Dan Anderson wrote:
> On Tue, 2 Sep 2008, Matthew Ahrens wrote:
>
>> How do we get 4% faster when running on the same endianness as the fs was
>> created on? (created on x86, run on amd64) Where are the bulk of the calls
>> to BSWAP_* happening in that case?
>> --matt
>>
>
> I've been wondering that too--this is just black box testing. I may have to
> look at the code.
>
> I've been setting up a retest that eliminates process overhead (see thread
> discussion) and disk read overhead (by running on swapfs).
> --
> This message posted from opensolaris.org
>
I suspect that if the 4% is "genuine", it may come from device drivers
which may perform endianness conversions. But I'm not sure that the
test results are bogus.
-- Garrett
> _______________________________________________
> crypto-discuss mailing list
> crypto-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/crypto-discuss
>
