Huge thanks to our Platinum Members Endace and LiveAction,
and our Silver Member Veeam, for supporting the Wireshark Foundation and project.

Wireshark-dev: Re: [Wireshark-dev] g_mem-chunk_destroy does not free memory for GLib > 2.8

From: "Anders Broman" <anders.broman@xxxxxxxxxxxx>
Date: Tue, 13 Oct 2009 03:55:57 +0200
Hi,
Should fixes to these probleems be backported to 1.2.3?
Regards
Anders 

-----Original Message-----
From: wireshark-dev-bounces@xxxxxxxxxxxxx
[mailto:wireshark-dev-bounces@xxxxxxxxxxxxx] On Behalf Of Bill Meier
Sent: den 11 oktober 2009 15:32
To: Developer support list for Wireshark
Subject: Re: [Wireshark-dev] g_mem-chunk_destroy does not free memory
for GLib > 2.8

Kovarththanan Rajaratnam wrote:
> Hey Bill,
> 
> 
> Is there any reason why we couldn't use se_alloc and friends? This 
> should be faster now that we're defaulting to _not_ use canaries for 
> seasonal memory.
> 

Using se_alloc is the conclusion I came to.... (when the code frees
GMemChunk memory per-capture).

However: packet-isakmp destroys GMemChunks more often. I haven't yet
looked at the other dissectors (and other code) using GMemChunks to see
if there are other cases wherein GMemChunks are destroyed more often
than every capture.

If so, it occurs to me that it might be quite simple to just allow
dissectors & etc  to init/alloc/free their own emem_chunks using
slightly generalized emem code.....

Bill


________________________________________________________________________
___
Sent via:    Wireshark-dev mailing list <wireshark-dev@xxxxxxxxxxxxx>
Archives:    http://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
 
mailto:wireshark-dev-request@xxxxxxxxxxxxx?subject=unsubscribe