Return to 1.1_changelog.html CVS log

Up to [Development] / xfs-website.orig

Updates for 1.1 release

<& xfsTemplate,top=>1,side=>1 &>

<!-- Start Project Content -->
<h2>
<b><font face="ARIAL NARROW,HELVETICA">XFS for Linux Release 1.1 Change Log</font></b></h2>


<p>
The following changes have been made to XFS for Linux
for the 1.1 release.
</p>

<h3>Kernel</h3>
Removed most synchronous transactions - faster deletes, fewer chances for null files after a crash!
<br>
Various error code return fixes
<br>
Restrict inodes to 32 bits on large filesystems (override w/ mount option)
<br>
Fixed concurrent file and mmapped I/O
<br>
Fixed dbench hangs on low memory systems
<br>
Fixed recovery of device special inodes
<br>
Fixed mount argument parsing
<br>
Various pagebuf reorganization, simplification, and cleanup
<br>
Fixed parallel direct and buffered I/O
<br>
Code merges from IRIX
<br>
Pagebuf merged into xfs
<br>
Fixed out-of-line extended attribute data
<br>
Fixed forced shutdown bug that overwrote superblock :(
<br>
Improved memory allocation when not in a transaction
<br>
Limit max file size to something Linux can handle
<br>
Some realtime device fixes (still not complete)
<br>
Cleaned up xfs_freeze path
<br>
Report filesystem name on duplicate UUID mount failure
<br>
Shrink xfs inode size
<br>
Fixed some direct I/O corner cases
<br>
Fixed mount with bad log or realtime device options
<br>
Make "osyncisdsync" the default on xfs filesystems
<br>
Restrict chown to file's owner, or someone w/ the right capability
<br>
Upgraded quota to Jan Kara's 32-bit VFS quota
<br>
Fixed memory leak in O_DIRECT read path
<br>
Use new reserved ea/acl syscall numbers
<br>
Fixed some sparc64 compile problems
<br>
Make xfs superblock coherent with block layer
<br>
Pagebuf use after free fix
<br>
Don't allow quota flag changes on read-only device
<br>
Make xfs metadata accesses refresh pages, keep them in the cache
<br>
Fixed sgid inheritence for root
<br>
Corrected utime permissions checking
<br>
Reduced xfs log memory usage
<br>
Fixed a bug in memory freeing
<br>
Delete nfs refcache sbdirty timer on unmount
<br>
Make nfs refcache sbdirty timer for each fs, not global
<br>
Fix for inode32 mount option on > 1TB filesystems


<h3>acl</h3>
Man page updates from Andreas
<br>
Test script updates from Andreas
<br>
Clean up the --default option handling in setfacl. The old
workarounds caused a bug for unusual input.
<br>
Changes to the --test output format setfacl generates: ACLs that
are not changed are now displayed as `*'.
<br>
Fixed a bug in setfacl/sequence.c:seq_delete_cmd()
<br>
Minor changes to test scripts
<br>
Apply several patches from Andreas, namely:
<br>
- man page fixes
<br>
- libacl code reformatting
<br>
- acl_from_text errno handling
<br>
Applied Andreas Gruenbacher's diffs
<br>
Fixed up chacl for deletion of access ACL to be in line with Andreas 
<br>
Incorporated the Debian packaging again
<br>
Reworked to use the new official system call API
<br>
Sync up with the XFS project, the SGI folk now use this source
<br>
Jumped to version 2 to allow XFS users to upgrade
(Rationale: the XFS ACL user tools were at version 1.1.X, and
packaging tools like rpm, dpkg, etc. must be presented with a
greater version number to allow an upgrade to proceed)
<br>
Added the chacl command to ease migration for existing XFS users,
and for compatibility with IRIX
<br>
Added a flag to allow acl_print to produce a single-line ACL, in
addition to the multi-line format
<br>
Extended attribute documentation has moved into the extended
attribute package from SGI ("attr"), this ACL package now deals
exclusively with ACLs
<br>
acl_from_text sometimes did not set errno when failing
<br>
Moved files and simplified #includes in libacl

<h3>attr</h3>
Add MIPS/MIPS64 system call numbers
<br>
Fixed build for architectures which don't have syscalls yet
<br>
Fixed the syscall number used on Sparc for fremovexattr(2)
<br>
Test script updates
<br>
Man page updates
<br>
A minor change to the test/run script
<br>
Added in ARM architecture system call numbers
<br>
Updates to the test output from Andreas
<br>
Added in S/390 system call numbers from Martin Schwidefsky
<br>
Revert IA64 syscall numbering after further mail with
David Mosberger (apparently sys_tkill will be moved)
See:	https://external-lists.valinux.com/archives/linux-ia64/2002-February/002990.html
<br>
Incorporated several documentation changes from Andreas,
including a script to convert from the aget format of
attribute backup file, to the new getfattr format
<br>
Fixed IA64 syscall numbering
<br>
Initial introduction of the new system call interface
<br>
Synced up with the ext2 project, incorporated get/set tools
<br>
New man pages for system calls, getfattr(1) and setfattr(1)
<br>
Made the attributes.h interface align properly with IRIX

<h3>DMAPI</h3>
The kernel-side of dmapi is now a module, and the device has
moved.  Change dmapi to use the dmapi device in its new
location of /proc/fs/xfs_dmapi.

<h3>xfsprogs</h3>
Fall back to BLKGETSIZE if BLKGETSIZE64 fails
<br>
Sync user/kernel headers and shared code
<br>
Major release to coincide with switch to new extended
attributes system call interfaces
<br>
Bumped version of libhandle, added new symbols to use
the reworked extended attributes handle ioctl interface
<br>
xfs_repair in no-modify mode opens the filesystem device
read-only now (fix from Chris Pascoe)
<br>
Sync up with recent (minor) changes to shared kernel code
<br>
Switch to using the BLKGETSIZE64 ioctl in libxfs, instead
of the (previously busted) BLKGETSIZE ioctl
<br>
Fixed xfs_repair option parsing for external logs
<br>
Added xfs_repair option parsing for realtime device
<br>
Fixed xfs_repair version (-V) option - should not
require an argument
<br>
Added -V option to usage string
<br>
Document verbose (-v) and -r options in manpage
<br>
Fixed mkfs.xfs buglet in overwriting signatures when run
on a regular file
<br>
mkfs.xfs overwrites pre-existing filesystem, swap, or md
driver signatures.
<br>
xfs_repair fix to prevent double insertion into the
uncertain_inode AVL trees ("avl_insert: duplicate range")
<br>
xfs_repair fix if the log is corrupted and we can't find
the head, don't exit - just proceed on with zeroing it
<br>
Use snprintf instead of sprintf throughout
<br>
Added text dump type to xfs_db (mkp)
<br>
Removed use of a temporary file in xfs_db when processing
commands on the command line - allows xfs_check to be run
on read-only root filesystems
<br>
Reenable the use of the BLKBSZSET ioctl, its baaack
<br>
Sync recent XFS kernel source changes back into libxfs
<br>
Fixed minor debian package version numbering issue
<br>
Added documentation for xfs_db(8) label/uuid commands
<br>
Automatic inode sizing code in mkfs.xfs has been removed
(restricting inodes to 32 bits) - Steve's recent kernel
changes mean this is no longer an issue
<br>
Fixed bug in mkfs.xfs size cross-check for realtime device

<h3>xfsdump/restore</h3>
Allow dumps of files greater than 2<sup>31-1</sup> bytes.
<br>
Reworked all code dealing with extended attributes to use
the new system calls (requires attr-2.0.0 or greater)
<br>
The attrctl-by-handle ioctl is history, replaced
by libhandle routines - more like what we have in IRIX
(requires xfsprogs-2.0.0 or greater)
<br>
Effectively no-op change (cleanup) - switch over to using
XFS_IOC_FSGEOMETRY instead of XFS_IOC_GETFSUUID ioctl, so
we can deprecate that "special" UUID ioctl in the kernel.
<br>
Added -q description to xfsdump/xfsrestore man pages and
usage text.
<br>
Change failed bulkstat WARNING to a TRACE message to that
it doesn't bother people.
<br>
Avoid a possible assertion failure for cumulative restores
with -B option.
<br>
Fixed xfsrestore so that cumulative restores (with -r)
will successfully delete removed directories whose
files have also been removed.
Previously, the files weren't removed until later, 
which meant that early directory removal failed.
SGI bug#844219.
<br>
Fixed xfsdump so that if an inode# is reused in the time
between building the inode map and pruning the inode map
(in phase 3 when some dirs are marked as not changed),
that it no longer aborts with an assertion failure.
SGI bug#846374.
<br>
Added new -B option to xfsrestore to correctly assign
ownership and permissions of the dump root directory
to the destination directory
<br>
Ported back IRIX changes primarily to xfsrestore for
improving performance when one has over a million files.
<br>
Some extra mlogs (messages) for dump estimates, 
dir tree diagnostics, type of dump format being used
<br>
Various fixes for restore with multiple threads
and extended attributes (note: multiple threads not
implemented on Linux yet)
<br>
Fix xfsdump to endian convert all of the record header
fields properly just prior to writing the header out
(in particular first_mark_offset).
This caused do_next_mark() assertion failures at some
sites.
<br>
Fixed xfsrestore so that it doesn't delete hardlinks
on alternate cumulative restores
<br>
Allow xfsdump to exclude files based on whether they have
a certain extended attribute set
<br>
Don't include /var/lib/xfsdump in the dump

<h3>misc</h3>
Updated documentation.

<br>


<& xfsTemplate,bottom=>1 &>