Change in libosmocore[master]: add osmo_classmark_* API

This is merely a historical archive of years 2008-2021, before the migration to mailman3.

A maintained and still updated list archive can be found at https://lists.osmocom.org/hyperkitty/list/gerrit-log@lists.osmocom.org/.

Neels Hofmeyr gerrit-no-reply at lists.osmocom.org
Mon Jan 21 18:19:21 UTC 2019


Neels Hofmeyr has posted comments on this change. ( https://gerrit.osmocom.org/12659 )

Change subject: add osmo_classmark_* API
......................................................................


Patch Set 2:

(6 comments)

https://gerrit.osmocom.org/#/c/12659/2/include/osmocom/gsm/protocol/gsm_04_08.h
File include/osmocom/gsm/protocol/gsm_04_08.h:

https://gerrit.osmocom.org/#/c/12659/2/include/osmocom/gsm/protocol/gsm_04_08.h@59
PS2, Line 59: osmo_classmark
> Since this is a part of GSM 04. […]
sure, can, but I preferred it shorter.
Hmm, ok, maybe there are other classmarks besides the 48 ones...


https://gerrit.osmocom.org/#/c/12659/2/include/osmocom/gsm/protocol/gsm_04_08.h@63
PS2, Line 63: classmark2
> (just wondering) we also have gsm48_classmark2, why not to use it here?
I'm not entirely sure, maybe anticipating that classmark2 might grow in a future spec?
But actually, no, if at all, cm3 would be extended.
Looks like I should use gsm48_classmark2 indeed.
Now I'm unsure whether cm2 might ever be < 3 bytes.
Since existing code seems to use the cm2_len already, I'm just going to keep that.


https://gerrit.osmocom.org/#/c/12659/2/src/gsm/gsm48.c
File src/gsm/gsm48.c:

https://gerrit.osmocom.org/#/c/12659/2/src/gsm/gsm48.c@1199
PS2, Line 1199: snprintf(cm1, sizeof(cm1), "no-cm1");
> PoC: https://pastebin. […]
sure, better.


https://gerrit.osmocom.org/#/c/12659/2/src/gsm/gsm48.c@1234
PS2, Line 1234: dst->classmark1_set
> Should we assign false otherwise?
no, it's just an update. We can only add classmarks, not remove them.


https://gerrit.osmocom.org/#/c/12659/2/src/gsm/gsm48.c@1238
PS2, Line 1238: sizeof(dst->classmark2)
> Why not classmark2_len?
to avoid having to check whether the len is > sizeof().
I can make it OSMO_MIN() if you prefer.


https://gerrit.osmocom.org/#/c/12659/2/src/gsm/gsm48.c@1303
PS2, Line 1303: return 0;
> (just wondering) shouldn't we indicate this 'out of range' case somehow?
we're in a libosmocore utility function, and I want to avoid adding logging.
The negative rc range is already used for "the missing classmark number (negative)",
so just return "no, A5/99 is not supported, go away".



-- 
To view, visit https://gerrit.osmocom.org/12659
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings

Gerrit-Project: libosmocore
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Ifd27bab0380f7ad0c44c719aa6c8bd62cf7b034c
Gerrit-Change-Number: 12659
Gerrit-PatchSet: 2
Gerrit-Owner: Neels Hofmeyr <nhofmeyr at sysmocom.de>
Gerrit-Reviewer: Jenkins Builder (1000002)
Gerrit-Reviewer: Neels Hofmeyr <nhofmeyr at sysmocom.de>
Gerrit-Reviewer: Vadim Yanitskiy <axilirator at gmail.com>
Gerrit-Comment-Date: Mon, 21 Jan 2019 18:19:21 +0000
Gerrit-HasComments: Yes
Gerrit-HasLabels: No
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osmocom.org/pipermail/gerrit-log/attachments/20190121/1ff6b082/attachment.htm>


More information about the gerrit-log mailing list