1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-24 13:33:37 +02:00
llvm-mirror/test/CodeGen/AMDGPU/llvm.AMDGPU.abs.ll
Marek Olsak d73d332555 AMDGPU/SI: select S_ABS_I32 when possible (v2)
v2: added more tests, moved the SALU->VALU conversion to a separate function

It looks like it's not possible to get subregisters in the S_ABS lowering
code, and I don't feel like guessing without testing what the correct code
would look like.

llvm-svn: 254095
2015-11-25 21:22:45 +00:00

48 lines
1.5 KiB
LLVM

; RUN: llc -march=amdgcn -mcpu=SI -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
; RUN: llc -march=r600 -mcpu=cypress -verify-machineinstrs < %s | FileCheck -check-prefix=EG -check-prefix=FUNC %s
declare i32 @llvm.AMDGPU.abs(i32) nounwind readnone
; Legacy name
declare i32 @llvm.AMDIL.abs.i32(i32) nounwind readnone
; FUNC-LABEL: {{^}}s_abs_i32:
; SI: s_abs_i32
; EG: SUB_INT
; EG: MAX_INT
define void @s_abs_i32(i32 addrspace(1)* %out, i32 %src) nounwind {
%abs = call i32 @llvm.AMDGPU.abs(i32 %src) nounwind readnone
store i32 %abs, i32 addrspace(1)* %out, align 4
ret void
}
; FUNC-LABEL: {{^}}v_abs_i32:
; SI: v_sub_i32_e32
; SI: v_max_i32_e32
; SI: s_endpgm
; EG: SUB_INT
; EG: MAX_INT
define void @v_abs_i32(i32 addrspace(1)* %out, i32 addrspace(1)* %src) nounwind {
%val = load i32, i32 addrspace(1)* %src, align 4
%abs = call i32 @llvm.AMDGPU.abs(i32 %val) nounwind readnone
store i32 %abs, i32 addrspace(1)* %out, align 4
ret void
}
; FUNC-LABEL: {{^}}abs_i32_legacy_amdil:
; SI: v_sub_i32_e32
; SI: v_max_i32_e32
; SI: s_endpgm
; EG: SUB_INT
; EG: MAX_INT
define void @abs_i32_legacy_amdil(i32 addrspace(1)* %out, i32 addrspace(1)* %src) nounwind {
%val = load i32, i32 addrspace(1)* %src, align 4
%abs = call i32 @llvm.AMDIL.abs.i32(i32 %val) nounwind readnone
store i32 %abs, i32 addrspace(1)* %out, align 4
ret void
}