on 01/09/2009 21:38 Edward Tomasz Napierala said the following:
> On 0901T1849, Markus Brueffer wrote:
> 
> [..]
> 
>> The real problem is not the kdelibs3 implementation but a regression in 
>> FreeBSD introduced with the NFSv4 ACL import (rev 194955).
>>
>> Before the import, acl_to_text checked if the passed acl_t was a NULL 
>> pointer 
>> and aborted. Although this still happens, the original POSIX.1e function is 
>> now called through a wrapper (acl_to_text_np) which determines what type of 
>> ACL is being passed and in the process dereferences the passed acl_t without 
>> checking if it is a NULL pointer.
>>
>> The attached patch should fix this (only compile tested). Other acl 
>> functions 
>> may have the same regression concerning POSIX.1e. From a very quick glance, 
>> at 
>> least acl_calc_mask does, maybe others. 
>>
>> trasz: Can this be fixed in time for 8-RELEASE?
> 
> I think so.  Already committed to HEAD.  Thanks for spotting this.

Markus, Edward,

big thanks!
I updated to the latest head and the problem went away.

-- 
Andriy Gapon
_______________________________________________
kde-freebsd mailing list
kde-freebsd@kde.org
https://mail.kde.org/mailman/listinfo/kde-freebsd
See also http://freebsd.kde.org/ for latest information

Reply via email to