[openssl-dev] [openssl-team] Discussion: design issue: async and -lpthread

[Paul Dale]

Somewhat tangentially related to this is the how thread locking in OpenSSL is slowing things up.

We've been doing some connection establishment performance analysis recently and have discovered a lot of waiting on locks is occurring.  By far the worst culprit is CRYPTO_LOCK_EVP_PKEY in CRYPTO_add_lock calls.  Changing these to gcc's atomic add operations (__sync_add_and_fetch) improved things significantly:

base OpenSSL        11935 connections/s    85% CPU utilisation
with Atomic Change  16465 connections/s    22% CPU utilisation

So fifty percent more connections for a quarter the CPU.  At this point a number of other locks are causing the slow down.

Now, I'm not sure if such a change would be interesting to the community or not, but there definitely  is room for significant gains in the multi threaded locking.  Ignoring the atomic operations and moving to a separate lock per reference count would likely save a an amount of blocking -- is this a suitable use for dynamic locks?


I also submitted a bug report and fix recently [openssl.org #4135] to do with threading, which will hopefully get included eventually.


Regards,

Pauli

-- 
Oracle
Dr Paul Dale | Cryptographer | Network Security & Encryption 
Phone +61 7 3031 7217
Oracle Australia

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mta.openssl.org/pipermail/openssl-dev/attachments/20151124/314c88a1/attachment.html>