File: neon-dot-product.ll

package info (click to toggle)
llvm-toolchain-14 1%3A14.0.6-12
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 1,496,180 kB
  • sloc: cpp: 5,593,972; ansic: 986,872; asm: 585,869; python: 184,223; objc: 72,530; lisp: 31,119; f90: 27,793; javascript: 9,780; pascal: 9,762; sh: 9,482; perl: 7,468; ml: 5,432; awk: 3,523; makefile: 2,538; xml: 953; cs: 573; fortran: 567
file content (82 lines) | stat: -rw-r--r-- 3,505 bytes parent folder | download | duplicates (26)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
; RUN: llc -mtriple armv8a-none-linux-gnu -mattr=+dotprod -float-abi=hard < %s | FileCheck %s

declare <2 x i32> @llvm.arm.neon.udot.v2i32.v8i8(<2 x i32>, <8 x i8>, <8 x i8>)
declare <4 x i32> @llvm.arm.neon.udot.v4i32.v16i8(<4 x i32>, <16 x i8>, <16 x i8>)
declare <2 x i32> @llvm.arm.neon.sdot.v2i32.v8i8(<2 x i32>, <8 x i8>, <8 x i8>)
declare <4 x i32> @llvm.arm.neon.sdot.v4i32.v16i8(<4 x i32>, <16 x i8>, <16 x i8>)

define <2 x i32> @test_vdot_u32(<2 x i32> %a, <8 x i8> %b, <8 x i8> %c) #0 {
entry:
; CHECK-LABEL: test_vdot_u32:
; CHECK: vudot.u8        d0, d1, d2
  %vdot1.i = call <2 x i32> @llvm.arm.neon.udot.v2i32.v8i8(<2 x i32> %a, <8 x i8> %b, <8 x i8> %c) #2
  ret <2 x i32> %vdot1.i
}

define <4 x i32> @test_vdotq_u32(<4 x i32> %a, <16 x i8> %b, <16 x i8> %c) #0 {
entry:
; CHECK-LABEL: test_vdotq_u32:
; CHECK: vudot.u8        q0, q1, q2
  %vdot1.i = call <4 x i32> @llvm.arm.neon.udot.v4i32.v16i8(<4 x i32> %a, <16 x i8> %b, <16 x i8> %c) #2
  ret <4 x i32> %vdot1.i
}

define <2 x i32> @test_vdot_s32(<2 x i32> %a, <8 x i8> %b, <8 x i8> %c) #0 {
entry:
; CHECK-LABEL: test_vdot_s32:
; CHECK: vsdot.s8        d0, d1, d2
  %vdot1.i = call <2 x i32> @llvm.arm.neon.sdot.v2i32.v8i8(<2 x i32> %a, <8 x i8> %b, <8 x i8> %c) #2
  ret <2 x i32> %vdot1.i
}

define <4 x i32> @test_vdotq_s32(<4 x i32> %a, <16 x i8> %b, <16 x i8> %c) #0 {
entry:
; CHECK-LABEL: test_vdotq_s32:
; CHECK: vsdot.s8        q0, q1, q2
  %vdot1.i = call <4 x i32> @llvm.arm.neon.sdot.v4i32.v16i8(<4 x i32> %a, <16 x i8> %b, <16 x i8> %c) #2
  ret <4 x i32> %vdot1.i
}

define <2 x i32> @test_vdot_lane_u32(<2 x i32> %a, <8 x i8> %b, <8 x i8> %c) {
entry:
; CHECK-LABEL: test_vdot_lane_u32:
; CHECK: vudot.u8        d0, d1, d2[1]
  %.cast = bitcast <8 x i8> %c to <2 x i32>
  %shuffle = shufflevector <2 x i32> %.cast, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
  %.cast5 = bitcast <2 x i32> %shuffle to <8 x i8>
  %vdot1.i = call <2 x i32> @llvm.arm.neon.udot.v2i32.v8i8(<2 x i32> %a, <8 x i8> %b, <8 x i8> %.cast5) #2
  ret <2 x i32> %vdot1.i
}

define <4 x i32> @test_vdotq_lane_u32(<4 x i32> %a, <16 x i8> %b, <8 x i8> %c) {
entry:
; CHECK-LABEL: test_vdotq_lane_u32:
; CHECK: vudot.u8        q0, q1, d4[1]
  %.cast = bitcast <8 x i8> %c to <2 x i32>
  %shuffle = shufflevector <2 x i32> %.cast, <2 x i32> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
  %.cast3 = bitcast <4 x i32> %shuffle to <16 x i8>
  %vdot1.i = call <4 x i32> @llvm.arm.neon.udot.v4i32.v16i8(<4 x i32> %a, <16 x i8> %b, <16 x i8> %.cast3) #2
  ret <4 x i32> %vdot1.i
}

define <2 x i32> @test_vdot_lane_s32(<2 x i32> %a, <8 x i8> %b, <8 x i8> %c) {
entry:
; CHECK-LABEL: test_vdot_lane_s32:
; CHECK: vsdot.s8        d0, d1, d2[1]
  %.cast = bitcast <8 x i8> %c to <2 x i32>
  %shuffle = shufflevector <2 x i32> %.cast, <2 x i32> undef, <2 x i32> <i32 1, i32 1>
  %.cast5 = bitcast <2 x i32> %shuffle to <8 x i8>
  %vdot1.i = call <2 x i32> @llvm.arm.neon.sdot.v2i32.v8i8(<2 x i32> %a, <8 x i8> %b, <8 x i8> %.cast5) #2
  ret <2 x i32> %vdot1.i
}

define <4 x i32> @test_vdotq_lane_s32(<4 x i32> %a, <16 x i8> %b, <8 x i8> %c) {
entry:
; CHECK-LABEL: test_vdotq_lane_s32:
; CHECK: vsdot.s8        q0, q1, d4[1]
  %.cast = bitcast <8 x i8> %c to <2 x i32>
  %shuffle = shufflevector <2 x i32> %.cast, <2 x i32> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
  %.cast3 = bitcast <4 x i32> %shuffle to <16 x i8>
  %vdot1.i = call <4 x i32> @llvm.arm.neon.sdot.v4i32.v16i8(<4 x i32> %a, <16 x i8> %b, <16 x i8> %.cast3) #2
  ret <4 x i32> %vdot1.i
}