Summary: | REGRESSION(r76177): All JavaScriptCore tests fail on ARM | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Csaba Osztrogonác <ossy> | ||||
Component: | JavaScriptCore | Assignee: | Nobody <webkit-unassigned> | ||||
Status: | RESOLVED FIXED | ||||||
Severity: | Blocker | CC: | commit-queue, galpeter, ggaren, koivisto, loki, oliver, zherczeg | ||||
Priority: | P1 | ||||||
Version: | 528+ (Nightly build) | ||||||
Hardware: | All | ||||||
OS: | All | ||||||
Bug Depends on: | |||||||
Bug Blocks: | 52622 | ||||||
Attachments: |
|
Description
Csaba Osztrogonác
2011-01-20 07:26:22 PST
Created attachment 79767 [details]
proposed fix
I'm not an OwnPtr expert, but shouldn't we need to get the approximateByteSize before releasing the OwnPtr?
Tested on an ARM build, with run-sunspider and run-javascriptcore-tests all runs ok.
Comment on attachment 79767 [details]
proposed fix
Perfect. Great fix. This depends on the compiler’s order of evaluation for arguments, so could behave differently for different compilers.
Comment on attachment 79767 [details] proposed fix Clearing flags on attachment: 79767 Committed r76376: <http://trac.webkit.org/changeset/76376> All reviewed patches have been landed. Closing bug. Good catch. Thanks for fixing this. |