Sektor van Skijlen

Ah, so that's clear. This is the fixed command set: ``` git clone https://github.com/ethouris/srt.git cd srt git checkout dev-fix-ignored-device-multicast ./configure make ```

Wait. What? How about taking the latest version from the official repository?

Except that you should clone the official repository, not my replica: https://github.com/Haivision/srt.git

There are sanitizers that could detect memory leaks and I think you have already tried some sanitizers during the work on thread problems.

The sender buffer is associated with the socket. Once you close the socket, all this memory should be freed. If it's not, it is a leak. Note of course that...

If we are talking about any "memory swelling" happening exactly here in the sender buffer, yes, this buffer grows dynamically, but not unlimited. The call to `sndBuffersLeft()` checks the limit...

If you suspect a leak, the best way to check it would be by using the LeakSanitizer: https://github.com/google/sanitizers/wiki/AddressSanitizerLeakSanitizer For adding any flags to the compiler command, there's `SRT_EXTRA_CFLAGS` variable used...

That won't be easy to fix. Probably a packet would have to be marked somehow once it has been qualified as ready to play, so that this state is not...

The problem isn't exactly in the drift of one single machine, but rather that drifts in the machines' clocks may cause a relative time difference between the sender and the...

I think a constructor added to `DynamicStruct` with a parameter (so that it doesn't cover the default constructor) could be called before these fields and that would clear it earlier...