Attention is currently required from: fixeria, laforge, pespin.
jolly has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bsc/+/34626?usp=email )
Change subject: ASCI: Add System Information 10 support ......................................................................
Patch Set 4:
(4 comments)
File src/osmo-bsc/system_information.c:
https://gerrit.osmocom.org/c/osmo-bsc/+/34626/comment/19049463_7a07c8f7 PS2, Line 1428: unsigned int save_cur_bit; : struct gsm_subscriber_connection *c;
both vars can be moved to the scope of use (for-loop)
I always want to separate variables from code. Even if it has no performance/memory gain, I want to have variables on top and code below.
https://gerrit.osmocom.org/c/osmo-bsc/+/34626/comment/3b3d2edf_6e44dbaa PS2, Line 1447: 32
Thanks for adding the comment. I still see a possible problem though. […]
SI 10 uses channel index with 5 bit. This means that it can address only the first 32 neighbor channels that are defined by SI5*. It does not make sense that there are more than 32 neighbor cells, because they cannot be reported by measurement report, which uses 5 bits as index too.
https://gerrit.osmocom.org/c/osmo-bsc/+/34626/comment/f43946fb_71499cbe PS2, Line 1499: /* Do spare padding. We cannot do it earlier, because encoding might corrupt it if differenctial cell info
So if you mean "differential", then please use the correct spelling.
Done
File src/osmo-bsc/system_information.c:
https://gerrit.osmocom.org/c/osmo-bsc/+/34626/comment/cc91560e_f8ce3bbd PS4, Line 1456: llist_for_each_entry(c, &conn->vgcs_chan.call->vgcs_call.chan_list, vgcs_chan.list) {
"struct gsm_bts *c_bts;" can be moved here (loop scope).
I always want to separate variables from code. Even if it has no performance/memory gain, I want to have variables on top and code below.