Bill:
Thanks, I would like very much to take you up on your offer, too. I wrote
my own a while back for precisely the same reason, but it was very Unix
specific (e.g., relied on signals, used pipes to collect results, etc.).
--Joe
>I'd appreciate a copy. We'd been meaning to get around to doing that;
>I'm glad you beat us to it :-).
>thanks much,
>Diana
>----------------------------
>Diana Smetters
>Member of the Research Staff
>Xerox PARC
>> [EMAIL PROTECTED] wrote:
>>
>> Hi,
>> I've modified speed.c ("openssl speed") to run with a variable number
>> of threads simultaneously. Why would I want to do this? Besides the
>> obvious use of testing on MP boxes, this is great for evaluating hardware
>> ENGINEs. If the hardware has multiple encryption chips, running with
>> multiple threads enables you to fully saturate the hardware card, so you
>> can really measure what the card is capable of doing.
>> At the moment, I've only completed this for RSA tests, because that is
>> what I am interested in right now. The other tests will run with multiple
>> threads, but the results won't be calculated correctly. I'll skip the gory
>> details, but I found a way to run with multiple threads, with minimal
>> modifications to the original code; the only problem is calculating the
>> results for each test.
>> Since this is probably for a "niche" market, I'm not sure if this is
>> something that needs to be rolled into the base code. However, if anyone
>> is interested, I can provide this for you to play with. Just send me a
>> note.
>>
>> Bill Hymas
>> IBM Global Services / IBM Linux Technology Center
>> Heidelberg, Germany
>>
______________________________________________________________________
OpenSSL Project http://www.openssl.org
Development Mailing List [EMAIL PROTECTED]
Automated List Manager [EMAIL PROTECTED]
