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
|
;=========================== begin_copyright_notice ============================
;
; Copyright (C) 2023 Intel Corporation
;
; SPDX-License-Identifier: MIT
;
;============================ end_copyright_notice =============================
; RUN: igc_opt -debugify --igc-gep-loop-strength-reduction -check-debugify -S < %s 2>&1 | FileCheck %s
;
; Pass targets only accesses unique for loop. If loop's induction variable is used
; after loop to access the same pointer, don't optimize it.
; Debug-info related check
; CHECK: CheckModuleDebugify: PASS
%struct.Foo = type { float, float }
define spir_kernel void @test(%struct.Foo addrspace(1)* %p, i32 %n) {
entry:
%cmp1 = icmp sgt i32 %n, 0
br i1 %cmp1, label %for.body.lr.ph, label %for.end
; CHECK-LABEL: for.body.lr.ph:
; CHECK-NOT: getelementptr
; CHECK: br label %for.body
for.body.lr.ph: ; preds = %entry
br label %for.body
; CHECK-LABEL: for.body:
; CHECK: %idxprom3 = phi i64 [ 0, %for.body.lr.ph ], [ %idxprom, %for.body ]
; CHECK: %i.02 = phi i32 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
; CHECK: %ptmp = getelementptr inbounds %struct.Foo, %struct.Foo addrspace(1)* %p, i64 %idxprom3
; CHECK: %a = getelementptr inbounds %struct.Foo, %struct.Foo addrspace(1)* %ptmp, i64 0, i32 0
; CHECK: %0 = load float, float addrspace(1)* %a, align 4
; CHECK: %add = fadd float %0, 1.000000e+00
; CHECK: store float %add, float addrspace(1)* %a, align 4
; CHECK: %inc = add nuw nsw i32 %i.02, 1
; CHECK: %cmp = icmp slt i32 %inc, %n
; CHECK: %idxprom = zext i32 %inc to i64
; CHECK: br i1 %cmp, label %for.body, label %for.cond.for.end_crit_edge
for.body: ; preds = %for.body.lr.ph, %for.body
%idxprom3 = phi i64 [ 0, %for.body.lr.ph ], [ %idxprom, %for.body ]
%i.02 = phi i32 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
%ptmp = getelementptr inbounds %struct.Foo, %struct.Foo addrspace(1)* %p, i64 %idxprom3
%a = getelementptr inbounds %struct.Foo, %struct.Foo addrspace(1)* %ptmp, i64 0, i32 0
%0 = load float, float addrspace(1)* %a, align 4
%add = fadd float %0, 1.000000e+00
store float %add, float addrspace(1)* %a, align 4
%inc = add nuw nsw i32 %i.02, 1
%cmp = icmp slt i32 %inc, %n
%idxprom = zext i32 %inc to i64
br i1 %cmp, label %for.body, label %for.cond.for.end_crit_edge
for.cond.for.end_crit_edge: ; preds = %for.body
%idxprom.lcssa4 = phi i64 [ %idxprom3, %for.body ]
br label %for.end
for.end: ; preds = %for.cond.for.end_crit_edge, %entry
%idxprom.lcssa = phi i64 [ %idxprom.lcssa4, %for.cond.for.end_crit_edge ], [ 0, %entry ]
%b = getelementptr inbounds %struct.Foo, %struct.Foo addrspace(1)* %p, i64 %idxprom.lcssa, i32 1
%1 = load float, float addrspace(1)* %b, align 4
%add3 = fadd float %1, 1.000000e+00
store float %add3, float addrspace(1)* %b, align 4
ret void
}
!igc.functions = !{!0}
!0 = !{void (%struct.Foo addrspace(1)*, i32)* @test, !1}
!1 = !{!2}
!2 = !{!"function_type", i32 0}
|
