argus 3.0.0 rc.28 memory leaks

[Carter Bullard]

Hey Gabriel,
Thanks, is posix_memalign() on all our platforms?
Carter

On Sep 15, 2006, at 2:27 PM, Gabriel L. Somlo wrote:

> On Mon, Sep 11, 2006 at 12:40:45PM -0400, Carter Bullard wrote:
>> Hey Gabriel,
>> So, ....,  I've added the concepts from your patch to argus, however,
>> there are a number of things that you are suggesting that are not
>> workable.   So, we'll need to test the new rc.29, which I should have
>> up on the server later today,  to see if it solves your memory
>> problem.
>
> Carter,
>
> Thanks for making those changes -- there are no more memory leaks that
> I can see (although I still end up swapping under certain traffic
> conditions -- need to look into that :) )
>
> Meanwhile, couple more things:
>
> 	1. there's some leftover files under argus-3.0.0.rc.29/argus/ :
> 		.argus.c.swp   .ArgusOutput.c.swp   .gdb_history
>
> 	2. ArgusDeleteList() deletes both a WfileStruct and a RfileStruct
> 	   the same way, which generates memory errors (RfileStruct
> 	   has no member 'filter', so looking for one generates an
> 	   invalid memory read operation).
> 	   This is fixed in the first attachment (argus-3.0.0.rc.29- 
> mem1.patch)
>
> 	3. ArgusMalloc(), ArgusCalloc(), and ArgusFree(); The second
> 	   attachment replaces the existing alignment magic with calls to
> 	   posix_memalign() which does the work for you in a standard way,
> 	   and simplifies the code a lot (I also got rid of the
> 	   #ifdef ARGUSMEMDEBUG ... #endif code - it wasn't used
> 	   anyway, and there are pretty good tools to debug that
> 	   should it be needed in the future).
>
> Let me know what you think.
>
> Thanks,
> Gabriel
> <argus-3.0.0.rc.29-mem1.patch>
> <argus-3.0.0.rc.29-mem2.patch>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://pairlist1.pair.net/pipermail/argus/attachments/20060915/2fc02767/attachment.html>