On 2020-08-19 00:27, Herbert Xu wrote:
On Wed, Aug 19, 2020 at 08:15:50AM +1000, Herbert Xu wrote:
On Tue, Aug 18, 2020 at 07:17:35AM -0700, Ben Greear wrote:
Is there any easy way to use your work to make shash fast for aesni? I
basically just want it to perform as well as it used to with my patch.
Yes. We could add a sync version of aesni that simply falls back
to aes-generic when simd is unavailable.
But I think before anyone attempts this we should explore making
mac80211 async like IPsec. Is there any fundamental reason why
that is not possible? Have the wireless people expressed any
objections to making this async before?
Ohh, is this still about a heavily-updated and rewritten version
of my old initial patch from 2014 for 3.16-wl?
<https://lore.kernel.org/linux-wireless/1518134.xFh23iA8q1@blech/>
Because back in 2016, I've asked this on linux-wireless:
| It would be a great if mac80211 would do to the encryption and
| decryption asynchronously. As this would work for other ciphers
| and also allows crypto offload to dedicated crypto hardware.
And the answer back then (same as now) was:
<https://lore.kernel.org/linux-wireless/1477168300.4123.8.camel@xxxxxxxxxxxxxxxx/>
>The only problem with that is that we'd essentially need a software
>queue for *all* frames, and release them from there in RX order after
>decrypt. That's surely doable, but so far nobody has thought it
>important enough since mostly HW crypto is used ...
Ben, keep up the good work!
Cheers,
Christian
