Skip to contentSkip to conversation

xen blkfront barrier updates

GIT PULL Jeremy Fitzhardinge 2 participants 2 messages Nov 11, 2010 – Nov 12, 2010

lore
Share & tools
lore
#0 lore raw

Hi Jens,

Here's the xen-blkfront barrier changes we discussed in Boston (its identical to the last posted version).

The only comment was Christoph pointing out EOPNOTSUPP is not used in the block interface, but blkfront is just using it as a local placeholder and it never gets returned up to the block layer.

I still haven't sorted out what's going on with the backends; we definitely need either a FLUSH operation or make empty WRITE_BARRIER work to flush disk caches.

Thanks,
J

The following changes since commit c8ddb2713c624f432fa5fe3c7ecffcdda46ea0d4:

Linux 2.6.37-rc1 (2010-11-01 07:54:12 -0400)

are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/jeremy/xen.git upstream/blkfront

Jeremy Fitzhardinge (4):
xen/blkfront: map REQ_FLUSH into a full barrier
xen/blkfront: change blk_shadow.request to proper pointer xen/blkfront: Implement FUA with BLKIF_OP_WRITE_BARRIER xen/blkfront: cope with backend that fail empty BLKIF_OP_WRITE_BARRIER requests

 drivers/block/xen-blkfront.c |   55 ++++++++++++++++++++++++-----------------
 1 files changed, 32 insertions(+), 23 deletions(-)
#1 lore raw
Quoted 5 lines · Jeremy Fitzhardinge

Hi Jens,

Here's the xen-blkfront barrier changes we discussed in Boston (its identical to the last posted version).

Quoted 7 lines

The only comment was Christoph pointing out EOPNOTSUPP is not used in the block interface, but blkfront is just using it as a local placeholder and it never gets returned up to the block layer.

I still haven't sorted out what's going on with the backends; we definitely need either a FLUSH operation or make empty WRITE_BARRIER work to flush disk caches.

Thanks Jeremy, I'll pull this in.

Signature · 1 line
Jens Axboe