SI parsing filter event timeout too short
I looked at the SI parsing code in sitp_si.c line 1992 (in rev 12899) and I think that the timeout used when a table is waiting for completion is too small. The timeout is set to g_sitp_si_update_poll_interval (= 1ms) and this means that if a table is not complete, the event queue will timeout immediately.
An example where this could happen is when waiting for SVCT table to be complete. If the table is not received, the event will timeout and the the handler will reset the timeout to 1ms until any table is received. This will create a lot of unnecessary filter checks.
The only reason for a short timeout was to reactivate the filter if the event was _last_ section found, but that is handled in the following code. So I think the corresponding timeout should be changed to normal timeout.
Should I open a bug for this?