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
|
;=========================== begin_copyright_notice ============================
;
; Copyright (C) 2023-2024 Intel Corporation
;
; SPDX-License-Identifier: MIT
;
;============================ end_copyright_notice =============================
; REQUIRES: opaque-ptr-fix, llvm-14-plus
; RUN: igc_opt --opaque-pointers -debugify --igc-gep-loop-strength-reduction -check-debugify -S < %s 2>&1 | FileCheck %s
;
; Kernel has two SCEV expressions:
; 1. {(zext i32 %delta to i64),+,1}<nuw><nsw><%for.body>
; 2. {(1 + (zext i32 (1 + %delta) to i64))<nuw><nsw>,+,1}<nuw><nsw><%for.body>
; Pass should strip zext instructions and calculate delta between two start addresses
; (zext i32 %delta to i64) and (1 + (zext i32 (1 + %delta) to i64)) to 2. This allows
; to combine two GEPs with single reduction.
; Debug-info related check
; CHECK: CheckModuleDebugify: PASS
define spir_kernel void @test(i32 addrspace(1)* %p, i32 %n, i32 %delta) {
entry:
%cmp1 = icmp slt i32 0, %n
br i1 %cmp1, label %for.body.lr.ph, label %for.end
; CHECK-LABEL: for.body.lr.ph:
; CHECK: [[ZEXT1:%.*]] = zext i32 %delta to i64
; CHECK: [[GEP_PHI1:%.*]] = getelementptr i32, i32 addrspace(1)* %p, i64 [[ZEXT1]]
; CHECK: br label %for.body
for.body.lr.ph: ; preds = %entry
br label %for.body
; CHECK-LABEL: for.body:
; CHECK: [[GEP1:%.*]] = phi i32 addrspace(1)* [ [[GEP_PHI1]], %for.body.lr.ph ], [ [[GEP_PHI2:%.*]], %for.body ]
; CHECK: %i.02 = phi i32 [ 0, %for.body.lr.ph ], [ %add7, %for.body ]
; CHECK-NOT: add{.*}{i32|i64}
; CHECK-NOT: zext
; CHECK-NOT: getelementptr inbounds i32, i32 addrspace(1)* %p
; CHECK: [[LOAD:%.*]] = load i32, i32 addrspace(1)* [[GEP1]], align 4
; CHECK-NOT: add{.*}{i32|i64}
; CHECK-NOT: zext
; CHECK: [[GEP2:%.*]] = getelementptr i32, i32 addrspace(1)* [[GEP1]], i64 2
; CHECK: store i32 [[LOAD]], i32 addrspace(1)* [[GEP2]], align 4
; CHECK: %add7 = add nuw nsw i32 %i.02, 1
; CHECK: %cmp = icmp slt i32 %add7, %n
; CHECK: [[GEP_PHI2]] = getelementptr i32, i32 addrspace(1)* [[GEP1]], i64 1
; CHECK: br i1 %cmp, label %for.body, label %for.cond.for.end_crit_edge
for.body: ; preds = %for.body.lr.ph, %for.body
%i.02 = phi i32 [ 0, %for.body.lr.ph ], [ %add7, %for.body ]
%ext1 = zext i32 %i.02 to i64
%ext2 = zext i32 %delta to i64
%add1 = add nuw nsw i64 %ext1, %ext2
%arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %p, i64 %add1
%0 = load i32, i32 addrspace(1)* %arrayidx, align 4
%add2 = add nuw nsw i32 %i.02, %delta
%add3 = add nuw nsw i32 %add2, 1
%ext3 = zext i32 %add3 to i64
%add4 = add nuw nsw i64 %ext3, 1
%arrayidx1 = getelementptr inbounds i32, i32 addrspace(1)* %p, i64 %add4
store i32 %0, i32 addrspace(1)* %arrayidx1, align 4
%add7 = add nuw nsw i32 %i.02, 1
%cmp = icmp slt i32 %add7, %n
br i1 %cmp, label %for.body, label %for.cond.for.end_crit_edge
for.cond.for.end_crit_edge: ; preds = %for.body
br label %for.end
for.end: ; preds = %for.cond.for.end_crit_edge, %entry
ret void
}
!igc.functions = !{!0}
!0 = !{void (i32 addrspace(1)*, i32, i32)* @test, !1}
!1 = !{!2}
!2 = !{!"function_type", i32 0}
|
