33181 - Wheel scrolling causes a memory violation

Status: VERIFIED WORKSFORME

(SeaMonkey :: General, defect, P3)

Description

[Robert Gee]

While viewing www.infoworld.com and scrolling the page using the Micron mouse
wheel, a memory exception occurred. Attached is an image of the actual message box.

Windows NT 4.0 SP4
128 MB
Diamond Multimedia Viper V770
Resolution 1024 x 768, 72 Hertz
Pointer device using msi8042
Mozilla Build 2000032308

Comment 1

[[e-mail] (gone - use [e-mail] instead)]

Please include the attachment.  Thanks

Comment 2

[Robert Gee]

Attachment: JPEG image of memory violation

Comment 3

[[e-mail] (gone - use [e-mail] instead)]

Have you enabled system stack dumps from Dr. Watson?  Or any other kind of 
debug program?  Perhaps you could post this here, in text form also?

Thanks

Comment 4

[Asa Dotzler [:asa]]

[e-mail] handles most mousewheel bugs.  I'm not sure this is really a
mouswheel bug but I'm copying him to take a look

Comment 5

[old account]

I will test this on Win32 when my build gets done.  A stack trace of some kind 
would be a HUGE help.

Comment 6

[old account]

Sorry, the checkin I just did a few minutes ago, which I marked as fixing this 
bug, was actually the fix for bug 32232.

Comment 7

[old account]

I'll take this bug.

Comment 8

[Robert Gee]

I'm in the process of changing jobs.  Whenever I get set up on my new computer, 
I'll launch Visual C++ and obtain some better statistics. 

Comment 9

[old account]

I tried to reproduce this just now (on Win98) and could not.  Some things that 
would be helpful:

- Where did you have the mouse pointer positioned when you were scrolling?
- Did it crash after the first scroll, or partway down the page?
- (Really Helpful) Turn on mousewheel logging as described in bug 20618 and 
attach the log.

If you had the pointer outside the window, you probably encountered bug 32232.

Comment 10

[old account]

Can you confirm whether this is a dup of bug 32232, please?

Thanks.

Comment 11

[Robert Gee]

Hi, I looked at bug 32232, and it does look similar.  However, I always view the 
browser in a maximized state so that the mouse is always inside the 
focused browser window.  I just received my new computer at work, so I should be 
able to test with the latest nightly build to verify the problem.  I hope to 
have an update by the April 3 - 7, 2000 timeframe.  Thanks for your patience.

Comment 12

[old account]

Hm, interestingly enough, I don't know if I've ever tested mousewheel on Win32
with the window maximized.  I might try that and see if anything strange
happens.

Comment 13

[old account]

Do you have anything more to report on this?  Unless you can verify that it's
still happening (and hopefully get a talkback report sent) I may have to close
this as WORKSFORME.

Comment 14

[Robert Gee]

No, I don't have any new information.  If it works for you in maximized mode on 
Windows, please go ahead and put the status to WORKSFORME.

Thanks,

Robert Gee

Comment 15

[old account]

works for me

Comment 16

[Doron Rosenberg (IBM)]

**Mass Spamm**
adding verifyme keyword to all worksforme bugs.

Comment 17

[[e-mail] (gone - use [e-mail] instead)]

VERIFIED WORKSFORME using Windows NT 200012908.

Comment 18

[Brian Ryner (not reading)]

transferring these to my netscape.com email.