[Larceny-users] Performance Of SHA-256 Implementation

[Ray Racine]

I coded up a somewhat naive literal translation of the the SHA-256 hash algo
along with the corresponding HMAC using Larceny (Funny In Head) ERR5RS.  I'm
actually using a slightly modified build from the latest trunk.

http://github.com/RayRacine/rl3/blob/master/rl3/crypto/hash/sha256.sls

Here is the transcript of an example from Amazon's API doc which requires a
SHA-256 HMAC.

(import (rnrs)
(rl3 crypto hash sha256)
(rl3 crypto base64)
(primitives time))

(define msg "GET\nwebservices.amazon.com
\n/onca/xml\nAWSAccessKeyId=00000000000000000000&ItemId=0679722769&Operation=ItemLookup&ResponseGroup=ItemAttributes%2COffers%2CImages%2CReviews&Service=AWSECommerceService&Timestamp=2009-01-01T12%3A00%3A00Z&Version=2009-01-06")
(define key "1234567890")
(time (base64-encode (hmac-sha256 key msg)))
> Words allocated: 4194032
Words reclaimed: 0
Elapsed time...: 998 ms (User: 981 ms; System: 1 ms)
Elapsed GC time: 0 ms (CPU: 2 in 4 collections.)
"Nace+U3Az4OhN7tISqgs1vdLBHBEijWcBeCqL5xN9xg="

The answer is correct, however, the 1 sec calculation time is a bit longer
than anticipated along with the 4 million allocated words.  I understand why
bitwise manipulations are going to be slower in dynamically typed language
with type tag bits.  Thought I'd toss it out in the hopes someone may spot a
low hanging fruit change in my implementation, recommend some compiler
options etc.   At a minimum it might make a decent benchmark for grading
(rnrs arithmetic bitwise) implementations.


If nothing obvious is noted I'll go with an FFI solution.


Ray

P.S. Noticed the Larceny SVN and mail list has been a bit quiet. Anything
new and notable coming down on the Larceny front? 64Bit? Native threads?

-- 
The object of life is not to be on the side of the majority, but to escape
finding oneself in the ranks of the insane. - Marcus Aurelius
-------------- next part --------------
HTML attachment scrubbed and removed