Message ID | 20210518151655.125153-1-clabbe@baylibre.com |
---|---|
Headers | show |
Series | crypto: add gemini/sl3516 crypto driver | expand |
On Tue, May 18, 2021 at 03:16:50PM +0000, Corentin Labbe wrote: > The gemini SL3516 SoC has a crypto IP. > This serie had support for it. > Please describe how this was tested. - Eric
On Tue, May 18, 2021 at 11:38:38PM +0200, Linus Walleij wrote: > On Tue, May 18, 2021 at 5:17 PM Corentin Labbe <clabbe@baylibre.com> wrote: > > > This patch adds documentation for Device-Tree bindings for the > > SL3516-ce cryptographic offloader driver. > > Signed-off-by: Corentin Labbe <clabbe@baylibre.com> > > Apart from misspelled subject "db-binding" I don't see any problems > so And try not to say 'dt binding' twice in the subject and 'documentation' is redundant as that's all bindings. > Reviewed-by: Linus Walleij <linus.walleij@linaro.org> Reviewed-by: Rob Herring <robh@kernel.org> > > Yours, > Linus Walleij
Le Wed, May 19, 2021 at 11:39:23AM -0700, Eric Biggers a écrit : > On Tue, May 18, 2021 at 03:16:50PM +0000, Corentin Labbe wrote: > > The gemini SL3516 SoC has a crypto IP. > > This serie had support for it. > > > > Please describe how this was tested. Sorry I forgot to write it. It was tested with CONFIG_CRYPTO_MANAGER_EXTRA_TESTS=y. Furthermore, it survives hours and gigs of write/read to a luks2 using xts(ecb-aes-sl3516). I forgot also to write on performance, which is quite good. On a luks2 partition (2To with bs=4096), a fsck comes from 14m26(without CE) to 8m48(with CE). So it is really usefull. With bs=512, the performance is similar with software. Regards