On Mon, 4 Mar 2002, Martin J. Bligh wrote:
> > Maybe that's more a sympthom that the rmap is doing
> > something silly with the lock acquired,
>
> It seems that we're reusing the pagemap_lru_lock for both the lru chain
> and the pte chain locking, which is hurting somewhat. Maybe a per-zone
> lock is enough to break this up (would also dispose of cross-node lock
> cacheline bouncing) ... I still think the two chains need to be seperated
> from each other though.
Absolutely agreed. I'll happily accept patches for this,
but unfortunately don't have the time to implement this
myself right now.
The reason the pagemap_lru_lock protects both the lru
lists and the pte_chain lists is that it was the easiest
way to get -rmap running on SMP and I had (and still have)
a pretty large TODO list of -rmap and unrelated things...
regards,
Rik
-- Will hack the VM for food.http://www.surriel.com/ http://distro.conectiva.com/
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu Mar 07 2002 - 21:00:40 EST