Summary: | Make maximumParallelReadCount static to fix lambda capture error in WebKit::NetworkCache::Storage::traverse() | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | John Wilander <wilander> | ||||||
Component: | WebKit2 | Assignee: | John Wilander <wilander> | ||||||
Status: | RESOLVED FIXED | ||||||||
Severity: | Normal | CC: | aestes, cdumez, cgarcia, ews-watchlist, mcatanzaro, ross.kirsling, thorton, webkit-bug-importer, Yousuke.Kimoto | ||||||
Priority: | P2 | Keywords: | InRadar | ||||||
Version: | WebKit Nightly Build | ||||||||
Hardware: | Unspecified | ||||||||
OS: | Unspecified | ||||||||
See Also: | https://bugs.webkit.org/show_bug.cgi?id=182679 | ||||||||
Attachments: |
|
Description
John Wilander
2018-02-14 10:51:59 PST
Created attachment 333820 [details]
Patch
Comment on attachment 333820 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=333820&action=review > Source/WebKit/NetworkProcess/cache/NetworkCacheStorage.cpp:911 > - traverseOperation.activeCondition.wait(lock, [&traverseOperation, maximumParallelReadCount] { > + traverseOperation.activeCondition.wait(lock, [&traverseOperation] { Seems like a clang bug if it's warning about maximumParallelReadCount being unused. It's clearly used! Yeah. I wonder if clang optimizes it and moved it into the lambda? This patch might be ok if GCC allows maximumParallelReadCount to be used without being captured. We should see what EWS says. Otherwise, we might need to either move the definition of maximumParallelReadCount into the lambda, or make it static. Created attachment 333823 [details]
Patch
WPE tree seems red. Comment on attachment 333823 [details]
Patch
Thanks, Tim and Andy!
Committed r228478: <https://trac.webkit.org/changeset/228478> |