Re: [RFC PATCH 0/1] dm-crypt excessive overhead

[Ignat Korchagin]

On Fri, Jun 19, 2020 at 7:39 PM Mikulas Patocka <mpatocka@xxxxxxxxxx> wrote:
>
>
>
> On Fri, 19 Jun 2020, Mike Snitzer wrote:
>
> > On Fri, Jun 19 2020 at 12:41pm -0400,
> > Ignat Korchagin <ignat@xxxxxxxxxxxxxx> wrote:
> >
> > > This is a follow up from the long-forgotten [1], but with some more convincing
> > > evidence. Consider the following script:
> > >
> > > [1]: https://www.spinics.net/lists/dm-crypt/msg07516.html
> > > [2]: https://blog.cloudflare.com/speeding-up-linux-disk-encryption/
> > >
> > > Ignat Korchagin (1):
> > >   Add DM_CRYPT_FORCE_INLINE flag to dm-crypt target
> > >
> > >  drivers/md/dm-crypt.c | 55 +++++++++++++++++++++++++++++++++----------
> > >  1 file changed, 43 insertions(+), 12 deletions(-)
> > >
> > > --
> > > 2.20.1
> > >
> >
> > Hi,
> >
> > I saw [2] and have been expecting something from cloudflare ever since.
> > Nice to see this submission.
> >
> > There is useful context in your 0th patch header.  I'll likely merge
> > parts of this patch header with the more terse 1/1 header (reality is
> > there only needed to be a single patch submission).
> >
> > Will review and stage accordingly if all looks fine to me.  Mikulas,
> > please have a look too.
> >
> > Thanks,
> > Mike
>
> +       if (test_bit(DM_CRYPT_FORCE_INLINE, &cc->flags)) {
> +               if (in_irq()) {
> +                       /* Crypto API will fail hard in hard IRQ context */
> +                       tasklet_init(&io->tasklet, kcryptd_crypt_tasklet, (unsigned long)&io->work);
> +                       tasklet_schedule(&io->tasklet);
> +               } else
> +                       kcryptd_crypt(&io->work);
> +       } else {
> +               INIT_WORK(&io->work, kcryptd_crypt);
> +               queue_work(cc->crypt_queue, &io->work);
> +       }
>
> I'm looking at this and I'd like to know why does the crypto API fail in
> hard-irq context and why does it work in tasklet context. What's the exact
> reason behind this?

This comes from
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/crypto/skcipher.c?id=5e857ce6eae7ca21b2055cca4885545e29228fe2#n433
And, I believe, it is there for the same reason kcryptd was introduced
in 2005 in dm-crypt:
"...because it would be very unwise to do decryption in an interrupt
context..." (that is, when other interrupts are disabled). In tasklet
however we can still service other interrupts even if we process a
large block.

>
>
> Mikulas