836990 - Assertions with transform-style: preserve-3d, overflow, position

Status: RESOLVED FIXED

(Core :: Layout, defect)

Description

[Jesse Ruderman]

Attachment: testcase

Similar symptoms to bug 831335.  Might be another regression from bug 827577.

###!!! ASSERTION: unexpected child list: 'Error', file ../../../layout/generic/nsBlockFrame.cpp, line 4733

###!!! ASSERTION: How did that happen?: 'aFrameItems.IsEmpty()', file ../../../layout/base/nsCSSFrameConstructor.cpp, line 1309

###!!! ASSERTION: Frames getting lost!: 'NS_SUCCEEDED(rv)', file ../../../layout/base/nsCSSFrameConstructor.cpp, line 1314

###!!! ASSERTION: Dangling child list.  Someone forgot to insert it?: '!FirstChild()', file ../../../layout/base/nsCSSFrameConstructor.cpp, line 652

Assertion failure: mPresArenaAllocCount == 0 (Some pres arena objects were not freed), at ../../../layout/base/nsPresShell.cpp:771

Comment 1

[Jesse Ruderman]

Attachment: stacks

Comment 2

[Boris Zbarsky [:bzbarsky]]

> Might be another regression from bug 827577.

Seems very likely.

Comment 3

[Matt Woodrow (:mattwoodrow)]

Rob, I think you should probably take this, I don't know this code at all.

Comment 4

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

Yeah, I don't know why I assigned this to you!

Comment 5

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

Attachment: fix

Comment 6

[Boris Zbarsky [:bzbarsky]]

Robert, I don't quite understand this change.  What causes the bug in this case?

Comment 7

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

In this testcase the fixed-pos container is the transformed overflow:auto element. The actual container frame is the scrollframe's child (the scrolled frame, an nsBlockFrame). That frame is *not* transformed, so ProcessFrameInsertions fails to put fixed-pos items on its abs-pos list.

Comment 8

[Boris Zbarsky [:bzbarsky]]

Comment on attachment 710132 [details] [diff] [review]
fix

Ah, I see.  r=me

Comment 9

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

Comment on attachment 710132 [details] [diff] [review]
fix

[Security approval request comment]
How easily could an exploit be constructed based on the patch?
Unknown.

Do comments in the patch, the check-in comment, or tests included in the patch paint a bulls-eye on the security problem?
Yes.

Which older supported branches are affected by this flaw?
Only trunk and Aurora are affected.

If not all supported branches, which bug introduced the flaw?
827577

Do you have backports for the affected branches? If not, how different, hard to create, and risky will they be?
This will be easy to put on Aurora.

How likely is this patch to cause regressions; how much testing does it need?
Dunno. It's a simple patch but this bit of the code has been regression-prone lately.

Comment 10

[Daniel Veditz [:dveditz]]

Comment on attachment 710132 [details] [diff] [review]
fix

sec-approval+

Comment 11

[Mats Palmgren (inactive)]

Could this be the underlying problem causing the crashes in bug 839263 / bug 837288 ?

Comment 12

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

https://hg.mozilla.org/integration/mozilla-inbound/rev/608f588cb3e9

(In reply to Mats Palmgren [:mats] from comment #11)
> Could this be the underlying problem causing the crashes in bug 839263 / bug
> 837288 ?

I think so. We'll find out shortly :-).

Comment 13

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

Comment on attachment 710132 [details] [diff] [review]
fix

[Approval Request Comment]
Bug caused by (feature/regressing bug #): 827577
User impact if declined: possible security-sensitive crashes
Testing completed (on m-c, etc.): just landed
Risk to taking this patch (and alternatives if risky): relatively low risk for regular browsing
String or UUID changes made by this patch: none

Comment 14

[Ed Morley [:emorley]]

https://hg.mozilla.org/mozilla-central/rev/608f588cb3e9

Comment 15

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/releases/mozilla-aurora/rev/184912994f15