[AMDGPU] frame index elimination hit assertion for scavenged nonreg (#130287)

Avoid spilling and fold the offset into `FIOp` when the scavenger fails
to find a free reg during frame index elimination.

Fixes #130120
This commit is contained in:
Pankaj Dwivedi 2025-03-17 19:33:07 +05:30 committed by GitHub
parent 19adc69029
commit 1c3a9a853c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 230 additions and 16 deletions

View File

@ -2769,12 +2769,15 @@ bool SIRegisterInfo::eliminateFrameIndex(MachineBasicBlock::iterator MI,
// available.
if (!TmpReg)
TmpReg = RS->scavengeRegisterBackwards(AMDGPU::SReg_32_XM0RegClass,
MI, false, 0);
BuildMI(*MBB, *MI, DL, TII->get(AMDGPU::S_LSHR_B32))
.addDef(TmpReg, RegState::Renamable)
.addReg(FrameReg)
.addImm(ST.getWavefrontSizeLog2())
.setOperandDead(3); // Set SCC dead
MI, /*RestoreAfter=*/false, 0,
/*AllowSpill=*/false);
if (TmpReg) {
BuildMI(*MBB, *MI, DL, TII->get(AMDGPU::S_LSHR_B32))
.addDef(TmpReg, RegState::Renamable)
.addReg(FrameReg)
.addImm(ST.getWavefrontSizeLog2())
.setOperandDead(3); // Set SCC dead
}
MaterializedReg = TmpReg;
}
@ -2802,18 +2805,20 @@ bool SIRegisterInfo::eliminateFrameIndex(MachineBasicBlock::iterator MI,
DstReg = TmpReg;
}
auto AddI32 = BuildMI(*MBB, *MI, DL, MI->getDesc())
.addDef(DstReg, RegState::Renamable)
.addReg(MaterializedReg, RegState::Kill)
.add(OtherOp);
if (DeadSCC)
AddI32.setOperandDead(3);
if (TmpReg) {
auto AddI32 = BuildMI(*MBB, *MI, DL, MI->getDesc())
.addDef(DstReg, RegState::Renamable)
.addReg(MaterializedReg, RegState::Kill)
.add(OtherOp);
if (DeadSCC)
AddI32.setOperandDead(3);
MaterializedReg = DstReg;
MaterializedReg = DstReg;
OtherOp.ChangeToRegister(MaterializedReg, false);
OtherOp.setIsKill(true);
OtherOp.setIsRenamable(true);
OtherOp.ChangeToRegister(MaterializedReg, false);
OtherOp.setIsKill(true);
OtherOp.setIsRenamable(true);
}
FIOp->ChangeToImmediate(Offset);
} else {
// If we don't have any other offset to apply, we can just directly

View File

@ -1509,3 +1509,84 @@ body: |
SI_RETURN implicit $vcc_lo, implicit $vcc_hi, implicit $sgpr0, implicit $sgpr1, implicit $sgpr2, implicit $sgpr3, implicit $sgpr4, implicit $sgpr5, implicit $sgpr6, implicit $sgpr7, implicit $sgpr8, implicit $sgpr9, implicit $sgpr10, implicit $sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $sgpr16, implicit $sgpr17, implicit $sgpr18, implicit $sgpr19, implicit $sgpr20, implicit $sgpr21, implicit $sgpr22, implicit $sgpr23, implicit $sgpr24, implicit $sgpr25, implicit $sgpr26, implicit $sgpr27, implicit $sgpr28, implicit $sgpr29
...
---
name: s_add_i32_frame_index_elimination_no_sgpr_dead_scc_offset_zero
tracksRegLiveness: true
stack:
- { id: 0, size: 1, alignment: 4, local-offset: 0 }
machineFunctionInfo:
scratchRSrcReg: '$sgpr0_sgpr1_sgpr2_sgpr3'
frameOffsetReg: '$sgpr33'
stackPtrOffsetReg: '$sgpr32'
body: |
bb.0:
liveins: $vcc_lo, $vcc_hi, $sgpr0, $sgpr1, $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $sgpr7, $sgpr8, $sgpr9, $sgpr10, $sgpr11, $sgpr12, $sgpr13, $sgpr14, $sgpr15, $sgpr16, $sgpr17, $sgpr18, $sgpr19, $sgpr20, $sgpr21, $sgpr22, $sgpr23, $sgpr24, $sgpr25, $sgpr26, $sgpr27, $sgpr28, $sgpr29, $sgpr30, $sgpr31, $sgpr32, $sgpr33, $sgpr34, $sgpr35, $sgpr36, $sgpr37, $sgpr38, $sgpr39, $sgpr40, $sgpr41, $sgpr42, $sgpr43, $sgpr44, $sgpr45, $sgpr46, $sgpr47, $sgpr48, $sgpr49, $sgpr50, $sgpr51, $sgpr52, $sgpr53, $sgpr54, $sgpr55, $sgpr56, $sgpr57, $sgpr58, $sgpr59, $sgpr60, $sgpr61, $sgpr62, $sgpr63, $sgpr64, $sgpr65, $sgpr66, $sgpr67, $sgpr68, $sgpr69, $sgpr70, $sgpr71, $sgpr72, $sgpr73, $sgpr74, $sgpr75, $sgpr76, $sgpr77, $sgpr78, $sgpr79, $sgpr80, $sgpr81, $sgpr82, $sgpr83, $sgpr84, $sgpr85, $sgpr86, $sgpr87, $sgpr88, $sgpr89, $sgpr90, $sgpr91, $sgpr92, $sgpr93, $sgpr94, $sgpr95
; MUBUFW64-LABEL: name: s_add_i32_frame_index_elimination_no_sgpr_dead_scc_offset_zero
; MUBUFW64: liveins: $vcc_lo, $vcc_hi, $sgpr0, $sgpr1, $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $sgpr7, $sgpr8, $sgpr9, $sgpr10, $sgpr11, $sgpr12, $sgpr13, $sgpr14, $sgpr15, $sgpr16, $sgpr17, $sgpr18, $sgpr19, $sgpr20, $sgpr21, $sgpr22, $sgpr23, $sgpr24, $sgpr25, $sgpr26, $sgpr27, $sgpr28, $sgpr29, $sgpr30, $sgpr31, $sgpr32, $sgpr33, $sgpr34, $sgpr35, $sgpr36, $sgpr37, $sgpr38, $sgpr39, $sgpr40, $sgpr41, $sgpr42, $sgpr43, $sgpr44, $sgpr45, $sgpr46, $sgpr47, $sgpr48, $sgpr49, $sgpr50, $sgpr51, $sgpr52, $sgpr53, $sgpr54, $sgpr55, $sgpr56, $sgpr57, $sgpr58, $sgpr59, $sgpr60, $sgpr61, $sgpr62, $sgpr63, $sgpr64, $sgpr65, $sgpr66, $sgpr67, $sgpr68, $sgpr69, $sgpr70, $sgpr71, $sgpr72, $sgpr73, $sgpr74, $sgpr75, $sgpr76, $sgpr77, $sgpr78, $sgpr79, $sgpr80, $sgpr81, $sgpr82, $sgpr83, $sgpr84, $sgpr85, $sgpr86, $sgpr87, $sgpr88, $sgpr89, $sgpr90, $sgpr91, $sgpr92, $sgpr93, $sgpr94, $sgpr95
; MUBUFW64-NEXT: {{ $}}
; MUBUFW64-NEXT: renamable $sgpr7 = COPY renamable $sgpr7
; MUBUFW64-NEXT: SI_RETURN implicit $vcc_lo, implicit $vcc_hi, implicit $sgpr0, implicit $sgpr1, implicit $sgpr2, implicit $sgpr3, implicit $sgpr4, implicit $sgpr5, implicit $sgpr6, implicit $sgpr7, implicit $sgpr8, implicit $sgpr9, implicit $sgpr10, implicit $sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $sgpr16, implicit $sgpr17, implicit $sgpr18, implicit $sgpr19, implicit $sgpr20, implicit $sgpr21, implicit $sgpr22, implicit $sgpr23, implicit $sgpr24, implicit $sgpr25, implicit $sgpr26, implicit $sgpr27, implicit $sgpr28, implicit $sgpr29, implicit $sgpr30, implicit $sgpr31, implicit $sgpr32, implicit $sgpr33, implicit $sgpr34, implicit $sgpr35, implicit $sgpr36, implicit $sgpr37, implicit $sgpr38, implicit $sgpr39, implicit $sgpr40, implicit $sgpr41, implicit $sgpr42, implicit $sgpr43, implicit $sgpr44, implicit $sgpr45, implicit $sgpr46, implicit $sgpr47, implicit $sgpr48, implicit $sgpr49, implicit $sgpr50, implicit $sgpr51, implicit $sgpr52, implicit $sgpr53, implicit $sgpr54, implicit $sgpr55, implicit $sgpr56, implicit $sgpr57, implicit $sgpr58, implicit $sgpr59, implicit $sgpr60, implicit $sgpr61, implicit $sgpr62, implicit $sgpr63, implicit $sgpr64, implicit $sgpr65, implicit $sgpr66, implicit $sgpr67, implicit $sgpr68, implicit $sgpr69, implicit $sgpr70, implicit $sgpr71, implicit $sgpr72, implicit $sgpr73, implicit $sgpr74, implicit $sgpr75, implicit $sgpr76, implicit $sgpr77, implicit $sgpr78, implicit $sgpr79, implicit $sgpr80, implicit $sgpr81, implicit $sgpr82, implicit $sgpr83, implicit $sgpr84, implicit $sgpr85, implicit $sgpr86, implicit $sgpr87, implicit $sgpr88, implicit $sgpr89, implicit $sgpr90, implicit $sgpr91, implicit $sgpr92, implicit $sgpr93, implicit $sgpr94, implicit $sgpr95
;
; MUBUFW32-LABEL: name: s_add_i32_frame_index_elimination_no_sgpr_dead_scc_offset_zero
; MUBUFW32: liveins: $vcc_lo, $vcc_hi, $sgpr0, $sgpr1, $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $sgpr7, $sgpr8, $sgpr9, $sgpr10, $sgpr11, $sgpr12, $sgpr13, $sgpr14, $sgpr15, $sgpr16, $sgpr17, $sgpr18, $sgpr19, $sgpr20, $sgpr21, $sgpr22, $sgpr23, $sgpr24, $sgpr25, $sgpr26, $sgpr27, $sgpr28, $sgpr29, $sgpr30, $sgpr31, $sgpr32, $sgpr33, $sgpr34, $sgpr35, $sgpr36, $sgpr37, $sgpr38, $sgpr39, $sgpr40, $sgpr41, $sgpr42, $sgpr43, $sgpr44, $sgpr45, $sgpr46, $sgpr47, $sgpr48, $sgpr49, $sgpr50, $sgpr51, $sgpr52, $sgpr53, $sgpr54, $sgpr55, $sgpr56, $sgpr57, $sgpr58, $sgpr59, $sgpr60, $sgpr61, $sgpr62, $sgpr63, $sgpr64, $sgpr65, $sgpr66, $sgpr67, $sgpr68, $sgpr69, $sgpr70, $sgpr71, $sgpr72, $sgpr73, $sgpr74, $sgpr75, $sgpr76, $sgpr77, $sgpr78, $sgpr79, $sgpr80, $sgpr81, $sgpr82, $sgpr83, $sgpr84, $sgpr85, $sgpr86, $sgpr87, $sgpr88, $sgpr89, $sgpr90, $sgpr91, $sgpr92, $sgpr93, $sgpr94, $sgpr95
; MUBUFW32-NEXT: {{ $}}
; MUBUFW32-NEXT: renamable $sgpr7 = COPY renamable $sgpr7
; MUBUFW32-NEXT: SI_RETURN implicit $vcc_lo, implicit $vcc_hi, implicit $sgpr0, implicit $sgpr1, implicit $sgpr2, implicit $sgpr3, implicit $sgpr4, implicit $sgpr5, implicit $sgpr6, implicit $sgpr7, implicit $sgpr8, implicit $sgpr9, implicit $sgpr10, implicit $sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $sgpr16, implicit $sgpr17, implicit $sgpr18, implicit $sgpr19, implicit $sgpr20, implicit $sgpr21, implicit $sgpr22, implicit $sgpr23, implicit $sgpr24, implicit $sgpr25, implicit $sgpr26, implicit $sgpr27, implicit $sgpr28, implicit $sgpr29, implicit $sgpr30, implicit $sgpr31, implicit $sgpr32, implicit $sgpr33, implicit $sgpr34, implicit $sgpr35, implicit $sgpr36, implicit $sgpr37, implicit $sgpr38, implicit $sgpr39, implicit $sgpr40, implicit $sgpr41, implicit $sgpr42, implicit $sgpr43, implicit $sgpr44, implicit $sgpr45, implicit $sgpr46, implicit $sgpr47, implicit $sgpr48, implicit $sgpr49, implicit $sgpr50, implicit $sgpr51, implicit $sgpr52, implicit $sgpr53, implicit $sgpr54, implicit $sgpr55, implicit $sgpr56, implicit $sgpr57, implicit $sgpr58, implicit $sgpr59, implicit $sgpr60, implicit $sgpr61, implicit $sgpr62, implicit $sgpr63, implicit $sgpr64, implicit $sgpr65, implicit $sgpr66, implicit $sgpr67, implicit $sgpr68, implicit $sgpr69, implicit $sgpr70, implicit $sgpr71, implicit $sgpr72, implicit $sgpr73, implicit $sgpr74, implicit $sgpr75, implicit $sgpr76, implicit $sgpr77, implicit $sgpr78, implicit $sgpr79, implicit $sgpr80, implicit $sgpr81, implicit $sgpr82, implicit $sgpr83, implicit $sgpr84, implicit $sgpr85, implicit $sgpr86, implicit $sgpr87, implicit $sgpr88, implicit $sgpr89, implicit $sgpr90, implicit $sgpr91, implicit $sgpr92, implicit $sgpr93, implicit $sgpr94, implicit $sgpr95
;
; FLATSCRW64-LABEL: name: s_add_i32_frame_index_elimination_no_sgpr_dead_scc_offset_zero
; FLATSCRW64: liveins: $vcc_lo, $vcc_hi, $sgpr0, $sgpr1, $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $sgpr7, $sgpr8, $sgpr9, $sgpr10, $sgpr11, $sgpr12, $sgpr13, $sgpr14, $sgpr15, $sgpr16, $sgpr17, $sgpr18, $sgpr19, $sgpr20, $sgpr21, $sgpr22, $sgpr23, $sgpr24, $sgpr25, $sgpr26, $sgpr27, $sgpr28, $sgpr29, $sgpr30, $sgpr31, $sgpr32, $sgpr33, $sgpr34, $sgpr35, $sgpr36, $sgpr37, $sgpr38, $sgpr39, $sgpr40, $sgpr41, $sgpr42, $sgpr43, $sgpr44, $sgpr45, $sgpr46, $sgpr47, $sgpr48, $sgpr49, $sgpr50, $sgpr51, $sgpr52, $sgpr53, $sgpr54, $sgpr55, $sgpr56, $sgpr57, $sgpr58, $sgpr59, $sgpr60, $sgpr61, $sgpr62, $sgpr63, $sgpr64, $sgpr65, $sgpr66, $sgpr67, $sgpr68, $sgpr69, $sgpr70, $sgpr71, $sgpr72, $sgpr73, $sgpr74, $sgpr75, $sgpr76, $sgpr77, $sgpr78, $sgpr79, $sgpr80, $sgpr81, $sgpr82, $sgpr83, $sgpr84, $sgpr85, $sgpr86, $sgpr87, $sgpr88, $sgpr89, $sgpr90, $sgpr91, $sgpr92, $sgpr93, $sgpr94, $sgpr95
; FLATSCRW64-NEXT: {{ $}}
; FLATSCRW64-NEXT: renamable $sgpr7 = COPY renamable $sgpr7
; FLATSCRW64-NEXT: SI_RETURN implicit $vcc_lo, implicit $vcc_hi, implicit $sgpr0, implicit $sgpr1, implicit $sgpr2, implicit $sgpr3, implicit $sgpr4, implicit $sgpr5, implicit $sgpr6, implicit $sgpr7, implicit $sgpr8, implicit $sgpr9, implicit $sgpr10, implicit $sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $sgpr16, implicit $sgpr17, implicit $sgpr18, implicit $sgpr19, implicit $sgpr20, implicit $sgpr21, implicit $sgpr22, implicit $sgpr23, implicit $sgpr24, implicit $sgpr25, implicit $sgpr26, implicit $sgpr27, implicit $sgpr28, implicit $sgpr29, implicit $sgpr30, implicit $sgpr31, implicit $sgpr32, implicit $sgpr33, implicit $sgpr34, implicit $sgpr35, implicit $sgpr36, implicit $sgpr37, implicit $sgpr38, implicit $sgpr39, implicit $sgpr40, implicit $sgpr41, implicit $sgpr42, implicit $sgpr43, implicit $sgpr44, implicit $sgpr45, implicit $sgpr46, implicit $sgpr47, implicit $sgpr48, implicit $sgpr49, implicit $sgpr50, implicit $sgpr51, implicit $sgpr52, implicit $sgpr53, implicit $sgpr54, implicit $sgpr55, implicit $sgpr56, implicit $sgpr57, implicit $sgpr58, implicit $sgpr59, implicit $sgpr60, implicit $sgpr61, implicit $sgpr62, implicit $sgpr63, implicit $sgpr64, implicit $sgpr65, implicit $sgpr66, implicit $sgpr67, implicit $sgpr68, implicit $sgpr69, implicit $sgpr70, implicit $sgpr71, implicit $sgpr72, implicit $sgpr73, implicit $sgpr74, implicit $sgpr75, implicit $sgpr76, implicit $sgpr77, implicit $sgpr78, implicit $sgpr79, implicit $sgpr80, implicit $sgpr81, implicit $sgpr82, implicit $sgpr83, implicit $sgpr84, implicit $sgpr85, implicit $sgpr86, implicit $sgpr87, implicit $sgpr88, implicit $sgpr89, implicit $sgpr90, implicit $sgpr91, implicit $sgpr92, implicit $sgpr93, implicit $sgpr94, implicit $sgpr95
;
; FLATSCRW32-LABEL: name: s_add_i32_frame_index_elimination_no_sgpr_dead_scc_offset_zero
; FLATSCRW32: liveins: $vcc_lo, $vcc_hi, $sgpr0, $sgpr1, $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $sgpr7, $sgpr8, $sgpr9, $sgpr10, $sgpr11, $sgpr12, $sgpr13, $sgpr14, $sgpr15, $sgpr16, $sgpr17, $sgpr18, $sgpr19, $sgpr20, $sgpr21, $sgpr22, $sgpr23, $sgpr24, $sgpr25, $sgpr26, $sgpr27, $sgpr28, $sgpr29, $sgpr30, $sgpr31, $sgpr32, $sgpr33, $sgpr34, $sgpr35, $sgpr36, $sgpr37, $sgpr38, $sgpr39, $sgpr40, $sgpr41, $sgpr42, $sgpr43, $sgpr44, $sgpr45, $sgpr46, $sgpr47, $sgpr48, $sgpr49, $sgpr50, $sgpr51, $sgpr52, $sgpr53, $sgpr54, $sgpr55, $sgpr56, $sgpr57, $sgpr58, $sgpr59, $sgpr60, $sgpr61, $sgpr62, $sgpr63, $sgpr64, $sgpr65, $sgpr66, $sgpr67, $sgpr68, $sgpr69, $sgpr70, $sgpr71, $sgpr72, $sgpr73, $sgpr74, $sgpr75, $sgpr76, $sgpr77, $sgpr78, $sgpr79, $sgpr80, $sgpr81, $sgpr82, $sgpr83, $sgpr84, $sgpr85, $sgpr86, $sgpr87, $sgpr88, $sgpr89, $sgpr90, $sgpr91, $sgpr92, $sgpr93, $sgpr94, $sgpr95
; FLATSCRW32-NEXT: {{ $}}
; FLATSCRW32-NEXT: renamable $sgpr7 = COPY renamable $sgpr7
; FLATSCRW32-NEXT: SI_RETURN implicit $vcc_lo, implicit $vcc_hi, implicit $sgpr0, implicit $sgpr1, implicit $sgpr2, implicit $sgpr3, implicit $sgpr4, implicit $sgpr5, implicit $sgpr6, implicit $sgpr7, implicit $sgpr8, implicit $sgpr9, implicit $sgpr10, implicit $sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $sgpr16, implicit $sgpr17, implicit $sgpr18, implicit $sgpr19, implicit $sgpr20, implicit $sgpr21, implicit $sgpr22, implicit $sgpr23, implicit $sgpr24, implicit $sgpr25, implicit $sgpr26, implicit $sgpr27, implicit $sgpr28, implicit $sgpr29, implicit $sgpr30, implicit $sgpr31, implicit $sgpr32, implicit $sgpr33, implicit $sgpr34, implicit $sgpr35, implicit $sgpr36, implicit $sgpr37, implicit $sgpr38, implicit $sgpr39, implicit $sgpr40, implicit $sgpr41, implicit $sgpr42, implicit $sgpr43, implicit $sgpr44, implicit $sgpr45, implicit $sgpr46, implicit $sgpr47, implicit $sgpr48, implicit $sgpr49, implicit $sgpr50, implicit $sgpr51, implicit $sgpr52, implicit $sgpr53, implicit $sgpr54, implicit $sgpr55, implicit $sgpr56, implicit $sgpr57, implicit $sgpr58, implicit $sgpr59, implicit $sgpr60, implicit $sgpr61, implicit $sgpr62, implicit $sgpr63, implicit $sgpr64, implicit $sgpr65, implicit $sgpr66, implicit $sgpr67, implicit $sgpr68, implicit $sgpr69, implicit $sgpr70, implicit $sgpr71, implicit $sgpr72, implicit $sgpr73, implicit $sgpr74, implicit $sgpr75, implicit $sgpr76, implicit $sgpr77, implicit $sgpr78, implicit $sgpr79, implicit $sgpr80, implicit $sgpr81, implicit $sgpr82, implicit $sgpr83, implicit $sgpr84, implicit $sgpr85, implicit $sgpr86, implicit $sgpr87, implicit $sgpr88, implicit $sgpr89, implicit $sgpr90, implicit $sgpr91, implicit $sgpr92, implicit $sgpr93, implicit $sgpr94, implicit $sgpr95
renamable $sgpr7 = S_ADD_I32 renamable $sgpr7, %stack.0, implicit-def dead $scc
SI_RETURN implicit $vcc_lo, implicit $vcc_hi, implicit $sgpr0, implicit $sgpr1, implicit $sgpr2, implicit $sgpr3, implicit $sgpr4, implicit $sgpr5, implicit $sgpr6, implicit $sgpr7, implicit $sgpr8, implicit $sgpr9, implicit $sgpr10, implicit $sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $sgpr16, implicit $sgpr17, implicit $sgpr18, implicit $sgpr19, implicit $sgpr20, implicit $sgpr21, implicit $sgpr22, implicit $sgpr23, implicit $sgpr24, implicit $sgpr25, implicit $sgpr26, implicit $sgpr27, implicit $sgpr28, implicit $sgpr29, implicit $sgpr30, implicit $sgpr31, implicit $sgpr32, implicit $sgpr33, implicit $sgpr34, implicit $sgpr35, implicit $sgpr36, implicit $sgpr37, implicit $sgpr38, implicit $sgpr39, implicit $sgpr40, implicit $sgpr41, implicit $sgpr42, implicit $sgpr43, implicit $sgpr44, implicit $sgpr45, implicit $sgpr46, implicit $sgpr47, implicit $sgpr48, implicit $sgpr49, implicit $sgpr50, implicit $sgpr51, implicit $sgpr52, implicit $sgpr53, implicit $sgpr54, implicit $sgpr55, implicit $sgpr56, implicit $sgpr57, implicit $sgpr58, implicit $sgpr59, implicit $sgpr60, implicit $sgpr61, implicit $sgpr62, implicit $sgpr63, implicit $sgpr64, implicit $sgpr65, implicit $sgpr66, implicit $sgpr67, implicit $sgpr68, implicit $sgpr69, implicit $sgpr70, implicit $sgpr71, implicit $sgpr72, implicit $sgpr73, implicit $sgpr74, implicit $sgpr75, implicit $sgpr76, implicit $sgpr77, implicit $sgpr78, implicit $sgpr79, implicit $sgpr80, implicit $sgpr81, implicit $sgpr82, implicit $sgpr83, implicit $sgpr84, implicit $sgpr85, implicit $sgpr86, implicit $sgpr87, implicit $sgpr88, implicit $sgpr89, implicit $sgpr90, implicit $sgpr91, implicit $sgpr92, implicit $sgpr93, implicit $sgpr94, implicit $sgpr95
...
---
name: s_add_i32_frame_index_elimination_no_sgpr_dead_scc_offset_non_zero
tracksRegLiveness: true
stack:
- { id: 0, type: default, size: 64, alignment: 16, stack-id: default }
- { id: 1, type: default, size: 4, alignment: 4, stack-id: default }
machineFunctionInfo:
scratchRSrcReg: '$sgpr0_sgpr1_sgpr2_sgpr3'
frameOffsetReg: '$sgpr33'
stackPtrOffsetReg: '$sgpr32'
body: |
bb.0:
liveins: $vcc_lo, $vcc_hi, $sgpr0, $sgpr1, $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $sgpr7, $sgpr8, $sgpr9, $sgpr10, $sgpr11, $sgpr12, $sgpr13, $sgpr14, $sgpr15, $sgpr16, $sgpr17, $sgpr18, $sgpr19, $sgpr20, $sgpr21, $sgpr22, $sgpr23, $sgpr24, $sgpr25, $sgpr26, $sgpr27, $sgpr28, $sgpr29, $sgpr30, $sgpr31, $sgpr32, $sgpr33, $sgpr34, $sgpr35, $sgpr36, $sgpr37, $sgpr38, $sgpr39, $sgpr40, $sgpr41, $sgpr42, $sgpr43, $sgpr44, $sgpr45, $sgpr46, $sgpr47, $sgpr48, $sgpr49, $sgpr50, $sgpr51, $sgpr52, $sgpr53, $sgpr54, $sgpr55, $sgpr56, $sgpr57, $sgpr58, $sgpr59, $sgpr60, $sgpr61, $sgpr62, $sgpr63, $sgpr64, $sgpr65, $sgpr66, $sgpr67, $sgpr68, $sgpr69, $sgpr70, $sgpr71, $sgpr72, $sgpr73, $sgpr74, $sgpr75, $sgpr76, $sgpr77, $sgpr78, $sgpr79, $sgpr80, $sgpr81, $sgpr82, $sgpr83, $sgpr84, $sgpr85, $sgpr86, $sgpr87, $sgpr88, $sgpr89, $sgpr90, $sgpr91, $sgpr92, $sgpr93, $sgpr94, $sgpr95
; MUBUFW64-LABEL: name: s_add_i32_frame_index_elimination_no_sgpr_dead_scc_offset_non_zero
; MUBUFW64: liveins: $vcc_lo, $vcc_hi, $sgpr0, $sgpr1, $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $sgpr7, $sgpr8, $sgpr9, $sgpr10, $sgpr11, $sgpr12, $sgpr13, $sgpr14, $sgpr15, $sgpr16, $sgpr17, $sgpr18, $sgpr19, $sgpr20, $sgpr21, $sgpr22, $sgpr23, $sgpr24, $sgpr25, $sgpr26, $sgpr27, $sgpr28, $sgpr29, $sgpr30, $sgpr31, $sgpr32, $sgpr33, $sgpr34, $sgpr35, $sgpr36, $sgpr37, $sgpr38, $sgpr39, $sgpr40, $sgpr41, $sgpr42, $sgpr43, $sgpr44, $sgpr45, $sgpr46, $sgpr47, $sgpr48, $sgpr49, $sgpr50, $sgpr51, $sgpr52, $sgpr53, $sgpr54, $sgpr55, $sgpr56, $sgpr57, $sgpr58, $sgpr59, $sgpr60, $sgpr61, $sgpr62, $sgpr63, $sgpr64, $sgpr65, $sgpr66, $sgpr67, $sgpr68, $sgpr69, $sgpr70, $sgpr71, $sgpr72, $sgpr73, $sgpr74, $sgpr75, $sgpr76, $sgpr77, $sgpr78, $sgpr79, $sgpr80, $sgpr81, $sgpr82, $sgpr83, $sgpr84, $sgpr85, $sgpr86, $sgpr87, $sgpr88, $sgpr89, $sgpr90, $sgpr91, $sgpr92, $sgpr93, $sgpr94, $sgpr95
; MUBUFW64-NEXT: {{ $}}
; MUBUFW64-NEXT: renamable $sgpr7 = S_ADD_I32 renamable $sgpr7, 64, implicit-def dead $scc
; MUBUFW64-NEXT: SI_RETURN implicit $vcc_lo, implicit $vcc_hi, implicit $sgpr0, implicit $sgpr1, implicit $sgpr2, implicit $sgpr3, implicit $sgpr4, implicit $sgpr5, implicit $sgpr6, implicit $sgpr7, implicit $sgpr8, implicit $sgpr9, implicit $sgpr10, implicit $sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $sgpr16, implicit $sgpr17, implicit $sgpr18, implicit $sgpr19, implicit $sgpr20, implicit $sgpr21, implicit $sgpr22, implicit $sgpr23, implicit $sgpr24, implicit $sgpr25, implicit $sgpr26, implicit $sgpr27, implicit $sgpr28, implicit $sgpr29, implicit $sgpr30, implicit $sgpr31, implicit $sgpr32, implicit $sgpr33, implicit $sgpr34, implicit $sgpr35, implicit $sgpr36, implicit $sgpr37, implicit $sgpr38, implicit $sgpr39, implicit $sgpr40, implicit $sgpr41, implicit $sgpr42, implicit $sgpr43, implicit $sgpr44, implicit $sgpr45, implicit $sgpr46, implicit $sgpr47, implicit $sgpr48, implicit $sgpr49, implicit $sgpr50, implicit $sgpr51, implicit $sgpr52, implicit $sgpr53, implicit $sgpr54, implicit $sgpr55, implicit $sgpr56, implicit $sgpr57, implicit $sgpr58, implicit $sgpr59, implicit $sgpr60, implicit $sgpr61, implicit $sgpr62, implicit $sgpr63, implicit $sgpr64, implicit $sgpr65, implicit $sgpr66, implicit $sgpr67, implicit $sgpr68, implicit $sgpr69, implicit $sgpr70, implicit $sgpr71, implicit $sgpr72, implicit $sgpr73, implicit $sgpr74, implicit $sgpr75, implicit $sgpr76, implicit $sgpr77, implicit $sgpr78, implicit $sgpr79, implicit $sgpr80, implicit $sgpr81, implicit $sgpr82, implicit $sgpr83, implicit $sgpr84, implicit $sgpr85, implicit $sgpr86, implicit $sgpr87, implicit $sgpr88, implicit $sgpr89, implicit $sgpr90, implicit $sgpr91, implicit $sgpr92, implicit $sgpr93, implicit $sgpr94, implicit $sgpr95
;
; MUBUFW32-LABEL: name: s_add_i32_frame_index_elimination_no_sgpr_dead_scc_offset_non_zero
; MUBUFW32: liveins: $vcc_lo, $vcc_hi, $sgpr0, $sgpr1, $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $sgpr7, $sgpr8, $sgpr9, $sgpr10, $sgpr11, $sgpr12, $sgpr13, $sgpr14, $sgpr15, $sgpr16, $sgpr17, $sgpr18, $sgpr19, $sgpr20, $sgpr21, $sgpr22, $sgpr23, $sgpr24, $sgpr25, $sgpr26, $sgpr27, $sgpr28, $sgpr29, $sgpr30, $sgpr31, $sgpr32, $sgpr33, $sgpr34, $sgpr35, $sgpr36, $sgpr37, $sgpr38, $sgpr39, $sgpr40, $sgpr41, $sgpr42, $sgpr43, $sgpr44, $sgpr45, $sgpr46, $sgpr47, $sgpr48, $sgpr49, $sgpr50, $sgpr51, $sgpr52, $sgpr53, $sgpr54, $sgpr55, $sgpr56, $sgpr57, $sgpr58, $sgpr59, $sgpr60, $sgpr61, $sgpr62, $sgpr63, $sgpr64, $sgpr65, $sgpr66, $sgpr67, $sgpr68, $sgpr69, $sgpr70, $sgpr71, $sgpr72, $sgpr73, $sgpr74, $sgpr75, $sgpr76, $sgpr77, $sgpr78, $sgpr79, $sgpr80, $sgpr81, $sgpr82, $sgpr83, $sgpr84, $sgpr85, $sgpr86, $sgpr87, $sgpr88, $sgpr89, $sgpr90, $sgpr91, $sgpr92, $sgpr93, $sgpr94, $sgpr95
; MUBUFW32-NEXT: {{ $}}
; MUBUFW32-NEXT: renamable $sgpr7 = S_ADD_I32 renamable $sgpr7, 64, implicit-def dead $scc
; MUBUFW32-NEXT: SI_RETURN implicit $vcc_lo, implicit $vcc_hi, implicit $sgpr0, implicit $sgpr1, implicit $sgpr2, implicit $sgpr3, implicit $sgpr4, implicit $sgpr5, implicit $sgpr6, implicit $sgpr7, implicit $sgpr8, implicit $sgpr9, implicit $sgpr10, implicit $sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $sgpr16, implicit $sgpr17, implicit $sgpr18, implicit $sgpr19, implicit $sgpr20, implicit $sgpr21, implicit $sgpr22, implicit $sgpr23, implicit $sgpr24, implicit $sgpr25, implicit $sgpr26, implicit $sgpr27, implicit $sgpr28, implicit $sgpr29, implicit $sgpr30, implicit $sgpr31, implicit $sgpr32, implicit $sgpr33, implicit $sgpr34, implicit $sgpr35, implicit $sgpr36, implicit $sgpr37, implicit $sgpr38, implicit $sgpr39, implicit $sgpr40, implicit $sgpr41, implicit $sgpr42, implicit $sgpr43, implicit $sgpr44, implicit $sgpr45, implicit $sgpr46, implicit $sgpr47, implicit $sgpr48, implicit $sgpr49, implicit $sgpr50, implicit $sgpr51, implicit $sgpr52, implicit $sgpr53, implicit $sgpr54, implicit $sgpr55, implicit $sgpr56, implicit $sgpr57, implicit $sgpr58, implicit $sgpr59, implicit $sgpr60, implicit $sgpr61, implicit $sgpr62, implicit $sgpr63, implicit $sgpr64, implicit $sgpr65, implicit $sgpr66, implicit $sgpr67, implicit $sgpr68, implicit $sgpr69, implicit $sgpr70, implicit $sgpr71, implicit $sgpr72, implicit $sgpr73, implicit $sgpr74, implicit $sgpr75, implicit $sgpr76, implicit $sgpr77, implicit $sgpr78, implicit $sgpr79, implicit $sgpr80, implicit $sgpr81, implicit $sgpr82, implicit $sgpr83, implicit $sgpr84, implicit $sgpr85, implicit $sgpr86, implicit $sgpr87, implicit $sgpr88, implicit $sgpr89, implicit $sgpr90, implicit $sgpr91, implicit $sgpr92, implicit $sgpr93, implicit $sgpr94, implicit $sgpr95
;
; FLATSCRW64-LABEL: name: s_add_i32_frame_index_elimination_no_sgpr_dead_scc_offset_non_zero
; FLATSCRW64: liveins: $vcc_lo, $vcc_hi, $sgpr0, $sgpr1, $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $sgpr7, $sgpr8, $sgpr9, $sgpr10, $sgpr11, $sgpr12, $sgpr13, $sgpr14, $sgpr15, $sgpr16, $sgpr17, $sgpr18, $sgpr19, $sgpr20, $sgpr21, $sgpr22, $sgpr23, $sgpr24, $sgpr25, $sgpr26, $sgpr27, $sgpr28, $sgpr29, $sgpr30, $sgpr31, $sgpr32, $sgpr33, $sgpr34, $sgpr35, $sgpr36, $sgpr37, $sgpr38, $sgpr39, $sgpr40, $sgpr41, $sgpr42, $sgpr43, $sgpr44, $sgpr45, $sgpr46, $sgpr47, $sgpr48, $sgpr49, $sgpr50, $sgpr51, $sgpr52, $sgpr53, $sgpr54, $sgpr55, $sgpr56, $sgpr57, $sgpr58, $sgpr59, $sgpr60, $sgpr61, $sgpr62, $sgpr63, $sgpr64, $sgpr65, $sgpr66, $sgpr67, $sgpr68, $sgpr69, $sgpr70, $sgpr71, $sgpr72, $sgpr73, $sgpr74, $sgpr75, $sgpr76, $sgpr77, $sgpr78, $sgpr79, $sgpr80, $sgpr81, $sgpr82, $sgpr83, $sgpr84, $sgpr85, $sgpr86, $sgpr87, $sgpr88, $sgpr89, $sgpr90, $sgpr91, $sgpr92, $sgpr93, $sgpr94, $sgpr95
; FLATSCRW64-NEXT: {{ $}}
; FLATSCRW64-NEXT: renamable $sgpr7 = S_ADD_I32 renamable $sgpr7, 64, implicit-def dead $scc
; FLATSCRW64-NEXT: SI_RETURN implicit $vcc_lo, implicit $vcc_hi, implicit $sgpr0, implicit $sgpr1, implicit $sgpr2, implicit $sgpr3, implicit $sgpr4, implicit $sgpr5, implicit $sgpr6, implicit $sgpr7, implicit $sgpr8, implicit $sgpr9, implicit $sgpr10, implicit $sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $sgpr16, implicit $sgpr17, implicit $sgpr18, implicit $sgpr19, implicit $sgpr20, implicit $sgpr21, implicit $sgpr22, implicit $sgpr23, implicit $sgpr24, implicit $sgpr25, implicit $sgpr26, implicit $sgpr27, implicit $sgpr28, implicit $sgpr29, implicit $sgpr30, implicit $sgpr31, implicit $sgpr32, implicit $sgpr33, implicit $sgpr34, implicit $sgpr35, implicit $sgpr36, implicit $sgpr37, implicit $sgpr38, implicit $sgpr39, implicit $sgpr40, implicit $sgpr41, implicit $sgpr42, implicit $sgpr43, implicit $sgpr44, implicit $sgpr45, implicit $sgpr46, implicit $sgpr47, implicit $sgpr48, implicit $sgpr49, implicit $sgpr50, implicit $sgpr51, implicit $sgpr52, implicit $sgpr53, implicit $sgpr54, implicit $sgpr55, implicit $sgpr56, implicit $sgpr57, implicit $sgpr58, implicit $sgpr59, implicit $sgpr60, implicit $sgpr61, implicit $sgpr62, implicit $sgpr63, implicit $sgpr64, implicit $sgpr65, implicit $sgpr66, implicit $sgpr67, implicit $sgpr68, implicit $sgpr69, implicit $sgpr70, implicit $sgpr71, implicit $sgpr72, implicit $sgpr73, implicit $sgpr74, implicit $sgpr75, implicit $sgpr76, implicit $sgpr77, implicit $sgpr78, implicit $sgpr79, implicit $sgpr80, implicit $sgpr81, implicit $sgpr82, implicit $sgpr83, implicit $sgpr84, implicit $sgpr85, implicit $sgpr86, implicit $sgpr87, implicit $sgpr88, implicit $sgpr89, implicit $sgpr90, implicit $sgpr91, implicit $sgpr92, implicit $sgpr93, implicit $sgpr94, implicit $sgpr95
;
; FLATSCRW32-LABEL: name: s_add_i32_frame_index_elimination_no_sgpr_dead_scc_offset_non_zero
; FLATSCRW32: liveins: $vcc_lo, $vcc_hi, $sgpr0, $sgpr1, $sgpr2, $sgpr3, $sgpr4, $sgpr5, $sgpr6, $sgpr7, $sgpr8, $sgpr9, $sgpr10, $sgpr11, $sgpr12, $sgpr13, $sgpr14, $sgpr15, $sgpr16, $sgpr17, $sgpr18, $sgpr19, $sgpr20, $sgpr21, $sgpr22, $sgpr23, $sgpr24, $sgpr25, $sgpr26, $sgpr27, $sgpr28, $sgpr29, $sgpr30, $sgpr31, $sgpr32, $sgpr33, $sgpr34, $sgpr35, $sgpr36, $sgpr37, $sgpr38, $sgpr39, $sgpr40, $sgpr41, $sgpr42, $sgpr43, $sgpr44, $sgpr45, $sgpr46, $sgpr47, $sgpr48, $sgpr49, $sgpr50, $sgpr51, $sgpr52, $sgpr53, $sgpr54, $sgpr55, $sgpr56, $sgpr57, $sgpr58, $sgpr59, $sgpr60, $sgpr61, $sgpr62, $sgpr63, $sgpr64, $sgpr65, $sgpr66, $sgpr67, $sgpr68, $sgpr69, $sgpr70, $sgpr71, $sgpr72, $sgpr73, $sgpr74, $sgpr75, $sgpr76, $sgpr77, $sgpr78, $sgpr79, $sgpr80, $sgpr81, $sgpr82, $sgpr83, $sgpr84, $sgpr85, $sgpr86, $sgpr87, $sgpr88, $sgpr89, $sgpr90, $sgpr91, $sgpr92, $sgpr93, $sgpr94, $sgpr95
; FLATSCRW32-NEXT: {{ $}}
; FLATSCRW32-NEXT: renamable $sgpr7 = S_ADD_I32 renamable $sgpr7, 64, implicit-def dead $scc
; FLATSCRW32-NEXT: SI_RETURN implicit $vcc_lo, implicit $vcc_hi, implicit $sgpr0, implicit $sgpr1, implicit $sgpr2, implicit $sgpr3, implicit $sgpr4, implicit $sgpr5, implicit $sgpr6, implicit $sgpr7, implicit $sgpr8, implicit $sgpr9, implicit $sgpr10, implicit $sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $sgpr16, implicit $sgpr17, implicit $sgpr18, implicit $sgpr19, implicit $sgpr20, implicit $sgpr21, implicit $sgpr22, implicit $sgpr23, implicit $sgpr24, implicit $sgpr25, implicit $sgpr26, implicit $sgpr27, implicit $sgpr28, implicit $sgpr29, implicit $sgpr30, implicit $sgpr31, implicit $sgpr32, implicit $sgpr33, implicit $sgpr34, implicit $sgpr35, implicit $sgpr36, implicit $sgpr37, implicit $sgpr38, implicit $sgpr39, implicit $sgpr40, implicit $sgpr41, implicit $sgpr42, implicit $sgpr43, implicit $sgpr44, implicit $sgpr45, implicit $sgpr46, implicit $sgpr47, implicit $sgpr48, implicit $sgpr49, implicit $sgpr50, implicit $sgpr51, implicit $sgpr52, implicit $sgpr53, implicit $sgpr54, implicit $sgpr55, implicit $sgpr56, implicit $sgpr57, implicit $sgpr58, implicit $sgpr59, implicit $sgpr60, implicit $sgpr61, implicit $sgpr62, implicit $sgpr63, implicit $sgpr64, implicit $sgpr65, implicit $sgpr66, implicit $sgpr67, implicit $sgpr68, implicit $sgpr69, implicit $sgpr70, implicit $sgpr71, implicit $sgpr72, implicit $sgpr73, implicit $sgpr74, implicit $sgpr75, implicit $sgpr76, implicit $sgpr77, implicit $sgpr78, implicit $sgpr79, implicit $sgpr80, implicit $sgpr81, implicit $sgpr82, implicit $sgpr83, implicit $sgpr84, implicit $sgpr85, implicit $sgpr86, implicit $sgpr87, implicit $sgpr88, implicit $sgpr89, implicit $sgpr90, implicit $sgpr91, implicit $sgpr92, implicit $sgpr93, implicit $sgpr94, implicit $sgpr95
renamable $sgpr7 = S_ADD_I32 renamable $sgpr7, %stack.1, implicit-def dead $scc
SI_RETURN implicit $vcc_lo, implicit $vcc_hi, implicit $sgpr0, implicit $sgpr1, implicit $sgpr2, implicit $sgpr3, implicit $sgpr4, implicit $sgpr5, implicit $sgpr6, implicit $sgpr7, implicit $sgpr8, implicit $sgpr9, implicit $sgpr10, implicit $sgpr11, implicit $sgpr12, implicit $sgpr13, implicit $sgpr14, implicit $sgpr15, implicit $sgpr16, implicit $sgpr17, implicit $sgpr18, implicit $sgpr19, implicit $sgpr20, implicit $sgpr21, implicit $sgpr22, implicit $sgpr23, implicit $sgpr24, implicit $sgpr25, implicit $sgpr26, implicit $sgpr27, implicit $sgpr28, implicit $sgpr29, implicit $sgpr30, implicit $sgpr31, implicit $sgpr32, implicit $sgpr33, implicit $sgpr34, implicit $sgpr35, implicit $sgpr36, implicit $sgpr37, implicit $sgpr38, implicit $sgpr39, implicit $sgpr40, implicit $sgpr41, implicit $sgpr42, implicit $sgpr43, implicit $sgpr44, implicit $sgpr45, implicit $sgpr46, implicit $sgpr47, implicit $sgpr48, implicit $sgpr49, implicit $sgpr50, implicit $sgpr51, implicit $sgpr52, implicit $sgpr53, implicit $sgpr54, implicit $sgpr55, implicit $sgpr56, implicit $sgpr57, implicit $sgpr58, implicit $sgpr59, implicit $sgpr60, implicit $sgpr61, implicit $sgpr62, implicit $sgpr63, implicit $sgpr64, implicit $sgpr65, implicit $sgpr66, implicit $sgpr67, implicit $sgpr68, implicit $sgpr69, implicit $sgpr70, implicit $sgpr71, implicit $sgpr72, implicit $sgpr73, implicit $sgpr74, implicit $sgpr75, implicit $sgpr76, implicit $sgpr77, implicit $sgpr78, implicit $sgpr79, implicit $sgpr80, implicit $sgpr81, implicit $sgpr82, implicit $sgpr83, implicit $sgpr84, implicit $sgpr85, implicit $sgpr86, implicit $sgpr87, implicit $sgpr88, implicit $sgpr89, implicit $sgpr90, implicit $sgpr91, implicit $sgpr92, implicit $sgpr93, implicit $sgpr94, implicit $sgpr95
...

View File

@ -0,0 +1,128 @@
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 < %s | FileCheck %s
define amdgpu_gfx [13 x i32] @issue130120() {
; CHECK-LABEL: issue130120:
; CHECK: ; %bb.0: ; %bb
; CHECK-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
; CHECK-NEXT: v_mov_b32_e32 v0, 0
; CHECK-NEXT: s_add_i32 s0, s32, 0xf0
; CHECK-NEXT: s_add_i32 s1, s32, 0xf4
; CHECK-NEXT: s_add_i32 s2, s32, 0xf8
; CHECK-NEXT: s_add_i32 s3, s32, 0xfc
; CHECK-NEXT: v_mov_b32_e32 v1, v0
; CHECK-NEXT: s_add_i32 s34, s32, 0x100
; CHECK-NEXT: s_add_i32 s35, s32, 0x104
; CHECK-NEXT: s_add_i32 s36, s32, 0x108
; CHECK-NEXT: s_add_i32 s37, s32, 0x110
; CHECK-NEXT: s_add_i32 s38, s32, 0x120
; CHECK-NEXT: s_or_b32 s39, s32, 4
; CHECK-NEXT: s_or_b32 s40, s32, 8
; CHECK-NEXT: s_or_b32 s41, s32, 12
; CHECK-NEXT: s_add_i32 s42, s32, 16
; CHECK-NEXT: s_add_i32 s43, s32, 20
; CHECK-NEXT: s_add_i32 s44, s32, 24
; CHECK-NEXT: s_mov_b32 s46, 1
; CHECK-NEXT: s_movk_i32 s45, 0x990
; CHECK-NEXT: s_mov_b32 s48, 0
; CHECK-NEXT: .LBB0_1: ; %bb3
; CHECK-NEXT: ; =>This Inner Loop Header: Depth=1
; CHECK-NEXT: s_cmp_eq_u32 s46, 0
; CHECK-NEXT: s_mov_b32 s49, s48
; CHECK-NEXT: s_mov_b32 s50, s48
; CHECK-NEXT: s_cselect_b32 s51, 0, s1
; CHECK-NEXT: s_cselect_b32 s55, 0, s35
; CHECK-NEXT: v_dual_mov_b32 v2, s48 :: v_dual_mov_b32 v3, s49
; CHECK-NEXT: s_cselect_b32 s52, 0, s2
; CHECK-NEXT: s_cselect_b32 s56, 0, s36
; CHECK-NEXT: s_cselect_b32 vcc_lo, 0, s43
; CHECK-NEXT: v_mov_b32_e32 v4, s50
; CHECK-NEXT: s_cselect_b32 s47, s45, 0xf0
; CHECK-NEXT: s_cselect_b32 s53, 0, s3
; CHECK-NEXT: s_cselect_b32 s54, 0, s34
; CHECK-NEXT: s_cselect_b32 s57, 0, s37
; CHECK-NEXT: s_cselect_b32 s58, 0, s38
; CHECK-NEXT: s_cselect_b32 s59, 0, s0
; CHECK-NEXT: s_cselect_b32 s60, 0, s39
; CHECK-NEXT: s_cselect_b32 s61, 0, s40
; CHECK-NEXT: s_cselect_b32 s62, 0, s41
; CHECK-NEXT: s_cselect_b32 s63, 0, s42
; CHECK-NEXT: s_cselect_b32 vcc_hi, 0, s44
; CHECK-NEXT: s_mov_b32 s46, s48
; CHECK-NEXT: scratch_store_b32 off, v0, s51
; CHECK-NEXT: scratch_store_b32 off, v0, s52
; CHECK-NEXT: scratch_store_b32 off, v0, s53
; CHECK-NEXT: scratch_store_b32 off, v0, s54
; CHECK-NEXT: scratch_store_b32 off, v0, s55
; CHECK-NEXT: scratch_store_b64 off, v[0:1], s56
; CHECK-NEXT: scratch_store_b32 off, v0, s57
; CHECK-NEXT: scratch_store_b32 off, v0, s47
; CHECK-NEXT: scratch_store_b96 off, v[2:4], s58
; CHECK-NEXT: scratch_store_b96 off, v[2:4], s59
; CHECK-NEXT: scratch_store_b32 off, v0, s60
; CHECK-NEXT: scratch_store_b32 off, v0, s61
; CHECK-NEXT: scratch_store_b32 off, v0, s62
; CHECK-NEXT: scratch_store_b32 off, v0, s63
; CHECK-NEXT: scratch_store_b32 off, v0, vcc_lo
; CHECK-NEXT: scratch_store_b32 off, v0, vcc_hi
; CHECK-NEXT: s_mov_b32 vcc_lo, exec_lo
; CHECK-NEXT: s_cbranch_vccnz .LBB0_1
; CHECK-NEXT: ; %bb.2: ; %DummyReturnBlock
; CHECK-NEXT: s_setpc_b64 s[30:31]
bb:
%i = alloca [8 x { i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, { <3 x float>, float, <3 x float>, float }, float, i32, i32, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, i32, i32, i32, i32, i32, i32, i32, i32 }], i32 0, align 16, addrspace(5)
%i1 = getelementptr [8 x { i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, { <3 x float>, float, <3 x float>, float }, float, i32, i32, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, i32, i32, i32, i32, i32, i32, i32, i32 }], ptr addrspace(5) %i, i32 0, i32 0, i32 20
%i2 = getelementptr [8 x { i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, { <3 x float>, float, <3 x float>, float }, float, i32, i32, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, i32, i32, i32, i32, i32, i32, i32, i32 }], ptr addrspace(5) %i, i32 0, i32 6, i32 20
br label %bb3
bb3: ; preds = %bb3, %bb
%i4 = phi i32 [ 1, %bb ], [ 0, %bb3 ]
%i5 = icmp eq i32 %i4, 0
%i6 = select i1 %i5, ptr addrspace(5) %i2, ptr addrspace(5) %i1
store i32 0, ptr addrspace(5) %i6, align 16
%i7 = getelementptr { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, ptr addrspace(5) %i1, i32 0, i32 1
%i8 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i7
store float 0.000000e+00, ptr addrspace(5) %i8, align 4
%i9 = getelementptr { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, ptr addrspace(5) %i1, i32 0, i32 2
%i10 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i9
store i32 0, ptr addrspace(5) %i10, align 8
%i11 = getelementptr { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, ptr addrspace(5) %i1, i32 0, i32 3
%i12 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i11
store i32 0, ptr addrspace(5) %i12, align 4
%i13 = getelementptr { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, ptr addrspace(5) %i1, i32 0, i32 4
%i14 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i13
store i32 0, ptr addrspace(5) %i14, align 16
%i15 = getelementptr { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, ptr addrspace(5) %i1, i32 0, i32 5
%i16 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i15
store i32 0, ptr addrspace(5) %i16, align 4
%i17 = getelementptr { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, ptr addrspace(5) %i1, i32 0, i32 6
%i18 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i17
store <2 x float> zeroinitializer, ptr addrspace(5) %i18, align 8
%i19 = getelementptr { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, ptr addrspace(5) %i1, i32 0, i32 7
%i20 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i19
store i32 0, ptr addrspace(5) %i20, align 16
%i21 = getelementptr { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, ptr addrspace(5) %i1, i32 0, i32 8
%i22 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i21
store <3 x float> zeroinitializer, ptr addrspace(5) %i22, align 16
%i23 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i1
store <3 x float> zeroinitializer, ptr addrspace(5) %i23, align 16
%i24 = getelementptr { i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, { <3 x float>, float, <3 x float>, float }, float, i32, i32, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, i32, i32, i32, i32, i32, i32, i32, i32 }, ptr addrspace(5) %i, i32 0, i32 1
%i25 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i24
store i32 0, ptr addrspace(5) %i25, align 4
%i26 = getelementptr { i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, { <3 x float>, float, <3 x float>, float }, float, i32, i32, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, i32, i32, i32, i32, i32, i32, i32, i32 }, ptr addrspace(5) %i, i32 0, i32 2
%i27 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i26
store i32 0, ptr addrspace(5) %i27, align 8
%i28 = getelementptr { i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, { <3 x float>, float, <3 x float>, float }, float, i32, i32, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, i32, i32, i32, i32, i32, i32, i32, i32 }, ptr addrspace(5) %i, i32 0, i32 3
%i29 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i28
store i32 0, ptr addrspace(5) %i29, align 4
%i30 = getelementptr { i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, { <3 x float>, float, <3 x float>, float }, float, i32, i32, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, i32, i32, i32, i32, i32, i32, i32, i32 }, ptr addrspace(5) %i, i32 0, i32 4
%i31 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i30
store i32 0, ptr addrspace(5) %i31, align 16
%i32 = getelementptr { i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, { <3 x float>, float, <3 x float>, float }, float, i32, i32, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, i32, i32, i32, i32, i32, i32, i32, i32 }, ptr addrspace(5) %i, i32 0, i32 5
%i33 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i32
store i32 0, ptr addrspace(5) %i33, align 4
%i34 = getelementptr { i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, { <3 x float>, float, <3 x float>, float }, float, i32, i32, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, { i32, float, i32, i32, i32, i32, <2 x float>, i32, <3 x float>, <3 x float> }, i32, i32, i32, i32, i32, i32, i32, i32, i32 }, ptr addrspace(5) %i, i32 0, i32 6
%i35 = select i1 %i5, ptr addrspace(5) null, ptr addrspace(5) %i34
store i32 0, ptr addrspace(5) %i35, align 8
br label %bb3
}