Patch Set 1:


> (1 comment)
File include/osmo-bts/gsm_data.h:

Line 97:                const struct bts_cm *cm; /* Table with supp. ch 
rate/mode combinations */
> if this is alway an array that needs to be as large as MAX_PCHAN_T, then wh
This pointer is not for a dynamically allocated struct. Its set by 
bts_model_init() on startup to a hardcoded table. (e.g. see l1_if.c in in 
osmo-bts-octphy) The problem I wanted to solve here is that I wanted to have an 
easy to fill out, static table inside the hardware specific part. The pointer 
is set once on initialization to give access to the data. If we want to have 
this as an array, we would have to copy the table on startup. If you think that 
makes it better understandable I can do this of course. Any ideas?
File src/common/bts.c:

Line 679: int bts_supports_cm(struct gsm_bts_role_bts *bts,
> Why not to use bool here?
I wanted it bo be uniform with bts_supports_cipher() (a few lines above) there 
they also used int.

Line 693:       for (i = 0;; i++) {
> Oops, I was mistaken. As the size of bts->
Yes, I will do this when we finally decided to turn into an array. 
At the moment it is a pointer to an array.

Line 702:               if (supported[i].pchan == _GSM_PCHAN_MAX)
> This could be moved to the 'for' condition...

To view, visit
To unsubscribe, visit

Gerrit-MessageType: comment
Gerrit-Change-Id: Id9b222b7ab19ece90591718bc562b3a8c5e02023
Gerrit-PatchSet: 1
Gerrit-Project: osmo-bts
Gerrit-Branch: master
Gerrit-Owner: dexter <>
Gerrit-Reviewer: Harald Welte <>
Gerrit-Reviewer: Jenkins Builder
Gerrit-Reviewer: Vadim Yanitskiy <>
Gerrit-Reviewer: dexter <>
Gerrit-HasComments: Yes

Reply via email to