summaryrefslogtreecommitdiffstats
path: root/drivers/md/raid5.c (follow)
Commit message (Expand)AuthorAgeFilesLines
* md/raid5: Make sure we clear R5_Discard when discard is finished.NeilBrown2012-11-211-2/+5
* md/raid5: move resolving of reconstruct_state earlier inNeilBrown2012-11-211-34/+34
* md/raid5: round discard alignment up to power of 2.NeilBrown2012-11-201-0/+4
* Merge tag 'md-3.7' of git://neil.brown.name/mdLinus Torvalds2012-10-131-22/+197
|\
| * md/raid5: be careful not to resize_stripes too big.NeilBrown2012-10-111-1/+2
| * Subject: [PATCH] md:change resync_mismatches to atomic64_t to avoid racesJianpeng Ma2012-10-111-2/+2
| * md/raid5: make sure to_read and to_write never go negative.NeilBrown2012-10-111-4/+1
| * md: When RAID5 is dirty, force reconstruct-write instead of read-modify-write.Alexander Lyakas2012-10-111-3/+16
| * md/raid5: protect debug message against NULL derefernce.NeilBrown2012-10-111-1/+1
| * md/raid5: add some missing locking in handle_failed_stripe.NeilBrown2012-10-111-0/+2
| * MD: raid5 avoid unnecessary zero page for trimShaohua Li2012-10-111-18/+17
| * MD: raid5 trim supportShaohua Li2012-10-111-3/+165
| * MD: change the parameter of md threadShaohua Li2012-10-111-1/+2
* | md/raid5: add missing spin_lock_init.NeilBrown2012-09-241-0/+1
* | md/raid5: fix calculate of 'degraded' when a replacement becomes active.NeilBrown2012-09-191-0/+4
* | Revert "md/raid5: For odirect-write performance, do not set STRIPE_PREREAD_AC...NeilBrown2012-09-191-1/+1
|/
* Merge tag 'md-3.6' of git://neil.brown.name/mdLinus Torvalds2012-08-021-16/+91
|\
| * raid5: raid5d handle stripe in batch wayShaohua Li2012-08-021-13/+32
| * raid5: make_request use batch stripe releaseShaohua Li2012-08-021-3/+59
* | Merge branch 'for-3.6/drivers' of git://git.kernel.dk/linux-blockLinus Torvalds2012-08-011-3/+2
|\|
| * md: remove plug_cnt feature of plugging.NeilBrown2012-07-311-3/+2
* | md/raid5: For odirect-write performance, do not set STRIPE_PREREAD_ACTIVE.majianpeng2012-07-311-1/+1
* | raid5: Add R5_ReadNoMerge flag which prevent bio from merging at block layermajianpeng2012-07-311-2/+12
* | raid5: add a per-stripe lockShaohua Li2012-07-191-16/+19
* | raid5: remove unnecessary bitmap write optimizationShaohua Li2012-07-191-12/+9
* | raid5: lockless access raid5 overrided bi_phys_segmentsShaohua Li2012-07-191-30/+32
* | raid5: reduce chance release_stripe() taking device_lockShaohua Li2012-07-191-34/+41
|/
* md: fix up plugging (again).NeilBrown2012-07-031-5/+1
* raid5: delayed stripe fixShaohua Li2012-07-031-1/+3
* md/raid456: When read error cannot be recovered, record bad blockmajianpeng2012-07-031-4/+11
* md: make 'name' arg to md_register_thread non-optional.NeilBrown2012-07-031-1/+3
* md/raid5: fix refcount problem when blocked_rdev is set.NeilBrown2012-07-031-2/+12
* md/raid5: In ops_run_io, inc nr_pending before calling md_wait_for_blocked_rdevmajianpeng2012-07-031-0/+6
* md/raid5: Do not add data_offset before call to is_badblockmajianpeng2012-07-031-2/+3
* md/raid5: prefer replacing failed devices over want-replacement devices.NeilBrown2012-07-031-5/+8
* md/raid5: improve removal of extra devices after reshape.NeilBrown2012-05-221-5/+5
* md/raid5: Allow reshape while a bitmap is present.NeilBrown2012-05-221-3/+3
* md: allow array to be resized while bitmap is present.NeilBrown2012-05-221-4/+10
* raid5: support sync requestShaohua Li2012-05-221-2/+10
* raid5: remove unused variablesShaohua Li2012-05-221-4/+0
* md/raid5: allow for change in data_offset while managing a reshape.NeilBrown2012-05-211-33/+76
* md/raid5: Use correct data_offset for all IO.NeilBrown2012-05-211-13/+59
* md: add possibility to change data-offset for devices.NeilBrown2012-05-211-3/+7
* md: allow a reshape operation to be reversed.NeilBrown2012-05-211-11/+12
* md/raid5: Fix a bug about judging if the operation is syncing or replacingmajianpeng2012-04-031-1/+3
* md/raid5: fix handling of bad blocks during recovery.NeilBrown2012-04-031-26/+29
* md: tidy up rdev_for_each usage.NeilBrown2012-03-191-4/+4
* md: allow re-add to failed arrays.NeilBrown2012-03-191-1/+1
* md/raid5: use atomic_dec_return() instead of atomic_dec() and atomic_read().majianpeng2012-03-131-4/+3
* md/raid5: removed unused 'added_devices' variable.NeilBrown2012-03-131-5/+2