summaryrefslogtreecommitdiffstats
Commit message (Expand)AuthorAgeFilesLines
* raid5-cache: use bio chainingChristoph Hellwig2015-11-011-22/+16
* raid5-cache: small log->seq cleanupChristoph Hellwig2015-11-011-2/+1
* raid5-cache: new helper: r5_reserve_log_entryChristoph Hellwig2015-11-011-11/+19
* raid5-cache: inline r5l_alloc_io_unit into r5l_new_metaChristoph Hellwig2015-11-011-18/+8
* raid5-cache: take rdev->data_offset into account early onChristoph Hellwig2015-11-011-5/+2
* raid5-cache: refactor bio allocationChristoph Hellwig2015-11-011-25/+20
* raid5-cache: clean up r5l_get_metaChristoph Hellwig2015-11-011-8/+4
* raid5-cache: simplify state machine when caches flushes are not neededChristoph Hellwig2015-11-011-4/+29
* raid5-cache: factor out a helper to run all stripes for an I/O unitChristoph Hellwig2015-11-011-10/+13
* raid5-cache: rename flushed_ios to finished_iosChristoph Hellwig2015-11-011-7/+7
* raid5-cache: free I/O units earlierChristoph Hellwig2015-11-011-89/+54
* raid5-cache: move reclaim stop to quiesceShaohua Li2015-11-013-12/+20
* md: show journal for journal disk in disk state sysfsShaohua Li2015-11-011-0/+4
* skip match_mddev_units check for special rolesSong Liu2015-11-011-2/+12
* raid5-cache: don't delay stripe captured in logShaohua Li2015-11-011-0/+5
* raid5-cache: check stripe finish out of orderShaohua Li2015-11-011-0/+5
* md: skip resync for raid array with journalShaohua Li2015-11-012-0/+5
* raid5-cache: optimize FLUSH IO with log enabledShaohua Li2015-11-013-2/+28
* raid5-cache: move functionality out of __r5l_set_io_unit_stateChristoph Hellwig2015-11-011-41/+38
* raid5-cache: fix a user-after-free bugShaohua Li2015-11-011-11/+7
* raid5-cache: switching to state machine for log disk cache flushShaohua Li2015-11-012-52/+71
* raid5: enable log for raid array with cache diskShaohua Li2015-11-012-0/+12
* raid5: don't allow resize/reshape with cache(log) supportShaohua Li2015-11-011-0/+14
* raid5: disable batch with log enabledShaohua Li2015-11-011-0/+4
* raid5-cache: use crc32c checksumShaohua Li2015-11-011-10/+10
* raid5: log recoveryShaohua Li2015-10-241-3/+240
* raid5: log reclaim supportShaohua Li2015-10-243-1/+267
* raid5: add basic stripe logShaohua Li2015-10-245-1/+680
* raid5: add a new state for stripe log handlingShaohua Li2015-10-242-0/+4
* raid5: export some functionsShaohua Li2015-10-242-51/+57
* md: override md superblock recovery_offset for journal deviceShaohua Li2015-10-243-2/+17
* MD: add a new disk role to present write journal deviceSong Liu2015-10-243-2/+29
* MD: replace special disk roles with macrosSong Liu2015-10-242-7/+11
* md-cluster: Call update_raid_disks() if another node --grow's raid_disksGoldwyn Rodrigues2015-10-242-10/+23
* md-cluster: remove mddev arg from add_resync_info()NeilBrown2015-10-241-3/+3
* md-cluster: don't cast void pointers when assigning them.NeilBrown2015-10-241-2/+2
* md-cluster: discard unused sb_mutex.NeilBrown2015-10-241-2/+0
* md-cluster: Fix warnings when build with CF=-D__CHECK_ENDIAN__Guoqing Jiang2015-10-241-10/+11
* md-cluster: metadata_update_finish: consistently use cmsg.raid_slot as le32NeilBrown2015-10-161-4/+5
* Merge branch 'md-next' of git://github.com/goldwynr/linux into for-nextNeilBrown2015-10-1411-214/+421
|\
| * md: check the return value for metadata_update_startGuoqing Jiang2015-10-121-6/+9
| * md-cluster: only call kick_rdev_from_array after remove disk successfullyGuoqing Jiang2015-10-121-6/+9
| * md-cluster: Add 'SUSE' as author for md-cluster.cGuoqing Jiang2015-10-121-0/+1
| * md-cluster: zero cmsg before it was sentGuoqing Jiang2015-10-121-3/+3
| * md-cluster: make sure the node do not receive it's own msgGuoqing Jiang2015-10-121-0/+3
| * md-cluster: remove unnecessary setting for slotGuoqing Jiang2015-10-121-2/+0
| * md-cluster: make other members of cluster_msg is handled by little endian funcsGuoqing Jiang2015-10-121-10/+14
| * md-cluster: Do not printk() every received messageGoldwyn Rodrigues2015-10-121-14/+0
| * md-cluster: Fix adding of new disk with new reload codeGoldwyn Rodrigues2015-10-123-35/+58
| * md-cluster: Perform resync/recovery under a DLM lockGoldwyn Rodrigues2015-10-124-9/+53
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-22 16:03:10 +0000