openoffice-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Patricia Shanahan <>
Subject Re: Profile.c bugs (was RE: Some thoughts on the learning curve)
Date Tue, 01 Mar 2016 04:11:55 GMT
On 2/29/2016 7:41 PM, Pedro Giffuni wrote:
> Hello;
> FWIW, I have looked at i126669 and Coverity confirms there is a bug there.
> CID 440168 (#1 of 1): Dereference after null check (FORWARD_NULL)9.
> var_deref_model: Passing null pointer "&pProfile->m_AccessLock" to
> function "pthread_mutex_unlock(pthread_mutex_t *)", which dereferences it.
> I suggest taking pthread_mutex_unlock outside the "if".
> I am attaching a patch: CC'd to orcmid jic the list rejects it.
> Of course it needs testing and you have to check the windows code too.

The current code does not unlock m_AccessLock in the non-null case until 
line 366, after doing more work on the profile. I am reasonably sure we 
should leave that alone. The question is what, if anything, should be 
done about m_AccessLock in the pProfile null case.


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message