> On Nov 14, 2017, at 7:52 AM, Arnold Schwaighofer <[email protected]> > wrote: > >> On Nov 14, 2017, at 12:33 AM, Slava Pestov via swift-dev >> <[email protected] <mailto:[email protected]>> wrote: >> > >>> On Nov 13, 2017, at 10:47 PM, Greg Parker via swift-dev >>> <[email protected] <mailto:[email protected]>> wrote: >>> >>> I see that in our Linux builds clang complains that it does not support the >>> swiftcall attribute. >>> >>> In file included from >>> /home/buildnode/jenkins/workspace/swift-PR-Linux/branch-master/swift/unittests/runtime/Metadata.cpp:13: >>> >>> /home/buildnode/jenkins/workspace/swift-PR-Linux/branch-master/swift/include/swift/Runtime/Metadata.h:3015:1: >>> warning: unknown attribute 'swiftcall' ignored [-Wunknown-attributes] >>> >>> How bad is this? Do we not use a custom calling convention on Linux >>> platforms today? >> >> This is while building the metadata unit tests — I’m assuming those are >> built with the host clang? Do we build the runtime with the Swift clang on >> Linux? If so, we should still be able to use swiftcall. > > Yes we build the runtime with the “just built clang”. We should switch over > the unit tests tests too. They definitely could call code in the runtime > where the ABI difference matters: for example heap object destructors use > swiftcc.
You're right. I filed https://bugs.swift.org/browse/SR-6387 <https://bugs.swift.org/browse/SR-6387> . -- Greg Parker [email protected] <mailto:[email protected]> Runtime Wrangler
_______________________________________________ swift-dev mailing list [email protected] https://lists.swift.org/mailman/listinfo/swift-dev
