[Live-devel] Struggling with a stange behavior

[Ulrich Teichert]

Hi,

>From: live-devel [mailto:live-devel-bounces at ns.live555.com] On Behalf Of Ross >Finlayson

>>Valgrind tells me that memory leaks (or something else that lost bytes) are found: >>60 bytes direct, 20.000 indirect bytes are lost in the thread,
>> in the H264BufferdPackedFactory::createNewPacket(...) by >>ReorederingPacketBuffer::getFreePacket() by SocketDescriptor::tcpReadHandler() by >>BasicTaskScheduler::singleStep().
>>I'm going crazy. Why this behaviour in this case?

>I don't know, but I wouldn't pay too much attention to the details in "valgrind" >reports; they're notoriously unreliable.

I recently suspected that an application of ours was leaking memory and ran
valgrind against it. After fixing the obvious leakages (singletons, lazyness..),
I only had false positives left in ALSA - this was on an ARM-Linux platform. All
other reports were IMHO correct. So I tend to disagree here ;-)

The problem with less obvious memory leaks is that valgrind can only report where
the memory gets allocated. To find the responsible object where to delete/free it
can be tricky, but only when it's not clear which object owns the memory.

HTH,
Uli