BTW, on OS/2 used CPU time of a thread is not available after a thread has ended, so right now cumulative times shown are wrong.
Best regards. Maurilio. Maurilio Longo wrote: > Przemyslaw, >> You can also try >> --thread=<n> --scale >> In such version each test is executed <n> times by one thread and then >> simultaneously once by <n> threads. It means that it exploits the worst >> possible cases in synchronization and scalability. It should also quite >> well test spinlocks on multi CPU machines. >> > I've attached this test, with thread=2, this is on a PIV HT, so no real dual > core. > >> BTW have you tested the speed difference between _gettid() and _hb_gettid() >> in OS2 GCC builds? Or maybe GCC _gettid() makes exactly the same job as our >> _hb_gettid()? >> > No speed test done, but I think there is only one way to have this info on > OS/2, that is, going through the Dos... API. > > Best regards. > > Maurilio. > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Harbour mailing list (attachment size limit: 40KB) > Harbour@harbour-project.org > http://lists.harbour-project.org/mailman/listinfo/harbour -- __________ | | | |__| Maurilio Longo |_|_|_|____| farmaconsult s.r.l. _______________________________________________ Harbour mailing list (attachment size limit: 40KB) Harbour@harbour-project.org http://lists.harbour-project.org/mailman/listinfo/harbour
