9e9907f1cf
Similar to 806761a762.
For IR files without a target triple, -mtriple= specifies the full
target triple while -march= merely sets the architecture part of the
default target triple, leaving a target triple which may not make sense,
e.g. amdgpu-apple-darwin.
Therefore, -march= is error-prone and not recommended for tests without
a target triple. The issue has been benign as we recognize
$unknown-apple-darwin as ELF instead of rejecting it outrightly.
This patch changes AMDGPU tests to not rely on the default
OS/environment components. Tests that need fixes are not changed:
```
LLVM :: CodeGen/AMDGPU/fabs.f64.ll
LLVM :: CodeGen/AMDGPU/fabs.ll
LLVM :: CodeGen/AMDGPU/floor.ll
LLVM :: CodeGen/AMDGPU/fneg-fabs.f64.ll
LLVM :: CodeGen/AMDGPU/fneg-fabs.ll
LLVM :: CodeGen/AMDGPU/r600-infinite-loop-bug-while-reorganizing-vector.ll
LLVM :: CodeGen/AMDGPU/schedule-if-2.ll
```
41 lines
2.5 KiB
LLVM
41 lines
2.5 KiB
LLVM
; RUN: llc -global-isel=0 < %s -mtriple=amdgcn -mcpu=gfx600 -verify-machineinstrs | FileCheck %s -check-prefixes=GCN,SI
|
|
; RUN: llc -global-isel=0 < %s -mtriple=amdgcn -mcpu=gfx700 -verify-machineinstrs | FileCheck %s -check-prefixes=GCN,GCNX3
|
|
|
|
; GCN-LABEL: {{^}}tbuffer_raw_load_immoffs_x3:
|
|
; SI: tbuffer_load_format_xyzw {{v\[[0-9]+:[0-9]+\]}}, off, {{s\[[0-9]+:[0-9]+\]}}, 0 format:[BUF_DATA_FORMAT_32_32_32_32,BUF_NUM_FORMAT_UINT] offset:42
|
|
; GCNX3: tbuffer_load_format_xyz {{v\[[0-9]+:[0-9]+\]}}, off, {{s\[[0-9]+:[0-9]+\]}}, 0 format:[BUF_DATA_FORMAT_32_32_32_32,BUF_NUM_FORMAT_UINT] offset:42
|
|
define amdgpu_vs <3 x float> @tbuffer_raw_load_immoffs_x3(<4 x i32> inreg) {
|
|
main_body:
|
|
%vdata = call <3 x i32> @llvm.amdgcn.raw.tbuffer.load.v3i32(<4 x i32> %0, i32 42, i32 0, i32 78, i32 0)
|
|
%vdata.f = bitcast <3 x i32> %vdata to <3 x float>
|
|
ret <3 x float> %vdata.f
|
|
}
|
|
|
|
|
|
; GCN-LABEL: {{^}}tbuffer_struct_load_immoffs_x3:
|
|
; GCN: v_mov_b32_e32 [[ZEROREG:v[0-9]+]], 0
|
|
; SI: tbuffer_load_format_xyzw {{v\[[0-9]+:[0-9]+\]}}, [[ZEROREG]], {{s\[[0-9]+:[0-9]+\]}}, 0 format:[BUF_DATA_FORMAT_32_32_32_32,BUF_NUM_FORMAT_UINT] idxen offset:42
|
|
; GCNX3: tbuffer_load_format_xyz {{v\[[0-9]+:[0-9]+\]}}, [[ZEROREG]], {{s\[[0-9]+:[0-9]+\]}}, 0 format:[BUF_DATA_FORMAT_32_32_32_32,BUF_NUM_FORMAT_UINT] idxen offset:42
|
|
define amdgpu_vs <3 x float> @tbuffer_struct_load_immoffs_x3(<4 x i32> inreg) {
|
|
main_body:
|
|
%vdata = call <3 x i32> @llvm.amdgcn.struct.tbuffer.load.v3i32(<4 x i32> %0, i32 0, i32 42, i32 0, i32 78, i32 0)
|
|
%vdata.f = bitcast <3 x i32> %vdata to <3 x float>
|
|
ret <3 x float> %vdata.f
|
|
}
|
|
|
|
|
|
; GCN-LABEL: {{^}}tbuffer_load_format_immoffs_x3:
|
|
; SI: tbuffer_load_format_xyzw {{v\[[0-9]+:[0-9]+\]}}, off, {{s\[[0-9]+:[0-9]+\]}}, 0 format:[BUF_DATA_FORMAT_32_32_32_32,BUF_NUM_FORMAT_UINT] offset:42
|
|
; GCNX3: tbuffer_load_format_xyz {{v\[[0-9]+:[0-9]+\]}}, off, {{s\[[0-9]+:[0-9]+\]}}, 0 format:[BUF_DATA_FORMAT_32_32_32_32,BUF_NUM_FORMAT_UINT] offset:42
|
|
define amdgpu_vs <3 x float> @tbuffer_load_format_immoffs_x3(<4 x i32> inreg) {
|
|
main_body:
|
|
%vdata = call <3 x i32> @llvm.amdgcn.tbuffer.load.v3i32(<4 x i32> %0, i32 0, i32 0, i32 0, i32 42, i32 14, i32 4, i1 0, i1 0)
|
|
%vdata.f = bitcast <3 x i32> %vdata to <3 x float>
|
|
ret <3 x float> %vdata.f
|
|
}
|
|
|
|
declare <3 x i32> @llvm.amdgcn.raw.tbuffer.load.v3i32(<4 x i32>, i32, i32, i32, i32)
|
|
declare <3 x i32> @llvm.amdgcn.struct.tbuffer.load.v3i32(<4 x i32>, i32, i32, i32, i32, i32)
|
|
declare <3 x i32> @llvm.amdgcn.tbuffer.load.v3i32(<4 x i32>, i32, i32, i32, i32, i32, i32, i1, i1)
|
|
|