Summary:
When we restore an SGPR value from scratch, we first load it into a
temporary VGPR and then use v_readlane_b32 to copy the value from the
VGPR back into an SGPR.
We weren't setting the kill flag on the VGPR in the v_readlane_b32
instruction, so the register scavenger wasn't able to re-use this
temp value later.
I wasn't able to create a lit test for this.
Reviewers: arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D19744
llvm-svn: 268287
.addMemOperand(MMO);
BuildMI(*MBB, MI, DL,
TII->getMCOpcodeFromPseudo(AMDGPU::V_READLANE_B32), SubReg)
- .addReg(TmpReg)
+ .addReg(TmpReg, RegState::Kill)
.addImm(0)
.addReg(MI->getOperand(0).getReg(), RegState::ImplicitDefine);
}