24244 - Upper part of long documents incorrectly drawn.

Status: VERIFIED WORKSFORME

(Core :: XUL, defect, P2)

Description

[jst]

Loading a long document paints the upper part of the document incorrectly,
scrolling down some paints that part of the document correctly. I have a
testcase available that I will attach.

I noticed this problem on

	http://slashdot.org/askslashdot/00/01/13/2330201.shtml

Comment 1

[jst]

Attachment: Testcase that shows the problem.

Comment 2

[jst]

I only se this problem on Linux (with GTK).

The test I just attached is a html file containing

some text 0
some text 1
some text 2
...
some text 1000

When I view this in mozilla (a couple of hours old) the document is painted
correctly when I scroll down so that the line "some text 9" is not displayed,
from there on down to the bottom of the document everything seems OK but nothing
above "some text 9" is painted correctly if even pained at all.

Could this be some 16 bit values wrapping around somewhere? If I remove 10 lines
from the testcase then mozilla paints the doc as it should!

Comment 3

[Peter Trudelle]

reassigning to pavlov as p2 for m14

Comment 4

[Stuart Parmenter]

This is working now.  I believe that Patrick Beard's view manager changes fixed
this.

Comment 5

[jst]

This is not working for me yet (about 1 hour old tree). I'll try this again
tomorrow after rebuilding and if it's not working I'll reopen this bug.

Comment 6

[Stuart Parmenter]

ok.  I was seeing this a day or so ago, but I didn't see it in my latest tree.

Comment 7

[jst]

I rebuilt yesterday (after adding my last comment) and the problem is still
there. Reopening this bug.

Comment 8

[leger]

Clearing FIXED resolution due to reopen.

Comment 9

[Peter Trudelle]

Putting on beta1 radar.

Comment 10

[Peter Trudelle]

resolving as wfm, [e-mail]: if you still see this then reopen with
sufficient information for us to reproduce (just saying I still see it is not
sufficient).

Comment 11

[jst]

This is indeed fixed in todays build, the test documents are both correctly
redrawn now :).

Comment 12

[Paul MacQuiddy]

thanks, jst! marking verified then