On Thu, Jun 01, 2023 at 03:10:18PM -0400, Mike Snitzer wrote:
>
> Strikes me as strange that open-coding skcipher_request_{alloc,free}
> is ideal, but dm-crypt is the only non-crypto consumer of
> MAX_CIPHER_BLOCKSIZE so really not worth standing up yet another
> interface wrapper.
It is pretty st
On Thu, Jun 01 2023 at 4:47P -0400,
Herbert Xu wrote:
> MAX_CIPHER_BLOCKSIZE is an internal implementation detail and should
> not be relied on by users of the Crypto API.
>
> Instead of storing the IV on the stack, allocate it together with
> the crypto request.
>
> Signed-off-by: Herbert Xu
