llvm/test/CodeGen/AArch64/aarch64-fold-lslfast.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=aarch64-linux-gnu -mattr=+lsl-fast | FileCheck %s

%struct.a = type [256 x i16]
%struct.b = type [256 x i32]
%struct.c = type [256 x i64]

declare void @foo()
define i16 @halfword(%struct.a* %ctx, i32 %xor72) nounwind {
; CHECK-LABEL: halfword:
; CHECK:       // %bb.0:
; CHECK-NEXT:    stp x30, x21, [sp, #-32]! // 16-byte Folded Spill
; CHECK-NEXT:    // kill: def $w1 killed $w1 def $x1
; CHECK-NEXT:    ubfx x21, x1, #9, #8
; CHECK-NEXT:    stp x20, x19, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT:    mov x19, x0
; CHECK-NEXT:    ldrh w20, [x0, x21, lsl #1]
; CHECK-NEXT:    bl foo
; CHECK-NEXT:    mov w0, w20
; CHECK-NEXT:    strh w20, [x19, x21, lsl #1]
; CHECK-NEXT:    ldp x20, x19, [sp, #16] // 16-byte Folded Reload
; CHECK-NEXT:    ldp x30, x21, [sp], #32 // 16-byte Folded Reload
; CHECK-NEXT:    ret
  %shr81 = lshr i32 %xor72, 9
  %conv82 = zext i32 %shr81 to i64
  %idxprom83 = and i64 %conv82, 255
  %arrayidx86 = getelementptr inbounds %struct.a, %struct.a* %ctx, i64 0, i64 %idxprom83
  %result = load i16, i16* %arrayidx86, align 2
  call void @foo()
  store i16 %result, i16* %arrayidx86, align 2
  ret i16 %result
}

define i32 @word(%struct.b* %ctx, i32 %xor72) nounwind {
; CHECK-LABEL: word:
; CHECK:       // %bb.0:
; CHECK-NEXT:    stp x30, x21, [sp, #-32]! // 16-byte Folded Spill
; CHECK-NEXT:    // kill: def $w1 killed $w1 def $x1
; CHECK-NEXT:    ubfx x21, x1, #9, #8
; CHECK-NEXT:    stp x20, x19, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT:    mov x19, x0
; CHECK-NEXT:    ldr w20, [x0, x21, lsl #2]
; CHECK-NEXT:    bl foo
; CHECK-NEXT:    mov w0, w20
; CHECK-NEXT:    str w20, [x19, x21, lsl #2]
; CHECK-NEXT:    ldp x20, x19, [sp, #16] // 16-byte Folded Reload
; CHECK-NEXT:    ldp x30, x21, [sp], #32 // 16-byte Folded Reload
; CHECK-NEXT:    ret
  %shr81 = lshr i32 %xor72, 9
  %conv82 = zext i32 %shr81 to i64
  %idxprom83 = and i64 %conv82, 255
  %arrayidx86 = getelementptr inbounds %struct.b, %struct.b* %ctx, i64 0, i64 %idxprom83
  %result = load i32, i32* %arrayidx86, align 4
  call void @foo()
  store i32 %result, i32* %arrayidx86, align 4
  ret i32 %result
}

define i64 @doubleword(%struct.c* %ctx, i32 %xor72) nounwind {
; CHECK-LABEL: doubleword:
; CHECK:       // %bb.0:
; CHECK-NEXT:    stp x30, x21, [sp, #-32]! // 16-byte Folded Spill
; CHECK-NEXT:    // kill: def $w1 killed $w1 def $x1
; CHECK-NEXT:    ubfx x21, x1, #9, #8
; CHECK-NEXT:    stp x20, x19, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT:    mov x19, x0
; CHECK-NEXT:    ldr x20, [x0, x21, lsl #3]
; CHECK-NEXT:    bl foo
; CHECK-NEXT:    mov x0, x20
; CHECK-NEXT:    str x20, [x19, x21, lsl #3]
; CHECK-NEXT:    ldp x20, x19, [sp, #16] // 16-byte Folded Reload
; CHECK-NEXT:    ldp x30, x21, [sp], #32 // 16-byte Folded Reload
; CHECK-NEXT:    ret
  %shr81 = lshr i32 %xor72, 9
  %conv82 = zext i32 %shr81 to i64
  %idxprom83 = and i64 %conv82, 255
  %arrayidx86 = getelementptr inbounds %struct.c, %struct.c* %ctx, i64 0, i64 %idxprom83
  %result = load i64, i64* %arrayidx86, align 8
  call void @foo()
  store i64 %result, i64* %arrayidx86, align 8
  ret i64 %result
}

define i64 @multi_use_non_memory(i64 %a, i64 %b) {
; CHECK-LABEL: multi_use_non_memory:
; CHECK:       // %bb.0: // %entry
; CHECK-NEXT:    lsl x8, x0, #3
; CHECK-NEXT:    lsl x9, x1, #3
; CHECK-NEXT:    cmp x8, x9
; CHECK-NEXT:    b.lt .LBB3_2
; CHECK-NEXT:  // %bb.1: // %falsebb
; CHECK-NEXT:    csel x0, x8, x9, gt
; CHECK-NEXT:    ret
; CHECK-NEXT:  .LBB3_2: // %truebb
; CHECK-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
; CHECK-NEXT:    .cfi_def_cfa_offset 16
; CHECK-NEXT:    .cfi_offset w30, -16
; CHECK-NEXT:    bl foo
entry:
  %mul1 = shl i64 %a, 3
  %mul2 = shl i64 %b, 3
  %cmp = icmp slt i64 %mul1, %mul2
  br i1 %cmp, label %truebb, label %falsebb
truebb:
  tail call void @foo()
  unreachable
falsebb:
  %cmp2 = icmp sgt i64 %mul1, %mul2
  br i1 %cmp2, label %exitbb, label %endbb
exitbb:
 ret i64 %mul1
endbb:
 ret i64 %mul2
}