Ethereal-dev: Re: [Ethereal-dev] Latest Ethereal profiling data ...

Note: This archive is from the project's previous web site, ethereal.com. This list is no longer active.

Date Index Thread Index Other Months All Mailing Lists
Date Prev Date Next Thread Prev Thread Next
From: Ian Schorr <spamcontrol2@xxxxxxxxxxx>
Date: Mon, 15 Dec 2003 16:41:53 -0500

On Dec 9, 2003, at 2:03 AM, Guy Harris wrote:


On Tue, Dec 02, 2003 at 03:20:25PM -0800, Guy Harris wrote:

Mainly called from "cell_size_request()", which is, in turn, mainly
called from "gtk_clist_optimal_column_width()", called mainly from
"gtk_clist_set_column_auto_resize()".  Not automatically sizing the
columns might be the right answer here.


Well, it made about a 10% difference when reading in one test capture
(without a color filter).
I've checked in a change to get rid of the column auto-resizing. A side effect is that it no longer picks the too-narrow column widths that I've 
seen it pick.
Of course, it makes a big difference whether GTK+2 is being used or not, for reasons you described quite a while ago.
There are serious performance gains either way, but I've seen up to 40% improvement with GTK+2 depending on the size of the trace - 500,000 frames used to take a VERY long time to auto-size (during which time Ethereal looked like it simply hung), now that delay is completely gone. 

Ian