christoph/fail
1
Fork 0
Code Actions Packages Releases Activity

jobserver: join remaining threads on shutdown

Browse Source 
To avoid accessing destroyed resources in CommThreads talking to clients,
we need to properly join them on shutdown.  The m_CommMutex becomes a
JobServer member to make sure it isn't destroyed before the JobServer
itself.

Change-Id: I35b9fb93ace08a7a9476650f8f5e93597a3a8aa0
This commit is contained in:
Horst Schirmeier
2014-01-17 17:35:34 +01:00
parent 8505ddbb04
commit 8671669053
2 changed files with 9 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
src/core/cpn/JobServer.cc
View File

@ -34,10 +34,6 @@ void JobServer::addParam(ExperimentData* exp)
volatile unsigned JobServer::m_DoneCount = 0;
#endif
#ifndef __puma
boost::mutex CommThread::m_CommMutex;
#endif
ExperimentData *JobServer::getDone()
{
#ifndef __puma
@ -52,7 +48,7 @@ ExperimentData *JobServer::getDone()
void JobServer::setNoMoreExperiments()
{
#ifndef __puma
boost::unique_lock<boost::mutex> lock(CommThread::m_CommMutex);
boost::unique_lock<boost::mutex> lock(m_CommMutex);
#endif
// currently not really necessary, as we only non-blockingly dequeue:
m_undoneJobs.setIsFinished();
@ -294,7 +290,7 @@ void CommThread::sendPendingExperimentData(Minion& minion)
// Prevent receiveExperimentResults from modifying (or indirectly, via
// getDone and the campaign, deleting) jobs in the m_runningJobs queue.
// (See details in receiveExperimentResults)
boost::unique_lock<boost::mutex> lock(m_CommMutex);
boost::unique_lock<boost::mutex> lock(m_js.m_CommMutex);
#endif
if ((temp_exp = m_js.m_runningJobs.pickone()) != NULL) { // 2nd priority
// (This picks one running job.)
@ -347,7 +343,7 @@ void CommThread::receiveExperimentResults(Minion& minion, FailControlMessage& ct
// by the campaign at any time.
// Additionally, receiving a result overwrites the job's contents. This
// already may cause breakage in sendPendingExperimentData (a).
boost::unique_lock<boost::mutex> lock(m_CommMutex);
boost::unique_lock<boost::mutex> lock(m_js.m_CommMutex);
#endif
for (i = 0; i < ctrlmsg.workloadid_size(); i++) {
if (m_js.m_runningJobs.remove(ctrlmsg.workloadid(i), exp)) { // ExperimentData* found