Przemyslaw, I've used your inlined asm inc/dec routines and now xHarbour MT is a little bit faster, but not so much.
xHarbour MT total application time: 203.21 total real time: 203.22 Best regards. Maurilio. Maurilio Longo wrote: > Przemyslaw, > > Przemyslaw Czerpak wrote: >> I expected worse results then Harbour ones but I didn't expect such >> big difference and such big MT overhead. xHarbour does not even use >> protection for complex variable reference counters in OS2 builds - >> Walter implemented them only for MS-Windows. > > Now that you reminded me about this I've checked and my local copy of xHarbour > has this change for OS/2 as well (I just had not enough time to commit it) but > it is done using a C runtime functions > > #define HB_ATOMIC_INC( x ) __atomic_increment_s32( (long *) &(x) ) > #define HB_ATOMIC_DEC( x ) __atomic_decrement_s32( (long *) &(x) ) > > instead of your inlined asm, so this could have a much higher overhead. > > I'll try to port your code back to xHarbour and see what changes speed wise. > > I did a full rebuild before testing, though. > > Best regards. > > Maurilio. > -- __________ | | | |__| Maurilio Longo |_|_|_|____| farmaconsult s.r.l. _______________________________________________ Harbour mailing list [email protected] http://lists.harbour-project.org/mailman/listinfo/harbour
