[Bug 1910] An option for architecture-specific SMP locks

bugzilla-daemon at rtems.org bugzilla-daemon at rtems.org
Wed Nov 23 16:20:57 UTC 2011


--- Comment #13 from Marta Rybczynska <marta.rybczynska at kalray.eu> 2011-11-23 10:20:41 CST ---
(In reply to comment #12)
> I wonder if we are missing some ingredient here to make the story complete. Are
> you using the _CPU_cache_flush_1_data_line in your target to do more work than
> just flushing a line from the data cache? If your cache_flush_1_data_line does
> some cpu-specific locking, then we need a new function and not to abuse an
> existing interface. If the cache flush is actually fixing some problem though,
> then this patch seems good to me.

The flush function doesn't do more stuff than described here: it makes sure the
value written to unlock the lock really goes to memory, because the
compare-and-swap in this case bypasses cache completely.

Configure bugmail: https://www.rtems.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.

More information about the bugs mailing list