Attention is currently required from: fixeria.
1 comment:
Commit Message:
o that we can dynamically choose
buffer size depending on the channel mode (speech/data).
I don't know if this is really an advantage. Sounds to me like we just put more load on the memory allocator and cause more memory fragmentation.
There's nothing wrong with a TRX allocating the maximum buffer size for all the burst buffers it needs in the "worst case" scenario once at start-up, and then using them as needed.
It's not like the total size of all those burst buffers is ever going to be of concern on any of our target platforms, right?
It's not super critical, after all we're not activating/deactivating hundreds or thousands of timeslots per second. But still something to think about.
To view, visit change 31672. To unsubscribe, or for help writing mail filters, visit settings.