On 5/15/23 05:48, Herbert Xu wrote:

On Fri, May 12, 2023 at 09:22:51PM +0100, Dmitry Safonov wrote:

quoted

TCP-AO, similarly to TCP-MD5, needs to allocate tfms on a slow-path,
which is setsockopt() and use crypto ahash requests on fast paths,
which are RX/TX softirqs. Also, it needs a temporary/scratch buffer
for preparing the hash.

Rework tcp_md5sig_pool in order to support other hashing algorithms
than MD5. It will make it possible to share pre-allocated crypto_ahash
descriptors and scratch area between all TCP hash users.

Internally tcp_sigpool preferences crypto_clone_ahash() API over
pre-allocating per-CPU crypto requests. Kudos to Herbert, who provided
this new crypto API [1]. Currently, there's still per-CPU crypto request
allocation fallback, that is needed for ciphers, that yet don't support
cloning (TCP-AO requires cmac(aes128) in RFC5925).

I was a little concerned over GFP_ATOMIC allocations of ahash and
crypto_request in RX/TX (see tcp_sigpool_start()), so I benchmarked both
"backends" with different algorithms, using patched version of iperf3[2].
On my laptop with i7-7600U @ 2.80GHz:

                         clone-tfm                per-CPU-requests
TCP-MD5                  2.25 Gbits/sec           2.30 Gbits/sec
TCP-AO(hmac(sha1))       2.53 Gbits/sec           2.54 Gbits/sec
TCP-AO(hmac(sha512))     1.67 Gbits/sec           1.64 Gbits/sec
TCP-AO(hmac(sha384))     1.77 Gbits/sec           1.80 Gbits/sec
TCP-AO(hmac(sha224))     1.29 Gbits/sec           1.30 Gbits/sec
TCP-AO(hmac(sha3-512))    481 Mbits/sec            480 Mbits/sec
TCP-AO(hmac(md5))        2.07 Gbits/sec           2.12 Gbits/sec
TCP-AO(hmac(rmd160))     1.01 Gbits/sec            995 Mbits/sec
TCP-AO(cmac(aes128))     [not supporetd yet]      2.11 Gbits/sec

So, it seems that my concerns don't have strong grounds and per-CPU
crypto_request allocation can be dropped/removed from tcp_sigpool once
ciphers get crypto_clone_ahash() support.

This support is now in the upstream kernel.  Please let me know
if you run into any issues using it.

Hi Herbert, thanks for your patches. Could you point me to the repo that
has ciphers clone-tfm support? I've looked in Torvald's/master, your
cryptodev-2.6.git and in linux-next, but I can't see anywhere in
cmac_create() something of inst->alg.base.clone_tfm = cmac_clone_tfm kind.

As I wrote two paragraphs above, it's required for TCP-AO to provide
cmac(aes128) support. Let me know if you have cmac clone-tfm somewhere
or if you're cooking it. On the cover-letter for this patch set, it's in
TODO.

Thanks,
         Dmitry