Attention is currently required from: pespin. fixeria has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bsc/+/27885 )

Change subject: paging: Submit up to 20 paging requests in a single work iteration ......................................................................

Patch Set 3:

(3 comments)

File src/osmo-bsc/paging.c:

https://gerrit.osmocom.org/c/osmo-bsc/+/27885/comment/e1e95e5a_b764d33f PS3, Line 210: struct gsm_paging_request, entry); Alignment needs to be fixed.

https://gerrit.osmocom.org/c/osmo-bsc/+/27885/comment/09ba706d_27343fe7 PS3, Line 243: } while (request != initial_request && num_paged < MAX_PAGE_REQ_PER_ITER); I am trying to understand how this is supposed to work if you have less than 20 penging requests awaiting in the queue. If we have just one request, then 'request == initial_request' is true, and we break the loop - so far so good. But if we have two? I may be missing something, but AFAIU the loop will schedule 20 paging messages for just 2 requests, 10 times each. Is this the case?

https://gerrit.osmocom.org/c/osmo-bsc/+/27885/comment/4e860eea_2f2b9bd9 PS3, Line 246: sched_next_paging This label name is confusing: when I saw the 'goto' above I thought you're somehow scheduling/sending paging in the absence of the OML link. The old name was cleaner IMO, so I would keep it. Or maybe 'sched_next_iter'?