This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [Patch/RFC] tr1::shared_ptr<> removal of lock, choosing thread safety


Paolo Carlini wrote:

+// { dg-options "-march=i586 -pthread" { target { {*-*-freebsd* *-*-netbsd* *-*-linux* alpha*-*-osf*} && { *i686*-*-* *i586*-*-* *athlon*-*-* *pentium4*-*-* *opteron*-*-* *k8*-*-* } } } }
+// { dg-options "-pthread" { target { {*-*-freebsd* *-*-netbsd* *-*-linux* alpha*-*-osf*} && { ! { *i686*-*-* *i586*-*-* *athlon*-*-* *pentium4*-*-* *opteron*-*-* *k8*-*-* } } } } }
+// { dg-options "-pthreads" { target *-*-solaris* } }


+// { dg-do run { target *-*-freebsd* *-*-netbsd* *-*-linux* *-*-solaris* *-*-cygwin *-*-darwin* alpha*-*-osf* } }
Are we really, really, sure we don't try to run the test on targets *not* providing the builtin atomics? What about old Sparc, for example? For now, outside the special x86 case, maybe better running the test only when _GLIBCXX_ATOMIC_BUILTINS is defined...

I'm thinking that maybe you can split this test in two: a first, generic, version, stressing whatever base class the target uses; a second one, stressing the lockfree base, and running for now only on a small, conservative set of targets.


Paolo.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]