Bug 128218 – [CSS Regions] Null dereference applying animation with CSS regions

• Component: CSS
• Status: RESOLVED FIXED

[Jon Honeycutt]

Created attachment 223174 [details]
Repro case

Crash when opening the attached test case, which applies a CSS animation when using CSS regions.

<rdar://problem/15983499>

[Radu Stavila]

Created attachment 223230 [details]
Patch

[Andrei Bucur]

Comment on attachment 223230 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=223230&action=review

> Source/WebCore/css/StyleResolver.cpp:3144
> +    if (!parentStyle)

Are we sure early returning is the right behavior here? What's the expected result in the case of regions? E.g. Maybe we don't properly zoom when we actually should do it. I see that the parent style is not used in the code below the checks so maybe we're supposed to actually zoom the thing. We should test and file a bug if zooming doesn't work correctly. Probably the same applies for the generic family change function.

> Source/WebCore/style/StyleResolveTree.cpp:455
> +        ASSERT((!child->renderer() || child->inNamedFlow()) || current.shadowRoot() || isInsertionPoint(current));

What certain situations are you talking about in the CL?

[Radu Stavila]

Comment on attachment 223230 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=223230&action=review

>> Source/WebCore/css/StyleResolver.cpp:3144
>> +    if (!parentStyle)
> 
> Are we sure early returning is the right behavior here? What's the expected result in the case of regions? E.g. Maybe we don't properly zoom when we actually should do it. I see that the parent style is not used in the code below the checks so maybe we're supposed to actually zoom the thing. We should test and file a bug if zooming doesn't work correctly. Probably the same applies for the generic family change function.

Yes, I believe early returning is the right behaviour because it follows the same pattern as the checkForGenericFamilyChange method, which is called just before checkForZoomChange. But I agree that zooming with regions should be tested and a bug filed if some problems are discovered.

>> Source/WebCore/style/StyleResolveTree.cpp:455
>> +        ASSERT((!child->renderer() || child->inNamedFlow()) || current.shadowRoot() || isInsertionPoint(current));
> 
> What certain situations are you talking about in the CL?

The case described in the test-case, animations applied to a region containing an element and a child of that element being flowed into a different flow.

[WebKit Commit Bot]

Comment on attachment 223230 [details]
Patch

Rejecting attachment 223230 [details] from review queue.

[e-mail] does not have reviewer permissions according to http://trac.webkit.org/browser/trunk/Tools/Scripts/webkitpy/common/config/contributors.json.

- If you do not have reviewer rights please read http://webkit.org/coding/contributing.html for instructions on how to use bugzilla flags.

- If you have reviewer rights please correct the error in Tools/Scripts/webkitpy/common/config/contributors.json by adding yourself to the file (no review needed).  The commit-queue restarts itself every 2 hours.  After restart the commit-queue will correctly respect your reviewer rights.

[Andrei Bucur]

(In reply to comment #4)
> (From update of attachment 223230 [details])
> Rejecting attachment 223230 [details] from review queue.
> 
> [e-mail] does not have reviewer permissions according to http://trac.webkit.org/browser/trunk/Tools/Scripts/webkitpy/common/config/contributors.json.
> 
> - If you do not have reviewer rights please read http://webkit.org/coding/contributing.html for instructions on how to use bugzilla flags.
> 
> - If you have reviewer rights please correct the error in Tools/Scripts/webkitpy/common/config/contributors.json by adding yourself to the file (no review needed).  The commit-queue restarts itself every 2 hours.  After restart the commit-queue will correctly respect your reviewer rights.

I suppose the duplicate entry for abucur in contributors.json confused the commit bot. Reapplying Simon's r+.

[WebKit Commit Bot]

Comment on attachment 223230 [details]
Patch

Rejecting attachment 223230 [details] from review queue.

[e-mail] does not have reviewer permissions according to http://trac.webkit.org/browser/trunk/Tools/Scripts/webkitpy/common/config/contributors.json.

- If you do not have reviewer rights please read http://webkit.org/coding/contributing.html for instructions on how to use bugzilla flags.

- If you have reviewer rights please correct the error in Tools/Scripts/webkitpy/common/config/contributors.json by adding yourself to the file (no review needed).  The commit-queue restarts itself every 2 hours.  After restart the commit-queue will correctly respect your reviewer rights.

[WebKit Commit Bot]

Comment on attachment 223230 [details]
Patch

Rejecting attachment 223230 [details] from review queue.

[e-mail] does not have reviewer permissions according to http://trac.webkit.org/browser/trunk/Tools/Scripts/webkitpy/common/config/contributors.json.

- If you do not have reviewer rights please read http://webkit.org/coding/contributing.html for instructions on how to use bugzilla flags.

- If you have reviewer rights please correct the error in Tools/Scripts/webkitpy/common/config/contributors.json by adding yourself to the file (no review needed).  The commit-queue restarts itself every 2 hours.  After restart the commit-queue will correctly respect your reviewer rights.

[WebKit Commit Bot]

Comment on attachment 223230 [details]
Patch

Clearing flags on attachment: 223230

Committed r163531: <http://trac.webkit.org/changeset/163531>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.