[PATCH 11/30] drbd: when receiving P_TRIM, zero-out partial unaligned chunks

2016-06-13 Thread Philipp Reisner
From: Lars Ellenberg We can avoid spurious data divergence caused by partially-ignored discards on certain backends with discard_zeroes_data=0, if we translate partial unaligned discard requests into explicit zero-out. The relevant use case is LVM/DM thin. If on different nodes, DRBD is backed

[PATCH 11/30] drbd: when receiving P_TRIM, zero-out partial unaligned chunks

2016-06-13 Thread Philipp Reisner
From: Lars Ellenberg We can avoid spurious data divergence caused by partially-ignored discards on certain backends with discard_zeroes_data=0, if we translate partial unaligned discard requests into explicit zero-out. The relevant use case is LVM/DM thin. If on different nodes, DRBD is backed

Re: [PATCH 11/30] drbd: when receiving P_TRIM, zero-out partial unaligned chunks

2016-04-25 Thread Lars Ellenberg
On Mon, Apr 25, 2016 at 02:07:38PM +0200, Philipp Reisner wrote: > + max_discard_sectors = min(q->limits.max_discard_sectors, (1U << 22)); > + max_discard_sectors -= max_discard_sectors % granularity; there On Mon, Apr 25, 2016 at 10:32:17PM +0200, Lars Ellenberg wrote: > On Mon, Apr

[PATCH 11/30] drbd: when receiving P_TRIM, zero-out partial unaligned chunks

2016-04-25 Thread Lars Ellenberg
On Mon, Apr 25, 2016 at 09:37:28AM -0700, Bart Van Assche wrote: > On 04/25/2016 05:13 AM, Philipp Reisner wrote: > >+while (nr_sectors >= granularity) { > >+nr = min_t(sector_t, nr_sectors, max_discard_sectors); > >+err |= blkdev_issue_discard(bdev, start, nr, GFP_NOIO,

Re: [Drbd-dev] [PATCH 11/30] drbd: when receiving P_TRIM, zero-out partial unaligned chunks

2016-04-25 Thread Bart Van Assche
On 04/25/2016 05:13 AM, Philipp Reisner wrote: + while (nr_sectors >= granularity) { + nr = min_t(sector_t, nr_sectors, max_discard_sectors); + err |= blkdev_issue_discard(bdev, start, nr, GFP_NOIO, 0); + nr_sectors -= nr; + start += n

[PATCH 11/30] drbd: when receiving P_TRIM, zero-out partial unaligned chunks

2016-04-25 Thread Philipp Reisner
From: Lars Ellenberg We can avoid spurious data divergence caused by partially-ignored discards on certain backends with discard_zeroes_data=0, if we translate partial unaligned discard requests into explicit zero-out. The relevant use case is LVM/DM thin. If on different nodes, DRBD is backed