Seems reasonable to me. I wonder, should this do something in the case where -nostdlib has been passed, so we're not linking against libc by default? Or should it error/warning/etc.?
One comment inline below. > On 2017-Feb-02, at 16:56, Matthias Braun via Phabricator > <revi...@reviews.llvm.org> wrote: > > MatzeB created this revision. > Herald added a subscriber: mcrosier. > > While there is nothing to do at link time to get pthreads support on > darwin, specifying -pthread is fine and should not produce a warning > about unused arguments. > > > Repository: > rL LLVM > > https://reviews.llvm.org/D29479 > > Files: > lib/Driver/Tools.cpp > test/Driver/darwin-ld-pthread.c > > > Index: test/Driver/darwin-ld-pthread.c > =================================================================== > --- /dev/null > +++ test/Driver/darwin-ld-pthread.c > @@ -0,0 +1,5 @@ > +// RUN: %clang -target x86_64-apple-darwin -c -o %t.o %s I *think* you can skip this command, and just pass /dev/null as the object file to the next driver command. > +// RUN: %clang -Wunused-command-line-argument -pthread -target > x86_64-apple-darwin -### %t.o -o %t.bin 2>&1 | > FileCheck %s > + > +// There is nothing to do at link time to get pthread support. But do not > warn. > +// CHECK-NOT: argument unused during compilation: '-pthread' > Index: lib/Driver/Tools.cpp > =================================================================== > --- lib/Driver/Tools.cpp > +++ lib/Driver/Tools.cpp > @@ -8718,6 +8718,9 @@ > } > } > > + // No need to do anything for pthreads. Claim argument to avoid warning. > + Args.ClaimAllArgs(options::OPT_pthread); > + > const char *Exec = Args.MakeArgString(getToolChain().GetLinkerPath()); > std::unique_ptr<Command> Cmd = > llvm::make_unique<Command>(JA, *this, Exec, CmdArgs, Inputs); > > > <D29479.86921.patch> _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
