diff --git a/test/CodeGen/AArch64/aarch64-load-ext.ll b/test/CodeGen/AArch64/aarch64-load-ext.ll index 1bbab3879dc..ec585264688 100644 --- a/test/CodeGen/AArch64/aarch64-load-ext.ll +++ b/test/CodeGen/AArch64/aarch64-load-ext.ll @@ -102,6 +102,56 @@ define <4 x i8> @test4(<4 x i8>* %v4i8_ptr) { ret <4 x i8> %v4i8 } +define <2 x i32> @fsext_v2i32(<2 x i8>* %a) { +; CHECK-LE-LABEL: fsext_v2i32: +; CHECK-LE: // %bb.0: +; CHECK-LE-NEXT: ldrsb w8, [x0] +; CHECK-LE-NEXT: ldrsb w9, [x0, #1] +; CHECK-LE-NEXT: fmov s0, w8 +; CHECK-LE-NEXT: mov v0.s[1], w9 +; CHECK-LE-NEXT: // kill: def $d0 killed $d0 killed $q0 +; CHECK-LE-NEXT: ret +; +; CHECK-BE-LABEL: fsext_v2i32: +; CHECK-BE: // %bb.0: +; CHECK-BE-NEXT: ldrsb w8, [x0] +; CHECK-BE-NEXT: ldrsb w9, [x0, #1] +; CHECK-BE-NEXT: fmov s0, w8 +; CHECK-BE-NEXT: mov v0.s[1], w9 +; CHECK-BE-NEXT: rev64 v0.2s, v0.2s +; CHECK-BE-NEXT: ret + %x = load <2 x i8>, <2 x i8>* %a + %y = sext <2 x i8> %x to <2 x i32> + ret <2 x i32> %y +} + +define <3 x i32> @fsext_v3i32(<3 x i8>* %a) { +; CHECK-LE-LABEL: fsext_v3i32: +; CHECK-LE: // %bb.0: +; CHECK-LE-NEXT: ldr s0, [x0] +; CHECK-LE-NEXT: zip1 v0.8b, v0.8b, v0.8b +; CHECK-LE-NEXT: ushll v0.4s, v0.4h, #0 +; CHECK-LE-NEXT: shl v0.4s, v0.4s, #24 +; CHECK-LE-NEXT: sshr v0.4s, v0.4s, #24 +; CHECK-LE-NEXT: ret +; +; CHECK-BE-LABEL: fsext_v3i32: +; CHECK-BE: // %bb.0: +; CHECK-BE-NEXT: ldr s0, [x0] +; CHECK-BE-NEXT: rev32 v0.8b, v0.8b +; CHECK-BE-NEXT: zip1 v0.8b, v0.8b, v0.8b +; CHECK-BE-NEXT: rev16 v0.8b, v0.8b +; CHECK-BE-NEXT: ushll v0.4s, v0.4h, #0 +; CHECK-BE-NEXT: shl v0.4s, v0.4s, #24 +; CHECK-BE-NEXT: sshr v0.4s, v0.4s, #24 +; CHECK-BE-NEXT: rev64 v0.4s, v0.4s +; CHECK-BE-NEXT: ext v0.16b, v0.16b, v0.16b, #8 +; CHECK-BE-NEXT: ret + %x = load <3 x i8>, <3 x i8>* %a + %y = sext <3 x i8> %x to <3 x i32> + ret <3 x i32> %y +} + define <4 x i32> @fsext_v4i32(<4 x i8>* %a) { ; CHECK-LE-LABEL: fsext_v4i32: ; CHECK-LE: // %bb.0: @@ -124,6 +174,31 @@ define <4 x i32> @fsext_v4i32(<4 x i8>* %a) { ret <4 x i32> %y } +define <8 x i32> @fsext_v8i32(<8 x i8>* %a) { +; CHECK-LE-LABEL: fsext_v8i32: +; CHECK-LE: // %bb.0: +; CHECK-LE-NEXT: ldr d0, [x0] +; CHECK-LE-NEXT: sshll v0.8h, v0.8b, #0 +; CHECK-LE-NEXT: sshll2 v1.4s, v0.8h, #0 +; CHECK-LE-NEXT: sshll v0.4s, v0.4h, #0 +; CHECK-LE-NEXT: ret +; +; CHECK-BE-LABEL: fsext_v8i32: +; CHECK-BE: // %bb.0: +; CHECK-BE-NEXT: ld1 { v0.8b }, [x0] +; CHECK-BE-NEXT: sshll v0.8h, v0.8b, #0 +; CHECK-BE-NEXT: sshll v1.4s, v0.4h, #0 +; CHECK-BE-NEXT: sshll2 v0.4s, v0.8h, #0 +; CHECK-BE-NEXT: rev64 v0.4s, v0.4s +; CHECK-BE-NEXT: rev64 v2.4s, v1.4s +; CHECK-BE-NEXT: ext v1.16b, v0.16b, v0.16b, #8 +; CHECK-BE-NEXT: ext v0.16b, v2.16b, v2.16b, #8 +; CHECK-BE-NEXT: ret + %x = load <8 x i8>, <8 x i8>* %a + %y = sext <8 x i8> %x to <8 x i32> + ret <8 x i32> %y +} + define <4 x i32> @fzext_v4i32(<4 x i8>* %a) { ; CHECK-LE-LABEL: fzext_v4i32: ; CHECK-LE: // %bb.0: @@ -172,6 +247,53 @@ define i32 @loadExti32(<4 x i8>* %ref) { ret i32 %conv } +define <2 x i16> @fsext_v2i16(<2 x i8>* %a) { +; CHECK-LE-LABEL: fsext_v2i16: +; CHECK-LE: // %bb.0: +; CHECK-LE-NEXT: ldrsb w8, [x0] +; CHECK-LE-NEXT: ldrsb w9, [x0, #1] +; CHECK-LE-NEXT: fmov s0, w8 +; CHECK-LE-NEXT: mov v0.s[1], w9 +; CHECK-LE-NEXT: // kill: def $d0 killed $d0 killed $q0 +; CHECK-LE-NEXT: ret +; +; CHECK-BE-LABEL: fsext_v2i16: +; CHECK-BE: // %bb.0: +; CHECK-BE-NEXT: ldrsb w8, [x0] +; CHECK-BE-NEXT: ldrsb w9, [x0, #1] +; CHECK-BE-NEXT: fmov s0, w8 +; CHECK-BE-NEXT: mov v0.s[1], w9 +; CHECK-BE-NEXT: rev64 v0.2s, v0.2s +; CHECK-BE-NEXT: ret + %x = load <2 x i8>, <2 x i8>* %a + %y = sext <2 x i8> %x to <2 x i16> + ret <2 x i16> %y +} + +define <3 x i16> @fsext_v3i16(<3 x i8>* %a) { +; CHECK-LE-LABEL: fsext_v3i16: +; CHECK-LE: // %bb.0: +; CHECK-LE-NEXT: ldr s0, [x0] +; CHECK-LE-NEXT: zip1 v0.8b, v0.8b, v0.8b +; CHECK-LE-NEXT: shl v0.4h, v0.4h, #8 +; CHECK-LE-NEXT: sshr v0.4h, v0.4h, #8 +; CHECK-LE-NEXT: ret +; +; CHECK-BE-LABEL: fsext_v3i16: +; CHECK-BE: // %bb.0: +; CHECK-BE-NEXT: ldr s0, [x0] +; CHECK-BE-NEXT: rev32 v0.8b, v0.8b +; CHECK-BE-NEXT: zip1 v0.8b, v0.8b, v0.8b +; CHECK-BE-NEXT: rev16 v0.8b, v0.8b +; CHECK-BE-NEXT: shl v0.4h, v0.4h, #8 +; CHECK-BE-NEXT: sshr v0.4h, v0.4h, #8 +; CHECK-BE-NEXT: rev64 v0.4h, v0.4h +; CHECK-BE-NEXT: ret + %x = load <3 x i8>, <3 x i8>* %a + %y = sext <3 x i8> %x to <3 x i16> + ret <3 x i16> %y +} + define <4 x i16> @fsext_v4i16(<4 x i8>* %a) { ; CHECK-LE-LABEL: fsext_v4i16: ; CHECK-LE: // %bb.0: @@ -192,6 +314,48 @@ define <4 x i16> @fsext_v4i16(<4 x i8>* %a) { ret <4 x i16> %y } +define <8 x i16> @fsext_v8i16(<8 x i8>* %a) { +; CHECK-LE-LABEL: fsext_v8i16: +; CHECK-LE: // %bb.0: +; CHECK-LE-NEXT: ldr d0, [x0] +; CHECK-LE-NEXT: sshll v0.8h, v0.8b, #0 +; CHECK-LE-NEXT: ret +; +; CHECK-BE-LABEL: fsext_v8i16: +; CHECK-BE: // %bb.0: +; CHECK-BE-NEXT: ld1 { v0.8b }, [x0] +; CHECK-BE-NEXT: sshll v0.8h, v0.8b, #0 +; CHECK-BE-NEXT: rev64 v0.8h, v0.8h +; CHECK-BE-NEXT: ext v0.16b, v0.16b, v0.16b, #8 +; CHECK-BE-NEXT: ret + %x = load <8 x i8>, <8 x i8>* %a + %y = sext <8 x i8> %x to <8 x i16> + ret <8 x i16> %y +} + +define <16 x i16> @fsext_v16i16(<16 x i8>* %a) { +; CHECK-LE-LABEL: fsext_v16i16: +; CHECK-LE: // %bb.0: +; CHECK-LE-NEXT: ldr q0, [x0] +; CHECK-LE-NEXT: sshll2 v1.8h, v0.16b, #0 +; CHECK-LE-NEXT: sshll v0.8h, v0.8b, #0 +; CHECK-LE-NEXT: ret +; +; CHECK-BE-LABEL: fsext_v16i16: +; CHECK-BE: // %bb.0: +; CHECK-BE-NEXT: ld1 { v0.16b }, [x0] +; CHECK-BE-NEXT: sshll v1.8h, v0.8b, #0 +; CHECK-BE-NEXT: sshll2 v0.8h, v0.16b, #0 +; CHECK-BE-NEXT: rev64 v0.8h, v0.8h +; CHECK-BE-NEXT: rev64 v2.8h, v1.8h +; CHECK-BE-NEXT: ext v1.16b, v0.16b, v0.16b, #8 +; CHECK-BE-NEXT: ext v0.16b, v2.16b, v2.16b, #8 +; CHECK-BE-NEXT: ret + %x = load <16 x i8>, <16 x i8>* %a + %y = sext <16 x i8> %x to <16 x i16> + ret <16 x i16> %y +} + define <4 x i16> @fzext_v4i16(<4 x i8>* %a) { ; CHECK-LE-LABEL: fzext_v4i16: ; CHECK-LE: // %bb.0: