|
|
 |
Work item list as of 3/29/2000
The current work item list for XFS on Linux are listed below.
The individual(s) working on the item are listed before the description
of each item. Some items are unassigned.
- Australian team (Ken M.)
-
Files we created outside the xfs directory need copyright insertion:
include/linux/*xfs*,
include/linux/page_buf.h,
fs/page_buf.c,
fs/page_buf_locking.c, and
fs/avl.c.
- unassigned
-
Documentation: we need man pages for all XFS commands ported from IRIX to
Linux, we also probably need a README of some type to say what works and does not
work, other documentation?
- Steve/Ted
-
Integrate any new fixes from the latest XFS in 6.5.8f into the Linux.
tree.
- Ted
-
Finish implementing the syssgi() commands.
Use /proc interface: To be used for growfs/bulkstat.
Primarily for admin commands.
- everyone
-
Compile warnings and cleanup. Kill dead code, fix compile warnings.
Each person should remove IRIX specific or any duplicate code as they see it.
- Russell
-
Build and maintain CVS tree on oss.sgi.com.
- Ananth
-
Have a kernel build work on XFS. This requires fixing the code that
converts holes into real space and correctly zeros pages/blocks.
Hole support tested more thoroughly. This has been tested somewhat.
Should be tested more especially after more coding is done above.
- Ananth
-
Direct I/O: We need to be able to turn the users pages into a pagebuf.
This involves having kiobufs point to user memory (already done for raw I/O),
and flushing any buffered user data that is dirty (since it will be
read directly from disk).
- Ananth/Steve/Russell
-
Delayed write support. The pagebuf cleaner and page cleaner need to
handle delaying writes. Steve and Russell will handle the meta data path
and Ananth will handle the normal I/O path.
- Ananth/Jim
-
Delalloc support: page cleaner and calls to cluster pages, map/allocate.
The page cleaner for user I/O is needed. Also,
xfs_iomap_write needs to know about delayed allocation (conversion and
creation).
- unassigned
-
Flush/invalidate/toss pages for an inode.
This is mostly there (pagebuf_inval) but more work may be needed
for direct I/O.
- Chait
-
Integrate kiobufs into disk drivers and provide interface
similar to ll_rw_block for XFS (and any other file system). This
should also work for volume managers like LVM, md, ...
- Linuxcare (Martin/Phil)
-
Modify the mount command/interface for XFS to allow multiple devices (log, data,
real-time). This has been mostly coded. We need some performance
numbers.
- unassigned
-
There are many "versions" and options currently supported by XFS on IRIX.
This includes things like version 1 directories and before unwritten extents.
We must decide which of all these will be supported on Linux.
- unassigned.
-
Replace the MR locks in XFS with the 2.3 new locks and possibly push
locking code into 2.3.
- unassigned
-
XFS's interface through ll_rw_block() must be able to handle multiple
block sizes so I/O can go bigger than 512 bytes.
- unassigned
-
Add the ability to run without the log. This is lower priority but
would be useful in comparing XFS to ext2 and other non-journalled file
systems. Stub out LOG/trans code for XFS. This would need to be
a mount and/or mkfs option.
- unassigned.
-
Have XFS operate as the root file system.
We should try /usr first before root.
- Australian team (Ken M.)
-
xfs_repair needs exercising and fixing.
- Australian team (Ken M.)
-
xfs_db has endian issues with block numbers that are large.
- Australian team (Ken M.)
-
dump/restore - the xfs specific ones and the generic ones.
- Australian team (Ken M.)
-
xfs_growfs needs to work with various volume managers.
This depends on /proc work see below.
- unassigned
-
Quotas: There are linux quotas and xfs quotas .....
- Ted
-
Extended attributes: There needs to be a kernel interface.
- Linuxcare (Martin/Phil)/Heinz(LVM)
-
Volume manager work: Allow XFS to sit on top of LVM and MD
extract stripe info from volume manager at mkfs time.
Support separate log, data, realtime subvolumes, or multiple devices.
- Linuxcare (Martin/Phil)
-
64 Bit LFS work: Allow user space to use 64 bit files. Demonstrate
a file larger than 4gig.
- unassigned
-
DMAPI: implement this in Linux.
- unassigned
-
Realtime: configure and test.
- unassigned
-
Access control: Linux and Irix do this in separate places, we have not
really integrated the two yet.
- unassigned
-
Sync: Are we syncing the right things at the right times?
- unassigned
-
Redo vnode/behavior code:
We could place the vnode and/or behavior in the linux inode and remove the
vnode management code. There are two separate reference counts at the moment.
For CXFS the interface could be made cleaner.
- Australian team (Ken M.)
-
Endian/architecture issues: have Linux XFS be able to read XFS file systems
created on IRIX/MIPS boxes and other big endian machines. The reverse
must work, too.
- Russell
-
Build RPMs of XFS. This depends on a relatively stable XFS. The first
RPM will be an alpha version.
- unassigned
-
Port GRIO to Linux->XFS.
- unassigned
-
XFS does not work on loopback devices. This should be fixed but it is
lower priority.
unassigned
Add global version number accross all xfs tools and
filesystem.
|