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
|
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -S -passes=slp-vectorizer,slp-vectorizer -mcpu=bdver1 < %s | FileCheck %s
target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"
@a = common local_unnamed_addr global [1 x i32] zeroinitializer, align 4
@b = common local_unnamed_addr global [1 x i32] zeroinitializer, align 4
define i32 @slp_schedule_bundle() local_unnamed_addr #0 {
; CHECK-LABEL: @slp_schedule_bundle(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP0:%.*]] = load <4 x i32>, ptr @b, align 4
; CHECK-NEXT: [[TMP1:%.*]] = lshr <4 x i32> [[TMP0]], <i32 31, i32 31, i32 31, i32 31>
; CHECK-NEXT: [[TMP2:%.*]] = xor <4 x i32> [[TMP1]], <i32 1, i32 1, i32 1, i32 1>
; CHECK-NEXT: store <4 x i32> [[TMP2]], ptr @a, align 4
; CHECK-NEXT: [[TMP3:%.*]] = load <2 x i32>, ptr getelementptr ([1 x i32], ptr @b, i64 4, i64 0), align 4
; CHECK-NEXT: [[TMP4:%.*]] = lshr <2 x i32> [[TMP3]], <i32 31, i32 31>
; CHECK-NEXT: [[TMP5:%.*]] = xor <2 x i32> [[TMP4]], <i32 1, i32 1>
; CHECK-NEXT: store <2 x i32> [[TMP5]], ptr getelementptr ([1 x i32], ptr @a, i64 4, i64 0), align 4
; CHECK-NEXT: ret i32 undef
;
entry:
%0 = load i32, ptr @b, align 4
%.lobit = lshr i32 %0, 31
%.lobit.not = xor i32 %.lobit, 1
store i32 %.lobit.not, ptr @a, align 4
%1 = load i32, ptr getelementptr inbounds ([1 x i32], ptr @b, i64 1, i64 0), align 4
%.lobit.1 = lshr i32 %1, 31
%.lobit.not.1 = xor i32 %.lobit.1, 1
store i32 %.lobit.not.1, ptr getelementptr inbounds ([1 x i32], ptr @a, i64 1, i64 0), align 4
%2 = load i32, ptr getelementptr ([1 x i32], ptr @b, i64 2, i64 0), align 4
%.lobit.2 = lshr i32 %2, 31
%.lobit.not.2 = xor i32 %.lobit.2, 1
store i32 %.lobit.not.2, ptr getelementptr ([1 x i32], ptr @a, i64 2, i64 0), align 4
%3 = load i32, ptr getelementptr ([1 x i32], ptr @b, i64 3, i64 0), align 4
%.lobit.3 = lshr i32 %3, 31
%.lobit.not.3 = xor i32 %.lobit.3, 1
store i32 %.lobit.not.3, ptr getelementptr ([1 x i32], ptr @a, i64 3, i64 0), align 4
%4 = load i32, ptr getelementptr ([1 x i32], ptr @b, i64 4, i64 0), align 4
%.lobit.4 = lshr i32 %4, 31
%.lobit.not.4 = xor i32 %.lobit.4, 1
store i32 %.lobit.not.4, ptr getelementptr ([1 x i32], ptr @a, i64 4, i64 0), align 4
%5 = load i32, ptr getelementptr ([1 x i32], ptr @b, i64 5, i64 0), align 4
%.lobit.5 = lshr i32 %5, 31
%.lobit.not.5 = xor i32 %.lobit.5, 1
store i32 %.lobit.not.5, ptr getelementptr ([1 x i32], ptr @a, i64 5, i64 0), align 4
ret i32 undef
}
|
