RE: [RFC] Convert pgtable cache to slab

From: Luck, Tony <tony.luck_at_intel.com>
Date: 2004-10-14 05:07:17
>With this patch I got a significant improvement in page fault time.
>Standard 2.4.x was about 700ns on average.  Generic 2.6.9rc4 is 3-4
>usec, whereas the slabified pgtcache drops us back down to 600-700 ns.

What was the benchmark you were running when you measured these?

>Tested on zx1 and sn2.

How does this scale on big sn2 systems? I assume that the answers are
going to be just fine as the slab allocator per-cpu lists should be
just as nice as local quicklists, but it would be nice to see some
data.

>The biggest caveat is that I've had to postpone setting up the gate
>page until the pgt slab has been initialized.  That wasn't an issue
>with the existing page-based allocation scheme.  David - how do you
>prefer I handle this?

I don't see the code for this postponement in this patch.

>Comments?

Overall looks nice ... less code, and goes faster too, what more
could we ask for!

Perhaps "zero_cache" isn't as descriptive a name as it might be (not
that I have any better suggestions :-(

-Tony
-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Received on Wed Oct 13 15:09:57 2004

This archive was generated by hypermail 2.1.8 : 2005-08-02 09:20:31 EST