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
| ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=aarch64-unknown-unknown -mcpu=cyclone -mattr=+slow-misaligned-128store | FileCheck %s --check-prefixes=CHECK,SPLITTING
; RUN: llc < %s -mtriple=aarch64-eabi -mattr=-slow-misaligned-128store | FileCheck %s --check-prefixes=CHECK,MISALIGNED
@g0 = external global <3 x float>, align 16
@g1 = external global <3 x float>, align 4
define void @blam() {
; SPLITTING-LABEL: blam:
; SPLITTING: // %bb.0:
; SPLITTING-NEXT: adrp x8, g1
; SPLITTING-NEXT: add x8, x8, :lo12:g1
; SPLITTING-NEXT: adrp x9, g0
; SPLITTING-NEXT: ldr q0, [x9, :lo12:g0]
; SPLITTING-NEXT: str d0, [x8]
; SPLITTING-NEXT: ret
;
; MISALIGNED-LABEL: blam:
; MISALIGNED: // %bb.0:
; MISALIGNED-NEXT: adrp x8, g0
; MISALIGNED-NEXT: ldr q0, [x8, :lo12:g0]
; MISALIGNED-NEXT: adrp x8, g1
; MISALIGNED-NEXT: add x8, x8, :lo12:g1
; MISALIGNED-NEXT: str d0, [x8]
; MISALIGNED-NEXT: ret
%tmp4 = getelementptr inbounds <3 x float>, <3 x float>* @g1, i64 0, i64 0
%tmp5 = load <3 x float>, <3 x float>* @g0, align 16
%tmp6 = extractelement <3 x float> %tmp5, i64 0
store float %tmp6, float* %tmp4
%tmp7 = getelementptr inbounds float, float* %tmp4, i64 1
%tmp8 = load <3 x float>, <3 x float>* @g0, align 16
%tmp9 = extractelement <3 x float> %tmp8, i64 1
store float %tmp9, float* %tmp7
ret void;
}
; PR21711 - Merge vector stores into wider vector stores.
; On Cyclone, the stores should not get merged into a 16-byte store because
; unaligned 16-byte stores are slow. This test would infinite loop when
; the fastness of unaligned accesses was not specified correctly.
define void @merge_vec_extract_stores(<4 x float> %v1, <2 x float>* %ptr) {
; SPLITTING-LABEL: merge_vec_extract_stores:
; SPLITTING: // %bb.0:
; SPLITTING-NEXT: ext v1.16b, v0.16b, v0.16b, #8
; SPLITTING-NEXT: str d0, [x0, #24]
; SPLITTING-NEXT: str d1, [x0, #32]
; SPLITTING-NEXT: ret
;
; MISALIGNED-LABEL: merge_vec_extract_stores:
; MISALIGNED: // %bb.0:
; MISALIGNED-NEXT: stur q0, [x0, #24]
; MISALIGNED-NEXT: ret
%idx0 = getelementptr inbounds <2 x float>, <2 x float>* %ptr, i64 3
%idx1 = getelementptr inbounds <2 x float>, <2 x float>* %ptr, i64 4
%shuffle0 = shufflevector <4 x float> %v1, <4 x float> undef, <2 x i32> <i32 0, i32 1>
%shuffle1 = shufflevector <4 x float> %v1, <4 x float> undef, <2 x i32> <i32 2, i32 3>
store <2 x float> %shuffle0, <2 x float>* %idx0, align 8
store <2 x float> %shuffle1, <2 x float>* %idx1, align 8
ret void
; FIXME: Ideally we would like to use a generic target for this test, but this relies
; on suppressing store pairs.
}
|