
This reverts commit 9c319d5bb40785c969d2af76535ca62448dfafa7. Some issues were discovered with the bootstrap builds, which seem like they were caused by this commit. I'm reverting to investigate.
561 lines
17 KiB
LLVM
561 lines
17 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc < %s -mtriple=aarch64-unknown-linux-gnu -mattr=+neon | FileCheck %s
|
|
; RUN: llc < %s -mtriple=aarch64-none-linux-gnu -mattr=+neon -global-isel -global-isel-abort=1 | FileCheck %s --check-prefix=GISEL
|
|
|
|
define i1 @test_redxor_v1i1(<1 x i1> %a) {
|
|
; CHECK-LABEL: test_redxor_v1i1:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: and w0, w0, #0x1
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v1i1:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: and w0, w0, #0x1
|
|
; GISEL-NEXT: ret
|
|
%or_result = call i1 @llvm.vector.reduce.xor.v1i1(<1 x i1> %a)
|
|
ret i1 %or_result
|
|
}
|
|
|
|
define i1 @test_redxor_v2i1(<2 x i1> %a) {
|
|
; CHECK-LABEL: test_redxor_v2i1:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: addp v0.2s, v0.2s, v0.2s
|
|
; CHECK-NEXT: fmov w8, s0
|
|
; CHECK-NEXT: and w0, w8, #0x1
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v2i1:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: // kill: def $d0 killed $d0 def $q0
|
|
; GISEL-NEXT: mov w8, v0.s[1]
|
|
; GISEL-NEXT: fmov w9, s0
|
|
; GISEL-NEXT: eor w8, w9, w8
|
|
; GISEL-NEXT: and w0, w8, #0x1
|
|
; GISEL-NEXT: ret
|
|
%or_result = call i1 @llvm.vector.reduce.xor.v2i1(<2 x i1> %a)
|
|
ret i1 %or_result
|
|
}
|
|
|
|
define i1 @test_redxor_v4i1(<4 x i1> %a) {
|
|
; CHECK-LABEL: test_redxor_v4i1:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: addv h0, v0.4h
|
|
; CHECK-NEXT: fmov w8, s0
|
|
; CHECK-NEXT: and w0, w8, #0x1
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v4i1:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: // kill: def $d0 killed $d0 def $q0
|
|
; GISEL-NEXT: umov w8, v0.h[0]
|
|
; GISEL-NEXT: umov w9, v0.h[1]
|
|
; GISEL-NEXT: umov w10, v0.h[2]
|
|
; GISEL-NEXT: umov w11, v0.h[3]
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: and w0, w8, #0x1
|
|
; GISEL-NEXT: ret
|
|
%or_result = call i1 @llvm.vector.reduce.xor.v4i1(<4 x i1> %a)
|
|
ret i1 %or_result
|
|
}
|
|
|
|
define i1 @test_redxor_v8i1(<8 x i1> %a) {
|
|
; CHECK-LABEL: test_redxor_v8i1:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: addv b0, v0.8b
|
|
; CHECK-NEXT: fmov w8, s0
|
|
; CHECK-NEXT: and w0, w8, #0x1
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v8i1:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: // kill: def $d0 killed $d0 def $q0
|
|
; GISEL-NEXT: umov w8, v0.b[0]
|
|
; GISEL-NEXT: umov w9, v0.b[1]
|
|
; GISEL-NEXT: umov w10, v0.b[2]
|
|
; GISEL-NEXT: umov w11, v0.b[3]
|
|
; GISEL-NEXT: umov w12, v0.b[4]
|
|
; GISEL-NEXT: umov w13, v0.b[5]
|
|
; GISEL-NEXT: umov w14, v0.b[6]
|
|
; GISEL-NEXT: umov w15, v0.b[7]
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w10, w12, w13
|
|
; GISEL-NEXT: eor w11, w14, w15
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: and w0, w8, #0x1
|
|
; GISEL-NEXT: ret
|
|
%or_result = call i1 @llvm.vector.reduce.xor.v8i1(<8 x i1> %a)
|
|
ret i1 %or_result
|
|
}
|
|
|
|
define i1 @test_redxor_v16i1(<16 x i1> %a) {
|
|
; CHECK-LABEL: test_redxor_v16i1:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: addv b0, v0.16b
|
|
; CHECK-NEXT: fmov w8, s0
|
|
; CHECK-NEXT: and w0, w8, #0x1
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v16i1:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: umov w8, v0.b[0]
|
|
; GISEL-NEXT: umov w9, v0.b[1]
|
|
; GISEL-NEXT: umov w10, v0.b[2]
|
|
; GISEL-NEXT: umov w11, v0.b[3]
|
|
; GISEL-NEXT: umov w12, v0.b[4]
|
|
; GISEL-NEXT: umov w13, v0.b[5]
|
|
; GISEL-NEXT: umov w14, v0.b[6]
|
|
; GISEL-NEXT: umov w15, v0.b[7]
|
|
; GISEL-NEXT: umov w16, v0.b[8]
|
|
; GISEL-NEXT: umov w17, v0.b[9]
|
|
; GISEL-NEXT: umov w18, v0.b[10]
|
|
; GISEL-NEXT: umov w0, v0.b[11]
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: umov w1, v0.b[12]
|
|
; GISEL-NEXT: umov w2, v0.b[13]
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w10, w12, w13
|
|
; GISEL-NEXT: umov w3, v0.b[14]
|
|
; GISEL-NEXT: eor w11, w14, w15
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: umov w4, v0.b[15]
|
|
; GISEL-NEXT: eor w12, w16, w17
|
|
; GISEL-NEXT: eor w13, w18, w0
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w14, w1, w2
|
|
; GISEL-NEXT: eor w10, w12, w13
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w15, w3, w4
|
|
; GISEL-NEXT: eor w11, w14, w15
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: and w0, w8, #0x1
|
|
; GISEL-NEXT: ret
|
|
%or_result = call i1 @llvm.vector.reduce.xor.v16i1(<16 x i1> %a)
|
|
ret i1 %or_result
|
|
}
|
|
|
|
define <16 x i1> @test_redxor_ins_v16i1(<16 x i1> %a) {
|
|
; CHECK-LABEL: test_redxor_ins_v16i1:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: addv b0, v0.16b
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_ins_v16i1:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: umov w8, v0.b[0]
|
|
; GISEL-NEXT: umov w9, v0.b[1]
|
|
; GISEL-NEXT: umov w10, v0.b[2]
|
|
; GISEL-NEXT: umov w11, v0.b[3]
|
|
; GISEL-NEXT: umov w12, v0.b[4]
|
|
; GISEL-NEXT: umov w13, v0.b[5]
|
|
; GISEL-NEXT: umov w14, v0.b[6]
|
|
; GISEL-NEXT: umov w15, v0.b[7]
|
|
; GISEL-NEXT: umov w16, v0.b[8]
|
|
; GISEL-NEXT: umov w17, v0.b[9]
|
|
; GISEL-NEXT: umov w18, v0.b[10]
|
|
; GISEL-NEXT: umov w0, v0.b[11]
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: umov w1, v0.b[12]
|
|
; GISEL-NEXT: umov w2, v0.b[13]
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w10, w12, w13
|
|
; GISEL-NEXT: umov w3, v0.b[14]
|
|
; GISEL-NEXT: eor w11, w14, w15
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: umov w4, v0.b[15]
|
|
; GISEL-NEXT: eor w12, w16, w17
|
|
; GISEL-NEXT: eor w13, w18, w0
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w14, w1, w2
|
|
; GISEL-NEXT: eor w10, w12, w13
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w15, w3, w4
|
|
; GISEL-NEXT: eor w11, w14, w15
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: fmov s0, w8
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i1 @llvm.vector.reduce.xor.v16i1(<16 x i1> %a)
|
|
%ins = insertelement <16 x i1> poison, i1 %xor_result, i64 0
|
|
ret <16 x i1> %ins
|
|
}
|
|
|
|
define i8 @test_redxor_v1i8(<1 x i8> %a) {
|
|
; CHECK-LABEL: test_redxor_v1i8:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: // kill: def $d0 killed $d0 def $q0
|
|
; CHECK-NEXT: umov w0, v0.b[0]
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v1i8:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: // kill: def $d0 killed $d0 def $q0
|
|
; GISEL-NEXT: umov w0, v0.b[0]
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i8 @llvm.vector.reduce.xor.v1i8(<1 x i8> %a)
|
|
ret i8 %xor_result
|
|
}
|
|
|
|
define i8 @test_redxor_v3i8(<3 x i8> %a) {
|
|
; CHECK-LABEL: test_redxor_v3i8:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: movi v0.2d, #0000000000000000
|
|
; CHECK-NEXT: mov v0.h[0], w0
|
|
; CHECK-NEXT: mov v0.h[1], w1
|
|
; CHECK-NEXT: fmov x8, d0
|
|
; CHECK-NEXT: mov v0.h[2], w2
|
|
; CHECK-NEXT: fmov x9, d0
|
|
; CHECK-NEXT: lsr x10, x9, #32
|
|
; CHECK-NEXT: lsr x9, x9, #16
|
|
; CHECK-NEXT: eor w8, w8, w10
|
|
; CHECK-NEXT: eor w0, w8, w9
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v3i8:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: eor w8, w0, w1
|
|
; GISEL-NEXT: eor w0, w8, w2
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i8 @llvm.vector.reduce.xor.v3i8(<3 x i8> %a)
|
|
ret i8 %xor_result
|
|
}
|
|
|
|
define i8 @test_redxor_v4i8(<4 x i8> %a) {
|
|
; CHECK-LABEL: test_redxor_v4i8:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: fmov x8, d0
|
|
; CHECK-NEXT: eor x8, x8, x8, lsr #32
|
|
; CHECK-NEXT: lsr x9, x8, #16
|
|
; CHECK-NEXT: eor w0, w8, w9
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v4i8:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: // kill: def $d0 killed $d0 def $q0
|
|
; GISEL-NEXT: umov w8, v0.h[0]
|
|
; GISEL-NEXT: umov w9, v0.h[1]
|
|
; GISEL-NEXT: umov w10, v0.h[2]
|
|
; GISEL-NEXT: umov w11, v0.h[3]
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w0, w8, w9
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i8 @llvm.vector.reduce.xor.v4i8(<4 x i8> %a)
|
|
ret i8 %xor_result
|
|
}
|
|
|
|
define i8 @test_redxor_v8i8(<8 x i8> %a) {
|
|
; CHECK-LABEL: test_redxor_v8i8:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: fmov x8, d0
|
|
; CHECK-NEXT: eor x8, x8, x8, lsr #32
|
|
; CHECK-NEXT: eor x8, x8, x8, lsr #16
|
|
; CHECK-NEXT: lsr x9, x8, #8
|
|
; CHECK-NEXT: eor w0, w8, w9
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v8i8:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: // kill: def $d0 killed $d0 def $q0
|
|
; GISEL-NEXT: umov w8, v0.b[0]
|
|
; GISEL-NEXT: umov w9, v0.b[1]
|
|
; GISEL-NEXT: umov w10, v0.b[2]
|
|
; GISEL-NEXT: umov w11, v0.b[3]
|
|
; GISEL-NEXT: umov w12, v0.b[4]
|
|
; GISEL-NEXT: umov w13, v0.b[5]
|
|
; GISEL-NEXT: umov w14, v0.b[6]
|
|
; GISEL-NEXT: umov w15, v0.b[7]
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w10, w12, w13
|
|
; GISEL-NEXT: eor w11, w14, w15
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w0, w8, w9
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i8 @llvm.vector.reduce.xor.v8i8(<8 x i8> %a)
|
|
ret i8 %xor_result
|
|
}
|
|
|
|
define i8 @test_redxor_v16i8(<16 x i8> %a) {
|
|
; CHECK-LABEL: test_redxor_v16i8:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: ext v1.16b, v0.16b, v0.16b, #8
|
|
; CHECK-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; CHECK-NEXT: fmov x8, d0
|
|
; CHECK-NEXT: eor x8, x8, x8, lsr #32
|
|
; CHECK-NEXT: eor x8, x8, x8, lsr #16
|
|
; CHECK-NEXT: lsr x9, x8, #8
|
|
; CHECK-NEXT: eor w0, w8, w9
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v16i8:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: mov d1, v0.d[1]
|
|
; GISEL-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; GISEL-NEXT: umov w8, v0.b[0]
|
|
; GISEL-NEXT: umov w9, v0.b[1]
|
|
; GISEL-NEXT: umov w10, v0.b[2]
|
|
; GISEL-NEXT: umov w11, v0.b[3]
|
|
; GISEL-NEXT: umov w12, v0.b[4]
|
|
; GISEL-NEXT: umov w13, v0.b[5]
|
|
; GISEL-NEXT: umov w14, v0.b[6]
|
|
; GISEL-NEXT: umov w15, v0.b[7]
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w10, w12, w13
|
|
; GISEL-NEXT: eor w11, w14, w15
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w0, w8, w9
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i8 @llvm.vector.reduce.xor.v16i8(<16 x i8> %a)
|
|
ret i8 %xor_result
|
|
}
|
|
|
|
define i8 @test_redxor_v32i8(<32 x i8> %a) {
|
|
; CHECK-LABEL: test_redxor_v32i8:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: eor v0.16b, v0.16b, v1.16b
|
|
; CHECK-NEXT: ext v1.16b, v0.16b, v0.16b, #8
|
|
; CHECK-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; CHECK-NEXT: fmov x8, d0
|
|
; CHECK-NEXT: eor x8, x8, x8, lsr #32
|
|
; CHECK-NEXT: eor x8, x8, x8, lsr #16
|
|
; CHECK-NEXT: lsr x9, x8, #8
|
|
; CHECK-NEXT: eor w0, w8, w9
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v32i8:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: eor v0.16b, v0.16b, v1.16b
|
|
; GISEL-NEXT: mov d1, v0.d[1]
|
|
; GISEL-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; GISEL-NEXT: umov w8, v0.b[0]
|
|
; GISEL-NEXT: umov w9, v0.b[1]
|
|
; GISEL-NEXT: umov w10, v0.b[2]
|
|
; GISEL-NEXT: umov w11, v0.b[3]
|
|
; GISEL-NEXT: umov w12, v0.b[4]
|
|
; GISEL-NEXT: umov w13, v0.b[5]
|
|
; GISEL-NEXT: umov w14, v0.b[6]
|
|
; GISEL-NEXT: umov w15, v0.b[7]
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w10, w12, w13
|
|
; GISEL-NEXT: eor w11, w14, w15
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w0, w8, w9
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i8 @llvm.vector.reduce.xor.v32i8(<32 x i8> %a)
|
|
ret i8 %xor_result
|
|
}
|
|
|
|
define i16 @test_redxor_v4i16(<4 x i16> %a) {
|
|
; CHECK-LABEL: test_redxor_v4i16:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: fmov x8, d0
|
|
; CHECK-NEXT: eor x8, x8, x8, lsr #32
|
|
; CHECK-NEXT: lsr x9, x8, #16
|
|
; CHECK-NEXT: eor w0, w8, w9
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v4i16:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: // kill: def $d0 killed $d0 def $q0
|
|
; GISEL-NEXT: umov w8, v0.h[0]
|
|
; GISEL-NEXT: umov w9, v0.h[1]
|
|
; GISEL-NEXT: umov w10, v0.h[2]
|
|
; GISEL-NEXT: umov w11, v0.h[3]
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w0, w8, w9
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i16 @llvm.vector.reduce.xor.v4i16(<4 x i16> %a)
|
|
ret i16 %xor_result
|
|
}
|
|
|
|
define i16 @test_redxor_v8i16(<8 x i16> %a) {
|
|
; CHECK-LABEL: test_redxor_v8i16:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: ext v1.16b, v0.16b, v0.16b, #8
|
|
; CHECK-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; CHECK-NEXT: fmov x8, d0
|
|
; CHECK-NEXT: eor x8, x8, x8, lsr #32
|
|
; CHECK-NEXT: lsr x9, x8, #16
|
|
; CHECK-NEXT: eor w0, w8, w9
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v8i16:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: mov d1, v0.d[1]
|
|
; GISEL-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; GISEL-NEXT: umov w8, v0.h[0]
|
|
; GISEL-NEXT: umov w9, v0.h[1]
|
|
; GISEL-NEXT: umov w10, v0.h[2]
|
|
; GISEL-NEXT: umov w11, v0.h[3]
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w0, w8, w9
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i16 @llvm.vector.reduce.xor.v8i16(<8 x i16> %a)
|
|
ret i16 %xor_result
|
|
}
|
|
|
|
define i16 @test_redxor_v16i16(<16 x i16> %a) {
|
|
; CHECK-LABEL: test_redxor_v16i16:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: eor v0.16b, v0.16b, v1.16b
|
|
; CHECK-NEXT: ext v1.16b, v0.16b, v0.16b, #8
|
|
; CHECK-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; CHECK-NEXT: fmov x8, d0
|
|
; CHECK-NEXT: eor x8, x8, x8, lsr #32
|
|
; CHECK-NEXT: lsr x9, x8, #16
|
|
; CHECK-NEXT: eor w0, w8, w9
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v16i16:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: eor v0.16b, v0.16b, v1.16b
|
|
; GISEL-NEXT: mov d1, v0.d[1]
|
|
; GISEL-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; GISEL-NEXT: umov w8, v0.h[0]
|
|
; GISEL-NEXT: umov w9, v0.h[1]
|
|
; GISEL-NEXT: umov w10, v0.h[2]
|
|
; GISEL-NEXT: umov w11, v0.h[3]
|
|
; GISEL-NEXT: eor w8, w8, w9
|
|
; GISEL-NEXT: eor w9, w10, w11
|
|
; GISEL-NEXT: eor w0, w8, w9
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i16 @llvm.vector.reduce.xor.v16i16(<16 x i16> %a)
|
|
ret i16 %xor_result
|
|
}
|
|
|
|
define i32 @test_redxor_v2i32(<2 x i32> %a) {
|
|
; CHECK-LABEL: test_redxor_v2i32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: fmov x8, d0
|
|
; CHECK-NEXT: lsr x9, x8, #32
|
|
; CHECK-NEXT: eor w0, w8, w9
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v2i32:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: // kill: def $d0 killed $d0 def $q0
|
|
; GISEL-NEXT: mov w8, v0.s[1]
|
|
; GISEL-NEXT: fmov w9, s0
|
|
; GISEL-NEXT: eor w0, w9, w8
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i32 @llvm.vector.reduce.xor.v2i32(<2 x i32> %a)
|
|
ret i32 %xor_result
|
|
}
|
|
|
|
define i32 @test_redxor_v4i32(<4 x i32> %a) {
|
|
; CHECK-LABEL: test_redxor_v4i32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: ext v1.16b, v0.16b, v0.16b, #8
|
|
; CHECK-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; CHECK-NEXT: fmov x8, d0
|
|
; CHECK-NEXT: lsr x9, x8, #32
|
|
; CHECK-NEXT: eor w0, w8, w9
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v4i32:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: mov d1, v0.d[1]
|
|
; GISEL-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; GISEL-NEXT: mov w8, v0.s[1]
|
|
; GISEL-NEXT: fmov w9, s0
|
|
; GISEL-NEXT: eor w0, w9, w8
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i32 @llvm.vector.reduce.xor.v4i32(<4 x i32> %a)
|
|
ret i32 %xor_result
|
|
}
|
|
|
|
define i32 @test_redxor_v8i32(<8 x i32> %a) {
|
|
; CHECK-LABEL: test_redxor_v8i32:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: eor v0.16b, v0.16b, v1.16b
|
|
; CHECK-NEXT: ext v1.16b, v0.16b, v0.16b, #8
|
|
; CHECK-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; CHECK-NEXT: fmov x8, d0
|
|
; CHECK-NEXT: lsr x9, x8, #32
|
|
; CHECK-NEXT: eor w0, w8, w9
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v8i32:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: eor v0.16b, v0.16b, v1.16b
|
|
; GISEL-NEXT: mov d1, v0.d[1]
|
|
; GISEL-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; GISEL-NEXT: mov w8, v0.s[1]
|
|
; GISEL-NEXT: fmov w9, s0
|
|
; GISEL-NEXT: eor w0, w9, w8
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i32 @llvm.vector.reduce.xor.v8i32(<8 x i32> %a)
|
|
ret i32 %xor_result
|
|
}
|
|
|
|
define i64 @test_redxor_v2i64(<2 x i64> %a) {
|
|
; CHECK-LABEL: test_redxor_v2i64:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: ext v1.16b, v0.16b, v0.16b, #8
|
|
; CHECK-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; CHECK-NEXT: fmov x0, d0
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v2i64:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: mov x8, v0.d[1]
|
|
; GISEL-NEXT: fmov x9, d0
|
|
; GISEL-NEXT: eor x0, x9, x8
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i64 @llvm.vector.reduce.xor.v2i64(<2 x i64> %a)
|
|
ret i64 %xor_result
|
|
}
|
|
|
|
define i64 @test_redxor_v4i64(<4 x i64> %a) {
|
|
; CHECK-LABEL: test_redxor_v4i64:
|
|
; CHECK: // %bb.0:
|
|
; CHECK-NEXT: eor v0.16b, v0.16b, v1.16b
|
|
; CHECK-NEXT: ext v1.16b, v0.16b, v0.16b, #8
|
|
; CHECK-NEXT: eor v0.8b, v0.8b, v1.8b
|
|
; CHECK-NEXT: fmov x0, d0
|
|
; CHECK-NEXT: ret
|
|
;
|
|
; GISEL-LABEL: test_redxor_v4i64:
|
|
; GISEL: // %bb.0:
|
|
; GISEL-NEXT: eor v0.16b, v0.16b, v1.16b
|
|
; GISEL-NEXT: mov x8, v0.d[1]
|
|
; GISEL-NEXT: fmov x9, d0
|
|
; GISEL-NEXT: eor x0, x9, x8
|
|
; GISEL-NEXT: ret
|
|
%xor_result = call i64 @llvm.vector.reduce.xor.v4i64(<4 x i64> %a)
|
|
ret i64 %xor_result
|
|
}
|
|
|
|
declare i1 @llvm.vector.reduce.xor.v1i1(<1 x i1>)
|
|
declare i1 @llvm.vector.reduce.xor.v2i1(<2 x i1>)
|
|
declare i1 @llvm.vector.reduce.xor.v4i1(<4 x i1>)
|
|
declare i1 @llvm.vector.reduce.xor.v8i1(<8 x i1>)
|
|
declare i1 @llvm.vector.reduce.xor.v16i1(<16 x i1>)
|
|
declare i64 @llvm.vector.reduce.xor.v2i64(<2 x i64>)
|
|
declare i64 @llvm.vector.reduce.xor.v4i64(<4 x i64>)
|
|
declare i32 @llvm.vector.reduce.xor.v2i32(<2 x i32>)
|
|
declare i32 @llvm.vector.reduce.xor.v4i32(<4 x i32>)
|
|
declare i32 @llvm.vector.reduce.xor.v8i32(<8 x i32>)
|
|
declare i16 @llvm.vector.reduce.xor.v4i16(<4 x i16>)
|
|
declare i16 @llvm.vector.reduce.xor.v8i16(<8 x i16>)
|
|
declare i16 @llvm.vector.reduce.xor.v16i16(<16 x i16>)
|
|
declare i8 @llvm.vector.reduce.xor.v1i8(<1 x i8>)
|
|
declare i8 @llvm.vector.reduce.xor.v3i8(<3 x i8>)
|
|
declare i8 @llvm.vector.reduce.xor.v4i8(<4 x i8>)
|
|
declare i8 @llvm.vector.reduce.xor.v8i8(<8 x i8>)
|
|
declare i8 @llvm.vector.reduce.xor.v16i8(<16 x i8>)
|
|
declare i8 @llvm.vector.reduce.xor.v32i8(<32 x i8>)
|