maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

[AMDGPU] Fix Livereg computation during epilogue insertion 

The LivePhysRegs calculated in order to find a scratch register in the
epilogue code wrongly uses 'LiveIns'. Instead, it should use the
'Liveout' sets.  For the liveness, also considering the operands of
the terminator (return) instruction which is the insertion point for
the scratch-exec-copy instruction.

Patch by Christudasan Devadasan

llvm-svn: 364470
This commit is contained in:
Matt Arsenault 2019-06-26 20:35:18 +00:00
parent 3d12971e1c
commit 6a87e0fc6a
2 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
llvm/lib/Target/AMDGPU/SIFrameLowering.cpp
View File

@ -678,7 +678,8 @@ void SIFrameLowering::emitEpilogue(MachineFunction &MF,
if (ScratchExecCopy == AMDGPU::NoRegister) {
// See emitPrologue
LivePhysRegs LiveRegs(*ST.getRegisterInfo());
LiveRegs.addLiveIns(MBB);
LiveRegs.addLiveOuts(MBB);
LiveRegs.stepBackward(*MBBI);
ScratchExecCopy
= findScratchNonCalleeSaveRegister(MF, LiveRegs,

1
llvm/test/CodeGen/AMDGPU/callee-special-input-sgprs.ll
View File

@ -489,6 +489,7 @@ define amdgpu_kernel void @kern_indirect_use_every_sgpr_input() #1 {
; GCN-NOT: s[8:9]
; GCN-NOT: s[10:11]
; GCN-NOT: s[12:13]
; GCN: s_or_saveexec_b64 s[6:7], -1
define hidden void @func_indirect_use_every_sgpr_input() #1 {
call void @use_every_sgpr_input()
ret void