6917066 zfs block picking can be improved

6826241 Sync write IOPS drops dramatically during TXG sync
6869229 zfs should switch to shiny new metaslabs more frequently

PSARC/2009/396 Tickless Kernel Architecture / lbolt decoupling
6860030 tickless clock requires a clock() decoupled lbolt / lbolt64
Portions contributed by Chad Mynhier <cmynhier@gmail.com>

6899159 injection isn't trashing pools
6899160 pool recovery info sometimes not appearing

6897958 ASSERT in metaslab_class_space_update() with 8+ exabyte pool
6898598 dsl needs to be more aware of dedup space

PSARC 2009/571 ZFS Deduplication Properties
6677093 zfs should have dedup capability

PSARC 2009/479 zpool recovery support
6667683 need a way to rollback to an uberblock from a previous txg
6885998 bad ASSERT() in traverse_zil_block()

6574286 removing a slog doesn't work
6856566 zpool import -F can cause panic
6863456 system panic by load_nvlist(spa, spa->spa_config_object, &nv) == 0 while running zfs test suite
6882947 dump_nvlist() should live in libnvpair

6596237 Stop looking and start ganging

6328632 zpool offline is a bit too conservative
6739487 ASSERT: txg <= spa_final_txg due to scrub/export race
6767129 ASSERT: cvd->vdev_isspare, in spa_vdev_detach()
6747698 checksum failures after offline -t / export / import / scrub
6745863 ZFS writes to disk after it has been offlined
6722540 50% slowdown on scrub/resilver with certain vdev configurations
6759999 resilver logic rewrites ditto blocks on both source and destination
6758107 I/O should never suspend during spa_load()
6776548 codereview(1) runs off the page when faced with multi-line comments
6761406 AMD errata 91 workaround doesn't work on 64-bit systems

6764914 vdev_reopen() can lead to failed allocations

6754011 SPA 3.0: lock breakup, i/o pipeline refactoring, device failure handling
6667208 zfs/zpool commands on failed pool should not hang
6430480 grabbing config lock as writer during I/O load can take excessively long

6354519 stack overflow in zfs due to zio pipeline
6533726 single-threaded checksum & parity calculations limit write bandwidth
6547248 ztest detects a future leak when there is none
6604198 zfs only using single cpu for compression (part II)

PSARC 2007/618 ZFS L2ARC
6536054 second tier ("external") ARC

PSARC 2007/567 zpool failmode property
6322646 ZFS should gracefully handle all devices failing (when writing)
6413847 vdev label write failure should be handled more gracefully
6417772 need nicer message on write failure
6417779 ZFS: I/O failure (write on ...) -- need to reallocate writes
6467927 Node gets into a panic loop when devices are fenced off
6565042 ZFS should gracefully handle all devices failing (when reading)
6596239 Stop issuing IOs to a vdev that is going to be removed

6569719 panic dangling dbufs (dn=ffffffff28814d30, dbuf=ffffffff20756008)
6573361 panic turnstile_block, unowned mutex
6584864 $MOS is not properly bounded by pool size
6585265 need bonus resize interface
6587723 BAD TRAP: type=e (#pf Page fault) occurred in module "zfs" due to a NULL pointer dereference
6589799 dangling dbuf after zinject
6594025 panic: dangling dbufs during shutdown

PSARC 2007/171 ZFS Separate Intent Log
6339640 Make ZIL use NVRAM when available.

6536526 waiting for read from read completion taskq caused zio pipeline stall

6495013 Loops and recursion in metaslab_ff_alloc can kill performance, even on a pool with lots of free data
6527607 zil allocation failures result in a leak

6527325 want more assertions in space map code

6341569 zio_alloc_blk() vdev distribution performs badly
6428639 large writes to zvol synchs too much, better cut down a little
6444692 Need to flush disk write cache for dmu_sync buffers
6465634 zvol: dmu_sync() should be issued in parallel
6468731 lwb_state_t can be nuked
6470042 parallel dmu_sync() isn't being used
6471679 stash blocksize in zvol_state_t rather than reading in every zvol_log_write
6472230 ZIL vdev management is inefficient
6473775 zil_commit changes in snv_48 make it hot for O_DSYNC workloads
6478388 ZIL replay takes too long causing issues while booting
6486390 zil_commit could push more transactions
6486496 zil_replay() useful debug

6463348 ZFS code could be more portable

6452923 really out of space panic even though ms_map.sm_space > 0

6447377 ZFS prefetch is inconsistant

6410711 intent log blocks don't get invited to pool parties

6410698 ZFS metadata needs to be more highly replicated (ditto blocks)
6410700 zdb should support reading raw blocks out of storage pool
6410709 ztest: spa config can change before pool export

6280668 pluggable block allocation policy
6399301 initial read of space maps is super slow
6407365 large-sector disk support in ZFS
6407366 ADVANCE_NOLOCK gathers MOS
6407367 three-way deadlock between db_mtx, dbuf_hash[], and ms_lock

PSARC 2006/077 zpool clear
PSARC 2006/139 FMA for ZFS
6284889 arc should replace the znode cache
6333006 DMU & DSL should not panic upon I/O error
6333092 concurrent reads to a file not scaling with number of readers
6338081 ZFS/FMA phase 1
6338386 need persistent error log
6341326 i/o error causes arc buf hash table corruption
6341639 zfs backup/restore should compute/verify checksum of backup stream
6348002 out of space due to changing properties
6354724 inaccurate error message from zfs restore
6354872 dmu_sync() blows predictive accounting
6355416 zpool scrubbing consumes all memory, system hung
6363995 df should only load libzfs when it encounters a ZFS filesystem
6366320 zfs backup/restore doesn't like signals
6368892 mount -m support needed for legacy mounts
6368902 boot archive fstat support needed for ZFS Mountroot
6369424 BFU complains when bfu'ing a ZFS root filesystem
6374062 mountroot support needed for ZFS
6376356 dirtying dbuf obj=43 lvl=0 blkid=0 but not tx_held
6378391 unused members of dmu_objset_stats_t
6378392 clean up zfs_cmd_t structure
6378685 buf_init should allocate its hash table more carefully
6378976 ziltest should be a first class citizen
6381086 zdb segfaults if there is a spa deferred-free bplist
6381203 deadlock due to i/o while assigning (tc_lock held)
6381209 freed space is not immediately available
6381344 'zpool clear'
6381345 FAULTED devices should really be UNAVAIL
6381346 import should mark devices as persistently unavailable
6383272 recursive mutex_enter() during log replay with zfs root
6386326 origin property is not displayed
6386354 libzfs does too much in its _init section, calls exit(1)
6386624 zpool should not complain about non-existent devices from libdiskmgt
6386910 spa needs to be i/o error hardened
6387735 need a mechanism to inject faults into ZFS
6387736 internal ZFS utilities should be placed in an ON-private package
6389928 libzfs should ship a lint library
6390609 malformed vdev config panics on zpool_create()
6390677 version number checking makes upgrades challenging
6390713 ztest hangs in zil_suspend()
6391873 metadata compression should be turned back on
6392113 ztest sometimes reports leaked blocks because ZIL isn't resilvered
6393004 minor memory leak in unique_insert()

PSARC 2002/240 ZFS
6338653 Integrate ZFS
PSARC 2004/652 - DKIOCFLUSH
5096886 Write caching disks need mechanism to flush cache to physical media

6917066 zfs block picking can be improved

6826241 Sync write IOPS drops dramatically during TXG sync
6869229 zfs should switch to shiny new metaslabs more frequently

PSARC/2009/396 Tickless Kernel Architecture / lbolt decoupling
6860030 tickless clock requires a clock() decoupled lbolt / lbolt64
Portions contributed by Chad Mynhier <cmynhier@gmail.com>

6899159 injection isn't trashing pools
6899160 pool recovery info sometimes not appearing

6897958 ASSERT in metaslab_class_space_update() with 8+ exabyte pool
6898598 dsl needs to be more aware of dedup space

PSARC 2009/571 ZFS Deduplication Properties
6677093 zfs should have dedup capability

PSARC 2009/479 zpool recovery support
6667683 need a way to rollback to an uberblock from a previous txg
6885998 bad ASSERT() in traverse_zil_block()

6574286 removing a slog doesn't work
6856566 zpool import -F can cause panic
6863456 system panic by load_nvlist(spa, spa->spa_config_object, &nv) == 0 while running zfs test suite
6882947 dump_nvlist() should live in libnvpair

6596237 Stop looking and start ganging

6328632 zpool offline is a bit too conservative
6739487 ASSERT: txg <= spa_final_txg due to scrub/export race
6767129 ASSERT: cvd->vdev_isspare, in spa_vdev_detach()
6747698 checksum failures after offline -t / export / import / scrub
6745863 ZFS writes to disk after it has been offlined
6722540 50% slowdown on scrub/resilver with certain vdev configurations
6759999 resilver logic rewrites ditto blocks on both source and destination
6758107 I/O should never suspend during spa_load()
6776548 codereview(1) runs off the page when faced with multi-line comments
6761406 AMD errata 91 workaround doesn't work on 64-bit systems

6764914 vdev_reopen() can lead to failed allocations

6754011 SPA 3.0: lock breakup, i/o pipeline refactoring, device failure handling
6667208 zfs/zpool commands on failed pool should not hang
6430480 grabbing config lock as writer during I/O load can take excessively long

6354519 stack overflow in zfs due to zio pipeline
6533726 single-threaded checksum & parity calculations limit write bandwidth
6547248 ztest detects a future leak when there is none
6604198 zfs only using single cpu for compression (part II)

PSARC 2007/618 ZFS L2ARC
6536054 second tier ("external") ARC

PSARC 2007/567 zpool failmode property
6322646 ZFS should gracefully handle all devices failing (when writing)
6413847 vdev label write failure should be handled more gracefully
6417772 need nicer message on write failure
6417779 ZFS: I/O failure (write on ...) -- need to reallocate writes
6467927 Node gets into a panic loop when devices are fenced off
6565042 ZFS should gracefully handle all devices failing (when reading)
6596239 Stop issuing IOs to a vdev that is going to be removed

6569719 panic dangling dbufs (dn=ffffffff28814d30, dbuf=ffffffff20756008)
6573361 panic turnstile_block, unowned mutex
6584864 $MOS is not properly bounded by pool size
6585265 need bonus resize interface
6587723 BAD TRAP: type=e (#pf Page fault) occurred in module "zfs" due to a NULL pointer dereference
6589799 dangling dbuf after zinject
6594025 panic: dangling dbufs during shutdown

PSARC 2007/171 ZFS Separate Intent Log
6339640 Make ZIL use NVRAM when available.

6536526 waiting for read from read completion taskq caused zio pipeline stall

6495013 Loops and recursion in metaslab_ff_alloc can kill performance, even on a pool with lots of free data
6527607 zil allocation failures result in a leak

6527325 want more assertions in space map code

6341569 zio_alloc_blk() vdev distribution performs badly
6428639 large writes to zvol synchs too much, better cut down a little
6444692 Need to flush disk write cache for dmu_sync buffers
6465634 zvol: dmu_sync() should be issued in parallel
6468731 lwb_state_t can be nuked
6470042 parallel dmu_sync() isn't being used
6471679 stash blocksize in zvol_state_t rather than reading in every zvol_log_write
6472230 ZIL vdev management is inefficient
6473775 zil_commit changes in snv_48 make it hot for O_DSYNC workloads
6478388 ZIL replay takes too long causing issues while booting
6486390 zil_commit could push more transactions
6486496 zil_replay() useful debug

6463348 ZFS code could be more portable

6452923 really out of space panic even though ms_map.sm_space > 0

6447377 ZFS prefetch is inconsistant

6410711 intent log blocks don't get invited to pool parties

6410698 ZFS metadata needs to be more highly replicated (ditto blocks)
6410700 zdb should support reading raw blocks out of storage pool
6410709 ztest: spa config can change before pool export

6280668 pluggable block allocation policy
6399301 initial read of space maps is super slow
6407365 large-sector disk support in ZFS
6407366 ADVANCE_NOLOCK gathers MOS
6407367 three-way deadlock between db_mtx, dbuf_hash[], and ms_lock

PSARC 2006/077 zpool clear
PSARC 2006/139 FMA for ZFS
6284889 arc should replace the znode cache
6333006 DMU & DSL should not panic upon I/O error
6333092 concurrent reads to a file not scaling with number of readers
6338081 ZFS/FMA phase 1
6338386 need persistent error log
6341326 i/o error causes arc buf hash table corruption
6341639 zfs backup/restore should compute/verify checksum of backup stream
6348002 out of space due to changing properties
6354724 inaccurate error message from zfs restore
6354872 dmu_sync() blows predictive accounting
6355416 zpool scrubbing consumes all memory, system hung
6363995 df should only load libzfs when it encounters a ZFS filesystem
6366320 zfs backup/restore doesn't like signals
6368892 mount -m support needed for legacy mounts
6368902 boot archive fstat support needed for ZFS Mountroot
6369424 BFU complains when bfu'ing a ZFS root filesystem
6374062 mountroot support needed for ZFS
6376356 dirtying dbuf obj=43 lvl=0 blkid=0 but not tx_held
6378391 unused members of dmu_objset_stats_t
6378392 clean up zfs_cmd_t structure
6378685 buf_init should allocate its hash table more carefully
6378976 ziltest should be a first class citizen
6381086 zdb segfaults if there is a spa deferred-free bplist
6381203 deadlock due to i/o while assigning (tc_lock held)
6381209 freed space is not immediately available
6381344 'zpool clear'
6381345 FAULTED devices should really be UNAVAIL
6381346 import should mark devices as persistently unavailable
6383272 recursive mutex_enter() during log replay with zfs root
6386326 origin property is not displayed
6386354 libzfs does too much in its _init section, calls exit(1)
6386624 zpool should not complain about non-existent devices from libdiskmgt
6386910 spa needs to be i/o error hardened
6387735 need a mechanism to inject faults into ZFS
6387736 internal ZFS utilities should be placed in an ON-private package
6389928 libzfs should ship a lint library
6390609 malformed vdev config panics on zpool_create()
6390677 version number checking makes upgrades challenging
6390713 ztest hangs in zil_suspend()
6391873 metadata compression should be turned back on
6392113 ztest sometimes reports leaked blocks because ZIL isn't resilvered
6393004 minor memory leak in unique_insert()

PSARC 2002/240 ZFS
6338653 Integrate ZFS
PSARC 2004/652 - DKIOCFLUSH
5096886 Write caching disks need mechanism to flush cache to physical media