Re: [Live-devel] Memory leak in OnDemandServerMediaSubsession

2022-01-10 Thread Ba Jinsheng
Streaming Media - development & use Subject: Re: [Live-devel] Memory leak in OnDemandServerMediaSubsession - External Email - > On Jan 10, 2022, at 9:18 PM, Ba Jinsheng wrote: > > Unfortunately, in my experimental environment, it looks unlikely that the > objects are delete

Re: [Live-devel] Memory leak in OnDemandServerMediaSubsession

2022-01-10 Thread Ross Finlayson
> On Jan 10, 2022, at 9:18 PM, Ba Jinsheng wrote: > > Unfortunately, in my experimental environment, it looks unlikely that the > objects are deleted, even after 60 seconds. If you are running the unmodified LIVE555 server code, then the function GenericMediaServer::ClientSession::liv

Re: [Live-devel] Memory leak in OnDemandServerMediaSubsession

2022-01-10 Thread Ba Jinsheng
: LIVE555 Streaming Media - development & use Subject: Re: [Live-devel] Memory leak in OnDemandServerMediaSubsession - External Email - > On Jan 10, 2022, at 7:54 PM, Ba Jinsheng > mailto:bajinsh...@u.nus.edu>> wrote: > > I am not sure if the design is

Re: [Live-devel] Memory leak in OnDemandServerMediaSubsession

2022-01-09 Thread Ross Finlayson
> On Jan 10, 2022, at 7:54 PM, Ba Jinsheng wrote: > > I am not sure if the design is like this, but it does incur unstopping memory > increasing. > > > OnDemandServerMediaSubsession::getStreamParameters() in > liveMedia/OnDemandServerMediaSubsession.cpp > This function will allocate a new S

Re: [Live-devel] Memory leak in OnDemandServerMediaSubsession

2022-01-09 Thread Ba Jinsheng
ive-devel On Behalf Of Ross Finlayson Sent: Monday, January 10, 2022 2:06 PM To: LIVE555 Streaming Media - development & use Subject: Re: [Live-devel] Memory leak in OnDemandServerMediaSubsession - External Email - Unfortunately, reports of memory leaks aren’t worthwhile fo

Re: [Live-devel] Memory leak in OnDemandServerMediaSubsession

2022-01-09 Thread Ross Finlayson
Unfortunately, reports of memory leaks aren’t worthwhile for me to investigate unless you can identify a specific object (class) that is allocated, and not later deallocated. Otherwise I can’t be sure that the heap isn’t just growing due to fragmentation - i.e., not a real memory leak. Ross F