297039 - Middle click takes long time to load on lengthy websites

Status: RESOLVED WORKSFORME

(Core :: Layout, defect)

Description

[Yuchao719]

User-Agent:       Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.8b2) Gecko/20050606 Firefox/1.0+
Build Identifier: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.8b2) Gecko/20050606 Firefox/1.0+

Middle click+scroll on a lengthy page such as the URL listed takes more than 3
seconds of waiting to take effect. Middle click+scroll takes 3 seconds to end.

Reproducible: Always

Steps to Reproduce:
1.Go to a lengthy text webpage such as:
http://64.233.167.104/search?q=cache:xCjuF_8ig4sJ:www.artsy.ca/stone.pdf+He+dashed+back+across+the+road,+hurried+up+to+his+office,+snapped+at+his+secretary+not+to+disturb+him,+seized+his+telephone,+and+had+almost+finished+dialing+his+home+number+when+he+changed+his+mind.+He+put+the+receiver+back+down+and+stroked+his+mustache,+thinking...+no,+he+was+being+stupid.+Potter+wasn%27t+such+an+unusual+name.+He+was+sure+there+were+lots+of+people+called+Potter+who+had+a+son+called+Harry.+Come+to+think+of+it,+he+wasn%27t+even+sure+his+nephew+was+called+Harry.+Picture+11+He%27d+never+even+seen+the+boy.+It+might+have+been+Harvey.+Or+Harold.+There+was+no+point&hl=en&client=firefox
2.Middle click the mouse and scroll for a while
3.Cancel middle click
Actual Results:  
Middle click+scroll takes a long time to both load and end itself. Quite chunky
scroll too, especially when using my an old 466MHz system.

Expected Results:  
Instant or slightly lagging middleclick+scroll as shown perfectly in Opera and IE. 

I'm not sure if this is a problem with 1GHz+ systems. "Veteran systems" like
mine should be respected. :)

Comment 1

[Martijn Wargers (dead)]

I don't see the bug with a 2004-11-25 build.
I can see the bug with a 2004-11-26 build (autoscroll image takes much longer to
appear, more cpu load).

I think this is a regression from fixing bug 270804.
I've backed out the patch from that bug in my (not a debug) build, and after
that the autoscroll image appeared much faster.

Comment 2

[Martijn Wargers (dead)]

Attachment: 7-zipped up testcase

A quick testcase. Without the patch from bug 270804, it took 25277ms.
(The result with the patch will have to wait, no time now).

Comment 3

[Martijn Wargers (dead)]

With the patch from bug 270804, it took 149525ms

Comment 4

[Yuchao719]

This bug doesn't appear on 1GHz+ systems. My computer is only a Celeron 466MHz

Comment 5

[Robert O'Callahan (:roc) (email my personal email if necessary)]

Is this for every page, or just some pages?

Comment 6

[Martijn Wargers (dead)]

No, not on every page, just some pages, and probably only noticable on slow
computers (this looks rather similar to bug 273293, although the cause is different)
Personally, I'm more worried about the loading performance, which is quite bad
on my 600MHz duron computer.

Comment 7

[Robert O'Callahan (:roc) (email my personal email if necessary)]

That page is has a huge number of absolutely positioned DIVs, which need to be
shrink-wrapped. I guess we're reflowing their lines many more times since the
patch in bug 270804 landed. I'll think about how this can be improved but it
doesn't seem to be high priority.

Comment 8

[Wayne Mery (:wsmwk)]

Yuchao719 do you still see a problem?

WFM Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9b4pre) Gecko/2008021504 on amd X2 1.8Ghz modest laptop.

autoscroll appears immediately and is responsive with no delays in all directions.

Comment 9

[Wayne Mery (:wsmwk)]

=> WFM per comment 8