6:42 p.m.
I recently started running XEmacs 21.5 (built from the Mercurial
repository¹) on my home machine.
Usually when at work I ssh to my home machine and use gnuclient to
access the always running XEmacs (and Gnus). My ADSL line is 8/2 mbit/s,
so the bandwidth is okay.
With 21.4 this was pretty snappy, but with 21.5 updating the frame is
quite slow - I can literally see the lines being drawn one by one, and
often miss things because the cursor moves slowly when I hit the arrow
keys (when I don't count, or miscount.)
Is this a side-effect of XFT and fancy fonts, or is it something else?
Is there anything (easy O:-)), to do about it?
Just curious,
Adam
¹ http://article.gmane.org/gmane.emacs.xemacs.beta/30317
--
"I think there are enough frivolous lawsuits in this Adam Sjøgren
country without people fighting over pop songs." asjo(a)koldfront.dk
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
3:01 a.m.
Adam Sjøgren writes:
With 21.4 this was pretty snappy, but with 21.5 updating the frame
is
quite slow - I can literally see the lines being drawn one by one, and
often miss things because the cursor moves slowly when I hit the arrow
keys (when I don't count, or miscount.)
Is this a side-effect of XFT and fancy fonts, or is it something else?
Probably something else if the cursor movement is slow. I've never
had a problem with that.
OTOH, frame update is way overengineered; the native widgets are
microcontrolled instead of being left to handle drawing themselves. I
think this is because the guy who implemented them was primarily
interested in the Windows platform which evidently doesn't have a
proper notion of widget. I suspect this causes a lot of roundtrips
and redisplay restarts.
Is there anything (easy O:-)), to do about it?
No. You could try building --without-widgets --with-xft. However,
I'm afraid that the basic flow of redisplay has been changed to
support widgets, so that all that will happen is you save redraws of
native buttons (rare in regular frames) and the tab widget.
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
5:39 p.m.
On Fri, 04 Sep 2009 11:01:39 +0900, Stephen wrote:
[...]
I suspect this causes a lot of roundtrips and redisplay restarts.
I just compared performance of pressing page down through a long file to
how GNU Emacs does over the same connection, using the same fancy font -
it is much faster (on par with XEmacs 21.4.)
> Is there anything (easy O:-)), to do about it?
No. You could try building --without-widgets --with-xft. However,
I'm afraid that the basic flow of redisplay has been changed to
support widgets, so that all that will happen is you save redraws of
native buttons (rare in regular frames) and the tab widget.
Ok, I will give that a try and see what happens. Thanks!
Best regards,
Adam
--
"I carry it over my shoulder like a lance Adam Sjøgren
A pole with a banner sayin'; T-To-Tolerance!!" asjo(a)koldfront.dk
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
7:09 p.m.
On Fri, 04 Sep 2009 18:39:25 +0200, Adam wrote:
On Fri, 04 Sep 2009 11:01:39 +0900, Stephen wrote:
> No. You could try building --without-widgets --with-xft.
However,
> I'm afraid that the basic flow of redisplay has been changed to
> support widgets, so that all that will happen is you save redraws of
> native buttons (rare in regular frames) and the tab widget.
Ok, I will give that a try and see what happens. Thanks!
That didn't really make any difference. The menubar updates quicker I
think, but everything else is slow. Using Gnus remotely is a pain.
Are there any, perhaps not as easy, things to do to try and remedy this?
I guess turning off xft won't make it any better?
I wonder how I can benchmark/profile this, I guess that would be the
place to start?
Best regards,
Adam
--
"Can you believe that? Hello? I'm an Information Adam Sjøgren
Architect, not a Knowledge Engineer!" asjo(a)koldfront.dk
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
4:29 a.m.
Adam Sjøgren writes:
I wonder how I can benchmark/profile this, I guess that would be the
place to start?
Well, the place to start is the Internals manual, the chapter on
redisplay. It's not terribly detailed and somewhat out of date, but
it gives an overview of how things work. Then look at
src/redisplay*.c and related files.
Profiling won't help much. You can see that the problem is the the
same content gets redrawn over and over again. The question that
needs to be answered is "why?" and I don't think profiling can answer
that because redraws are event-driven. The question is what are the
events that are causing redraws, and why are they so frequent?
My hypothesis is based on the fact that redisplay can be "cancelled"
or "aborted" (I forget the exact word used) if it detects that the
display has changed since it started, somehow. This is *supposed* to
be a reaction to X events initiated by the user, but what actually
happens is that redisplay goes behind the toolkit's back to map and
unmap windows, which generates events, which causes redisplay to
abort. I'm pretty sure that this is what is causing the problem.
There's some crufty code which tries to reognize events induced by
redisplay itself, but it pretty clearly doesn't work.
You might try putting a counter in places where such an "abort" can
happen. Use the DEFVAR_INT macro to define it, then you can just do
redisplay_abort_counter++ in those places, and check it from Lisp.
For comparison, you could see if you can find (or build) a 21.1 XEmacs
(21.4 won't do, it already has the cruft in redisplay). Of course it
won't have Xft, but you can roughly compare the number of aborts. If
my hypothesis is correct, the difference should be spectacular.
You could also use xevent to capture the events going to the window
and see what they are.
_______________________________________________
XEmacs-Beta mailing list
XEmacs-Beta(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-beta
5340
days inactive
5346
days old
4 comments
2 participants
participants (2)
-
Adam Sjøgren -
Stephen J. Turnbull