mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-26 12:43:36 +01:00
ee0d5cd952
This adds support for the new 32-bit vector float instructions of z14. This includes: - Enabling the instructions for the assembler/disassembler. - CodeGen for the instructions, including new LLVM intrinsics. - Scheduler description support for the instructions. - Update to the vector cost function calculations. In general, CodeGen support for the new v4f32 instructions closely matches support for the existing v2f64 instructions. llvm-svn: 308195
24 lines
609 B
LLVM
24 lines
609 B
LLVM
; Test f32 and v4f32 square root on z14.
|
|
;
|
|
; RUN: llc < %s -mtriple=s390x-linux-gnu -mcpu=z14 | FileCheck %s
|
|
|
|
declare float @llvm.sqrt.f32(float)
|
|
declare <4 x float> @llvm.sqrt.v4f32(<4 x float>)
|
|
|
|
define <4 x float> @f1(<4 x float> %val) {
|
|
; CHECK-LABEL: f1:
|
|
; CHECK: vfsqsb %v24, %v24
|
|
; CHECK: br %r14
|
|
%ret = call <4 x float> @llvm.sqrt.v4f32(<4 x float> %val)
|
|
ret <4 x float> %ret
|
|
}
|
|
|
|
define float @f2(<4 x float> %val) {
|
|
; CHECK-LABEL: f2:
|
|
; CHECK: wfsqsb %f0, %v24
|
|
; CHECK: br %r14
|
|
%scalar = extractelement <4 x float> %val, i32 0
|
|
%ret = call float @llvm.sqrt.f32(float %scalar)
|
|
ret float %ret
|
|
}
|