Attention is currently required from: neels, fixeria, pespin.
11 comments:
Patchset:
I hope I've addressed all your comments. […]
Done
File src/osmo-bsc/bsc_vty.c:
Patch Set #3, Line 3089: "Enable BTS setup ramping to limit the number of BTS which configure within a time window.\n"
... the amount of BTS to configure ...
Done
Patch Set #3, Line 3090: "Limit number of BTS.\n"
amount. […]
Done
File src/osmo-bsc/bts_setup_ramp.c:
Patch Set #4, Line 86: { 0, NULL },
unaligned
Done
Patch Set #4, Line 101: net->bts_setup_ramp.active
Use bts_setup_ramp_active()?
because it's a single line. If ramp_active() is checking more, i'll use it here too.
Patch Set #4, Line 107: net->bts_setup_ramp.active
Use !bts_setup_ramp_active()?
same
Patch Set #4, Line 164: net->bts_setup_ramp.active
Why not using the API bts_setup_ramp_active() for that?
same
Patch Set #4, Line 201: INIT_LLIST_HEAD(&bts->bts_setup_ramp.list);
exactly why we want to have a well defined value for empty llist_entries.
Done
File src/osmo-bsc/nm_bb_transc_fsm.c:
Patch Set #1, Line 345: case NM_EV_RAMP_GO:
I don't see why you'd need to add it in all FSM states. […]
If I do it as you ask, I would have to check in which state the fsm is to prevent the LOGL_ERROR line about an invalid event. Further I don't like to call config_loop() when the MO is already in enabled mode. In theory it could go there if there is a bug in the BTS (e.g. ipaccess).
File src/osmo-bsc/nm_bts_ramp.c:
Patch Set #1, Line 63: if (llist_entry_empty(&bts->bts_ramp.list))
I don't like to have this llist set to itself. (INIT_LLIST_HEAD) […]
Done
Patch Set #1, Line 86: net->bts_ramp.count >= net->bts_ramp.max_bts)
Fixed. […]
Done
To view, visit change 29788. To unsubscribe, or for help writing mail filters, visit settings.