360636 - [reflow branch] ASSERTION: no common ancestor at all???: 'parent', file nsLayoutUtils.cpp, line 345

Status: RESOLVED WORKSFORME

(Core :: Layout, defect)

Description

[Mats Palmgren (inactive)]

###!!! ASSERTION: no common ancestor at all???: 'parent', file nsLayoutUtils.cpp, line 345

STEPS TO REPRODUCE
1. load the attached testcase
2. click the Reload button

The assertion occurs in a Firefox debug build from REFLOW_20061031_BRANCH
a few hours old. The assertion does not occur in a Firefox trunk build.
(I searched for bugs that has been fixed on trunk since 20061031 but
couldn't find anything).

Note: the testcase uses a remote stylesheet, the assertion does not
occur for me if I embed it inline using <style>.

Comment 1

[Mats Palmgren (inactive)]

Attachment: Testcase #1

Comment 2

[Mats Palmgren (inactive)]

FWIW, the testcase was reduced from the Minefield start page.

Comment 3

[Boris Zbarsky [:bzbarsky]]

This seems to be worksforme....

Comment 4

[Mats Palmgren (inactive)]

Attachment: Stack

Comment 5

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

Attachment: fix?

See if this fixes it (I don't have a reflow branch build yet)

Comment 6

[Mats Palmgren (inactive)]

(In reply to comment #5)
> See if this fixes it (I don't have a reflow branch build yet)

It does not fix it (same stack as before).
The bug also occurs on trunk now after the reflow branch landed (same stack).

Comment 7

[Mats Palmgren (inactive)]

FWIW, I've noticed that reloading a page with generated content triggers
the assertion... try reloading for example:
http://meyerweb.com/eric/css/tests/css2/sec12-04-01.htm
http://meyerweb.com/eric/css/tests/css2/sec12-05a.htm
http://meyerweb.com/eric/css/tests/css2/sec12-05b.htm

Comment 8

[[e-mail]]

The last time I bothered to debug this (my debugging setup is a bit screwey after I made the mistake of installing VC8SP1) one of the content nodes was a text node that didn't appear to have a parent, which I thought was odd.

Comment 9

[Martijn Wargers (dead)]

Is this still an issue? I don't get this assertion with a recent trunk debug build.

Comment 10

[Martijn Wargers (dead)]

I still don't see the assertions in my debug build.
Mats, can we resolve this bug to WORKSFORME?

Comment 11

[Mats Palmgren (inactive)]

I have seen this assertion recently, I don't have simple STR for it though...

Comment 13

[[e-mail]]

I don't know if it's the same one but I get one every time I switch to the CSS Style Rules viewer in DOM Inspector.

Comment 14

[Bob Clary [:bc] (inactive)]

I can't seem to reproduce this on winxp/linux minefield from today, but can get a very similar stack in bug 373946. Works for me or Dupe?

Comment 15

[Bob Clary [:bc] (inactive)]

Attachment: stack

Stack using Neil's steps in comment 13. Note iframe related calls similar though not identical to bug 373946. I'll test roc's patch in attachment 247769 [details] [diff] [review] on both bugs next.

Comment 16

[Bob Clary [:bc] (inactive)]

Oops, missed that the patch didn't work (and doesn't apply). If you need anything else, please re-add qawanted.

Comment 17

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Is this still present?  Are there current steps to reproduce?

Comment 18

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

Testcase does not assert on Mac. Matthew G, does it assert for you?

Comment 19

[Matthew Gregan [:kinetik]]

I'm can't reproduce this assert with a current trunk build on Linux using the testcase, the steps in comment #7, or the using DOMi/CSS view suggested in comment #13.