From cbeaa8b0e2b0c001059741d957620a925f76a802 Mon Sep 17 00:00:00 2001
From: Scott Linder <scott@scottlinder.com>
Date: Thu, 21 Jun 2018 18:48:48 +0000
Subject: [PATCH] [AMDGPU] Fix bug with tracking processed blocks in
 SIInsertWaitcnts

BlockWaitcntProcessedSet was not being cleared between calls, so it was
producing incorrect counts in cases where MBB addresses happened to coincide
across multiple calls.

Differential Revision: https://reviews.llvm.org/D48391



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@335268 91177308-0d34-0410-b5e6-96231b3b80d8
---
 lib/Target/AMDGPU/SIInsertWaitcnts.cpp | 1 +
 1 file changed, 1 insertion(+)

diff --git a/lib/Target/AMDGPU/SIInsertWaitcnts.cpp b/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
index fb24d528c33..7f750989376 100644
--- a/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
+++ b/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
@@ -1868,6 +1868,7 @@ bool SIInsertWaitcnts::runOnMachineFunction(MachineFunction &MF) {
   BlockVisitedSet.clear();
   VCCZBugHandledSet.clear();
   LoopWaitcntDataMap.clear();
+  BlockWaitcntProcessedSet.clear();
 
   // Walk over the blocks in reverse post-dominator order, inserting
   // s_waitcnt where needed.