Bug 76927

Summary: [Chromium] Inconsistent initialization of CCSettings in WebLayerTreeView.
Product: WebKit Reporter: David Reveman <reveman>
Component: WebKit Misc.Assignee: David Reveman <reveman>
Status: RESOLVED FIXED    
Severity: Normal CC: jamesr, piman, webkit.review.bot
Priority: P2    
Version: 528+ (Nightly build)   
Hardware: Unspecified   
OS: Unspecified   
Bug Depends on:    
Bug Blocks: 76937    
Attachments:
Description Flags
Patch
none
Patch
none
Patch none

David Reveman
Reported 2012-01-24 11:03:39 PST
perTilePainting and partialTextureUpdates are currently not initialized properly in WebLayerTreeView::Settings::operator CCSettings().
Attachments
Patch (1.67 KB, patch)
2012-01-24 11:08 PST, David Reveman
no flags
Patch (1.91 KB, patch)
2012-02-13 16:06 PST, David Reveman
no flags
Patch (1.64 KB, patch)
2012-02-13 21:28 PST, David Reveman
no flags
David Reveman
Comment 1 2012-01-24 11:08:16 PST
James Robinson
Comment 2 2012-01-24 11:24:36 PST
Comment on attachment 123774 [details] Patch Test?
David Reveman
Comment 3 2012-01-24 13:10:45 PST
This is not as bad as I thought. All settings are initialized properly, there's just some inconsistency about which ones gets set in WebLayerTreeView.cpp. I'd like to fix the inconsistency with this patch prior to landing: https://bugs.webkit.org/show_bug.cgi?id=76937
David Reveman
Comment 4 2012-02-13 16:06:31 PST
Created attachment 126854 [details] Patch Rebase
James Robinson
Comment 5 2012-02-13 20:39:05 PST
Comment on attachment 126854 [details] Patch Wait, CCSettings already has a constructor that sets all of these values to their defaults. Why do we bother manually resetting these to their defaults again in WLTV?
David Reveman
Comment 6 2012-02-13 21:13:12 PST
(In reply to comment #5) > (From update of attachment 126854 [details]) > Wait, CCSettings already has a constructor that sets all of these values to their defaults. Why do we bother manually resetting these to their defaults again in WLTV? I don't know. Maybe to be explicit about what settings are not yet exposed here? This patch just makes sure we're consistent. I'm happy to have it just remove all this unnecessary initialization instead. Would you prefer that?
David Reveman
Comment 7 2012-02-13 21:28:46 PST
Created attachment 126903 [details] Patch Instead remove unnecessary initialization
James Robinson
Comment 8 2012-02-13 21:37:57 PST
Yeah, I think so - it makes more sense to me if WLTV only has to know about the values that it cares about and can leave everything else alone.
James Robinson
Comment 9 2012-02-13 21:38:45 PST
Comment on attachment 126903 [details] Patch I prefer this version. I'm somewhat curious if Antoine has any specific opinion. I definitely don't want to have to update two files every time we add a new setting to CCSettings unless we want to also expose it via the WLTV API.
Antoine Labour
Comment 10 2012-02-13 22:09:01 PST
Just FYI, I think showFPSCounter & showPlatformLayerTree have been accidentally disabled with a merge gone wrong, they used to work (although triggering a harmless assert in debug). I just haven't gone back and check that they actually work... In general, we don't need to expose anything in WLTV::Settings unless we want to use it, and we can rely on the CCSettings constructor to properly initialize things.
WebKit Review Bot
Comment 11 2012-02-14 02:36:14 PST
Comment on attachment 126903 [details] Patch Clearing flags on attachment: 126903 Committed r107695: <http://trac.webkit.org/changeset/107695>
WebKit Review Bot
Comment 12 2012-02-14 02:36:21 PST
All reviewed patches have been landed. Closing bug.
Note You need to log in before you can comment on or make changes to this bug.