mainlining ....

[OGAWA Hirofumi]

Lars Segerlund <lars.segerlund at gmail.com> writes:

>  Quick question, what requirements do we have for mainlining ?
>
>  Perhaps a rough tasklist or goal would be nice to have, I can think
> this as a way to motivate developers, much more than being mainline
> itself.

This is my rough todo list (well, this might be rather reminder though),
although I don't think all are necessary for mainlining.  And I have
more items on todo, however others are too random or some ideas.

* Future
  - [ ] page base bitmap write
  - [ ] cleanup
  - [ ] backup of web site
  - [ ] build/stress/performace test
  - [ ] delta triggers
  - [ ] alloc goals
  - [ ] fsync/fdatasync()
  - [ ] write optimize for logblock format
    - current logblocks can't read at once
      We can use extent for each delta
  - [ ] inode->i_blocks (maybe)
  - [ ] update_time() handler for FIFO, device, etc.
  - [ ] new shard index  and "."/".."
    - shard.c
  - [ ] enospc
  - [ ] mmap
    - support read/write mmap
    - mmap write returns proper signal with si_errno on allocation failure
  - [ ] xattr fixes
  - [ ] atime
  - [ ] memory deadlock audit
  - [ ] IO/OOM error handling
  - [ ] otable compaction - btree merge/rebalance
  - [ ] fiemap() (maybe)
  - [ ] SEEK_HOLE/SEEK_DATA
  - [ ] swapfile
  - [ ] auto defrag
  - [ ] debug
  - [ ] snapshots
  - [ ] fsck
    - What are uptags? (ml)
  - [ ] ->migratepage
* test
  - [ ] phoronix-test-suite
  - [ ] ztest
    http://hub.opensolaris.org/bin/view/Community+Group+zfs/ztest

Thanks.
-- 
OGAWA Hirofumi <hirofumi at mail.parknet.co.jp>