diff options
author | Hans Wennborg <hans@hanshq.net> | 2018-01-30 15:29:20 +0000 |
---|---|---|
committer | Hans Wennborg <hans@hanshq.net> | 2018-01-30 15:29:20 +0000 |
commit | 7d30102ce03bcf32928e3d0a35ae533638f5b905 (patch) | |
tree | 724a0cfe07463031948177ec9d2448e783ceda3a /lib | |
parent | 0ab6eba6b84235523cbb11f4c2a0b98016d1df75 (diff) |
Merging r323706:
------------------------------------------------------------------------
r323706 | mareko | 2018-01-30 00:19:10 +0100 (Tue, 30 Jan 2018) | 15 lines
AMDGPU: Allow a SGPR for the conditional KILL operand
Patch by: Bas Nieuwenhuizen
Just use the _e64 variant if needed. This should be possible as per
def : Pat <
(int_amdgcn_kill (i1 (setcc f32:$src, InlineFPImm<f32>:$imm, cond:$cond))),
(SI_KILL_F32_COND_IMM_PSEUDO $src, (bitcast_fpimm_to_i32 $imm), (cond_as_i32imm $cond))
> ;
I don't think we can get an immediate for the other operand for which we
need the second 32-bit word.
https://reviews.llvm.org/D42302
------------------------------------------------------------------------
git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_60@323772 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib')
-rw-r--r-- | lib/Target/AMDGPU/SIInsertSkips.cpp | 54 |
1 files changed, 31 insertions, 23 deletions
diff --git a/lib/Target/AMDGPU/SIInsertSkips.cpp b/lib/Target/AMDGPU/SIInsertSkips.cpp index a2f844d7854..eb7277b7a5b 100644 --- a/lib/Target/AMDGPU/SIInsertSkips.cpp +++ b/lib/Target/AMDGPU/SIInsertSkips.cpp @@ -210,65 +210,73 @@ void SIInsertSkips::kill(MachineInstr &MI) { switch (MI.getOperand(2).getImm()) { case ISD::SETOEQ: case ISD::SETEQ: - Opcode = AMDGPU::V_CMPX_EQ_F32_e32; + Opcode = AMDGPU::V_CMPX_EQ_F32_e64; break; case ISD::SETOGT: case ISD::SETGT: - Opcode = AMDGPU::V_CMPX_LT_F32_e32; + Opcode = AMDGPU::V_CMPX_LT_F32_e64; break; case ISD::SETOGE: case ISD::SETGE: - Opcode = AMDGPU::V_CMPX_LE_F32_e32; + Opcode = AMDGPU::V_CMPX_LE_F32_e64; break; case ISD::SETOLT: case ISD::SETLT: - Opcode = AMDGPU::V_CMPX_GT_F32_e32; + Opcode = AMDGPU::V_CMPX_GT_F32_e64; break; case ISD::SETOLE: case ISD::SETLE: - Opcode = AMDGPU::V_CMPX_GE_F32_e32; + Opcode = AMDGPU::V_CMPX_GE_F32_e64; break; case ISD::SETONE: case ISD::SETNE: - Opcode = AMDGPU::V_CMPX_LG_F32_e32; + Opcode = AMDGPU::V_CMPX_LG_F32_e64; break; case ISD::SETO: - Opcode = AMDGPU::V_CMPX_O_F32_e32; + Opcode = AMDGPU::V_CMPX_O_F32_e64; break; case ISD::SETUO: - Opcode = AMDGPU::V_CMPX_U_F32_e32; + Opcode = AMDGPU::V_CMPX_U_F32_e64; break; case ISD::SETUEQ: - Opcode = AMDGPU::V_CMPX_NLG_F32_e32; + Opcode = AMDGPU::V_CMPX_NLG_F32_e64; break; case ISD::SETUGT: - Opcode = AMDGPU::V_CMPX_NGE_F32_e32; + Opcode = AMDGPU::V_CMPX_NGE_F32_e64; break; case ISD::SETUGE: - Opcode = AMDGPU::V_CMPX_NGT_F32_e32; + Opcode = AMDGPU::V_CMPX_NGT_F32_e64; break; case ISD::SETULT: - Opcode = AMDGPU::V_CMPX_NLE_F32_e32; + Opcode = AMDGPU::V_CMPX_NLE_F32_e64; break; case ISD::SETULE: - Opcode = AMDGPU::V_CMPX_NLT_F32_e32; + Opcode = AMDGPU::V_CMPX_NLT_F32_e64; break; case ISD::SETUNE: - Opcode = AMDGPU::V_CMPX_NEQ_F32_e32; + Opcode = AMDGPU::V_CMPX_NEQ_F32_e64; break; default: llvm_unreachable("invalid ISD:SET cond code"); } - // TODO: Allow this: - if (!MI.getOperand(0).isReg() || - !TRI->isVGPR(MBB.getParent()->getRegInfo(), - MI.getOperand(0).getReg())) - llvm_unreachable("SI_KILL operand should be a VGPR"); - - BuildMI(MBB, &MI, DL, TII->get(Opcode)) - .add(MI.getOperand(1)) - .add(MI.getOperand(0)); + assert(MI.getOperand(0).isReg()); + + if (TRI->isVGPR(MBB.getParent()->getRegInfo(), + MI.getOperand(0).getReg())) { + Opcode = AMDGPU::getVOPe32(Opcode); + BuildMI(MBB, &MI, DL, TII->get(Opcode)) + .add(MI.getOperand(1)) + .add(MI.getOperand(0)); + } else { + BuildMI(MBB, &MI, DL, TII->get(Opcode)) + .addReg(AMDGPU::VCC, RegState::Define) + .addImm(0) // src0 modifiers + .add(MI.getOperand(1)) + .addImm(0) // src1 modifiers + .add(MI.getOperand(0)) + .addImm(0); // omod + } break; } case AMDGPU::SI_KILL_I1_TERMINATOR: { |