Messages in this thread | | | Date | Fri, 9 Jan 2009 20:18:14 +1100 | From | Herbert Xu <> | Subject | Re: [RFC PATCH crypto 4/4] AES-NI: Add support to Intel AES-NI instructions for x86_64 platform |
| |
On Fri, Jan 09, 2009 at 04:54:33PM +0800, Huang Ying wrote: > > - cryptd thread is not per-CPU, so I think there will be some > unnecessary cache inter-CPU migration. Why not use a dedicate workqueue
Well it shouldn't be hard to make cryptd per-cpu.
> or just system events workqueue?
That's actually bad because the system events queue is a shared resource which is often subject to starvation problems. On the one hand we may be starved by others and we may also starve others by doing too much crypto.
> - with cryptd(__*-aes-aesni), we need 4 internal tfms for each external > tfm allocation request. For example, for one external cbc(aes) tfm > allocation request, we need one cbc(aes) ablkcipher tfm, one > cryptd(cbc-aes-aesni) tfm, and two cbc-aes-aesni tfm. Do we use too much > memory? And we need to call aesni_set_key() twice.
Not at all, tfms are just "shell" objects and they were designed to be used in thie way. Calling setkey twice is an issue but it's not a show-stopper. We have the same problem in other places to so this something that we can potentially optimise.
Cheers, -- Visit Openswan at http://www.openswan.org/ Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au> Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
| |