From ef490e308ccce8e6df85144784a0f4580f5ac6a1 Mon Sep 17 00:00:00 2001 From: Aleric Inglewood Date: Sat, 5 Feb 2011 15:58:07 +0100 Subject: Introduces a LLThreadLocalData class that can be accessed through the static LLThread::tldata(). Currently this object contains two (public) thread-local objects: a LLAPRRootPool and a LLVolatileAPRPool. The first is the general memory pool used by this thread (and this thread alone), while the second is intended for short lived memory allocations (needed for APR). The advantages of not mixing those two is that the latter is used most frequently, and as a result of it's nature can be destroyed and reconstructed on a "regular" basis. This patch adds LLAPRPool (completely replacing the old one), which is a wrapper around apr_pool_t* and has complete thread-safity checking. Whenever an apr call requires memory for some resource, a memory pool in the form of an LLAPRPool object can be created with the same life-time as this resource; assuring clean up of the memory no sooner, but also not much later than the life-time of the resource that needs the memory. Many, many function calls and constructors had the pool parameter simply removed (it is no longer the concern of the developer, if you don't write code that actually does an libapr call then you are no longer bothered with memory pools at all). However, I kept the notion of short-lived and long-lived allocations alive (see my remark in the jira here: https://jira.secondlife.com/browse/STORM-864?focusedCommentId=235356&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-235356 which requires that the LLAPRFile API needs to allow the user to specify how long they think a file will stay open. By choosing 'short_lived' as default for the constructor that immediately opens a file, the number of instances where this needs to be specified is drastically reduced however (obviously, any automatic LLAPRFile is short lived). *** Addressed Boroondas remarks in https://codereview.secondlife.com/r/99/ regarding (doxygen) comments. This patch effectively only changes comments. Includes some 'merge' stuff that ended up in llvocache.cpp (while starting as a bug fix, now only resulting in a cleanup). *** Added comment 'The use of apr_pool_t is OK here'. Added this comment on every line where apr_pool_t is correctly being used. This should make it easier to spot (future) errors where someone started to use apr_pool_t; you can just grep all sources for 'apr_pool_t' and immediately see where it's being used while LLAPRPool should have been used. Note that merging this patch is very easy: If there are no other uses of apr_pool_t in the code (one grep) and it compiles, then it will work. *** Second Merge (needed to remove 'delete mCreationMutex' from LLImageDecodeThread::~LLImageDecodeThread). *** Added back #include . Apparently that is needed on libapr version 1.2.8., the version used by Linden Lab, for calls to apr_queue_*. This is a bug in libapr (we also include , that is fixed in (at least) 1.3.7. Note that 1.2.8 is VERY old. Even 1.3.x is old. *** License fixes (GPL -> LGPL). And typo in comments. Addresses merov's comments on the review board. *** Added Merov's compile fixes for windows. --- indra/newview/llwatchdog.cpp | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'indra/newview/llwatchdog.cpp') diff --git a/indra/newview/llwatchdog.cpp b/indra/newview/llwatchdog.cpp index 1694126802..d982ca5020 100644 --- a/indra/newview/llwatchdog.cpp +++ b/indra/newview/llwatchdog.cpp @@ -178,8 +178,8 @@ void LLWatchdog::init(killer_event_callback func) mKillerCallback = func; if(!mSuspectsAccessMutex && !mTimer) { - mSuspectsAccessMutex = new LLMutex(NULL); - mTimer = new LLWatchdogTimerThread(); + mSuspectsAccessMutex = new LLMutex; + mTimer = new LLWatchdogTimerThread; mTimer->setSleepTime(WATCHDOG_SLEEP_TIME_USEC / 1000); mLastClockCount = LLTimer::getTotalTime(); -- cgit v1.2.3 From b88c7166f4323d9035bc78f517ff51396e80c123 Mon Sep 17 00:00:00 2001 From: Roxie Linden Date: Fri, 15 Jul 2011 12:53:09 -0700 Subject: STORM-1355 Memory issues from UI for very large groups This change is not guaranteed to fix this issue as the issue is difficult to repro, but there was a sketchy case group member responses come back from the simulator in message packets. For very large numbers of members, there may be a large number of packets received. The member data is placed in a structure of type LLGroupMgrGroupData, based on the group id. The problem is, if the user refreshes the group before the entire contents of the previous request comes back, response packets from the previous request will be intermingled with the responses from the refresh. Both the request call and the response handler would create the group data structure, if the structure wasn't already there. There may be a case where a response from the previous request causes creation of the group data, populating it with the contents of the response, and the responses from the second request would use that group data structure. Also, cleaned up some comments and variable names to be consistent --- indra/newview/llwatchdog.cpp | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'indra/newview/llwatchdog.cpp') diff --git a/indra/newview/llwatchdog.cpp b/indra/newview/llwatchdog.cpp index 1694126802..4f582fc2db 100644 --- a/indra/newview/llwatchdog.cpp +++ b/indra/newview/llwatchdog.cpp @@ -126,8 +126,8 @@ void LLWatchdogTimeout::start(const std::string& state) // Order of operation is very impmortant here. // After LLWatchdogEntry::start() is called // LLWatchdogTimeout::isAlive() will be called asynchronously. - mTimer.start(); ping(state); + mTimer.start(); LLWatchdogEntry::start(); } -- cgit v1.2.3