dm integrity: support larger block sizes
authorMikulas Patocka <mpatocka@redhat.com>
Tue, 18 Apr 2017 20:51:52 +0000 (16:51 -0400)
committerMike Snitzer <snitzer@redhat.com>
Mon, 24 Apr 2017 16:04:33 +0000 (12:04 -0400)
commit9d609f85b7eb96697ae22ad1d47a3a3920174ba8
tree25610e72c2f7eff159c44c6afc8a88d080ee7447
parent56b67a4f292f14548f4046979d46d07bcf8ba971
dm integrity: support larger block sizes

The DM integrity block size can now be 512, 1k, 2k or 4k.  Using larger
blocks reduces metadata handling overhead.  The block size can be
configured at table load time using the "block_size:<value>" option;
where <value> is expressed in bytes (defult is still 512 bytes).

It is safe to use larger block sizes with DM integrity, because the
DM integrity journal makes sure that the whole block is updated
atomically even if the underlying device doesn't support atomic writes
of that size (e.g. 4k block ontop of a 512b device).

Depends-on: 2859323e ("block: fix blk_integrity_register to use template's interval_exp if not 0")
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Documentation/device-mapper/dm-integrity.txt
drivers/md/dm-integrity.c