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
```
21 lines
734 B
LLVM
21 lines
734 B
LLVM
; RUN: llc < %s -mtriple=amdgcn -mcpu=tahiti -verify-machineinstrs | FileCheck %s
|
|
; RUN: llc < %s -mtriple=amdgcn -mcpu=tonga -verify-machineinstrs | FileCheck %s
|
|
|
|
; CHECK: {{^}}fconst_f64:
|
|
; CHECK-DAG: s_mov_b32 {{s[0-9]+}}, 0x40140000
|
|
; CHECK-DAG: s_mov_b32 {{s[0-9]+}}, 0
|
|
|
|
define amdgpu_kernel void @fconst_f64(ptr addrspace(1) %out, ptr addrspace(1) %in) {
|
|
%tid = call i32 @llvm.amdgcn.workitem.id.x()
|
|
%gep = getelementptr inbounds double, ptr addrspace(1) %in, i32 %tid
|
|
%r1 = load double, ptr addrspace(1) %gep
|
|
%r2 = fadd double %r1, 5.000000e+00
|
|
store double %r2, ptr addrspace(1) %out
|
|
ret void
|
|
}
|
|
|
|
declare i32 @llvm.amdgcn.workitem.id.x() #1
|
|
|
|
attributes #0 = { nounwind }
|
|
attributes #1 = { nounwind readnone }
|