Attention is currently required from: fixeria.
pespin has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bsc/+/28328 )
Change subject: lchan_select: implement dynamic selection mode for assignment
......................................................................
Patch Set 2:
(2 comments)
File src/osmo-bsc/lchan_select.c:
https://gerrit.osmocom.org/c/osmo-bsc/+/28328/comment/592c48e1_f025f9ca
PS2, Line 229: /* Sort by TRX power is descending order (if needed) */
"in descending"?
https://gerrit.osmocom.org/c/osmo-bsc/+/28328/comment/d947b786_f89e82cc
PS2, Line 231: qsort(&trx_list[0], bts->num_trx, sizeof(trx), &qsort_func);
I'm also not liking the fact that you seem to be basically mixing logic for different alloc algorithms in different steps.
I'd really go for having a func ptr or switch statement with different per-algo functions doing each whatever. Otherwise the whole thing looks too complex imho.
--
To view, visit https://gerrit.osmocom.org/c/osmo-bsc/+/28328
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings
Gerrit-Project: osmo-bsc
Gerrit-Branch: master
Gerrit-Change-Id: I1b7a0d706976b73cc5c30a8714b830811addfe8d
Gerrit-Change-Number: 28328
Gerrit-PatchSet: 2
Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de>
Gerrit-Reviewer: Jenkins Builder
Gerrit-CC: laforge <laforge(a)osmocom.org>
Gerrit-CC: pespin <pespin(a)sysmocom.de>
Gerrit-Attention: fixeria <vyanitskiy(a)sysmocom.de>
Gerrit-Comment-Date: Mon, 20 Jun 2022 09:53:46 +0000
Gerrit-HasComments: Yes
Gerrit-Has-Labels: No
Gerrit-MessageType: comment
Attention is currently required from: fixeria.
pespin has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bsc/+/28327 )
Change subject: lchan_select: allow different alloc order for assignment and handover
......................................................................
Patch Set 1:
(2 comments)
File src/osmo-bsc/bts_vty.c:
https://gerrit.osmocom.org/c/osmo-bsc/+/28327/comment/49642fe8_60ccf49c
PS1, Line 541: DEFUN_ATTR(cfg_bts_challoc_mode_all,
iiuc this should be marked as DEPRECATED in favor of "mode set-all"
File src/osmo-bsc/lchan_select.c:
https://gerrit.osmocom.org/c/osmo-bsc/+/28327/comment/caa02afd_45e33dd8
PS1, Line 254: chan_alloc_reverse = bts->chan_alloc_chan_req_reverse;
you could put these in an array chan_alloc_mode[] since you already have an enum. Not important though.
--
To view, visit https://gerrit.osmocom.org/c/osmo-bsc/+/28327
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings
Gerrit-Project: osmo-bsc
Gerrit-Branch: master
Gerrit-Change-Id: I3ae73b36ee9433cc768376b56f0765e5f416162f
Gerrit-Change-Number: 28327
Gerrit-PatchSet: 1
Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: laforge <laforge(a)osmocom.org>
Gerrit-CC: pespin <pespin(a)sysmocom.de>
Gerrit-Attention: fixeria <vyanitskiy(a)sysmocom.de>
Gerrit-Comment-Date: Mon, 20 Jun 2022 09:39:51 +0000
Gerrit-HasComments: Yes
Gerrit-Has-Labels: No
Gerrit-MessageType: comment
Attention is currently required from: neels, fixeria.
pespin has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bsc/+/28326 )
Change subject: lchan_select: prepare a list of timeslots once, iterate over it
......................................................................
Patch Set 1:
(1 comment)
Patchset:
PS1:
I'm not really liking this new approach/path, looks like unnecessarily dropping performance here generating a full array each time, then iterating it. I think this can be probably still be done without having to generate a full list of TS every time a new lchan needs to be selected.
Let's merge if everyone else is happy with this, just sharing my thoughts here.
--
To view, visit https://gerrit.osmocom.org/c/osmo-bsc/+/28326
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings
Gerrit-Project: osmo-bsc
Gerrit-Branch: master
Gerrit-Change-Id: I7ccc56856bfd40fd7c63b7437736de60c2b516ff
Gerrit-Change-Number: 28326
Gerrit-PatchSet: 1
Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: laforge <laforge(a)osmocom.org>
Gerrit-Reviewer: neels <nhofmeyr(a)sysmocom.de>
Gerrit-CC: pespin <pespin(a)sysmocom.de>
Gerrit-Attention: neels <nhofmeyr(a)sysmocom.de>
Gerrit-Attention: fixeria <vyanitskiy(a)sysmocom.de>
Gerrit-Comment-Date: Mon, 20 Jun 2022 09:28:28 +0000
Gerrit-HasComments: Yes
Gerrit-Has-Labels: No
Gerrit-MessageType: comment
Attention is currently required from: fixeria.
pespin has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bsc/+/28325 )
Change subject: bts_chan_load(): also calculate per-TRX channel load
......................................................................
Patch Set 1:
(2 comments)
File src/osmo-bsc/chan_alloc.c:
https://gerrit.osmocom.org/c/osmo-bsc/+/28325/comment/33aeb96b_80de2535
PS1, Line 47: struct load_counter *ll = &trx->lchan_load;
I wonder what ll means here.
https://gerrit.osmocom.org/c/osmo-bsc/+/28325/comment/47af5698_b0b46c82
PS1, Line 85: ll->total++;
You can simply do:
ll->total += pl->total;
ll->used += pl->used;
after the loop, instead of doing +1 N times.
--
To view, visit https://gerrit.osmocom.org/c/osmo-bsc/+/28325
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings
Gerrit-Project: osmo-bsc
Gerrit-Branch: master
Gerrit-Change-Id: I220145238c23135f7e68ca2d474764312ffb66c5
Gerrit-Change-Number: 28325
Gerrit-PatchSet: 1
Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: laforge <laforge(a)osmocom.org>
Gerrit-CC: pespin <pespin(a)sysmocom.de>
Gerrit-Attention: fixeria <vyanitskiy(a)sysmocom.de>
Gerrit-Comment-Date: Mon, 20 Jun 2022 09:13:25 +0000
Gerrit-HasComments: Yes
Gerrit-Has-Labels: No
Gerrit-MessageType: comment