[Asterisk-code-review] RFC: ao2 reference storage location logging. (asterisk[master])

Matt Jordan asteriskteam at digium.com
Thu Jul 7 16:37:32 CDT 2016


Matt Jordan has posted comments on this change.

Change subject: RFC: ao2 reference storage location logging.
......................................................................


Patch Set 2:

I don't have any issue with what is attempting to be accomplished here. My concerns are the following:

(1) The plethora of new API calls requires everyone to learn the 'new' way to use ao2 in order to get better tracking. The API proposed - while there isn't anything wrong - puts the ao2 API in the camp of many older APIs that have '_ex' or '_s' prepended/appended everywhere. Ideally, we'd have a way to accomplish the goals here without requiring all API calls that modify a reference count from being renamed. I get that there may not be an easy mechanism for this, since the pairing of reference count bumps can be difficult to match up (and may not always be known apriori), but some of the simpler ones may be able to be handled by simply knowing the pointer address of the object being modified - or the address of the function that performed the manipulation.

(2) Since this is in master, I think it's reasonable to propose a breaking API change in ao2. While this is painful, as it will absolutely break the crap out of third party modules, that is - for better or worse - what master is for. It's at least worth debating, given the scope of the change - although I'd hate to think of all of the lines of code that would have to be updated to take advantage of this.

-- 
To view, visit https://gerrit.asterisk.org/3141
To unsubscribe, visit https://gerrit.asterisk.org/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: Iacb0a51cefaa98c83eab18aa2b7d18850bb33951
Gerrit-PatchSet: 2
Gerrit-Project: asterisk
Gerrit-Branch: master
Gerrit-Owner: Corey Farrell <git at cfware.com>
Gerrit-Reviewer: Anonymous Coward #1000019
Gerrit-Reviewer: Matt Jordan <mjordan at digium.com>
Gerrit-HasComments: No



More information about the asterisk-code-review mailing list