[amdgpu] Fix scoreboard updating on `s_waitcnt_vscnt`.

Summary: - Other counters are accidentally cleared.

Subscribers: arsenm, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, hiraditya, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D71866
This commit is contained in:
Michael Liao 2019-12-25 00:47:18 -05:00
parent 8be5a0fe12
commit 79d401905f
2 changed files with 18 additions and 1 deletions

View File

@ -1081,7 +1081,7 @@ bool SIInsertWaitcnts::generateWaitcntInstBefore(
assert(II->getOpcode() == AMDGPU::S_WAITCNT_VSCNT);
assert(II->getOperand(0).getReg() == AMDGPU::SGPR_NULL);
ScoreBrackets.applyWaitcnt(
AMDGPU::Waitcnt(0, 0, 0, II->getOperand(1).getImm()));
AMDGPU::Waitcnt(~0u, ~0u, ~0u, II->getOperand(1).getImm()));
}
}
}

View File

@ -0,0 +1,17 @@
# RUN: llc -march=amdgcn -mcpu=gfx1010 -verify-machineinstrs -run-pass si-insert-waitcnts -o - %s | FileCheck -check-prefix=GFX10 %s
# GFX10-LABEL: waitcnt-vscnt
# GFX10: GLOBAL_ATOMIC_ADD_RTN
# GFX10-NEXT: S_WAITCNT 49279
---
name: waitcnt-vscnt
machineFunctionInfo:
isEntryFunction: true
body: |
bb.0:
liveins: $sgpr0_sgpr1
$sgpr4 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0, 0 :: (dereferenceable invariant load 4 from `i32 addrspace(4)* undef`)
S_WAITCNT_VSCNT undef $sgpr_null, 0
$vgpr0 = GLOBAL_ATOMIC_ADD_RTN $vgpr0_vgpr1, $vgpr2, 0, 0, implicit $exec :: (load store syncscope("agent") seq_cst 4, addrspace 1)
S_CMP_LG_U32 killed $sgpr4, 0, implicit-def $scc
...