linux/drivers/block/drbd, branch v3.0.79

drbd: fix for deadlock when using automatic split-brain-recovery

2013-05-19T17:04:48Z

commit 7c689e63a847316c1b2500f86891b0a574ce7e69 upstream. With an automatic after split-brain recovery policy of "after-sb-1pri call-pri-lost-after-sb", when trying to drbd_set_role() to R_SECONDARY, we run into a deadlock. This was first recognized and supposedly fixed by 2009-06-10 "Fixed a deadlock when using automatic split brain recovery when both nodes are" replacing drbd_set_role() with drbd_change_state() in that code-path, but the first hunk of that patch forgets to remove the drbd_set_role(). We apparently only ever tested the "two primaries" case. Signed-off-by: Philipp Reisner Signed-off-by: Lars Ellenberg Signed-off-by: Jens Axboe Signed-off-by: Jonghwan Choi Signed-off-by: Greg Kroah-Hartman

drbd: add missing part_round_stats to _drbd_start_io_acct

2013-01-21T19:44:59Z

commit 72585d2428fa3a0daab02ebad1f41e5ef517dbaa upstream. Without this, iostat frequently sees bogus svctime and >= 100% "utilization". Signed-off-by: Philipp Reisner Signed-off-by: Lars Ellenberg Cc: Raoul Bhatia Signed-off-by: Greg Kroah-Hartman

Merge branch 'for-3.0-important' of git://git.drbd.org/linux-2.6-drbd into for-linus

2011-06-30T08:10:50Z

drbd: we should write meta data updates with FLUSH FUA

2011-06-30T07:23:46Z

We used to write these with BIO_RW_BARRIER aka REQ_HARDBARRIER (unless disabled in the configuration). The correct semantic now would be to write with FLUSH/FUA. For example, with activity log transactions, FUA alone is not enough, we need the corresponding bitmap update (and all related application updates) on stable storage as well. Signed-off-by: Philipp Reisner Signed-off-by: Lars Ellenberg

drbd: when receive times out on meta socket, also check last receive time on data socket

2011-06-30T07:23:44Z

If we have an asymetrically congested network, we may send P_PING, but due to congestion, the corresponding P_PING_ACK would time out, and we would drop a (congested, but otherwise) healthy connection ("PingAck did not arrive in time.") Signed-off-by: Philipp Reisner Signed-off-by: Lars Ellenberg

drbd: account bitmap IO during resync as resync-(related-)-io

2011-06-30T07:23:43Z

If we have a good resync rate, we will frequently update the on-disk bitmap, which, if not accounted for as resync io, may let an otherwise idle device appear to be "busy", and cause us to throttle resync. Signed-off-by: Philipp Reisner Signed-off-by: Lars Ellenberg

drbd: don't cond_resched_lock with IRQs disabled

2011-06-30T07:23:42Z

The last commit, drbd: add missing spinlock to bitmap receive, introduced a cond_resched_lock(), where the lock in question is taken with irqs disabled. As we must not schedule with IRQs disabled, and cond_resched_lock_irq() does not exist, yet, we re-aquire the spin_lock_irq() for each bitmap page processed in turn. Signed-off-by: Philipp Reisner Signed-off-by: Lars Ellenberg

drbd: add missing spinlock to bitmap receive

2011-06-30T07:23:41Z

During bitmap exchange, when using the RLE bitmap compression scheme, we have a code path that can set the whole bitmap at once. To avoid holding spin_lock_irq() for too long, we used to lock out other bitmap modifications during bitmap exchange by other means, and then, knowing we have exclusive access to the bitmap, modify it without the spinlock, and with IRQs enabled. Since we now allow local IO to continue, potentially setting additional bits during the bitmap receive phase, this is no longer true, and we get uncoordinated updates of bitmap members, causing bm_set to no longer accurately reflect the total number of set bits. To actually see this, you'd need to have a large bitmap, use RLE bitmap compression, and have busy IO during sync handshake and bitmap exchange. Fix this by taking the spin_lock_irq() in this code path as well, but calling cond_resched_lock() after each page worth of bits processed. Signed-off-by: Philipp Reisner Signed-off-by: Lars Ellenberg

drbd: Use the correct max_bio_size when creating resync requests

2011-06-30T07:23:40Z

Signed-off-by: Philipp Reisner Signed-off-by: Lars Ellenberg

Merge branch 'for-2.6.40/drivers' of git://git.kernel.dk/linux-2.6-block

2011-05-25T16:15:35Z

* 'for-2.6.40/drivers' of git://git.kernel.dk/linux-2.6-block: (110 commits) loop: handle on-demand devices correctly loop: limit 'max_part' module param to DISK_MAX_PARTS drbd: fix warning drbd: fix warning drbd: Fix spelling drbd: fix schedule in atomic drbd: Take a more conservative approach when deciding max_bio_size drbd: Fixed state transitions after async outdate-peer-handler returned drbd: Disallow the peer_disk_state to be D_OUTDATED while connected drbd: Fix for the connection problems on high latency links drbd: fix potential activity log refcount imbalance in error path drbd: Only downgrade the disk state in case of disk failures drbd: fix disconnect/reconnect loop, if ping-timeout == ping-int drbd: fix potential distributed deadlock lru_cache.h: fix comments referring to ts_ instead of lc_ drbd: Fix for application IO with the on-io-error=pass-on policy xen/p2m: Add EXPORT_SYMBOL_GPL to the M2P override functions. xen/p2m/m2p/gnttab: Support GNTMAP_host_map in the M2P override. xen/blkback: don't fail empty barrier requests xen/blkback: fix xenbus_transaction_start() hang caused by double xenbus_transaction_end() ...