[asterisk-dev] [asterisk-commits] may: trunk r306499 - /trunk/addons/chan_ooh323.c
Kevin P. Fleming
kpfleming at digium.com
Tue Feb 8 06:52:25 CST 2011
On 02/07/2011 10:17 AM, Richard Mudgett wrote:
>> On 02/05/2011 11:16 PM, SVN commits to the Asterisk project wrote:
>>> Author: may
>>> Date: Sat Feb 5 16:16:07 2011
>>> New Revision: 306499
>>>
>>> URL: http://svnview.digium.com/svn/asterisk?view=rev&rev=306499
>>> Log:
>>> fix trivial issue after dvossel patch, initial zero fill user and
>>> peer
>>> structure before cap structure allocated.
>>>
>>> Modified:
>>> trunk/addons/chan_ooh323.c
>>>
>>> Modified: trunk/addons/chan_ooh323.c
>>> URL:
>>> http://svnview.digium.com/svn/asterisk/trunk/addons/chan_ooh323.c?view=diff&rev=306499&r1=306498&r2=306499
>>> ==============================================================================
>>> --- trunk/addons/chan_ooh323.c (original)
>>> +++ trunk/addons/chan_ooh323.c Sat Feb 5 16:16:07 2011
>>> @@ -2179,11 +2179,11 @@
>>>
>>> user = ast_calloc(1,sizeof(struct ooh323_user));
>>> if (user) {
>>> + memset(user, 0, sizeof(struct ooh323_user));
>>> if (!(user->cap = ast_format_cap_alloc_nolock())) {
>>> ast_free(user);
>>> return NULL;
>>> }
>>> - memset(user, 0, sizeof(struct ooh323_user));
>>> ast_mutex_init(&user->lock);
>>> ast_copy_string(user->name, name, sizeof(user->name));
>>> ast_format_cap_copy(user->cap, gCap);
>>
>> This cannot be the real fix for whatever problem you are seeing... you
>> are allocating the structure using ast_calloc(), which returns
>> zero-filled memory. Setting it to zero again isn't going to accomplish
>> much :-)
>>
> Yes it does fix a problem. The struct was memset to zero after the ast_format_cap_alloc_nolock() thus a memory leak.
True... but the memset() was never necessary in the first place and
should be removed :-)
--
Kevin P. Fleming
Digium, Inc. | Director of Software Technologies
445 Jan Davis Drive NW - Huntsville, AL 35806 - USA
skype: kpfleming | jabber: kfleming at digium.com
Check us out at www.digium.com & www.asterisk.org
More information about the asterisk-dev
mailing list