_undef: +;CHECK-LABEL: test_vrev32Q16_undef: ;CHECK: vrev32.16 - %tmp1 = load <8 x i16>* %A + %tmp1 = load <8 x i16>, <8 x i16>* %A %tmp2 = shufflevector <8 x i16> %tmp1, <8 x i16> undef, <8 x i32> ret <8 x i16> %tmp2 } @@ -133,10 +133,10 @@ define <8 x i16> @test_vrev32Q16_undef(<8 x i16>* %A) nounwind { ; A vcombine feeding a VREV should not obscure things. Radar 8597007. define void @test_with_vcombine(<4 x float>* %v) nounwind { -;CHECK: test_with_vcombine: +;CHECK-LABEL: test_with_vcombine: ;CHECK-NOT: vext ;CHECK: vrev64.32 - %tmp1 = load <4 x float>* %v, align 16 + %tmp1 = load <4 x float>, <4 x float>* %v, align 16 %tmp2 = bitcast <4 x float> %tmp1 to <2 x double> %tmp3 = extractelement <2 x double> %tmp2, i32 0 %tmp4 = bitcast double %tmp3 to <2 x float> @@ -148,14 +148,14 @@ define void @test_with_vcombine(<4 x float>* %v) nounwind { ret void } -; vrev <4 x i16> should use VREV32 and not VREV64 +; The type <2 x i16> is legalized to <2 x i32> and need to be trunc-stored +; to <2 x i16> when stored to memory. define void @test_vrev64(<4 x i16>* nocapture %source, <2 x i16>* nocapture %dst) nounwind ssp { -; CHECK: test_vrev64: -; CHECK: vext.16 -; CHECK: vrev32.16 +; CHECK-LABEL: test_vrev64: +; CHECK: vst1.32 entry: %0 = bitcast <4 x i16>* %source to <8 x i16>* - %tmp2 = load <8 x i16>* %0, align 4 + %tmp2 = load <8 x i16>, <8 x i16>* %0, align 4 %tmp3 = extractelement <8 x i16> %tmp2, i32 6 %tmp5 = insertelement <2 x i16> undef, i16 %tmp3, i32 0 %tmp9 = extractelement <8 x i16> %tmp2, i32 5 @@ -163,3 +163,26 @@ entry: store <2 x i16> %tmp11, <2 x i16>* %dst, align 4 ret void } + +; Test vrev of float4 +define void @float_vrev64(float* nocapture %source, <4 x float>* nocapture %dest) nounwind noinline ssp { +; CHECK: float_vrev64 +; CHECK: vext.32 +; CHECK: vrev64.32 +entry: + %0 = bitcast float* %source to <4 x float>* + %tmp2 = load <4 x float>, <4 x float>* %0, align 4 + %tmp5 = shufflevector <4 x float> , <4 x float> %tmp2, <4 x i32> + %arrayidx8 = getelementptr inbounds <4 x float>, <4 x float>* %dest, i32 11 + store <4 x float> %tmp5, <4 x float>* %arrayidx8, align 4 + ret void +} + +define <4 x i32> @test_vrev32_bswap(<4 x i32> %source) nounwind { +; CHECK-LABEL: test_vrev32_bswap: +; CHECK: vrev32.8 + %bswap = call <4 x i32> @llvm.bswap.v4i32(<4 x i32> %source) + ret <4 x i32> %bswap +} + +declare <4 x i32> @llvm.bswap.v4i32(<4 x i32>) nounwind readnone