On 24 March 2017 at 15:26, Fabien DESSENNE <fabien.dessenne@xxxxxx> wrote:
> On 24/03/17 05:28, PrasannaKumar Muralidharan wrote:
>> On 21 March 2017 at 20:43, Fabien Dessenne <fabien.dessenne@xxxxxx> wrote:
>>> This module registers a CRC32 ("Ethernet") and a CRC32C (Castagnoli)
>>> algorithm that make use of the STMicroelectronics STM32 crypto hardware.
>>>
>>> Theses algorithms are compatible with the little-endian generic ones.
>>> Both algorithms use ~0 as default seed (key).
>>> With CRC32C the output is xored with ~0.
>>>
>>> Using TCRYPT CRC32C speed test, this shows up to 900% speedup compared
>>> to the crc32c-generic algorithm.
>> Comparing with crc3c-generic alogrithm does not sound like a good
>> metric for someone who has to decide between hw crypto or not.
>> Wouldn't it be better if the comparison is between crc32 using NEON
>> with hw crypto module? It will help in choosing between hw crypto or
>> arch optimised crc routiene.
>
> The STM32 microcontrollers are based on ARM Cortex-M7 (or older core)
> that do not have NEON support.
I was not aware of the absence of NEON support. Sorry for the noise.
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
