1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-22 18:54:02 +01:00
llvm-mirror/test/CodeGen/AMDGPU/ffloor.ll
Matt Arsenault dd9ab77318 AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernel
Currently the default C calling convention functions are treated
the same as compute kernels. Make this explicit so the default
calling convention can be changed to a non-kernel.

Converted with perl -pi -e 's/define void/define amdgpu_kernel void/'
on the relevant test directories (and undoing in one place that actually
wanted a non-kernel).

llvm-svn: 298444
2017-03-21 21:39:51 +00:00

50 lines
1.6 KiB
LLVM

; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=FUNC %s
; RUN: llc -march=r600 -mcpu=redwood < %s | FileCheck -check-prefix=R600 -check-prefix=FUNC %s
; FUNC-LABEL: {{^}}floor_f32:
; SI: v_floor_f32_e32
; R600: FLOOR
define amdgpu_kernel void @floor_f32(float addrspace(1)* %out, float %in) {
%tmp = call float @llvm.floor.f32(float %in) #0
store float %tmp, float addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}floor_v2f32:
; SI: v_floor_f32_e32
; SI: v_floor_f32_e32
define amdgpu_kernel void @floor_v2f32(<2 x float> addrspace(1)* %out, <2 x float> %in) {
%tmp = call <2 x float> @llvm.floor.v2f32(<2 x float> %in) #0
store <2 x float> %tmp, <2 x float> addrspace(1)* %out
ret void
}
; FUNC-LABEL: {{^}}floor_v4f32:
; SI: v_floor_f32_e32
; SI: v_floor_f32_e32
; SI: v_floor_f32_e32
; SI: v_floor_f32_e32
; R600: FLOOR
; R600: FLOOR
; R600: FLOOR
; R600: FLOOR
define amdgpu_kernel void @floor_v4f32(<4 x float> addrspace(1)* %out, <4 x float> %in) {
%tmp = call <4 x float> @llvm.floor.v4f32(<4 x float> %in) #0
store <4 x float> %tmp, <4 x float> addrspace(1)* %out
ret void
}
; Function Attrs: nounwind readonly
declare float @llvm.floor.f32(float) #0
; Function Attrs: nounwind readonly
declare <2 x float> @llvm.floor.v2f32(<2 x float>) #0
; Function Attrs: nounwind readonly
declare <4 x float> @llvm.floor.v4f32(<4 x float>) #0
attributes #0 = { nounwind readnone }