maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/AMDGPU/empty-function.ll

22 lines
585 B
LLVM
Raw Normal View History

AMDGPU: Remove -mcpu=SI Leftover from before amdgcn/r600 split. llvm-svn: 310277
2017-08-08 02:30:35 +08:00
; RUN: llc -march=amdgcn -mcpu=tahiti -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
R600/SI: Enable all tests that pass on VI without changes llvm-svn: 227214
2015-01-28 01:27:15 +08:00
; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -check-prefix=SI %s
R600: Fix assert on empty function If a function is just an unreachable, this would hit a "this is not a MachO target" assertion because of setting HasSubsectionViaSymbols. llvm-svn: 221920
2014-11-14 04:07:40 +08:00
; Make sure we don't assert on empty functions
; SI: .text
Call EmitFunctionHeader just before EmitFunctionBody. This avoids switching to .AMDGPU.config and back and hardcoding the section it switches back to. llvm-svn: 232479
2015-03-17 22:34:42 +08:00
; SI-LABEL: {{^}}empty_function_ret:
R600: Fix assert on empty function If a function is just an unreachable, this would hit a "this is not a MachO target" assertion because of setting HasSubsectionViaSymbols. llvm-svn: 221920
2014-11-14 04:07:40 +08:00
; SI: s_endpgm
; SI: codeLenInByte = 4
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
2017-03-22 05:39:51 +08:00
define amdgpu_kernel void @empty_function_ret() #0 {
R600: Fix assert on empty function If a function is just an unreachable, this would hit a "this is not a MachO target" assertion because of setting HasSubsectionViaSymbols. llvm-svn: 221920
2014-11-14 04:07:40 +08:00
ret void
}
; SI: .text
Call EmitFunctionHeader just before EmitFunctionBody. This avoids switching to .AMDGPU.config and back and hardcoding the section it switches back to. llvm-svn: 232479
2015-03-17 22:34:42 +08:00
; SI-LABEL: {{^}}empty_function_unreachable:
R600: Fix assert on empty function If a function is just an unreachable, this would hit a "this is not a MachO target" assertion because of setting HasSubsectionViaSymbols. llvm-svn: 221920
2014-11-14 04:07:40 +08:00
; SI: codeLenInByte = 0
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
2017-03-22 05:39:51 +08:00
define amdgpu_kernel void @empty_function_unreachable() #0 {
R600: Fix assert on empty function If a function is just an unreachable, this would hit a "this is not a MachO target" assertion because of setting HasSubsectionViaSymbols. llvm-svn: 221920
2014-11-14 04:07:40 +08:00
unreachable
}
attributes #0 = { nounwind }