On 12/9/20 11:30 PM, Ard Biesheuvel wrote:
On Thu, 10 Dec 2020 at 04:01, Ben Greear <greearb@xxxxxxxxxxxxxxx> wrote:
On 12/9/20 6:43 PM, Herbert Xu wrote:
On Thu, Dec 10, 2020 at 01:18:12AM +0100, Ard Biesheuvel wrote:
One thing I realized just now is that in the current situation, all
the synchronous skciphers already degrade like this.
I.e., in Ben's case, without the special ccm implementation, ccm(aes)
will resolve to ccm(ctr(aesni),cbcmac(aesni)), which is instantiated
as a sync skcipher using the ctr and ccm/cbcmac templates built on top
of the AES-NI cipher (not skcipher). This cipher will also fall back
to suboptimal scalar code if the SIMD is in use in process context.
Sure, your patch is not making it any worse. But I don't think
the extra code is worth it considering that you're still going to
be running into that slow fallback path all the time.
How can we test this assumption? I see 3x performance gain, so it is not hitting
the fallback path much in my case. What traffic pattern and protocol do you think
will cause the slow fallback path to happen often enough to make this patch not
helpful?
Is there a way to verify Herbert's assertion that TX and RX tend to be
handled by the same core? I am not a networking guy, but that seems
dubious to me.
You could add a pr_warn_ratelimited() inside the fallback path and see
if it ever gets called at all under various loads.
Even if it does sometimes use the same core, if performance is better and
CPU usage is lower, why would it even matter?
Anyway, looks like Herbert is dead set against this code in hopes that he
can force other subsystems to re-write their code. If you come up with
some other variant that Herbert will accept, let me know and I'll test it.
Otherwise, I will just add your patch to my kernel and carry on.
Thanks,
Ben
--
Ben Greear <greearb@xxxxxxxxxxxxxxx>
Candela Technologies Inc http://www.candelatech.com
