| Summary: | [V8] Rename isolated() to getWorld(), rename worldForEnteredContextIfIsolated() to worldForEnteredContext() | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Kentaro Hara <haraken> | ||||||
| Component: | WebCore JavaScript | Assignee: | Kentaro Hara <haraken> | ||||||
| Status: | RESOLVED FIXED | ||||||||
| Severity: | Normal | CC: | abarth, dcarney, dglazkov, japhet, mkwst+watchlist, webkit.review.bot | ||||||
| Priority: | P2 | ||||||||
| Version: | 528+ (Nightly build) | ||||||||
| Hardware: | Unspecified | ||||||||
| OS: | Unspecified | ||||||||
| Attachments: |
|
||||||||
|
Description
Kentaro Hara
2013-02-06 03:54:27 PST
Created attachment 186816 [details]
Patch
Comment on attachment 186816 [details] Patch Attachment 186816 [details] did not pass cr-linux-debug-ews (chromium-xvfb): Output: http://queues.webkit.org/results/16388281 Comment on attachment 186816 [details]
Patch
Looks like the cr-linux-debug bot is unhappy with this patch.
Created attachment 187009 [details]
patch for landing
Comment on attachment 187009 [details]
patch for landing
A patch which this patch depends on was rolled out.
Committed r142424: <http://trac.webkit.org/changeset/142424> This change makes no sense. The function on DOMWrapperWorld only returns an isolated world. It does that for performance. This leads to semantic ambiguity like in V8GCController::minorGCPrologue:
DOMWrapperWorld* world = worldForEnteredContextWithoutContextCheck();
if (world && world->isMainWorld()) {
the branch will never be taken and minor gc is disabled because of this.
Please consider this reverting this renaming.
ah... This was a big mistake. Let me revert them. Because we're limiting the number of nodes handled by each minor GC to 10000, I couldn't detect the performance regression by Dromaeo/dom-modify... |