[didier <dgautheron@xxxxxxxx>]

Ronnie Sahlberg wrote:
> ----- Original Message ----- 
> From: "Guy Harris"
> Sent: Tuesday, December 02, 2003 8:11 AM
> Subject: Re: [Ethereal-dev] Ethereal optimizations (proposals)
>
>
>
> > That's the major problem - another potential problem is that
> > "g_node_append()" is, I think, linear in the length of the list to
> > which it's appending, which means that there's some N^2 behavior; this
> > shows up in some very big packets that take many many seconds to be
> > disssected.
>
>
> I have seen that as well, some huge reassembled PDUs take an unproportional
> long time to parse.
>
> I will look at that one tonight.

3 things
- there's a bug in reassembling code with lost frame.
- an unfinished fix for reassemble.c
- with huge capture, by huge I mean capture where ethereal memory usage 
is > physical memory, I've seen system load > 40%

using
mallopt (M_MMAP_MAX, 0);
mallopt (M_TOP_PAD, 10*1024*1024);
mallopt (M_TRIM_THRESHOLD, 20*1024*1024);
help a lot, system < 10%

Didier

Attachment: reassemble.patch.gz
Description: Unix tar archive