[llvm] f53bfa3 - [AArch64] Add And/Or/XOr test coverage for GISel. NFC
David Green via llvm-commits
llvm-commits at lists.llvm.org
Fri Aug 23 01:19:34 PDT 2024
Author: David Green
Date: 2024-08-23T09:19:25+01:00
New Revision: f53bfa39a7dae444650a9c0e16d52301a733f5fc
URL: https://github.com/llvm/llvm-project/commit/f53bfa39a7dae444650a9c0e16d52301a733f5fc
DIFF: https://github.com/llvm/llvm-project/commit/f53bfa39a7dae444650a9c0e16d52301a733f5fc.diff
LOG: [AArch64] Add And/Or/XOr test coverage for GISel. NFC
Added:
llvm/test/CodeGen/AArch64/andorxor.ll
Modified:
Removed:
################################################################################
diff --git a/llvm/test/CodeGen/AArch64/andorxor.ll b/llvm/test/CodeGen/AArch64/andorxor.ll
new file mode 100644
index 00000000000000..efa4be707ceda9
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/andorxor.ll
@@ -0,0 +1,1522 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
+; RUN: llc -mtriple=aarch64-none-eabi -verify-machineinstrs %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-SD
+; RUN: llc -mtriple=aarch64-none-eabi -global-isel -global-isel-abort=2 -verify-machineinstrs %s -o - 2>&1 | FileCheck %s --check-prefixes=CHECK,CHECK-GI
+
+; CHECK-GI: warning: Instruction selection used fallback path for and_v2i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for or_v2i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for xor_v2i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for and_v3i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for or_v3i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for xor_v3i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for and_v4i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for or_v4i128
+; CHECK-GI-NEXT: warning: Instruction selection used fallback path for xor_v4i128
+
+define i8 @and_i8(i8 %a, i8 %b) {
+; CHECK-LABEL: and_i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and w0, w0, w1
+; CHECK-NEXT: ret
+entry:
+ %s = and i8 %a, %b
+ ret i8 %s
+}
+
+define i8 @or_i8(i8 %a, i8 %b) {
+; CHECK-LABEL: or_i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr w0, w0, w1
+; CHECK-NEXT: ret
+entry:
+ %s = or i8 %a, %b
+ ret i8 %s
+}
+
+define i8 @xor_i8(i8 %a, i8 %b) {
+; CHECK-LABEL: xor_i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor w0, w0, w1
+; CHECK-NEXT: ret
+entry:
+ %s = xor i8 %a, %b
+ ret i8 %s
+}
+
+define i16 @and_i16(i16 %a, i16 %b) {
+; CHECK-LABEL: and_i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and w0, w0, w1
+; CHECK-NEXT: ret
+entry:
+ %s = and i16 %a, %b
+ ret i16 %s
+}
+
+define i16 @or_i16(i16 %a, i16 %b) {
+; CHECK-LABEL: or_i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr w0, w0, w1
+; CHECK-NEXT: ret
+entry:
+ %s = or i16 %a, %b
+ ret i16 %s
+}
+
+define i16 @xor_i16(i16 %a, i16 %b) {
+; CHECK-LABEL: xor_i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor w0, w0, w1
+; CHECK-NEXT: ret
+entry:
+ %s = xor i16 %a, %b
+ ret i16 %s
+}
+
+define i32 @and_i32(i32 %a, i32 %b) {
+; CHECK-LABEL: and_i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and w0, w0, w1
+; CHECK-NEXT: ret
+entry:
+ %s = and i32 %a, %b
+ ret i32 %s
+}
+
+define i32 @or_i32(i32 %a, i32 %b) {
+; CHECK-LABEL: or_i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr w0, w0, w1
+; CHECK-NEXT: ret
+entry:
+ %s = or i32 %a, %b
+ ret i32 %s
+}
+
+define i32 @xor_i32(i32 %a, i32 %b) {
+; CHECK-LABEL: xor_i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor w0, w0, w1
+; CHECK-NEXT: ret
+entry:
+ %s = xor i32 %a, %b
+ ret i32 %s
+}
+
+define i64 @and_i64(i64 %a, i64 %b) {
+; CHECK-LABEL: and_i64:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and x0, x0, x1
+; CHECK-NEXT: ret
+entry:
+ %s = and i64 %a, %b
+ ret i64 %s
+}
+
+define i64 @or_i64(i64 %a, i64 %b) {
+; CHECK-LABEL: or_i64:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr x0, x0, x1
+; CHECK-NEXT: ret
+entry:
+ %s = or i64 %a, %b
+ ret i64 %s
+}
+
+define i64 @xor_i64(i64 %a, i64 %b) {
+; CHECK-LABEL: xor_i64:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor x0, x0, x1
+; CHECK-NEXT: ret
+entry:
+ %s = xor i64 %a, %b
+ ret i64 %s
+}
+
+define i128 @and_i128(i128 %a, i128 %b) {
+; CHECK-SD-LABEL: and_i128:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: and x1, x1, x3
+; CHECK-SD-NEXT: and x0, x0, x2
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: and_i128:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: and x0, x0, x2
+; CHECK-GI-NEXT: and x1, x1, x3
+; CHECK-GI-NEXT: ret
+entry:
+ %s = and i128 %a, %b
+ ret i128 %s
+}
+
+define i128 @or_i128(i128 %a, i128 %b) {
+; CHECK-SD-LABEL: or_i128:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: orr x1, x1, x3
+; CHECK-SD-NEXT: orr x0, x0, x2
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: or_i128:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: orr x0, x0, x2
+; CHECK-GI-NEXT: orr x1, x1, x3
+; CHECK-GI-NEXT: ret
+entry:
+ %s = or i128 %a, %b
+ ret i128 %s
+}
+
+define i128 @xor_i128(i128 %a, i128 %b) {
+; CHECK-SD-LABEL: xor_i128:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: eor x1, x1, x3
+; CHECK-SD-NEXT: eor x0, x0, x2
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: xor_i128:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: eor x0, x0, x2
+; CHECK-GI-NEXT: eor x1, x1, x3
+; CHECK-GI-NEXT: ret
+entry:
+ %s = xor i128 %a, %b
+ ret i128 %s
+}
+
+define void @and_v2i8(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: and_v2i8:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: ld1 { v0.b }[0], [x0]
+; CHECK-SD-NEXT: ld1 { v1.b }[0], [x1]
+; CHECK-SD-NEXT: add x8, x0, #1
+; CHECK-SD-NEXT: add x9, x1, #1
+; CHECK-SD-NEXT: ld1 { v0.b }[4], [x8]
+; CHECK-SD-NEXT: ld1 { v1.b }[4], [x9]
+; CHECK-SD-NEXT: and v0.8b, v0.8b, v1.8b
+; CHECK-SD-NEXT: mov w8, v0.s[1]
+; CHECK-SD-NEXT: fmov w9, s0
+; CHECK-SD-NEXT: strb w9, [x0]
+; CHECK-SD-NEXT: strb w8, [x0, #1]
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: and_v2i8:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldr b0, [x0]
+; CHECK-GI-NEXT: ldr b1, [x0, #1]
+; CHECK-GI-NEXT: ldr b2, [x1]
+; CHECK-GI-NEXT: ldr b3, [x1, #1]
+; CHECK-GI-NEXT: mov v0.s[1], v1.s[0]
+; CHECK-GI-NEXT: mov v2.s[1], v3.s[0]
+; CHECK-GI-NEXT: and v0.8b, v0.8b, v2.8b
+; CHECK-GI-NEXT: mov s1, v0.s[1]
+; CHECK-GI-NEXT: str b0, [x0]
+; CHECK-GI-NEXT: str b1, [x0, #1]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <2 x i8>, ptr %p1
+ %e = load <2 x i8>, ptr %p2
+ %s = and <2 x i8> %d, %e
+ store <2 x i8> %s, ptr %p1
+ ret void
+}
+
+define void @or_v2i8(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: or_v2i8:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: ld1 { v0.b }[0], [x0]
+; CHECK-SD-NEXT: ld1 { v1.b }[0], [x1]
+; CHECK-SD-NEXT: add x8, x0, #1
+; CHECK-SD-NEXT: add x9, x1, #1
+; CHECK-SD-NEXT: ld1 { v0.b }[4], [x8]
+; CHECK-SD-NEXT: ld1 { v1.b }[4], [x9]
+; CHECK-SD-NEXT: orr v0.8b, v0.8b, v1.8b
+; CHECK-SD-NEXT: mov w8, v0.s[1]
+; CHECK-SD-NEXT: fmov w9, s0
+; CHECK-SD-NEXT: strb w9, [x0]
+; CHECK-SD-NEXT: strb w8, [x0, #1]
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: or_v2i8:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldr b0, [x0]
+; CHECK-GI-NEXT: ldr b1, [x0, #1]
+; CHECK-GI-NEXT: ldr b2, [x1]
+; CHECK-GI-NEXT: ldr b3, [x1, #1]
+; CHECK-GI-NEXT: mov v0.s[1], v1.s[0]
+; CHECK-GI-NEXT: mov v2.s[1], v3.s[0]
+; CHECK-GI-NEXT: orr v0.8b, v0.8b, v2.8b
+; CHECK-GI-NEXT: mov s1, v0.s[1]
+; CHECK-GI-NEXT: str b0, [x0]
+; CHECK-GI-NEXT: str b1, [x0, #1]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <2 x i8>, ptr %p1
+ %e = load <2 x i8>, ptr %p2
+ %s = or <2 x i8> %d, %e
+ store <2 x i8> %s, ptr %p1
+ ret void
+}
+
+define void @xor_v2i8(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: xor_v2i8:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: ld1 { v0.b }[0], [x0]
+; CHECK-SD-NEXT: ld1 { v1.b }[0], [x1]
+; CHECK-SD-NEXT: add x8, x0, #1
+; CHECK-SD-NEXT: add x9, x1, #1
+; CHECK-SD-NEXT: ld1 { v0.b }[4], [x8]
+; CHECK-SD-NEXT: ld1 { v1.b }[4], [x9]
+; CHECK-SD-NEXT: eor v0.8b, v0.8b, v1.8b
+; CHECK-SD-NEXT: mov w8, v0.s[1]
+; CHECK-SD-NEXT: fmov w9, s0
+; CHECK-SD-NEXT: strb w9, [x0]
+; CHECK-SD-NEXT: strb w8, [x0, #1]
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: xor_v2i8:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldr b0, [x0]
+; CHECK-GI-NEXT: ldr b1, [x0, #1]
+; CHECK-GI-NEXT: ldr b2, [x1]
+; CHECK-GI-NEXT: ldr b3, [x1, #1]
+; CHECK-GI-NEXT: mov v0.s[1], v1.s[0]
+; CHECK-GI-NEXT: mov v2.s[1], v3.s[0]
+; CHECK-GI-NEXT: eor v0.8b, v0.8b, v2.8b
+; CHECK-GI-NEXT: mov s1, v0.s[1]
+; CHECK-GI-NEXT: str b0, [x0]
+; CHECK-GI-NEXT: str b1, [x0, #1]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <2 x i8>, ptr %p1
+ %e = load <2 x i8>, ptr %p2
+ %s = xor <2 x i8> %d, %e
+ store <2 x i8> %s, ptr %p1
+ ret void
+}
+
+define void @and_v3i8(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: and_v3i8:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: sub sp, sp, #16
+; CHECK-SD-NEXT: .cfi_def_cfa_offset 16
+; CHECK-SD-NEXT: ldr s0, [x0]
+; CHECK-SD-NEXT: ldr s1, [x1]
+; CHECK-SD-NEXT: zip1 v0.8b, v0.8b, v0.8b
+; CHECK-SD-NEXT: zip1 v1.8b, v1.8b, v0.8b
+; CHECK-SD-NEXT: and v0.8b, v0.8b, v1.8b
+; CHECK-SD-NEXT: uzp1 v1.8b, v0.8b, v0.8b
+; CHECK-SD-NEXT: umov w8, v0.h[2]
+; CHECK-SD-NEXT: str s1, [sp, #12]
+; CHECK-SD-NEXT: ldrh w9, [sp, #12]
+; CHECK-SD-NEXT: strb w8, [x0, #2]
+; CHECK-SD-NEXT: strh w9, [x0]
+; CHECK-SD-NEXT: add sp, sp, #16
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: and_v3i8:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldrb w8, [x0]
+; CHECK-GI-NEXT: ldrb w9, [x0, #1]
+; CHECK-GI-NEXT: ldrb w10, [x1]
+; CHECK-GI-NEXT: ldrb w11, [x1, #1]
+; CHECK-GI-NEXT: fmov s0, w8
+; CHECK-GI-NEXT: fmov s1, w9
+; CHECK-GI-NEXT: fmov s2, w10
+; CHECK-GI-NEXT: fmov s3, w11
+; CHECK-GI-NEXT: ldrb w8, [x0, #2]
+; CHECK-GI-NEXT: ldrb w9, [x1, #2]
+; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
+; CHECK-GI-NEXT: fmov s1, w8
+; CHECK-GI-NEXT: mov v2.h[1], v3.h[0]
+; CHECK-GI-NEXT: fmov s3, w9
+; CHECK-GI-NEXT: mov v0.h[2], v1.h[0]
+; CHECK-GI-NEXT: mov v2.h[2], v3.h[0]
+; CHECK-GI-NEXT: and v0.8b, v0.8b, v2.8b
+; CHECK-GI-NEXT: mov h1, v0.h[1]
+; CHECK-GI-NEXT: mov h2, v0.h[2]
+; CHECK-GI-NEXT: str b0, [x0]
+; CHECK-GI-NEXT: str b1, [x0, #1]
+; CHECK-GI-NEXT: str b2, [x0, #2]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <3 x i8>, ptr %p1
+ %e = load <3 x i8>, ptr %p2
+ %s = and <3 x i8> %d, %e
+ store <3 x i8> %s, ptr %p1
+ ret void
+}
+
+define void @or_v3i8(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: or_v3i8:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: sub sp, sp, #16
+; CHECK-SD-NEXT: .cfi_def_cfa_offset 16
+; CHECK-SD-NEXT: ldr s0, [x0]
+; CHECK-SD-NEXT: ldr s1, [x1]
+; CHECK-SD-NEXT: zip1 v0.8b, v0.8b, v0.8b
+; CHECK-SD-NEXT: zip1 v1.8b, v1.8b, v0.8b
+; CHECK-SD-NEXT: orr v0.8b, v0.8b, v1.8b
+; CHECK-SD-NEXT: uzp1 v1.8b, v0.8b, v0.8b
+; CHECK-SD-NEXT: umov w8, v0.h[2]
+; CHECK-SD-NEXT: str s1, [sp, #12]
+; CHECK-SD-NEXT: ldrh w9, [sp, #12]
+; CHECK-SD-NEXT: strb w8, [x0, #2]
+; CHECK-SD-NEXT: strh w9, [x0]
+; CHECK-SD-NEXT: add sp, sp, #16
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: or_v3i8:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldrb w8, [x0]
+; CHECK-GI-NEXT: ldrb w9, [x0, #1]
+; CHECK-GI-NEXT: ldrb w10, [x1]
+; CHECK-GI-NEXT: ldrb w11, [x1, #1]
+; CHECK-GI-NEXT: fmov s0, w8
+; CHECK-GI-NEXT: fmov s1, w9
+; CHECK-GI-NEXT: fmov s2, w10
+; CHECK-GI-NEXT: fmov s3, w11
+; CHECK-GI-NEXT: ldrb w8, [x0, #2]
+; CHECK-GI-NEXT: ldrb w9, [x1, #2]
+; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
+; CHECK-GI-NEXT: fmov s1, w8
+; CHECK-GI-NEXT: mov v2.h[1], v3.h[0]
+; CHECK-GI-NEXT: fmov s3, w9
+; CHECK-GI-NEXT: mov v0.h[2], v1.h[0]
+; CHECK-GI-NEXT: mov v2.h[2], v3.h[0]
+; CHECK-GI-NEXT: orr v0.8b, v0.8b, v2.8b
+; CHECK-GI-NEXT: mov h1, v0.h[1]
+; CHECK-GI-NEXT: mov h2, v0.h[2]
+; CHECK-GI-NEXT: str b0, [x0]
+; CHECK-GI-NEXT: str b1, [x0, #1]
+; CHECK-GI-NEXT: str b2, [x0, #2]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <3 x i8>, ptr %p1
+ %e = load <3 x i8>, ptr %p2
+ %s = or <3 x i8> %d, %e
+ store <3 x i8> %s, ptr %p1
+ ret void
+}
+
+define void @xor_v3i8(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: xor_v3i8:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: sub sp, sp, #16
+; CHECK-SD-NEXT: .cfi_def_cfa_offset 16
+; CHECK-SD-NEXT: ldr s0, [x0]
+; CHECK-SD-NEXT: ldr s1, [x1]
+; CHECK-SD-NEXT: zip1 v0.8b, v0.8b, v0.8b
+; CHECK-SD-NEXT: zip1 v1.8b, v1.8b, v0.8b
+; CHECK-SD-NEXT: eor v0.8b, v0.8b, v1.8b
+; CHECK-SD-NEXT: uzp1 v1.8b, v0.8b, v0.8b
+; CHECK-SD-NEXT: umov w8, v0.h[2]
+; CHECK-SD-NEXT: str s1, [sp, #12]
+; CHECK-SD-NEXT: ldrh w9, [sp, #12]
+; CHECK-SD-NEXT: strb w8, [x0, #2]
+; CHECK-SD-NEXT: strh w9, [x0]
+; CHECK-SD-NEXT: add sp, sp, #16
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: xor_v3i8:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldrb w8, [x0]
+; CHECK-GI-NEXT: ldrb w9, [x0, #1]
+; CHECK-GI-NEXT: ldrb w10, [x1]
+; CHECK-GI-NEXT: ldrb w11, [x1, #1]
+; CHECK-GI-NEXT: fmov s0, w8
+; CHECK-GI-NEXT: fmov s1, w9
+; CHECK-GI-NEXT: fmov s2, w10
+; CHECK-GI-NEXT: fmov s3, w11
+; CHECK-GI-NEXT: ldrb w8, [x0, #2]
+; CHECK-GI-NEXT: ldrb w9, [x1, #2]
+; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
+; CHECK-GI-NEXT: fmov s1, w8
+; CHECK-GI-NEXT: mov v2.h[1], v3.h[0]
+; CHECK-GI-NEXT: fmov s3, w9
+; CHECK-GI-NEXT: mov v0.h[2], v1.h[0]
+; CHECK-GI-NEXT: mov v2.h[2], v3.h[0]
+; CHECK-GI-NEXT: eor v0.8b, v0.8b, v2.8b
+; CHECK-GI-NEXT: mov h1, v0.h[1]
+; CHECK-GI-NEXT: mov h2, v0.h[2]
+; CHECK-GI-NEXT: str b0, [x0]
+; CHECK-GI-NEXT: str b1, [x0, #1]
+; CHECK-GI-NEXT: str b2, [x0, #2]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <3 x i8>, ptr %p1
+ %e = load <3 x i8>, ptr %p2
+ %s = xor <3 x i8> %d, %e
+ store <3 x i8> %s, ptr %p1
+ ret void
+}
+
+define void @and_v4i8(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: and_v4i8:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: ldr s0, [x0]
+; CHECK-SD-NEXT: ldr s1, [x1]
+; CHECK-SD-NEXT: ushll v0.8h, v0.8b, #0
+; CHECK-SD-NEXT: ushll v1.8h, v1.8b, #0
+; CHECK-SD-NEXT: and v0.8b, v0.8b, v1.8b
+; CHECK-SD-NEXT: uzp1 v0.8b, v0.8b, v0.8b
+; CHECK-SD-NEXT: str s0, [x0]
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: and_v4i8:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldr w8, [x0]
+; CHECK-GI-NEXT: ldr w9, [x1]
+; CHECK-GI-NEXT: fmov s0, w8
+; CHECK-GI-NEXT: fmov s1, w9
+; CHECK-GI-NEXT: mov b2, v0.b[1]
+; CHECK-GI-NEXT: mov b3, v1.b[1]
+; CHECK-GI-NEXT: mov b4, v0.b[2]
+; CHECK-GI-NEXT: mov b5, v0.b[3]
+; CHECK-GI-NEXT: mov b6, v1.b[3]
+; CHECK-GI-NEXT: mov v0.b[1], v2.b[0]
+; CHECK-GI-NEXT: mov b2, v1.b[2]
+; CHECK-GI-NEXT: mov v1.b[1], v3.b[0]
+; CHECK-GI-NEXT: mov v0.b[2], v4.b[0]
+; CHECK-GI-NEXT: mov v1.b[2], v2.b[0]
+; CHECK-GI-NEXT: mov v0.b[3], v5.b[0]
+; CHECK-GI-NEXT: mov v1.b[3], v6.b[0]
+; CHECK-GI-NEXT: ushll v0.8h, v0.8b, #0
+; CHECK-GI-NEXT: ushll v1.8h, v1.8b, #0
+; CHECK-GI-NEXT: and v0.8b, v0.8b, v1.8b
+; CHECK-GI-NEXT: mov h1, v0.h[1]
+; CHECK-GI-NEXT: mov h2, v0.h[2]
+; CHECK-GI-NEXT: mov h3, v0.h[3]
+; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
+; CHECK-GI-NEXT: mov v0.h[2], v2.h[0]
+; CHECK-GI-NEXT: mov v0.h[3], v3.h[0]
+; CHECK-GI-NEXT: xtn v0.8b, v0.8h
+; CHECK-GI-NEXT: fmov w8, s0
+; CHECK-GI-NEXT: str w8, [x0]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <4 x i8>, ptr %p1
+ %e = load <4 x i8>, ptr %p2
+ %s = and <4 x i8> %d, %e
+ store <4 x i8> %s, ptr %p1
+ ret void
+}
+
+define void @or_v4i8(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: or_v4i8:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: ldr s0, [x0]
+; CHECK-SD-NEXT: ldr s1, [x1]
+; CHECK-SD-NEXT: ushll v0.8h, v0.8b, #0
+; CHECK-SD-NEXT: ushll v1.8h, v1.8b, #0
+; CHECK-SD-NEXT: orr v0.8b, v0.8b, v1.8b
+; CHECK-SD-NEXT: uzp1 v0.8b, v0.8b, v0.8b
+; CHECK-SD-NEXT: str s0, [x0]
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: or_v4i8:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldr w8, [x0]
+; CHECK-GI-NEXT: ldr w9, [x1]
+; CHECK-GI-NEXT: fmov s0, w8
+; CHECK-GI-NEXT: fmov s1, w9
+; CHECK-GI-NEXT: mov b2, v0.b[1]
+; CHECK-GI-NEXT: mov b3, v1.b[1]
+; CHECK-GI-NEXT: mov b4, v0.b[2]
+; CHECK-GI-NEXT: mov b5, v0.b[3]
+; CHECK-GI-NEXT: mov b6, v1.b[3]
+; CHECK-GI-NEXT: mov v0.b[1], v2.b[0]
+; CHECK-GI-NEXT: mov b2, v1.b[2]
+; CHECK-GI-NEXT: mov v1.b[1], v3.b[0]
+; CHECK-GI-NEXT: mov v0.b[2], v4.b[0]
+; CHECK-GI-NEXT: mov v1.b[2], v2.b[0]
+; CHECK-GI-NEXT: mov v0.b[3], v5.b[0]
+; CHECK-GI-NEXT: mov v1.b[3], v6.b[0]
+; CHECK-GI-NEXT: ushll v0.8h, v0.8b, #0
+; CHECK-GI-NEXT: ushll v1.8h, v1.8b, #0
+; CHECK-GI-NEXT: orr v0.8b, v0.8b, v1.8b
+; CHECK-GI-NEXT: mov h1, v0.h[1]
+; CHECK-GI-NEXT: mov h2, v0.h[2]
+; CHECK-GI-NEXT: mov h3, v0.h[3]
+; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
+; CHECK-GI-NEXT: mov v0.h[2], v2.h[0]
+; CHECK-GI-NEXT: mov v0.h[3], v3.h[0]
+; CHECK-GI-NEXT: xtn v0.8b, v0.8h
+; CHECK-GI-NEXT: fmov w8, s0
+; CHECK-GI-NEXT: str w8, [x0]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <4 x i8>, ptr %p1
+ %e = load <4 x i8>, ptr %p2
+ %s = or <4 x i8> %d, %e
+ store <4 x i8> %s, ptr %p1
+ ret void
+}
+
+define void @xor_v4i8(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: xor_v4i8:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: ldr s0, [x0]
+; CHECK-SD-NEXT: ldr s1, [x1]
+; CHECK-SD-NEXT: ushll v0.8h, v0.8b, #0
+; CHECK-SD-NEXT: ushll v1.8h, v1.8b, #0
+; CHECK-SD-NEXT: eor v0.8b, v0.8b, v1.8b
+; CHECK-SD-NEXT: uzp1 v0.8b, v0.8b, v0.8b
+; CHECK-SD-NEXT: str s0, [x0]
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: xor_v4i8:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldr w8, [x0]
+; CHECK-GI-NEXT: ldr w9, [x1]
+; CHECK-GI-NEXT: fmov s0, w8
+; CHECK-GI-NEXT: fmov s1, w9
+; CHECK-GI-NEXT: mov b2, v0.b[1]
+; CHECK-GI-NEXT: mov b3, v1.b[1]
+; CHECK-GI-NEXT: mov b4, v0.b[2]
+; CHECK-GI-NEXT: mov b5, v0.b[3]
+; CHECK-GI-NEXT: mov b6, v1.b[3]
+; CHECK-GI-NEXT: mov v0.b[1], v2.b[0]
+; CHECK-GI-NEXT: mov b2, v1.b[2]
+; CHECK-GI-NEXT: mov v1.b[1], v3.b[0]
+; CHECK-GI-NEXT: mov v0.b[2], v4.b[0]
+; CHECK-GI-NEXT: mov v1.b[2], v2.b[0]
+; CHECK-GI-NEXT: mov v0.b[3], v5.b[0]
+; CHECK-GI-NEXT: mov v1.b[3], v6.b[0]
+; CHECK-GI-NEXT: ushll v0.8h, v0.8b, #0
+; CHECK-GI-NEXT: ushll v1.8h, v1.8b, #0
+; CHECK-GI-NEXT: eor v0.8b, v0.8b, v1.8b
+; CHECK-GI-NEXT: mov h1, v0.h[1]
+; CHECK-GI-NEXT: mov h2, v0.h[2]
+; CHECK-GI-NEXT: mov h3, v0.h[3]
+; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
+; CHECK-GI-NEXT: mov v0.h[2], v2.h[0]
+; CHECK-GI-NEXT: mov v0.h[3], v3.h[0]
+; CHECK-GI-NEXT: xtn v0.8b, v0.8h
+; CHECK-GI-NEXT: fmov w8, s0
+; CHECK-GI-NEXT: str w8, [x0]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <4 x i8>, ptr %p1
+ %e = load <4 x i8>, ptr %p2
+ %s = xor <4 x i8> %d, %e
+ store <4 x i8> %s, ptr %p1
+ ret void
+}
+
+define <8 x i8> @and_v8i8(<8 x i8> %d, <8 x i8> %e) {
+; CHECK-LABEL: and_v8i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and v0.8b, v0.8b, v1.8b
+; CHECK-NEXT: ret
+entry:
+ %s = and <8 x i8> %d, %e
+ ret <8 x i8> %s
+}
+
+define <8 x i8> @or_v8i8(<8 x i8> %d, <8 x i8> %e) {
+; CHECK-LABEL: or_v8i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr v0.8b, v0.8b, v1.8b
+; CHECK-NEXT: ret
+entry:
+ %s = or <8 x i8> %d, %e
+ ret <8 x i8> %s
+}
+
+define <8 x i8> @xor_v8i8(<8 x i8> %d, <8 x i8> %e) {
+; CHECK-LABEL: xor_v8i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor v0.8b, v0.8b, v1.8b
+; CHECK-NEXT: ret
+entry:
+ %s = xor <8 x i8> %d, %e
+ ret <8 x i8> %s
+}
+
+define <16 x i8> @and_v16i8(<16 x i8> %d, <16 x i8> %e) {
+; CHECK-LABEL: and_v16i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = and <16 x i8> %d, %e
+ ret <16 x i8> %s
+}
+
+define <16 x i8> @or_v16i8(<16 x i8> %d, <16 x i8> %e) {
+; CHECK-LABEL: or_v16i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = or <16 x i8> %d, %e
+ ret <16 x i8> %s
+}
+
+define <16 x i8> @xor_v16i8(<16 x i8> %d, <16 x i8> %e) {
+; CHECK-LABEL: xor_v16i8:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = xor <16 x i8> %d, %e
+ ret <16 x i8> %s
+}
+
+define <32 x i8> @and_v32i8(<32 x i8> %d, <32 x i8> %e) {
+; CHECK-SD-LABEL: and_v32i8:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: and v1.16b, v1.16b, v3.16b
+; CHECK-SD-NEXT: and v0.16b, v0.16b, v2.16b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: and_v32i8:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: and v0.16b, v0.16b, v2.16b
+; CHECK-GI-NEXT: and v1.16b, v1.16b, v3.16b
+; CHECK-GI-NEXT: ret
+entry:
+ %s = and <32 x i8> %d, %e
+ ret <32 x i8> %s
+}
+
+define <32 x i8> @or_v32i8(<32 x i8> %d, <32 x i8> %e) {
+; CHECK-SD-LABEL: or_v32i8:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: orr v1.16b, v1.16b, v3.16b
+; CHECK-SD-NEXT: orr v0.16b, v0.16b, v2.16b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: or_v32i8:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: orr v0.16b, v0.16b, v2.16b
+; CHECK-GI-NEXT: orr v1.16b, v1.16b, v3.16b
+; CHECK-GI-NEXT: ret
+entry:
+ %s = or <32 x i8> %d, %e
+ ret <32 x i8> %s
+}
+
+define <32 x i8> @xor_v32i8(<32 x i8> %d, <32 x i8> %e) {
+; CHECK-SD-LABEL: xor_v32i8:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: eor v1.16b, v1.16b, v3.16b
+; CHECK-SD-NEXT: eor v0.16b, v0.16b, v2.16b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: xor_v32i8:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: eor v0.16b, v0.16b, v2.16b
+; CHECK-GI-NEXT: eor v1.16b, v1.16b, v3.16b
+; CHECK-GI-NEXT: ret
+entry:
+ %s = xor <32 x i8> %d, %e
+ ret <32 x i8> %s
+}
+
+define void @and_v2i16(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: and_v2i16:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: ld1 { v0.h }[0], [x0]
+; CHECK-SD-NEXT: ld1 { v1.h }[0], [x1]
+; CHECK-SD-NEXT: add x8, x0, #2
+; CHECK-SD-NEXT: add x9, x1, #2
+; CHECK-SD-NEXT: ld1 { v0.h }[2], [x8]
+; CHECK-SD-NEXT: ld1 { v1.h }[2], [x9]
+; CHECK-SD-NEXT: and v0.8b, v0.8b, v1.8b
+; CHECK-SD-NEXT: mov w8, v0.s[1]
+; CHECK-SD-NEXT: fmov w9, s0
+; CHECK-SD-NEXT: strh w9, [x0]
+; CHECK-SD-NEXT: strh w8, [x0, #2]
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: and_v2i16:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldr h0, [x0]
+; CHECK-GI-NEXT: ldr h1, [x0, #2]
+; CHECK-GI-NEXT: ldr h2, [x1]
+; CHECK-GI-NEXT: ldr h3, [x1, #2]
+; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
+; CHECK-GI-NEXT: mov v2.h[1], v3.h[0]
+; CHECK-GI-NEXT: ushll v0.4s, v0.4h, #0
+; CHECK-GI-NEXT: ushll v1.4s, v2.4h, #0
+; CHECK-GI-NEXT: and v0.8b, v0.8b, v1.8b
+; CHECK-GI-NEXT: mov s1, v0.s[1]
+; CHECK-GI-NEXT: str h0, [x0]
+; CHECK-GI-NEXT: str h1, [x0, #2]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <2 x i16>, ptr %p1
+ %e = load <2 x i16>, ptr %p2
+ %s = and <2 x i16> %d, %e
+ store <2 x i16> %s, ptr %p1
+ ret void
+}
+
+define void @or_v2i16(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: or_v2i16:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: ld1 { v0.h }[0], [x0]
+; CHECK-SD-NEXT: ld1 { v1.h }[0], [x1]
+; CHECK-SD-NEXT: add x8, x0, #2
+; CHECK-SD-NEXT: add x9, x1, #2
+; CHECK-SD-NEXT: ld1 { v0.h }[2], [x8]
+; CHECK-SD-NEXT: ld1 { v1.h }[2], [x9]
+; CHECK-SD-NEXT: orr v0.8b, v0.8b, v1.8b
+; CHECK-SD-NEXT: mov w8, v0.s[1]
+; CHECK-SD-NEXT: fmov w9, s0
+; CHECK-SD-NEXT: strh w9, [x0]
+; CHECK-SD-NEXT: strh w8, [x0, #2]
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: or_v2i16:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldr h0, [x0]
+; CHECK-GI-NEXT: ldr h1, [x0, #2]
+; CHECK-GI-NEXT: ldr h2, [x1]
+; CHECK-GI-NEXT: ldr h3, [x1, #2]
+; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
+; CHECK-GI-NEXT: mov v2.h[1], v3.h[0]
+; CHECK-GI-NEXT: ushll v0.4s, v0.4h, #0
+; CHECK-GI-NEXT: ushll v1.4s, v2.4h, #0
+; CHECK-GI-NEXT: orr v0.8b, v0.8b, v1.8b
+; CHECK-GI-NEXT: mov s1, v0.s[1]
+; CHECK-GI-NEXT: str h0, [x0]
+; CHECK-GI-NEXT: str h1, [x0, #2]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <2 x i16>, ptr %p1
+ %e = load <2 x i16>, ptr %p2
+ %s = or <2 x i16> %d, %e
+ store <2 x i16> %s, ptr %p1
+ ret void
+}
+
+define void @xor_v2i16(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: xor_v2i16:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: ld1 { v0.h }[0], [x0]
+; CHECK-SD-NEXT: ld1 { v1.h }[0], [x1]
+; CHECK-SD-NEXT: add x8, x0, #2
+; CHECK-SD-NEXT: add x9, x1, #2
+; CHECK-SD-NEXT: ld1 { v0.h }[2], [x8]
+; CHECK-SD-NEXT: ld1 { v1.h }[2], [x9]
+; CHECK-SD-NEXT: eor v0.8b, v0.8b, v1.8b
+; CHECK-SD-NEXT: mov w8, v0.s[1]
+; CHECK-SD-NEXT: fmov w9, s0
+; CHECK-SD-NEXT: strh w9, [x0]
+; CHECK-SD-NEXT: strh w8, [x0, #2]
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: xor_v2i16:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldr h0, [x0]
+; CHECK-GI-NEXT: ldr h1, [x0, #2]
+; CHECK-GI-NEXT: ldr h2, [x1]
+; CHECK-GI-NEXT: ldr h3, [x1, #2]
+; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
+; CHECK-GI-NEXT: mov v2.h[1], v3.h[0]
+; CHECK-GI-NEXT: ushll v0.4s, v0.4h, #0
+; CHECK-GI-NEXT: ushll v1.4s, v2.4h, #0
+; CHECK-GI-NEXT: eor v0.8b, v0.8b, v1.8b
+; CHECK-GI-NEXT: mov s1, v0.s[1]
+; CHECK-GI-NEXT: str h0, [x0]
+; CHECK-GI-NEXT: str h1, [x0, #2]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <2 x i16>, ptr %p1
+ %e = load <2 x i16>, ptr %p2
+ %s = xor <2 x i16> %d, %e
+ store <2 x i16> %s, ptr %p1
+ ret void
+}
+
+define void @and_v3i16(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: and_v3i16:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: ldr x8, [x0]
+; CHECK-SD-NEXT: ldr x9, [x1]
+; CHECK-SD-NEXT: and x8, x8, x9
+; CHECK-SD-NEXT: fmov d0, x8
+; CHECK-SD-NEXT: str w8, [x0]
+; CHECK-SD-NEXT: add x8, x0, #4
+; CHECK-SD-NEXT: st1 { v0.h }[2], [x8]
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: and_v3i16:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldr h0, [x0]
+; CHECK-GI-NEXT: ldr h1, [x0, #2]
+; CHECK-GI-NEXT: add x8, x0, #2
+; CHECK-GI-NEXT: ldr h2, [x1]
+; CHECK-GI-NEXT: ldr h3, [x1, #2]
+; CHECK-GI-NEXT: add x9, x0, #4
+; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
+; CHECK-GI-NEXT: ldr h1, [x0, #4]
+; CHECK-GI-NEXT: mov v2.h[1], v3.h[0]
+; CHECK-GI-NEXT: ldr h3, [x1, #4]
+; CHECK-GI-NEXT: mov v0.h[2], v1.h[0]
+; CHECK-GI-NEXT: mov v2.h[2], v3.h[0]
+; CHECK-GI-NEXT: and v0.8b, v0.8b, v2.8b
+; CHECK-GI-NEXT: str h0, [x0]
+; CHECK-GI-NEXT: st1 { v0.h }[1], [x8]
+; CHECK-GI-NEXT: st1 { v0.h }[2], [x9]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <3 x i16>, ptr %p1
+ %e = load <3 x i16>, ptr %p2
+ %s = and <3 x i16> %d, %e
+ store <3 x i16> %s, ptr %p1
+ ret void
+}
+
+define void @or_v3i16(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: or_v3i16:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: ldr x8, [x0]
+; CHECK-SD-NEXT: ldr x9, [x1]
+; CHECK-SD-NEXT: orr x8, x8, x9
+; CHECK-SD-NEXT: fmov d0, x8
+; CHECK-SD-NEXT: str w8, [x0]
+; CHECK-SD-NEXT: add x8, x0, #4
+; CHECK-SD-NEXT: st1 { v0.h }[2], [x8]
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: or_v3i16:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldr h0, [x0]
+; CHECK-GI-NEXT: ldr h1, [x0, #2]
+; CHECK-GI-NEXT: add x8, x0, #2
+; CHECK-GI-NEXT: ldr h2, [x1]
+; CHECK-GI-NEXT: ldr h3, [x1, #2]
+; CHECK-GI-NEXT: add x9, x0, #4
+; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
+; CHECK-GI-NEXT: ldr h1, [x0, #4]
+; CHECK-GI-NEXT: mov v2.h[1], v3.h[0]
+; CHECK-GI-NEXT: ldr h3, [x1, #4]
+; CHECK-GI-NEXT: mov v0.h[2], v1.h[0]
+; CHECK-GI-NEXT: mov v2.h[2], v3.h[0]
+; CHECK-GI-NEXT: orr v0.8b, v0.8b, v2.8b
+; CHECK-GI-NEXT: str h0, [x0]
+; CHECK-GI-NEXT: st1 { v0.h }[1], [x8]
+; CHECK-GI-NEXT: st1 { v0.h }[2], [x9]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <3 x i16>, ptr %p1
+ %e = load <3 x i16>, ptr %p2
+ %s = or <3 x i16> %d, %e
+ store <3 x i16> %s, ptr %p1
+ ret void
+}
+
+define void @xor_v3i16(ptr %p1, ptr %p2) {
+; CHECK-SD-LABEL: xor_v3i16:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: ldr x8, [x0]
+; CHECK-SD-NEXT: ldr x9, [x1]
+; CHECK-SD-NEXT: eor x8, x8, x9
+; CHECK-SD-NEXT: fmov d0, x8
+; CHECK-SD-NEXT: str w8, [x0]
+; CHECK-SD-NEXT: add x8, x0, #4
+; CHECK-SD-NEXT: st1 { v0.h }[2], [x8]
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: xor_v3i16:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: ldr h0, [x0]
+; CHECK-GI-NEXT: ldr h1, [x0, #2]
+; CHECK-GI-NEXT: add x8, x0, #2
+; CHECK-GI-NEXT: ldr h2, [x1]
+; CHECK-GI-NEXT: ldr h3, [x1, #2]
+; CHECK-GI-NEXT: add x9, x0, #4
+; CHECK-GI-NEXT: mov v0.h[1], v1.h[0]
+; CHECK-GI-NEXT: ldr h1, [x0, #4]
+; CHECK-GI-NEXT: mov v2.h[1], v3.h[0]
+; CHECK-GI-NEXT: ldr h3, [x1, #4]
+; CHECK-GI-NEXT: mov v0.h[2], v1.h[0]
+; CHECK-GI-NEXT: mov v2.h[2], v3.h[0]
+; CHECK-GI-NEXT: eor v0.8b, v0.8b, v2.8b
+; CHECK-GI-NEXT: str h0, [x0]
+; CHECK-GI-NEXT: st1 { v0.h }[1], [x8]
+; CHECK-GI-NEXT: st1 { v0.h }[2], [x9]
+; CHECK-GI-NEXT: ret
+entry:
+ %d = load <3 x i16>, ptr %p1
+ %e = load <3 x i16>, ptr %p2
+ %s = xor <3 x i16> %d, %e
+ store <3 x i16> %s, ptr %p1
+ ret void
+}
+
+define <4 x i16> @and_v4i16(<4 x i16> %d, <4 x i16> %e) {
+; CHECK-LABEL: and_v4i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and v0.8b, v0.8b, v1.8b
+; CHECK-NEXT: ret
+entry:
+ %s = and <4 x i16> %d, %e
+ ret <4 x i16> %s
+}
+
+define <4 x i16> @or_v4i16(<4 x i16> %d, <4 x i16> %e) {
+; CHECK-LABEL: or_v4i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr v0.8b, v0.8b, v1.8b
+; CHECK-NEXT: ret
+entry:
+ %s = or <4 x i16> %d, %e
+ ret <4 x i16> %s
+}
+
+define <4 x i16> @xor_v4i16(<4 x i16> %d, <4 x i16> %e) {
+; CHECK-LABEL: xor_v4i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor v0.8b, v0.8b, v1.8b
+; CHECK-NEXT: ret
+entry:
+ %s = xor <4 x i16> %d, %e
+ ret <4 x i16> %s
+}
+
+define <8 x i16> @and_v8i16(<8 x i16> %d, <8 x i16> %e) {
+; CHECK-LABEL: and_v8i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = and <8 x i16> %d, %e
+ ret <8 x i16> %s
+}
+
+define <8 x i16> @or_v8i16(<8 x i16> %d, <8 x i16> %e) {
+; CHECK-LABEL: or_v8i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = or <8 x i16> %d, %e
+ ret <8 x i16> %s
+}
+
+define <8 x i16> @xor_v8i16(<8 x i16> %d, <8 x i16> %e) {
+; CHECK-LABEL: xor_v8i16:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = xor <8 x i16> %d, %e
+ ret <8 x i16> %s
+}
+
+define <16 x i16> @and_v16i16(<16 x i16> %d, <16 x i16> %e) {
+; CHECK-SD-LABEL: and_v16i16:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: and v1.16b, v1.16b, v3.16b
+; CHECK-SD-NEXT: and v0.16b, v0.16b, v2.16b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: and_v16i16:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: and v0.16b, v0.16b, v2.16b
+; CHECK-GI-NEXT: and v1.16b, v1.16b, v3.16b
+; CHECK-GI-NEXT: ret
+entry:
+ %s = and <16 x i16> %d, %e
+ ret <16 x i16> %s
+}
+
+define <16 x i16> @or_v16i16(<16 x i16> %d, <16 x i16> %e) {
+; CHECK-SD-LABEL: or_v16i16:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: orr v1.16b, v1.16b, v3.16b
+; CHECK-SD-NEXT: orr v0.16b, v0.16b, v2.16b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: or_v16i16:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: orr v0.16b, v0.16b, v2.16b
+; CHECK-GI-NEXT: orr v1.16b, v1.16b, v3.16b
+; CHECK-GI-NEXT: ret
+entry:
+ %s = or <16 x i16> %d, %e
+ ret <16 x i16> %s
+}
+
+define <16 x i16> @xor_v16i16(<16 x i16> %d, <16 x i16> %e) {
+; CHECK-SD-LABEL: xor_v16i16:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: eor v1.16b, v1.16b, v3.16b
+; CHECK-SD-NEXT: eor v0.16b, v0.16b, v2.16b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: xor_v16i16:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: eor v0.16b, v0.16b, v2.16b
+; CHECK-GI-NEXT: eor v1.16b, v1.16b, v3.16b
+; CHECK-GI-NEXT: ret
+entry:
+ %s = xor <16 x i16> %d, %e
+ ret <16 x i16> %s
+}
+
+define <2 x i32> @and_v2i32(<2 x i32> %d, <2 x i32> %e) {
+; CHECK-LABEL: and_v2i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and v0.8b, v0.8b, v1.8b
+; CHECK-NEXT: ret
+entry:
+ %s = and <2 x i32> %d, %e
+ ret <2 x i32> %s
+}
+
+define <2 x i32> @or_v2i32(<2 x i32> %d, <2 x i32> %e) {
+; CHECK-LABEL: or_v2i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr v0.8b, v0.8b, v1.8b
+; CHECK-NEXT: ret
+entry:
+ %s = or <2 x i32> %d, %e
+ ret <2 x i32> %s
+}
+
+define <2 x i32> @xor_v2i32(<2 x i32> %d, <2 x i32> %e) {
+; CHECK-LABEL: xor_v2i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor v0.8b, v0.8b, v1.8b
+; CHECK-NEXT: ret
+entry:
+ %s = xor <2 x i32> %d, %e
+ ret <2 x i32> %s
+}
+
+define <3 x i32> @and_v3i32(<3 x i32> %d, <3 x i32> %e) {
+; CHECK-LABEL: and_v3i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = and <3 x i32> %d, %e
+ ret <3 x i32> %s
+}
+
+define <3 x i32> @or_v3i32(<3 x i32> %d, <3 x i32> %e) {
+; CHECK-LABEL: or_v3i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = or <3 x i32> %d, %e
+ ret <3 x i32> %s
+}
+
+define <3 x i32> @xor_v3i32(<3 x i32> %d, <3 x i32> %e) {
+; CHECK-LABEL: xor_v3i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = xor <3 x i32> %d, %e
+ ret <3 x i32> %s
+}
+
+define <4 x i32> @and_v4i32(<4 x i32> %d, <4 x i32> %e) {
+; CHECK-LABEL: and_v4i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = and <4 x i32> %d, %e
+ ret <4 x i32> %s
+}
+
+define <4 x i32> @or_v4i32(<4 x i32> %d, <4 x i32> %e) {
+; CHECK-LABEL: or_v4i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = or <4 x i32> %d, %e
+ ret <4 x i32> %s
+}
+
+define <4 x i32> @xor_v4i32(<4 x i32> %d, <4 x i32> %e) {
+; CHECK-LABEL: xor_v4i32:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = xor <4 x i32> %d, %e
+ ret <4 x i32> %s
+}
+
+define <8 x i32> @and_v8i32(<8 x i32> %d, <8 x i32> %e) {
+; CHECK-SD-LABEL: and_v8i32:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: and v1.16b, v1.16b, v3.16b
+; CHECK-SD-NEXT: and v0.16b, v0.16b, v2.16b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: and_v8i32:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: and v0.16b, v0.16b, v2.16b
+; CHECK-GI-NEXT: and v1.16b, v1.16b, v3.16b
+; CHECK-GI-NEXT: ret
+entry:
+ %s = and <8 x i32> %d, %e
+ ret <8 x i32> %s
+}
+
+define <8 x i32> @or_v8i32(<8 x i32> %d, <8 x i32> %e) {
+; CHECK-SD-LABEL: or_v8i32:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: orr v1.16b, v1.16b, v3.16b
+; CHECK-SD-NEXT: orr v0.16b, v0.16b, v2.16b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: or_v8i32:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: orr v0.16b, v0.16b, v2.16b
+; CHECK-GI-NEXT: orr v1.16b, v1.16b, v3.16b
+; CHECK-GI-NEXT: ret
+entry:
+ %s = or <8 x i32> %d, %e
+ ret <8 x i32> %s
+}
+
+define <8 x i32> @xor_v8i32(<8 x i32> %d, <8 x i32> %e) {
+; CHECK-SD-LABEL: xor_v8i32:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: eor v1.16b, v1.16b, v3.16b
+; CHECK-SD-NEXT: eor v0.16b, v0.16b, v2.16b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: xor_v8i32:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: eor v0.16b, v0.16b, v2.16b
+; CHECK-GI-NEXT: eor v1.16b, v1.16b, v3.16b
+; CHECK-GI-NEXT: ret
+entry:
+ %s = xor <8 x i32> %d, %e
+ ret <8 x i32> %s
+}
+
+define <2 x i64> @and_v2i64(<2 x i64> %d, <2 x i64> %e) {
+; CHECK-LABEL: and_v2i64:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = and <2 x i64> %d, %e
+ ret <2 x i64> %s
+}
+
+define <2 x i64> @or_v2i64(<2 x i64> %d, <2 x i64> %e) {
+; CHECK-LABEL: or_v2i64:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = or <2 x i64> %d, %e
+ ret <2 x i64> %s
+}
+
+define <2 x i64> @xor_v2i64(<2 x i64> %d, <2 x i64> %e) {
+; CHECK-LABEL: xor_v2i64:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor v0.16b, v0.16b, v1.16b
+; CHECK-NEXT: ret
+entry:
+ %s = xor <2 x i64> %d, %e
+ ret <2 x i64> %s
+}
+
+define <3 x i64> @and_v3i64(<3 x i64> %d, <3 x i64> %e) {
+; CHECK-SD-LABEL: and_v3i64:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: and v0.8b, v0.8b, v3.8b
+; CHECK-SD-NEXT: and v1.8b, v1.8b, v4.8b
+; CHECK-SD-NEXT: and v2.8b, v2.8b, v5.8b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: and_v3i64:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: // kill: def $d0 killed $d0 def $q0
+; CHECK-GI-NEXT: // kill: def $d3 killed $d3 def $q3
+; CHECK-GI-NEXT: // kill: def $d1 killed $d1 def $q1
+; CHECK-GI-NEXT: // kill: def $d4 killed $d4 def $q4
+; CHECK-GI-NEXT: fmov x8, d2
+; CHECK-GI-NEXT: fmov x9, d5
+; CHECK-GI-NEXT: mov v0.d[1], v1.d[0]
+; CHECK-GI-NEXT: mov v3.d[1], v4.d[0]
+; CHECK-GI-NEXT: and x8, x8, x9
+; CHECK-GI-NEXT: fmov d2, x8
+; CHECK-GI-NEXT: and v0.16b, v0.16b, v3.16b
+; CHECK-GI-NEXT: mov d1, v0.d[1]
+; CHECK-GI-NEXT: // kill: def $d0 killed $d0 killed $q0
+; CHECK-GI-NEXT: ret
+entry:
+ %s = and <3 x i64> %d, %e
+ ret <3 x i64> %s
+}
+
+define <3 x i64> @or_v3i64(<3 x i64> %d, <3 x i64> %e) {
+; CHECK-SD-LABEL: or_v3i64:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: orr v0.8b, v0.8b, v3.8b
+; CHECK-SD-NEXT: orr v1.8b, v1.8b, v4.8b
+; CHECK-SD-NEXT: orr v2.8b, v2.8b, v5.8b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: or_v3i64:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: // kill: def $d0 killed $d0 def $q0
+; CHECK-GI-NEXT: // kill: def $d3 killed $d3 def $q3
+; CHECK-GI-NEXT: // kill: def $d1 killed $d1 def $q1
+; CHECK-GI-NEXT: // kill: def $d4 killed $d4 def $q4
+; CHECK-GI-NEXT: fmov x8, d2
+; CHECK-GI-NEXT: fmov x9, d5
+; CHECK-GI-NEXT: mov v0.d[1], v1.d[0]
+; CHECK-GI-NEXT: mov v3.d[1], v4.d[0]
+; CHECK-GI-NEXT: orr x8, x8, x9
+; CHECK-GI-NEXT: fmov d2, x8
+; CHECK-GI-NEXT: orr v0.16b, v0.16b, v3.16b
+; CHECK-GI-NEXT: mov d1, v0.d[1]
+; CHECK-GI-NEXT: // kill: def $d0 killed $d0 killed $q0
+; CHECK-GI-NEXT: ret
+entry:
+ %s = or <3 x i64> %d, %e
+ ret <3 x i64> %s
+}
+
+define <3 x i64> @xor_v3i64(<3 x i64> %d, <3 x i64> %e) {
+; CHECK-SD-LABEL: xor_v3i64:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: eor v0.8b, v0.8b, v3.8b
+; CHECK-SD-NEXT: eor v1.8b, v1.8b, v4.8b
+; CHECK-SD-NEXT: eor v2.8b, v2.8b, v5.8b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: xor_v3i64:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: // kill: def $d0 killed $d0 def $q0
+; CHECK-GI-NEXT: // kill: def $d3 killed $d3 def $q3
+; CHECK-GI-NEXT: // kill: def $d1 killed $d1 def $q1
+; CHECK-GI-NEXT: // kill: def $d4 killed $d4 def $q4
+; CHECK-GI-NEXT: fmov x8, d2
+; CHECK-GI-NEXT: fmov x9, d5
+; CHECK-GI-NEXT: mov v0.d[1], v1.d[0]
+; CHECK-GI-NEXT: mov v3.d[1], v4.d[0]
+; CHECK-GI-NEXT: eor x8, x8, x9
+; CHECK-GI-NEXT: fmov d2, x8
+; CHECK-GI-NEXT: eor v0.16b, v0.16b, v3.16b
+; CHECK-GI-NEXT: mov d1, v0.d[1]
+; CHECK-GI-NEXT: // kill: def $d0 killed $d0 killed $q0
+; CHECK-GI-NEXT: ret
+entry:
+ %s = xor <3 x i64> %d, %e
+ ret <3 x i64> %s
+}
+
+define <4 x i64> @and_v4i64(<4 x i64> %d, <4 x i64> %e) {
+; CHECK-SD-LABEL: and_v4i64:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: and v1.16b, v1.16b, v3.16b
+; CHECK-SD-NEXT: and v0.16b, v0.16b, v2.16b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: and_v4i64:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: and v0.16b, v0.16b, v2.16b
+; CHECK-GI-NEXT: and v1.16b, v1.16b, v3.16b
+; CHECK-GI-NEXT: ret
+entry:
+ %s = and <4 x i64> %d, %e
+ ret <4 x i64> %s
+}
+
+define <4 x i64> @or_v4i64(<4 x i64> %d, <4 x i64> %e) {
+; CHECK-SD-LABEL: or_v4i64:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: orr v1.16b, v1.16b, v3.16b
+; CHECK-SD-NEXT: orr v0.16b, v0.16b, v2.16b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: or_v4i64:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: orr v0.16b, v0.16b, v2.16b
+; CHECK-GI-NEXT: orr v1.16b, v1.16b, v3.16b
+; CHECK-GI-NEXT: ret
+entry:
+ %s = or <4 x i64> %d, %e
+ ret <4 x i64> %s
+}
+
+define <4 x i64> @xor_v4i64(<4 x i64> %d, <4 x i64> %e) {
+; CHECK-SD-LABEL: xor_v4i64:
+; CHECK-SD: // %bb.0: // %entry
+; CHECK-SD-NEXT: eor v1.16b, v1.16b, v3.16b
+; CHECK-SD-NEXT: eor v0.16b, v0.16b, v2.16b
+; CHECK-SD-NEXT: ret
+;
+; CHECK-GI-LABEL: xor_v4i64:
+; CHECK-GI: // %bb.0: // %entry
+; CHECK-GI-NEXT: eor v0.16b, v0.16b, v2.16b
+; CHECK-GI-NEXT: eor v1.16b, v1.16b, v3.16b
+; CHECK-GI-NEXT: ret
+entry:
+ %s = xor <4 x i64> %d, %e
+ ret <4 x i64> %s
+}
+
+define <2 x i128> @and_v2i128(<2 x i128> %d, <2 x i128> %e) {
+; CHECK-LABEL: and_v2i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: and x2, x2, x6
+; CHECK-NEXT: and x0, x0, x4
+; CHECK-NEXT: and x1, x1, x5
+; CHECK-NEXT: and x3, x3, x7
+; CHECK-NEXT: ret
+entry:
+ %s = and <2 x i128> %d, %e
+ ret <2 x i128> %s
+}
+
+define <2 x i128> @or_v2i128(<2 x i128> %d, <2 x i128> %e) {
+; CHECK-LABEL: or_v2i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: orr x2, x2, x6
+; CHECK-NEXT: orr x0, x0, x4
+; CHECK-NEXT: orr x1, x1, x5
+; CHECK-NEXT: orr x3, x3, x7
+; CHECK-NEXT: ret
+entry:
+ %s = or <2 x i128> %d, %e
+ ret <2 x i128> %s
+}
+
+define <2 x i128> @xor_v2i128(<2 x i128> %d, <2 x i128> %e) {
+; CHECK-LABEL: xor_v2i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: eor x2, x2, x6
+; CHECK-NEXT: eor x0, x0, x4
+; CHECK-NEXT: eor x1, x1, x5
+; CHECK-NEXT: eor x3, x3, x7
+; CHECK-NEXT: ret
+entry:
+ %s = xor <2 x i128> %d, %e
+ ret <2 x i128> %s
+}
+
+define <3 x i128> @and_v3i128(<3 x i128> %d, <3 x i128> %e) {
+; CHECK-LABEL: and_v3i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: ldp x8, x9, [sp]
+; CHECK-NEXT: and x0, x0, x6
+; CHECK-NEXT: ldp x11, x10, [sp, #16]
+; CHECK-NEXT: and x1, x1, x7
+; CHECK-NEXT: and x2, x2, x8
+; CHECK-NEXT: and x3, x3, x9
+; CHECK-NEXT: and x4, x4, x11
+; CHECK-NEXT: and x5, x5, x10
+; CHECK-NEXT: ret
+entry:
+ %s = and <3 x i128> %d, %e
+ ret <3 x i128> %s
+}
+
+define <3 x i128> @or_v3i128(<3 x i128> %d, <3 x i128> %e) {
+; CHECK-LABEL: or_v3i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: ldp x8, x9, [sp]
+; CHECK-NEXT: orr x0, x0, x6
+; CHECK-NEXT: ldp x11, x10, [sp, #16]
+; CHECK-NEXT: orr x1, x1, x7
+; CHECK-NEXT: orr x2, x2, x8
+; CHECK-NEXT: orr x3, x3, x9
+; CHECK-NEXT: orr x4, x4, x11
+; CHECK-NEXT: orr x5, x5, x10
+; CHECK-NEXT: ret
+entry:
+ %s = or <3 x i128> %d, %e
+ ret <3 x i128> %s
+}
+
+define <3 x i128> @xor_v3i128(<3 x i128> %d, <3 x i128> %e) {
+; CHECK-LABEL: xor_v3i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: ldp x8, x9, [sp]
+; CHECK-NEXT: eor x0, x0, x6
+; CHECK-NEXT: ldp x11, x10, [sp, #16]
+; CHECK-NEXT: eor x1, x1, x7
+; CHECK-NEXT: eor x2, x2, x8
+; CHECK-NEXT: eor x3, x3, x9
+; CHECK-NEXT: eor x4, x4, x11
+; CHECK-NEXT: eor x5, x5, x10
+; CHECK-NEXT: ret
+entry:
+ %s = xor <3 x i128> %d, %e
+ ret <3 x i128> %s
+}
+
+define <4 x i128> @and_v4i128(<4 x i128> %d, <4 x i128> %e) {
+; CHECK-LABEL: and_v4i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: ldp x9, x8, [sp, #32]
+; CHECK-NEXT: ldp x11, x10, [sp]
+; CHECK-NEXT: ldp x13, x12, [sp, #16]
+; CHECK-NEXT: ldp x15, x14, [sp, #48]
+; CHECK-NEXT: and x4, x4, x9
+; CHECK-NEXT: and x0, x0, x11
+; CHECK-NEXT: and x1, x1, x10
+; CHECK-NEXT: and x5, x5, x8
+; CHECK-NEXT: and x2, x2, x13
+; CHECK-NEXT: and x3, x3, x12
+; CHECK-NEXT: and x6, x6, x15
+; CHECK-NEXT: and x7, x7, x14
+; CHECK-NEXT: ret
+entry:
+ %s = and <4 x i128> %d, %e
+ ret <4 x i128> %s
+}
+
+define <4 x i128> @or_v4i128(<4 x i128> %d, <4 x i128> %e) {
+; CHECK-LABEL: or_v4i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: ldp x9, x8, [sp, #32]
+; CHECK-NEXT: ldp x11, x10, [sp]
+; CHECK-NEXT: ldp x13, x12, [sp, #16]
+; CHECK-NEXT: ldp x15, x14, [sp, #48]
+; CHECK-NEXT: orr x4, x4, x9
+; CHECK-NEXT: orr x0, x0, x11
+; CHECK-NEXT: orr x1, x1, x10
+; CHECK-NEXT: orr x5, x5, x8
+; CHECK-NEXT: orr x2, x2, x13
+; CHECK-NEXT: orr x3, x3, x12
+; CHECK-NEXT: orr x6, x6, x15
+; CHECK-NEXT: orr x7, x7, x14
+; CHECK-NEXT: ret
+entry:
+ %s = or <4 x i128> %d, %e
+ ret <4 x i128> %s
+}
+
+define <4 x i128> @xor_v4i128(<4 x i128> %d, <4 x i128> %e) {
+; CHECK-LABEL: xor_v4i128:
+; CHECK: // %bb.0: // %entry
+; CHECK-NEXT: ldp x9, x8, [sp, #32]
+; CHECK-NEXT: ldp x11, x10, [sp]
+; CHECK-NEXT: ldp x13, x12, [sp, #16]
+; CHECK-NEXT: ldp x15, x14, [sp, #48]
+; CHECK-NEXT: eor x4, x4, x9
+; CHECK-NEXT: eor x0, x0, x11
+; CHECK-NEXT: eor x1, x1, x10
+; CHECK-NEXT: eor x5, x5, x8
+; CHECK-NEXT: eor x2, x2, x13
+; CHECK-NEXT: eor x3, x3, x12
+; CHECK-NEXT: eor x6, x6, x15
+; CHECK-NEXT: eor x7, x7, x14
+; CHECK-NEXT: ret
+entry:
+ %s = xor <4 x i128> %d, %e
+ ret <4 x i128> %s
+}
More information about the llvm-commits
mailing list