mirror of
https://github.com/RPCSX/llvm.git
synced 2025-01-07 12:30:44 +00:00
7eba65d30c
For some reason there are both of these available, except for scalar 64-bit compares which only has u64. I'm not sure why there are both (I'm guessing it's for the one bit inputs we don't use), but for consistency always using the unsigned one. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@282832 91177308-0d34-0410-b5e6-96231b3b80d8
63 lines
2.3 KiB
LLVM
63 lines
2.3 KiB
LLVM
; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=SI -check-prefix=FUNC %s
|
|
; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=VI -check-prefix=FUNC %s
|
|
|
|
; FIXME: This should be merged with sint_to_fp.ll, but s_sint_to_fp_v2i64 crashes on r600
|
|
|
|
; FUNC-LABEL: {{^}}s_sint_to_fp_i64_to_f32:
|
|
define void @s_sint_to_fp_i64_to_f32(float addrspace(1)* %out, i64 %in) #0 {
|
|
%result = sitofp i64 %in to float
|
|
store float %result, float addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}v_sint_to_fp_i64_to_f32:
|
|
; GCN: {{buffer|flat}}_load_dwordx2
|
|
|
|
; GCN: v_ashrrev_i32_e32 v{{[0-9]+}}, 31, v{{[0-9]+}}
|
|
; GCN: v_xor_b32
|
|
|
|
; GCN: v_ffbh_u32
|
|
; GCN: v_ffbh_u32
|
|
; GCN: v_cndmask
|
|
; GCN: v_cndmask
|
|
|
|
; GCN-DAG: v_cmp_eq_u64
|
|
; GCN-DAG: v_cmp_lt_u64
|
|
|
|
; GCN: v_xor_b32_e32 v{{[0-9]+}}, 0x80000000, v{{[0-9]+}}
|
|
; GCN: v_cndmask_b32_e{{32|64}} [[SIGN_SEL:v[0-9]+]],
|
|
; GCN: {{buffer|flat}}_store_dword {{.*}}[[SIGN_SEL]]
|
|
define void @v_sint_to_fp_i64_to_f32(float addrspace(1)* %out, i64 addrspace(1)* %in) #0 {
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%in.gep = getelementptr i64, i64 addrspace(1)* %in, i32 %tid
|
|
%out.gep = getelementptr float, float addrspace(1)* %out, i32 %tid
|
|
%val = load i64, i64 addrspace(1)* %in.gep
|
|
%result = sitofp i64 %val to float
|
|
store float %result, float addrspace(1)* %out.gep
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}s_sint_to_fp_v2i64:
|
|
; GCN-NOT: v_and_b32_e32 v{{[0-9]+}}, -1,
|
|
define void @s_sint_to_fp_v2i64(<2 x float> addrspace(1)* %out, <2 x i64> %in) #0{
|
|
%result = sitofp <2 x i64> %in to <2 x float>
|
|
store <2 x float> %result, <2 x float> addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}v_sint_to_fp_v4i64:
|
|
define void @v_sint_to_fp_v4i64(<4 x float> addrspace(1)* %out, <4 x i64> addrspace(1)* %in) #0 {
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%in.gep = getelementptr <4 x i64>, <4 x i64> addrspace(1)* %in, i32 %tid
|
|
%out.gep = getelementptr <4 x float>, <4 x float> addrspace(1)* %out, i32 %tid
|
|
%value = load <4 x i64>, <4 x i64> addrspace(1)* %in.gep
|
|
%result = sitofp <4 x i64> %value to <4 x float>
|
|
store <4 x float> %result, <4 x float> addrspace(1)* %out.gep
|
|
ret void
|
|
}
|
|
|
|
declare i32 @llvm.amdgcn.workitem.id.x() #1
|
|
|
|
attributes #0 = { nounwind }
|
|
attributes #1 = { nounwind readnone }
|