Dears,

I have see some performance regression with the benchmark through
"openssl speed sha1" command (example).

The reason is quite simple, the test methodology has changed in 1.0.1.

(simplified form)

Before:
EVP_MD_CTX_init()
EVP_MD_CTX_set_flags()
loop() {
        EVP_DigestInit_ex()
        EVP_DigestUpdate()
        EVP_DigestFinal_ex()
}
EVP_MD_CTX_cleanup()

After:
loop() {
        EVP_MD_CTX_init()
        EVP_MD_CTX_set_flags()
        EVP_DigestInit_ex()
        EVP_DigestUpdate()
        EVP_DigestFinal_ex()
        EVP_MD_CTX_cleanup()
}

Can it be possible to rool back on the previous algorithm to have comparable 
result ?

Note: The performance lost is about 20% in my case.

Regards

-- 
Alexandre Martins
NETASQ -- We secure IT

Attachment: smime.p7s
Description: S/MIME cryptographic signature

Reply via email to