Brian Shire wrote:
> 
> On Mar 28, 2008, at 6:35 AM, Gopal V wrote:
>> Looks like the regression is in the memleak patches by shire. I didn't
>> pay enough attention when he committed it, but those leaks were ignored
>> or rather left in-limbo because of past bugs. Other parts of
>>
>> http://news.php.net/php.pecl.cvs/10041
>>
>> b0rks up this older "fix" :)
>>
>> http://cvs.php.net/viewvc.cgi/pecl/apc/apc_main.c?r1=3.98&r2=3.99
>>
>> The order in which extensions are unloaded and the way they use
>> memory makes it really hard for APC to zero-copy data without
>> leaks - memcpy'ing opcodes into local memory is a huge perf-hit,
>> but the workaround with huge refcounts introduces such leaks.
>>
>> Maybe I should make it clearer in code.
>>
> 
> Ah, thanks Gopal, I didn't realize callbacks where happening in the 
> rshutdown's of other modules.   If there where a hook into the function 
> cleanup I assume it would be an easy fix to clean this up (although it's 
> technically cosmetic only, but does help people spot *actual* new 
> "leaks").  It seems like the classes could also have the same pitfalls 
> if it happened to be used by the callback function, is this a possible 
> issue or am missing something there?

Classes have static members and leaving them as leaks results in
weird issues

http://pecl.php.net/bugs/bug.php?id=8526

Yes, a session handler with a class static function wouldn't work
with APC on - but that's a "known bug", while the bug above was
a straight out SEGV :)

Cheers,
Gopal
-- 
Strange, I've fixed that bug multiple times.