mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2025-02-01 05:01:59 +01:00
81a9bfe915
This switches to the workaround that HSA defaults to for the mesa path. This should be applied to the 4.0 branch. Patch by Vedran Miletić <vedran@miletic.net> llvm-svn: 292982
95 lines
3.3 KiB
LLVM
95 lines
3.3 KiB
LLVM
; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=GCN-NOHSA -check-prefix=FUNC %s
|
|
; RUN: llc -mtriple=amdgcn-amdhsa -mcpu=kaveri -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=GCN-HSA -check-prefix=FUNC %s
|
|
; RUN: llc -march=amdgcn -mcpu=tonga -mattr=-flat-for-global -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -check-prefix=GCN-NOHSA -check-prefix=FUNC %s
|
|
|
|
; FUNC-LABEL: {{^}}global_load_f64:
|
|
; GCN-NOHSA: buffer_load_dwordx2 [[VAL:v\[[0-9]+:[0-9]+\]]]
|
|
; GCN-NOHSA: buffer_store_dwordx2 [[VAL]]
|
|
|
|
; GCN-HSA: flat_load_dwordx2 [[VAL:v\[[0-9]+:[0-9]+\]]]
|
|
; GCN-HSA: flat_store_dwordx2 {{v\[[0-9]+:[0-9]+\]}}, [[VAL]]
|
|
define void @global_load_f64(double addrspace(1)* %out, double addrspace(1)* %in) #0 {
|
|
%ld = load double, double addrspace(1)* %in
|
|
store double %ld, double addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}global_load_v2f64:
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
define void @global_load_v2f64(<2 x double> addrspace(1)* %out, <2 x double> addrspace(1)* %in) #0 {
|
|
entry:
|
|
%ld = load <2 x double>, <2 x double> addrspace(1)* %in
|
|
store <2 x double> %ld, <2 x double> addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}global_load_v3f64:
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
define void @global_load_v3f64(<3 x double> addrspace(1)* %out, <3 x double> addrspace(1)* %in) #0 {
|
|
entry:
|
|
%ld = load <3 x double>, <3 x double> addrspace(1)* %in
|
|
store <3 x double> %ld, <3 x double> addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}global_load_v4f64:
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
define void @global_load_v4f64(<4 x double> addrspace(1)* %out, <4 x double> addrspace(1)* %in) #0 {
|
|
entry:
|
|
%ld = load <4 x double>, <4 x double> addrspace(1)* %in
|
|
store <4 x double> %ld, <4 x double> addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}global_load_v8f64:
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
define void @global_load_v8f64(<8 x double> addrspace(1)* %out, <8 x double> addrspace(1)* %in) #0 {
|
|
entry:
|
|
%ld = load <8 x double>, <8 x double> addrspace(1)* %in
|
|
store <8 x double> %ld, <8 x double> addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
; FUNC-LABEL: {{^}}global_load_v16f64:
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
; GCN-NOHSA: buffer_load_dwordx4
|
|
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
; GCN-HSA: flat_load_dwordx4
|
|
define void @global_load_v16f64(<16 x double> addrspace(1)* %out, <16 x double> addrspace(1)* %in) #0 {
|
|
entry:
|
|
%ld = load <16 x double>, <16 x double> addrspace(1)* %in
|
|
store <16 x double> %ld, <16 x double> addrspace(1)* %out
|
|
ret void
|
|
}
|
|
|
|
attributes #0 = { nounwind }
|