Received-SPF: pass (google.com: domain of linux-crypto-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20201201194556.5220-1-ardb@kernel.org> <20201201215722.GA31941@gondor.apana.org.au>
 <CAMj1kXHb27ugTWuQZhPD0DvjtgYC8t_pj+igqK7dNfh+WsUS4w@mail.gmail.com>
 <20201201220431.GA32072@gondor.apana.org.au> <CAMj1kXGO+kbZ+2VmUQKxLYos2nR5vqZKjengxPxPjSXudG-zLw@mail.gmail.com>
 <20201201221628.GA32130@gondor.apana.org.au> <CAMj1kXFrLiHfv1S1AM=5pc1J9gWwZVuoGvmFoTT0-+oREoojTA@mail.gmail.com>
 <20201201231158.GA32274@gondor.apana.org.au> <CAMj1kXE2RULwwxAGRTeACQVCpYoeuY3LmMK0hw4BOQo1gH5d8Q@mail.gmail.com>
 <20201201233024.GB32382@gondor.apana.org.au>
In-Reply-To: <20201201233024.GB32382@gondor.apana.org.au>
From:   Ard Biesheuvel <ardb@kernel.org>
Date:   Wed, 2 Dec 2020 00:41:36 +0100
Message-ID: <CAMj1kXEfRCNuaz_sX29CQ=JsUF6niYbYceXUjy9cq3=eF77mvg@mail.gmail.com>
Subject: Re: [PATCH v2] crypto: aesni - add ccm(aes) algorithm implementation
To:     Herbert Xu <herbert@gondor.apana.org.au>
Cc:     Linux Crypto Mailing List <linux-crypto@vger.kernel.org>,
        Ben Greear <greearb@candelatech.com>,
        Steve deRosier <derosier@cal-sierra.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, 2 Dec 2020 at 00:30, Herbert Xu <herbert@gondor.apana.org.au> wrote:
>
> On Wed, Dec 02, 2020 at 12:24:47AM +0100, Ard Biesheuvel wrote:
> >
> > True. But the fallback only gets executed if the scheduler is stupid
> > enough to schedule the TX task onto the core that is overloaded doing
> > RX softirqs. So in the general case, both TX and RX will be using
> > AES-NI instructions (unless the CCMP is done in hardware which is the
> > most common case by far)
>
> I don't think this makes sense.  TX is typically done in response
> to RX so the natural alignment is for it to be on the same CPU.
>

You just explained that TX typically runs in process context, whereas
RX is handled in softirq context. So how exactly are these going to
end up on the same core?

> > Wireless is very different. Wifi uses a medium that is fundamentally
> > shared, and so the load it can induce is bounded. There is no way a
> > wifi interface is going to saturate a 64-bit AES-NI core doing CCMP in
> > software.
>
> This sounds pretty tenuous.  In any case, even if wireless itself
> doesn't get you, there could be loads running on top of it, for
> example, IPsec.
>

Yes, but IPsec will not use the synchronous interface.

> > Given the above, can't we be pragmatic here? This code addresses a
> > niche use case, which is not affected by the general concerns
> > regarding async crypto.
>
> We already have a framework for acceleration that works properly
> in aesni, I don't want to see us introduce another broken model
> within the same driver.
>
> So either just leave the whole thing along or do it properly by
> making wireless async.
>

Fair enough. But it is unfortunate that we cannot support Ben's use
case without a lot of additional work that serves no purpose
otherwise.