> On 30.05.2019, at 12:14, David Sterba wrote:
>
> On Wed, May 29, 2019 at 09:32:59PM +0200, David Gstir wrote:
>> If you aim for using as many of the hardware drivers as possible,
>> it might be better to use the ahash API, since some drivers
>> (eg. CAAM on NXP's i.MX) only implement that API
On Wed, May 29, 2019 at 09:32:59PM +0200, David Gstir wrote:
> If you aim for using as many of the hardware drivers as possible,
> it might be better to use the ahash API, since some drivers
> (eg. CAAM on NXP's i.MX) only implement that API and not shash.
> Looking at crypto_ahash_init_tfm(...) in
Hi Johannes!
> On 22.05.2019, at 10:19, Johannes Thumshirn wrote:
>
> Currently btrfs_csum_data() relied on the crc32c() wrapper around the crypto
> framework for calculating the CRCs.
>
> As we have our own crypto_shash structure in the fs_info now, we can
> directly call into the crypto frame
On Wed, May 22, 2019 at 11:33:39AM +0300, Nikolay Borisov wrote:
> > + memset(csum, 0xff, btrfs_super_csum_size(sctx->fs_info->super_copy));
David can you fold this in?
diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c
index 65bd4999c7ad..d97039338952 100644
--- a/fs/btrfs/scrub.c
+++ b/fs/btrfs
On Wed, May 22, 2019 at 11:33:39AM +0300, Nikolay Borisov wrote:
>
> > @@ -1799,16 +1801,22 @@ static int scrub_checksum_data(struct scrub_block
> > *sblock)
> > if (!sblock->pagev[0]->have_csum)
> > return 0;
> >
> > + shash->tfm = fs_info->csum_shash;
> > + shash->flags =
On 22.05.19 г. 11:19 ч., Johannes Thumshirn wrote:
> Currently btrfs_csum_data() relied on the crc32c() wrapper around the crypto
> framework for calculating the CRCs.
>
> As we have our own crypto_shash structure in the fs_info now, we can
> directly call into the crypto framework without goin
