[llvm] r312337 - AMDGPU: IMPLICIT_DEFs and DBG_VALUEs do not contribute to wait states
Nicolai Haehnle via llvm-commits
llvm-commits at lists.llvm.org
Fri Sep 1 09:56:32 PDT 2017
Author: nha
Date: Fri Sep 1 09:56:32 2017
New Revision: 312337
URL: http://llvm.org/viewvc/llvm-project?rev=312337&view=rev
Log:
AMDGPU: IMPLICIT_DEFs and DBG_VALUEs do not contribute to wait states
Summary:
This fixes a bug that was exposed on gfx9 in various
GL45-CTS.shaders.loops.*_iterations.select_iteration_count_fragment tests,
e.g. GL45-CTS.shaders.loops.do_while_uniform_iterations.select_iteration_count_fragment
Reviewers: arsenm
Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, t-tye, llvm-commits
Differential Revision: https://reviews.llvm.org/D36193
Added:
llvm/trunk/test/CodeGen/AMDGPU/hazard.mir
Modified:
llvm/trunk/lib/Target/AMDGPU/GCNHazardRecognizer.cpp
Modified: llvm/trunk/lib/Target/AMDGPU/GCNHazardRecognizer.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Target/AMDGPU/GCNHazardRecognizer.cpp?rev=312337&r1=312336&r2=312337&view=diff
==============================================================================
--- llvm/trunk/lib/Target/AMDGPU/GCNHazardRecognizer.cpp (original)
+++ llvm/trunk/lib/Target/AMDGPU/GCNHazardRecognizer.cpp Fri Sep 1 09:56:32 2017
@@ -218,12 +218,17 @@ void GCNHazardRecognizer::RecedeCycle()
int GCNHazardRecognizer::getWaitStatesSince(
function_ref<bool(MachineInstr *)> IsHazard) {
- int WaitStates = -1;
+ int WaitStates = 0;
for (MachineInstr *MI : EmittedInstrs) {
+ if (MI) {
+ if (IsHazard(MI))
+ return WaitStates;
+
+ unsigned Opcode = MI->getOpcode();
+ if (Opcode == AMDGPU::DBG_VALUE || Opcode == AMDGPU::IMPLICIT_DEF)
+ continue;
+ }
++WaitStates;
- if (!MI || !IsHazard(MI))
- continue;
- return WaitStates;
}
return std::numeric_limits<int>::max();
}
Added: llvm/trunk/test/CodeGen/AMDGPU/hazard.mir
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/AMDGPU/hazard.mir?rev=312337&view=auto
==============================================================================
--- llvm/trunk/test/CodeGen/AMDGPU/hazard.mir (added)
+++ llvm/trunk/test/CodeGen/AMDGPU/hazard.mir Fri Sep 1 09:56:32 2017
@@ -0,0 +1,31 @@
+# RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs -run-pass post-RA-hazard-rec %s -o - | FileCheck -check-prefix=GCN -check-prefix=VI %s
+# RUN: llc -march=amdgcn -mcpu=gfx900 -verify-machineinstrs -run-pass post-RA-hazard-rec %s -o - | FileCheck -check-prefix=GCN -check-prefix=GFX9 %s
+
+# GCN: bb.0.entry:
+# GCN: %m0 = S_MOV_B32
+# GFX9: S_NOP 0
+# VI-NOT: S_NOP_0
+# GCN: V_INTERP_P1_F32
+
+---
+name: hazard_implicit_def
+alignment: 0
+exposesReturnsTwice: false
+legalized: false
+regBankSelected: false
+selected: false
+tracksRegLiveness: true
+registers:
+liveins:
+ - { reg: '%sgpr7', virtual-reg: '' }
+ - { reg: '%vgpr4', virtual-reg: '' }
+body: |
+ bb.0.entry:
+ liveins: %sgpr7, %vgpr4
+
+ %m0 = S_MOV_B32 killed %sgpr7
+ %vgpr5 = IMPLICIT_DEF
+ %vgpr0 = V_INTERP_P1_F32 killed %vgpr4, 0, 0, implicit %m0, implicit %exec
+ SI_RETURN_TO_EPILOG killed %vgpr5, killed %vgpr0
+
+...
More information about the llvm-commits
mailing list