On Thu, Aug 31, 2017 at 10:17:43AM -0500, Eric Blake wrote:
> On 08/31/2017 06:05 AM, Daniel P. Berrange wrote:
> > Instead of sector offset, take the bytes offset when encrypting
> > or decrypting data.
> >
> > Signed-off-by: Daniel P. Berrange <berra...@redhat.com>
> > ---
> > block/crypto.c | 8 ++++----
> > block/qcow.c | 7 +++++--
> > block/qcow2-cluster.c | 8 +++-----
> > block/qcow2.c | 4 ++--
> > crypto/block-luks.c | 12 ++++++++----
> > crypto/block-qcow.c | 12 ++++++++----
> > crypto/block.c | 14 ++++++++------
> > crypto/blockpriv.h | 4 ++--
> > include/crypto/block.h | 14 ++++++++------
> > 9 files changed, 48 insertions(+), 35 deletions(-)
> >
> > diff --git a/block/crypto.c b/block/crypto.c
> > index 40daa77188..6b8d88efbc 100644
> > --- a/block/crypto.c
> > +++ b/block/crypto.c
> > @@ -426,8 +426,8 @@ block_crypto_co_preadv(BlockDriverState *bs, uint64_t
> > offset, uint64_t bytes,
> > goto cleanup;
> > }
> >
> > - if (qcrypto_block_decrypt(crypto->block, sector_num, cipher_data,
> > - cur_bytes, NULL) < 0) {
> > + if (qcrypto_block_decrypt(crypto->block, offset + bytes_done,
> > + cipher_data, cur_bytes, NULL) < 0) {
>
> How close are we to getting rid of even needing 'sector_num' as a variable?
I thought there was more usage, but I just realized we can in fact
remove it in this patch.
>
> > +++ b/block/qcow.c
> > @@ -456,7 +456,9 @@ static uint64_t get_cluster_offset(BlockDriverState *bs,
> > if (i < n_start || i >= n_end) {
> > Error *err = NULL;
> > memset(s->cluster_data, 0x00, 512);
> > - if (qcrypto_block_encrypt(s->crypto,
> > start_sect + i,
> > + if (qcrypto_block_encrypt(s->crypto,
> > + start_sect + i *
> > + BDRV_SECTOR_SIZE,
>
> Umm, not the same. You want (start_sect + i) * BDRV_SECTOR_SIZE.
Heh, oppps.
Regards,
Daniel
--
|: https://berrange.com -o- https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org -o- https://fstop138.berrange.com :|
|: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
