F11 Preview Hard Lock

[Adam Jackson]

On Wed, 2009-04-29 at 15:04 -0500, Chris Adams wrote:
> Once upon a time, Matthew Woehlke <mw_triad at users.sourceforge.net> said:
> > Adam Jackson wrote:
> > >On Wed, 2009-04-29 at 10:28 -0400, Adam Jackson wrote:
> > >>Check out frame 0 [1].  The server is being asked to do a blit that's
> > >>15000 pixels tall.  Who knows why, it could easily be a bug in firefox,
> > >>but it could also be something the web page really did ask for.
> > >
> > >I should note that this really is the web page's fault:
> > >
> > >http://www.wthitv.com/images/bg_module.png
> > >
> > >So that's pretty awesome.  By which I mean, let's drink until we can't
> > >feel feelings anymore.
> > 
> > Indeed. Someone needs to go back to CSS school.
> > 
> > (Anyone sent the webmaster a "your site hard-locks my computer" mail, 
> > yet? :-) )
> 
> Well, to be honest, nothing a silly webmaster does should lock the
> computer.  Firefox should probably be smarter about this.

Ennh.  Firefox doesn't have any way of knowing what the acceleration
size limit is for a given driver.  It could chunk up pixmaps into 2k
blocks, which is nice and conservative, but getting rendering correct
once you do that is entirely non-trivial.  Easy for blits, very hard for
blends and transformations.

There's a project to handle this transparently in the X server.  It
doesn't work yet, but I gave a talk about it at LCA this year:

http://mirror.linux.org.au/pub/linux.conf.au/2009/Friday/76.ogg

and I've got a Summer of Code student looking at it this summer, so,
maybe someday we'll have a real window system.

Also, just to be pedantic, it's not a lockup, just really really slow,
and much slower than it needs to be because of a kernel bug.  Should be
able to get it to be merely a performance hiccup by F11.

- ajax
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part
URL: <http://listman.redhat.com/archives/fedora-devel-list/attachments/20090429/363c2cdd/attachment.sig>