Summary: | Cannot compile JavaScriptCore/runtime/VMTraps.cpp on FreeBSD because std::pair has a non-trivial copy constructor | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Ting-Wei Lan <lantw44> | ||||
Component: | JavaScriptCore | Assignee: | JF Bastien <jfbastien> | ||||
Status: | RESOLVED FIXED | ||||||
Severity: | Normal | CC: | buildbot, commit-queue, Hironori.Fujii, jfbastien, keith_miller, mark.lam, msaboff, saam, ysuzuki | ||||
Priority: | P2 | ||||||
Version: | WebKit Local Build | ||||||
Hardware: | Unspecified | ||||||
OS: | Unspecified | ||||||
Attachments: |
|
Description
Ting-Wei Lan
2017-04-15 08:26:25 PDT
It is surprising to see the constexpr version being instantiated here. (In reply to JF Bastien from comment #1) > It is surprising to see the constexpr version being instantiated here. JSC::VM*, WTF::StackBounds, JSC::VMTraps::Error are all trivially destructible, so WTF::ExpectedDetail::BaseSelect is WTF::ExpectedDetail::ConstexprBase here. Created attachment 307486 [details] patch (In reply to Ting-Wei Lan from comment #2) > (In reply to JF Bastien from comment #1) > > It is surprising to see the constexpr version being instantiated here. > > JSC::VM*, WTF::StackBounds, JSC::VMTraps::Error are all trivially > destructible, so WTF::ExpectedDetail::BaseSelect is > WTF::ExpectedDetail::ConstexprBase here. Derp, right you are! Here's a patch that works around the issue. Comment on attachment 307486 [details]
patch
r=me if EWS bots build and are happy.
Comment on attachment 307486 [details]
patch
Tests are broken for unrelated reason. Fetch or somesuch.
cq+
Comment on attachment 307486 [details] patch Clearing flags on attachment: 307486 Committed r215522: <http://trac.webkit.org/changeset/215522> All reviewed patches have been landed. Closing bug. |