333493 - Hang with sub - display: table-cell and div - display: table-column

Status: VERIFIED FIXED

(Core :: Layout: Tables, defect)

Description

[Martijn Wargers (dead)]

See upcoming testcase, which hangs current trunk Mozilla build.
Doesn't hang in 2006-04-07 build, hangs in 2006-04-08 build, I think a regression from bug 325984.

Basically, the testcase that causes the hang, consists of this:
<sub style="display: table-cell;"></sub>
<div style="display: table-column;"></div>

Comment 1

[Martijn Wargers (dead)]

Attachment: testcase

Comment 2

[Jesse Ruderman]

I think I've hit this too, while playing with bug 306939.  I got a fairly similar testcase.

Comment 3

[Bernd]

Yes I work on this  ;-) , it was only a very busy week for me.

Comment 4

[Bernd]

Hi Boris,

here I am again with the same mess as in bug 325984 and bug 329900
and you told me more than once:
>I _think_ I follow this.  :(  My eyes bleed, though.  :(

>We really need a better system for this stuff.  :(

And basically what I currently see is what we (I) know since a long time:
http://lxr.mozilla.org/seamonkey/source/layout/tables/nsTableFrame.cpp#385
>385 // XXX this needs to be cleaned up so that the frame constructor breaks out col group
>386 // frames into a separate child list.

Knowing my glacial speed, I am considering to backout 325984. Boris how long would the current trunk state of affairs be tolerable?

Comment 5

[Boris Zbarsky [:bzbarsky]]

At least a month, I'd guess.  I don't think we're shipping anything till then no matter how much I want to.

Possibly more like 6 weeks or more....

Comment 6

[Bernd]

Attachment: patch

Comment 7

[Bernd]

fix checked in

Comment 8

[Bernd]

Comment on attachment 219045 [details] [diff] [review]
patch

This is the cleanup of bug 325984 which has a branch approval.

Comment 9

[Bernd]

Attachment: cumulative 1.8/1.8.0 branch patch

this patch accumulates bug 33493 bug 325984 and bug 329900

Comment 10

[Daniel Veditz [:dveditz]]

Comment on attachment 219045 [details] [diff] [review]
patch

removing branch approvals from this patch in favor of the regression-fix cummulative patch in bug 333493

Comment 11

[Daniel Veditz [:dveditz]]

Comment on attachment 226173 [details] [diff] [review]
cumulative 1.8/1.8.0 branch patch

Requesting 1.8 approval from bz. The regression history of this series of bugs is scaring us off getting it in 1.8.0.5, please get landed on the 1.8 branch (when approved by bz) for baking and for now we'll assume this one is for 1.8.0.6

Comment 12

[Bernd]

fixed on the 1.8 branch

Comment 13

[Daniel Veditz [:dveditz]]

We really do want bug 329900, no known regressions since this one in April on the trunk I think we'll try for 1.8.0.5 after all.

Comment 14

[Daniel Veditz [:dveditz]]

Comment on attachment 226173 [details] [diff] [review]
cumulative 1.8/1.8.0 branch patch

approved for 1.8.0 branch, a=dveditz for drivers

Comment 15

[Bernd]

the cumulative patch got checked into 1.8.0.5

Comment 16

[Adam Guthrie]

Verified FIXED using Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.0.5) Gecko/20060621 Firefox/1.5.0.5. 

No hang when viewing the testcase.