At 05:53 PM 4/23/02 -0400, Robert Love wrote:
>On Tue, 2002-04-23 at 17:51, Dieter Nützel wrote:
>
>> Page coloring for 2.4.18+ isn't preempt save?
>>
>> It gave ~10% speedup for memory intensive apps on my single 1 GHz Athlon II
>> SlotA (0,18µm, L2 512K) but look the system hard from time to time. Nothing
>> in the logs.
>>
>> I've changed the patch for 2.4.19-pre7 + vm3 + latest rml-O(1) + preempt.
>
>Beats me. Some of the implementations of page colouring I have seen are
>not even SMP-safe.
>
>"Don't do that"
While I haven't tested on an SMP system, the patch should at least be
SMP safe. As for preempt-safe, I noticed in Dieter's previous tests that
the free list had been sufficiently fragmented that there were no high-
order groups of pages left available. Could this cause a problem?
I can't track this list carefully enough to apply and test all the patches
mentioned.
jasonp
-
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 : Tue Apr 23 2002 - 22:00:37 EST