summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorArmin Le Grand <Armin.Le.Grand@cib.de>2016-07-07 10:16:02 +0200
committerArmin Le Grand <Armin.Le.Grand@cib.de>2018-08-09 13:20:02 +0200
commit949ecd96e8bec1709b89bca6177a1fbba6cc6f94 (patch)
tree1620defee71ceab5df27c8f226083e55bee978e0
parent2f91c89baae9b0cf9e11cff02cf56338ebdda177 (diff)
Related: tdf#50613 use an own instance of ThreadPool
Using the global ThreadPool (getSharedOptimalPool()) can lead to problems when more than one usage executes and one of them already calls waitUntilEmpty() what of course influences the other usage. Thus I added an own instance of ThreadPool for async loading of chart models in writer Change-Id: I4bea64af0d36e87081abec95c75574966d0fe5b9
-rw-r--r--sw/source/core/ole/ndole.cxx39
1 files changed, 35 insertions, 4 deletions
diff --git a/sw/source/core/ole/ndole.cxx b/sw/source/core/ole/ndole.cxx
index 903d96814bc1..1be81749133e 100644
--- a/sw/source/core/ole/ndole.cxx
+++ b/sw/source/core/ole/ndole.cxx
@@ -676,7 +676,36 @@ void SwOLENode::SetChanged()
}
}
//////////////////////////////////////////////////////////////////////////////
+// due to some problems in test cases with the SharedOptimalPool, use
+// an own instance of comphelper::ThreadPool. Problem is that other
+// usages of getSharedOptimalPool() may interfere if more than one
+// pool user calls waitUntilEmpty().
+//
+// It gets created on-demand and will be available during LO's
+// lifetime for loading chart models used in writer in parallel. It
+// would be possible to add a usage count, then trigger a timer and
+// clean it up (due to lifetime issues), but that's probably overkill.
+// It gets created on demand, is ready for global reuse and makes no
+// harm (not much ressources needed)
+static comphelper::ThreadPool* pLocalPool = 0;
+
+comphelper::ThreadPool* getLocalThreadPool()
+{
+ if (pLocalPool)
+ {
+ return pLocalPool;
+ }
+
+ if (0 == comphelper::ThreadPool::getSharedOptimalPool().getWorkerCount())
+ {
+ return nullptr;
+ }
+
+ pLocalPool = new comphelper::ThreadPool(comphelper::ThreadPool::getSharedOptimalPool().getWorkerCount());
+ return pLocalPool;
+}
+/// Holder for local data for a parallely-executed task to load a chart model
class DeflateData
{
private:
@@ -734,8 +763,7 @@ public:
}
};
-//////////////////////////////////////////////////////////////////////////////
-
+/// Task for parallely-executed task to load a chart model
class DeflateThread : public comphelper::ThreadTask
{
// the data to work on
@@ -1080,11 +1108,14 @@ drawinglayer::primitive2d::Primitive2DContainer SwOLEObj::tryToGetChartContentAs
if(aXModel.is())
{
+ // loaded using own instance of comphelper::ThreadPool,
+ // see getLocalThreadPool(). Disable via bool below if
+ // trouble surfaces somewhere
static bool bAnynchronousLoadingAllowed = true;
if(bSynchron ||
!bAnynchronousLoadingAllowed ||
- 0 == comphelper::ThreadPool::getSharedOptimalPool().getWorkerCount())
+ nullptr == getLocalThreadPool())
{
// load chart synchron in this Thread
m_aPrimitive2DSequence = ChartHelper::tryToGetChartContentAsPrimitive2DSequence(
@@ -1100,7 +1131,7 @@ drawinglayer::primitive2d::Primitive2DContainer SwOLEObj::tryToGetChartContentAs
{
m_pDeflateData = new DeflateData(aXModel);
DeflateThread* pNew = new DeflateThread(*m_pDeflateData);
- comphelper::ThreadPool::getSharedOptimalPool().pushTask(pNew);
+ getLocalThreadPool()->pushTask(pNew);
}
}
}