292986 - The page in the URL is slow as hell

Status: RESOLVED INVALID

(SeaMonkey :: General, defect)

Description

[Diego Gonzalez]

User-Agent:       Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8b2) Gecko/20050422 Epiphany/1.7.0
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8b2) Gecko/20050422 Epiphany/1.7.0

The page http://www.automaticjack.tv/ is very very slow t navigate, epiphany
blocks several seconds and scrolling is also slow. I'm using epiphany on a PIII
733 MHz, but it shouldn't be *that* slow.

Reproducible: Always

Steps to Reproduce:
1. Go to the URL given above
2.
3.

Comment 1

[Chris Blore]

Seems fine for me in Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b2)
Gecko/20050505 Firefox/1.0+.

Comment 2

[Thomas K. (:tom)]

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8b2) Gecko/20050504
Firefox/1.0+

Fine here.  Also:

http://www.gnome.org/projects/epiphany/
Epiphany is the web browser for the GNOME desktop.   Powered by the Gecko
engine, Epiphany displays webpages with the same speed and accuracy as Mozilla
Firefox.

However, Epiphany is not Firefox, nor a Mozilla.org official build.  Can you
reproduce the problem on the Mozilla Suite or Firefox?  Downloaded as a binary
from Mozilla.org of course...

Comment 3

[Wolf [:wolf]]

Bugzilla is for bugs in Mozilla products only. Epiphany is not a mozilla.org
product.

--> INVALID.

Reopen if you can reproduce using a Mozilla product, thanks.

Comment 4

[Diego Gonzalez]

Yes, I can reproduce it with Firefox from Debian, but it might turn out that as
Firefox from Debian is not an official build (eventhough firefox from debian is
powered by gecko and by the same source as firefox from mozilla.org) the bug
will be rejected as invalid again. 

By the way I have no intentetion of downloading the official build, I can live
using konqueror on that page.

I have reported bugs several times and is the first time that a bug is rejected
because because the frontend is different when the problem is clearly on the
engine. Hope you don't loose any of those 50.000.000 user you just got.

Comment 5

[Andrew Schultz]

This is almost certainly a flash problem.  The page has 8 of them (including
scrolling banners) -- on a 733MHz machine, they could easily bog down.

resolving INVALID.
reopen if you if you produce a non-flash testcase with perf problems.

Comment 6

[Boris Zbarsky [:bzbarsky]]

So profiling scrolling on that page shows that most of the time is taken in
|sContextStack|, called from |g_timeout_dispatch|.

Any idea what that is?

Comment 7

[Andrew Schultz]

the only instance of sContextStack in Mozilla code is a static variable 
http://lxr.mozilla.org/seamonkey/source/modules/plugin/base/src/nsJSNPRuntime.cpp#73

would jprof pick up frames in flash's lib?

Comment 8

[Boris Zbarsky [:bzbarsky]]

Possibly, yes.  It certainly picks up frames in X libs...