On Sun, 2014-01-19 at 14:21 +0200, Sagi Grimberg wrote:
> On 1/19/2014 4:44 AM, Nicholas A. Bellinger wrote:
> > From: Nicholas Bellinger <n...@linux-iscsi.org>
> >
> > This patch adds blk_integrity passthrough support for block_device
> > backends using IBLOCK.
>
> Nice!
>
> > This includes iblock_alloc_bip() + setup of bio_integrity_payload
> > information that attaches to the leading struct bio once bio_list
> > is populated during fast-path iblock_execute_rw() I/O dispatch.
> >
> > It also updates setup in iblock_configure_device() to detect modes
> > of protection + se dev->dev_attrib.pi_prot_type accordingly, along
> > with creating required bio_set integrity mempools.
> >
> > Cc: Martin K. Petersen <martin.peter...@oracle.com>
> > Cc: Christoph Hellwig <h...@lst.de>
> > Cc: Hannes Reinecke <h...@suse.de>
> > Cc: Sagi Grimberg <sa...@mellanox.com>
> > Cc: Or Gerlitz <ogerl...@mellanox.com>
> > Signed-off-by: Nicholas Bellinger <n...@linux-iscsi.org>
> > ---
> > drivers/target/Kconfig | 1 +
> > drivers/target/target_core_iblock.c | 91
> > ++++++++++++++++++++++++++++++++++-
> > 2 files changed, 90 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/target/Kconfig b/drivers/target/Kconfig
> > index 50aad2e..dc2d84a 100644
> > --- a/drivers/target/Kconfig
> > +++ b/drivers/target/Kconfig
> > @@ -14,6 +14,7 @@ if TARGET_CORE
> >
> > config TCM_IBLOCK
> > tristate "TCM/IBLOCK Subsystem Plugin for Linux/BLOCK"
> > + select BLK_DEV_INTEGRITY
> > help
> > Say Y here to enable the TCM/IBLOCK subsystem plugin for non-buffered
> > access to Linux/Block devices using BIO
> > diff --git a/drivers/target/target_core_iblock.c
> > b/drivers/target/target_core_iblock.c
> > index 15d9121..293d9b0 100644
> > --- a/drivers/target/target_core_iblock.c
> > +++ b/drivers/target/target_core_iblock.c
> > @@ -91,6 +91,7 @@ static int iblock_configure_device(struct se_device *dev)
> > struct iblock_dev *ib_dev = IBLOCK_DEV(dev);
> > struct request_queue *q;
> > struct block_device *bd = NULL;
> > + struct blk_integrity *bi;
> > fmode_t mode;
> > int ret = -ENOMEM;
> >
> > @@ -155,8 +156,40 @@ static int iblock_configure_device(struct se_device
> > *dev)
> > if (blk_queue_nonrot(q))
> > dev->dev_attrib.is_nonrot = 1;
> >
> > + bi = bdev_get_integrity(bd);
> > + if (bi) {
> > + struct bio_set *bs = ib_dev->ibd_bio_set;
> > +
> > + if (!strcmp(bi->name, "T10-DIF-TYPE3-IP") ||
> > + !strcmp(bi->name, "T10-DIF-TYPE1-IP")) {
> > + pr_err("IBLOCK export of blk_integrity: %s not"
> > + " supported\n", bi->name);
> > + ret = -ENOSYS;
> > + goto out_blkdev_put;
> > + }
>
> Please remind me why we ignore IP-CSUM guard type again?
> MKP, will this be irrelevant for the initiator as well? if so, I don't
> see a reason to expose this in RDMA verbs.
My understanding is that this was used for pre-production prototyping,
and is not supported by any real backend storage hardware.
--nab
--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
