linux-block.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Damien Le Moal <damien.lemoal@wdc.com>
To: dm-devel@redhat.com, Mike Snitzer <snitzer@redhat.com>,
	linux-block@vger.kernel.org, Jens Axboe <axboe@kernel.dk>,
	linux-nvme@lists.infradead.org, Christoph Hellwig <hch@lst.de>,
	linux-scsi@vger.kernel.org,
	"Martin K . Petersen" <martin.petersen@oracle.com>,
	linux-fsdevel@vger.kernel.org, linux-btrfs@vger.kernel.org,
	David Sterba <dsterba@suse.com>,
	Josef Bacik <josef@toxicpanda.com>
Cc: Johannes Thumshirn <johannes.thumshirn@wdc.com>,
	Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>,
	Naohiro Aota <naohiro.aota@wdc.com>
Subject: [PATCH 2/4] dm crypt: Fix zoned block device support
Date: Fri, 16 Apr 2021 12:05:26 +0900	[thread overview]
Message-ID: <20210416030528.757513-3-damien.lemoal@wdc.com> (raw)
In-Reply-To: <20210416030528.757513-1-damien.lemoal@wdc.com>

Zone append BIOs (REQ_OP_ZONE_APPEND) always specify the start sector of
the zone to be written instead of the actual location sector to write.
The write location is determined by the device and returned to the host
upon completion of the operation. This interface, while simple and
efficient for writing into sequential zones of a zoned block device, is
incompatible with the use of sector values to calculate a cypher block
IV. All data written in a zone end up using the same IV values
corresponding to the first sectors of the zone, but read operation will
specify any sector within the zone, resulting in an IV mismatch between
encryption and decryption.

Using a single sector value (e.g. the zone start sector) for all read
and writes into a zone can solve this problem, but at the cost of
weakening the cypher chosen by the user. Instead, to solve this
problem, explicitly disable support for zone append operations using
the zone_append_not_supported field of struct dm_target if the IV mode
used is sector-based, that is for all IVs modes except null and random.

The cypher flag CRYPT_IV_NO_SECTORS iis introduced to indicate that the
cypher does not use sector values. This flag is set in
crypt_ctr_ivmode() for the null and random IV modes and checked in
crypt_ctr() to set to true zone_append_not_supported if
CRYPT_IV_NO_SECTORS is not set for the chosen cypher.

Reported-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
---
 drivers/md/dm-crypt.c | 48 +++++++++++++++++++++++++++++++++++--------
 1 file changed, 39 insertions(+), 9 deletions(-)

diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index b0ab080f2567..0a44bc0ff960 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -137,6 +137,7 @@ enum cipher_flags {
 	CRYPT_MODE_INTEGRITY_AEAD,	/* Use authenticated mode for cipher */
 	CRYPT_IV_LARGE_SECTORS,		/* Calculate IV from sector_size, not 512B sectors */
 	CRYPT_ENCRYPT_PREPROCESS,	/* Must preprocess data for encryption (elephant) */
+	CRYPT_IV_NO_SECTORS,		/* IV calculation does not use sectors */
 };
 
 /*
@@ -2750,9 +2751,10 @@ static int crypt_ctr_ivmode(struct dm_target *ti, const char *ivmode)
 	}
 
 	/* Choose ivmode, see comments at iv code. */
-	if (ivmode == NULL)
+	if (ivmode == NULL) {
 		cc->iv_gen_ops = NULL;
-	else if (strcmp(ivmode, "plain") == 0)
+		set_bit(CRYPT_IV_NO_SECTORS, &cc->cipher_flags);
+	} else if (strcmp(ivmode, "plain") == 0)
 		cc->iv_gen_ops = &crypt_iv_plain_ops;
 	else if (strcmp(ivmode, "plain64") == 0)
 		cc->iv_gen_ops = &crypt_iv_plain64_ops;
@@ -2762,9 +2764,10 @@ static int crypt_ctr_ivmode(struct dm_target *ti, const char *ivmode)
 		cc->iv_gen_ops = &crypt_iv_essiv_ops;
 	else if (strcmp(ivmode, "benbi") == 0)
 		cc->iv_gen_ops = &crypt_iv_benbi_ops;
-	else if (strcmp(ivmode, "null") == 0)
+	else if (strcmp(ivmode, "null") == 0) {
 		cc->iv_gen_ops = &crypt_iv_null_ops;
-	else if (strcmp(ivmode, "eboiv") == 0)
+		set_bit(CRYPT_IV_NO_SECTORS, &cc->cipher_flags);
+	} else if (strcmp(ivmode, "eboiv") == 0)
 		cc->iv_gen_ops = &crypt_iv_eboiv_ops;
 	else if (strcmp(ivmode, "elephant") == 0) {
 		cc->iv_gen_ops = &crypt_iv_elephant_ops;
@@ -2791,6 +2794,7 @@ static int crypt_ctr_ivmode(struct dm_target *ti, const char *ivmode)
 		cc->key_extra_size = cc->iv_size + TCW_WHITENING_SIZE;
 	} else if (strcmp(ivmode, "random") == 0) {
 		cc->iv_gen_ops = &crypt_iv_random_ops;
+		set_bit(CRYPT_IV_NO_SECTORS, &cc->cipher_flags);
 		/* Need storage space in integrity fields. */
 		cc->integrity_iv_size = cc->iv_size;
 	} else {
@@ -3281,14 +3285,31 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	}
 	cc->start = tmpll;
 
-	/*
-	 * For zoned block devices, we need to preserve the issuer write
-	 * ordering. To do so, disable write workqueues and force inline
-	 * encryption completion.
-	 */
 	if (bdev_is_zoned(cc->dev->bdev)) {
+		/*
+		 * For zoned block devices, we need to preserve the issuer write
+		 * ordering. To do so, disable write workqueues and force inline
+		 * encryption completion.
+		 */
 		set_bit(DM_CRYPT_NO_WRITE_WORKQUEUE, &cc->flags);
 		set_bit(DM_CRYPT_WRITE_INLINE, &cc->flags);
+
+		/*
+		 * All zone append writes to a zone of a zoned block device will
+		 * have the same BIO sector (the start of the zone). When the
+		 * cypher IV mode uses sector values, all data targeting a
+		 * zone will be encrypted using the first sector numbers of the
+		 * zone. This will not result in write errors but will
+		 * cause most reads to fail as reads will use the sector values
+		 * for the actual data location, resulting in IV mismatch.
+		 * To avoid this problem, allow zone append operations only for
+		 * cyphers with an IV mode not using sector values (null and
+		 * random IVs).
+		 */
+		if (!test_bit(CRYPT_IV_NO_SECTORS, &cc->cipher_flags)) {
+			DMWARN("Zone append is not supported with sector-based IV cyphers");
+			ti->zone_append_not_supported = true;
+		}
 	}
 
 	if (crypt_integrity_aead(cc) || cc->integrity_iv_size) {
@@ -3356,6 +3377,15 @@ static int crypt_map(struct dm_target *ti, struct bio *bio)
 	struct dm_crypt_io *io;
 	struct crypt_config *cc = ti->private;
 
+	/*
+	 * For zoned targets using a sector based IV, zone append is not
+	 * supported. We should not see any such operation in that case.
+	 * In the unlikely case we do, warn and fail the request.
+	 */
+	if (WARN_ON_ONCE(bio_op(bio) == REQ_OP_ZONE_APPEND &&
+			 !test_bit(CRYPT_IV_NO_SECTORS, &cc->cipher_flags)))
+		return DM_MAPIO_KILL;
+
 	/*
 	 * If bio is REQ_PREFLUSH or REQ_OP_DISCARD, just bypass crypt queues.
 	 * - for REQ_PREFLUSH device-mapper core ensures that no IO is in-flight
-- 
2.30.2


  parent reply	other threads:[~2021-04-16  3:05 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-16  3:05 [PATCH 0/4] Fix dm-crypt zoned block device support Damien Le Moal
2021-04-16  3:05 ` [PATCH 1/4] dm: Introduce zone append support control Damien Le Moal
2021-04-16  6:51   ` Johannes Thumshirn
2021-04-16  3:05 ` Damien Le Moal [this message]
2021-04-16  7:13   ` [PATCH 2/4] dm crypt: Fix zoned block device support Johannes Thumshirn
2021-04-16  7:30     ` Damien Le Moal
2021-04-16  9:10       ` Johannes Thumshirn
2021-04-16  3:05 ` [PATCH 3/4] btrfs: zoned: fail mount if the device does not support zone append Damien Le Moal
2021-04-16 16:17   ` David Sterba
2021-04-19  9:28     ` Christoph Hellwig
2021-04-19  9:35       ` Damien Le Moal
2021-04-19  9:39         ` hch
2021-04-19  9:46           ` Damien Le Moal
2021-04-19 16:48             ` David Sterba
2021-04-19  9:39       ` Johannes Thumshirn
2021-04-16  3:05 ` [PATCH 4/4] zonefs: fix synchronous write to sequential zone files Damien Le Moal
2021-04-16  6:50   ` Johannes Thumshirn

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210416030528.757513-3-damien.lemoal@wdc.com \
    --to=damien.lemoal@wdc.com \
    --cc=axboe@kernel.dk \
    --cc=dm-devel@redhat.com \
    --cc=dsterba@suse.com \
    --cc=hch@lst.de \
    --cc=johannes.thumshirn@wdc.com \
    --cc=josef@toxicpanda.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=naohiro.aota@wdc.com \
    --cc=shinichiro.kawasaki@wdc.com \
    --cc=snitzer@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).