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
|
; Test strict vector addition on z14.
;
; RUN: llc < %s -mtriple=s390x-linux-gnu -mcpu=z14 | FileCheck %s
declare float @llvm.experimental.constrained.fadd.f32(float, float, metadata, metadata)
declare <4 x float> @llvm.experimental.constrained.fadd.v4f32(<4 x float>, <4 x float>, metadata, metadata)
declare <1 x fp128> @llvm.experimental.constrained.fadd.v1f128(<1 x fp128>, <1 x fp128>, metadata, metadata)
; Test a v4f32 addition.
define <4 x float> @f1(<4 x float> %dummy, <4 x float> %val1,
<4 x float> %val2) strictfp {
; CHECK-LABEL: f1:
; CHECK: vfasb %v24, %v26, %v28
; CHECK: br %r14
%ret = call <4 x float> @llvm.experimental.constrained.fadd.v4f32(
<4 x float> %val1, <4 x float> %val2,
metadata !"round.dynamic",
metadata !"fpexcept.strict") strictfp
ret <4 x float> %ret
}
; Test an f32 addition that uses vector registers.
define float @f2(<4 x float> %val1, <4 x float> %val2) strictfp {
; CHECK-LABEL: f2:
; CHECK: wfasb %f0, %v24, %v26
; CHECK: br %r14
%scalar1 = extractelement <4 x float> %val1, i32 0
%scalar2 = extractelement <4 x float> %val2, i32 0
%ret = call float @llvm.experimental.constrained.fadd.f32(
float %scalar1, float %scalar2,
metadata !"round.dynamic",
metadata !"fpexcept.strict") strictfp
ret float %ret
}
; Test a v1f128 addition.
define <1 x fp128> @f3(<1 x fp128> %dummy, <1 x fp128> %val1,
<1 x fp128> %val2) strictfp {
; CHECK-LABEL: f3:
; CHECK: wfaxb %v24, %v26, %v28
; CHECK: br %r14
%ret = call <1 x fp128> @llvm.experimental.constrained.fadd.v1f128(
<1 x fp128> %val1, <1 x fp128> %val2,
metadata !"round.dynamic",
metadata !"fpexcept.strict") strictfp
ret <1 x fp128> %ret
}
|
