| Summary: | Fix memory leak for the Pasteboard singleton | ||||||
|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Mariusz Grzegorczyk <mariusz.g> | ||||
| Component: | WebCore Misc. | Assignee: | Nobody <webkit-unassigned> | ||||
| Status: | RESOLVED INVALID | ||||||
| Severity: | Normal | CC: | gyuyoung.kim, mifenton, mjs, mrowe, rakuco, rwlbuis, tonikitoo, webkit.review.bot | ||||
| Priority: | P2 | ||||||
| Version: | 528+ (Nightly build) | ||||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Attachments: |
|
||||||
Created attachment 168165 [details]
patch
Comment on attachment 168165 [details] patch Attachment 168165 [details] did not pass mac-ews (mac): Output: http://queues.webkit.org/results/14252500 Comment on attachment 168165 [details]
patch
No.
Nowhere in WebKit do we free memory at exit; the OS is much better at doing that. In fact - we have warnings in place to catch this behavior and that's why the Mac build is failing.
|
All ports have generalPasteboard() implementation looks similar to: Pasteboard* Pasteboard::generalPasteboard() { static Pasteboard* pasteboard = new Pasteboard(); return pasteboard; } Because of this, memory is not deallocated at application exit.