diff --git a/llvm/test/CodeGen/LoongArch/lasx/bitreverse.ll b/llvm/test/CodeGen/LoongArch/lasx/bitreverse.ll new file mode 100644 index 0000000000000..3d0d232fcca68 --- /dev/null +++ b/llvm/test/CodeGen/LoongArch/lasx/bitreverse.ll @@ -0,0 +1,107 @@ +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5 +; RUN: llc --mtriple=loongarch64 -mattr=+lasx --verify-machineinstrs < %s \ +; RUN: | FileCheck %s + +declare <32 x i8> @llvm.bitreverse.v32i8(<32 x i8>) + +define <32 x i8> @test_bitreverse_v32i8(<32 x i8> %a) nounwind { +; CHECK-LABEL: test_bitreverse_v32i8: +; CHECK: # %bb.0: +; CHECK-NEXT: xvslli.b $xr1, $xr0, 4 +; CHECK-NEXT: xvsrli.b $xr0, $xr0, 4 +; CHECK-NEXT: xvor.v $xr0, $xr0, $xr1 +; CHECK-NEXT: xvandi.b $xr1, $xr0, 51 +; CHECK-NEXT: xvslli.b $xr1, $xr1, 2 +; CHECK-NEXT: xvsrli.b $xr0, $xr0, 2 +; CHECK-NEXT: xvandi.b $xr0, $xr0, 51 +; CHECK-NEXT: xvor.v $xr0, $xr0, $xr1 +; CHECK-NEXT: xvandi.b $xr1, $xr0, 85 +; CHECK-NEXT: xvslli.b $xr1, $xr1, 1 +; CHECK-NEXT: xvsrli.b $xr0, $xr0, 1 +; CHECK-NEXT: xvandi.b $xr0, $xr0, 85 +; CHECK-NEXT: xvor.v $xr0, $xr0, $xr1 +; CHECK-NEXT: ret + %b = call <32 x i8> @llvm.bitreverse.v32i8(<32 x i8> %a) + ret <32 x i8> %b +} + +declare <16 x i16> @llvm.bitreverse.v16i16(<16 x i16>) + +define <16 x i16> @test_bitreverse_v16i16(<16 x i16> %a) nounwind { +; CHECK-LABEL: test_bitreverse_v16i16: +; CHECK: # %bb.0: +; CHECK-NEXT: xvshuf4i.b $xr0, $xr0, 177 +; CHECK-NEXT: xvsrli.h $xr1, $xr0, 4 +; CHECK-NEXT: xvrepli.b $xr2, 15 +; CHECK-NEXT: xvand.v $xr1, $xr1, $xr2 +; CHECK-NEXT: xvand.v $xr0, $xr0, $xr2 +; CHECK-NEXT: xvslli.h $xr0, $xr0, 4 +; CHECK-NEXT: xvor.v $xr0, $xr1, $xr0 +; CHECK-NEXT: xvsrli.h $xr1, $xr0, 2 +; CHECK-NEXT: xvrepli.b $xr2, 51 +; CHECK-NEXT: xvand.v $xr1, $xr1, $xr2 +; CHECK-NEXT: xvand.v $xr0, $xr0, $xr2 +; CHECK-NEXT: xvslli.h $xr0, $xr0, 2 +; CHECK-NEXT: xvor.v $xr0, $xr1, $xr0 +; CHECK-NEXT: xvsrli.h $xr1, $xr0, 1 +; CHECK-NEXT: xvrepli.b $xr2, 85 +; CHECK-NEXT: xvand.v $xr1, $xr1, $xr2 +; CHECK-NEXT: xvand.v $xr0, $xr0, $xr2 +; CHECK-NEXT: xvslli.h $xr0, $xr0, 1 +; CHECK-NEXT: xvor.v $xr0, $xr1, $xr0 +; CHECK-NEXT: ret + %b = call <16 x i16> @llvm.bitreverse.v16i16(<16 x i16> %a) + ret <16 x i16> %b +} + +declare <8 x i32> @llvm.bitreverse.v8i32(<8 x i32>) + +define <8 x i32> @test_bitreverse_v8i32(<8 x i32> %a) nounwind { +; CHECK-LABEL: test_bitreverse_v8i32: +; CHECK: # %bb.0: +; CHECK-NEXT: xvshuf4i.b $xr0, $xr0, 27 +; CHECK-NEXT: xvsrli.w $xr1, $xr0, 4 +; CHECK-NEXT: xvrepli.b $xr2, 15 +; CHECK-NEXT: xvand.v $xr1, $xr1, $xr2 +; CHECK-NEXT: xvand.v $xr0, $xr0, $xr2 +; CHECK-NEXT: xvslli.w $xr0, $xr0, 4 +; CHECK-NEXT: xvor.v $xr0, $xr1, $xr0 +; CHECK-NEXT: xvsrli.w $xr1, $xr0, 2 +; CHECK-NEXT: xvrepli.b $xr2, 51 +; CHECK-NEXT: xvand.v $xr1, $xr1, $xr2 +; CHECK-NEXT: xvand.v $xr0, $xr0, $xr2 +; CHECK-NEXT: xvslli.w $xr0, $xr0, 2 +; CHECK-NEXT: xvor.v $xr0, $xr1, $xr0 +; CHECK-NEXT: xvsrli.w $xr1, $xr0, 1 +; CHECK-NEXT: xvrepli.b $xr2, 85 +; CHECK-NEXT: xvand.v $xr1, $xr1, $xr2 +; CHECK-NEXT: xvand.v $xr0, $xr0, $xr2 +; CHECK-NEXT: xvslli.w $xr0, $xr0, 1 +; CHECK-NEXT: xvor.v $xr0, $xr1, $xr0 +; CHECK-NEXT: ret + %b = call <8 x i32> @llvm.bitreverse.v8i32(<8 x i32> %a) + ret <8 x i32> %b +} + +declare <4 x i64> @llvm.bitreverse.v4i64(<4 x i64>) + +define <4 x i64> @test_bitreverse_v4i64(<4 x i64> %a) nounwind { +; CHECK-LABEL: test_bitreverse_v4i64: +; CHECK: # %bb.0: +; CHECK-NEXT: xvpickve2gr.d $a0, $xr0, 0 +; CHECK-NEXT: bitrev.d $a0, $a0 +; CHECK-NEXT: xvinsgr2vr.d $xr1, $a0, 0 +; CHECK-NEXT: xvpickve2gr.d $a0, $xr0, 1 +; CHECK-NEXT: bitrev.d $a0, $a0 +; CHECK-NEXT: xvinsgr2vr.d $xr1, $a0, 1 +; CHECK-NEXT: xvpickve2gr.d $a0, $xr0, 2 +; CHECK-NEXT: bitrev.d $a0, $a0 +; CHECK-NEXT: xvinsgr2vr.d $xr1, $a0, 2 +; CHECK-NEXT: xvpickve2gr.d $a0, $xr0, 3 +; CHECK-NEXT: bitrev.d $a0, $a0 +; CHECK-NEXT: xvinsgr2vr.d $xr1, $a0, 3 +; CHECK-NEXT: xvori.b $xr0, $xr1, 0 +; CHECK-NEXT: ret + %b = call <4 x i64> @llvm.bitreverse.v4i64(<4 x i64> %a) + ret <4 x i64> %b +} diff --git a/llvm/test/CodeGen/LoongArch/lsx/bitreverse.ll b/llvm/test/CodeGen/LoongArch/lsx/bitreverse.ll new file mode 100644 index 0000000000000..93624c8dd6a96 --- /dev/null +++ b/llvm/test/CodeGen/LoongArch/lsx/bitreverse.ll @@ -0,0 +1,101 @@ +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5 +; RUN: llc --mtriple=loongarch64 -mattr=+lsx --verify-machineinstrs < %s \ +; RUN: | FileCheck %s + +declare <16 x i8> @llvm.bitreverse.v16i8(<16 x i8>) + +define <16 x i8> @test_bitreverse_v16i8(<16 x i8> %a) nounwind { +; CHECK-LABEL: test_bitreverse_v16i8: +; CHECK: # %bb.0: +; CHECK-NEXT: vslli.b $vr1, $vr0, 4 +; CHECK-NEXT: vsrli.b $vr0, $vr0, 4 +; CHECK-NEXT: vor.v $vr0, $vr0, $vr1 +; CHECK-NEXT: vandi.b $vr1, $vr0, 51 +; CHECK-NEXT: vslli.b $vr1, $vr1, 2 +; CHECK-NEXT: vsrli.b $vr0, $vr0, 2 +; CHECK-NEXT: vandi.b $vr0, $vr0, 51 +; CHECK-NEXT: vor.v $vr0, $vr0, $vr1 +; CHECK-NEXT: vandi.b $vr1, $vr0, 85 +; CHECK-NEXT: vslli.b $vr1, $vr1, 1 +; CHECK-NEXT: vsrli.b $vr0, $vr0, 1 +; CHECK-NEXT: vandi.b $vr0, $vr0, 85 +; CHECK-NEXT: vor.v $vr0, $vr0, $vr1 +; CHECK-NEXT: ret + %b = call <16 x i8> @llvm.bitreverse.v16i8(<16 x i8> %a) + ret <16 x i8> %b +} + +declare <8 x i16> @llvm.bitreverse.v8i16(<8 x i16>) + +define <8 x i16> @test_bitreverse_v8i16(<8 x i16> %a) nounwind { +; CHECK-LABEL: test_bitreverse_v8i16: +; CHECK: # %bb.0: +; CHECK-NEXT: vshuf4i.b $vr0, $vr0, 177 +; CHECK-NEXT: vsrli.h $vr1, $vr0, 4 +; CHECK-NEXT: vrepli.b $vr2, 15 +; CHECK-NEXT: vand.v $vr1, $vr1, $vr2 +; CHECK-NEXT: vand.v $vr0, $vr0, $vr2 +; CHECK-NEXT: vslli.h $vr0, $vr0, 4 +; CHECK-NEXT: vor.v $vr0, $vr1, $vr0 +; CHECK-NEXT: vsrli.h $vr1, $vr0, 2 +; CHECK-NEXT: vrepli.b $vr2, 51 +; CHECK-NEXT: vand.v $vr1, $vr1, $vr2 +; CHECK-NEXT: vand.v $vr0, $vr0, $vr2 +; CHECK-NEXT: vslli.h $vr0, $vr0, 2 +; CHECK-NEXT: vor.v $vr0, $vr1, $vr0 +; CHECK-NEXT: vsrli.h $vr1, $vr0, 1 +; CHECK-NEXT: vrepli.b $vr2, 85 +; CHECK-NEXT: vand.v $vr1, $vr1, $vr2 +; CHECK-NEXT: vand.v $vr0, $vr0, $vr2 +; CHECK-NEXT: vslli.h $vr0, $vr0, 1 +; CHECK-NEXT: vor.v $vr0, $vr1, $vr0 +; CHECK-NEXT: ret + %b = call <8 x i16> @llvm.bitreverse.v8i16(<8 x i16> %a) + ret <8 x i16> %b +} + +declare <4 x i32> @llvm.bitreverse.v4i32(<4 x i32>) + +define <4 x i32> @test_bitreverse_v4i32(<4 x i32> %a) nounwind { +; CHECK-LABEL: test_bitreverse_v4i32: +; CHECK: # %bb.0: +; CHECK-NEXT: vshuf4i.b $vr0, $vr0, 27 +; CHECK-NEXT: vsrli.w $vr1, $vr0, 4 +; CHECK-NEXT: vrepli.b $vr2, 15 +; CHECK-NEXT: vand.v $vr1, $vr1, $vr2 +; CHECK-NEXT: vand.v $vr0, $vr0, $vr2 +; CHECK-NEXT: vslli.w $vr0, $vr0, 4 +; CHECK-NEXT: vor.v $vr0, $vr1, $vr0 +; CHECK-NEXT: vsrli.w $vr1, $vr0, 2 +; CHECK-NEXT: vrepli.b $vr2, 51 +; CHECK-NEXT: vand.v $vr1, $vr1, $vr2 +; CHECK-NEXT: vand.v $vr0, $vr0, $vr2 +; CHECK-NEXT: vslli.w $vr0, $vr0, 2 +; CHECK-NEXT: vor.v $vr0, $vr1, $vr0 +; CHECK-NEXT: vsrli.w $vr1, $vr0, 1 +; CHECK-NEXT: vrepli.b $vr2, 85 +; CHECK-NEXT: vand.v $vr1, $vr1, $vr2 +; CHECK-NEXT: vand.v $vr0, $vr0, $vr2 +; CHECK-NEXT: vslli.w $vr0, $vr0, 1 +; CHECK-NEXT: vor.v $vr0, $vr1, $vr0 +; CHECK-NEXT: ret + %b = call <4 x i32> @llvm.bitreverse.v4i32(<4 x i32> %a) + ret <4 x i32> %b +} + +declare <2 x i64> @llvm.bitreverse.v2i64(<2 x i64>) + +define <2 x i64> @test_bitreverse_v2i64(<2 x i64> %a) nounwind { +; CHECK-LABEL: test_bitreverse_v2i64: +; CHECK: # %bb.0: +; CHECK-NEXT: vpickve2gr.d $a0, $vr0, 0 +; CHECK-NEXT: bitrev.d $a0, $a0 +; CHECK-NEXT: vinsgr2vr.d $vr1, $a0, 0 +; CHECK-NEXT: vpickve2gr.d $a0, $vr0, 1 +; CHECK-NEXT: bitrev.d $a0, $a0 +; CHECK-NEXT: vinsgr2vr.d $vr1, $a0, 1 +; CHECK-NEXT: vori.b $vr0, $vr1, 0 +; CHECK-NEXT: ret + %b = call <2 x i64> @llvm.bitreverse.v2i64(<2 x i64> %a) + ret <2 x i64> %b +}