1 xxxx-xx-xx Oracle, Inc.
6 Description: MMP might sleep negative time
8 -------------------------------------------------------------------------------
10 2010-10-29 Oracle, Inc.
15 Description: Bring upstream patch for ldiskfs.
19 Description: add support for SLES11 SP1
21 -------------------------------------------------------------------------------
23 2010-07-31 Oracle, Inc.
28 Description: Reduce group prealloc size and skip groups with little free space.
32 Description: Fix issue with proc_remove.
36 Description: add mount option to disable mb_cache since it can cause slowdown.
40 Description: Disable delayed allocation by default for ext4-based ldiskfs on
45 Description: A mount failure can corrupt the slab. This is a bug in the latest
46 RHEL5.5 kernel and only ext4-based ldiskfs is impacted.
48 -------------------------------------------------------------------------------
50 2010-04-30 Oracle, Inc.
54 -------------------------------------------------------------------------------
56 2010-01-29 Sun Microsystems, Inc.
60 -------------------------------------------------------------------------------
62 2009-11-05 Sun Microsystems, Inc.
66 Frequency : in recovery
68 Description: ext3-corrupted-orphans-2.6.patch added.
69 Details : Backport of ext3 patch to handle corrupted orphans.
71 -------------------------------------------------------------------------------
73 2009-07-31 Sun Microsystems, Inc.
77 Frequency : in recovery
79 Description: don't mix llog inodes with normal.
80 Details : allocate inodes for log in last inode group
82 -------------------------------------------------------------------------------
84 2009-04-20 Sun Microsystems, Inc.
89 Description: minor fixes and cleanups
90 Details : use EXT_UNSET_BLOCK to avoid confusion with EXT_MAX_BLOCK.
91 Initialize 'ix' variable in extents patch to stop compiler warning.
95 Description: update FIEMAP ioctl to match upstream kernel version
96 Details : the FIEMAP block-mapping ioctl had a prototype version in
97 ldiskfs 3.0.7 but this release updates it to match the
98 interface in the upstream kernel, with a new ioctl number.
101 Frequency : only if MMP is active and detects filesystem is in use
103 Description: if MMP startup fails, an oops is triggered
104 Details : if ldiskfs mounting doesn't succeed the error handling doesn't
105 clean up the MMP data correctly, causing an oops.
107 -------------------------------------------------------------------------------
109 2009-04-06 Sun Microsystems, Inc.
115 Description: fix racy locking of mballoc block bitmaps causing BUG
116 Details : The locking of the mballoc buddy bitmap and the in-memory
117 block bitmap was using two different spin locks in some
118 cases. This made it possible to incorrectly access the
119 mballoc bitmap while another process was modifying it,
120 causing a sanity assertion to fail. While no on-disk corruption
121 was reported, there was some risk of this happening.
123 -------------------------------------------------------------------------------
125 2009-02-07 Sun Microsystems, Inc.
128 Severity : enhancement
130 Description: Get RAID stripe size from superblock
131 Details : RAID striping parameters are now saved in the superblock itself,
132 so we should use these parameters instead of having to specify
133 a mount option each time.
136 Frequency : only if server is running on unsupported big-endian machine
138 Description: Disable big-endian ldiskfs server support.
139 Details : The ldiskfs code is not tested on big-endian machines, and
140 there are known compatibility problems in the extents code
141 when running for most of the kernels. Print an error message
142 and refuse to mount, in case anyone tests this. For existing
143 filesystems that might have been created in this way it is
144 possible to mount with the "bigendian_extents" option to
148 Frequency : only with software RAID-5
150 Description: MMP block reads may fail to detect another user of the filesystem
151 Details : with software RAID it is possible that the RAID driver will
152 cache the MMP block and not refetch it from disk. Force the
153 read to invalidate the RAID page cache and go directly to disk.
155 -------------------------------------------------------------------------------
157 2008-08-31 Sun Microsystems, Inc.
160 Severity : enhancement
162 Description: Interoperability at server side (Disk interoperability)
163 Details : Exported some ldiskfs functions which are required for iop
167 Description: OSS crashes frequently, e2fsck does not fix
168 Details : Add an extra check to ldiskfs extents code for the condition,
169 eh_entries = 0 & eh_depth != 0
173 Description: migrate ldiskfs - ldiskfs2 (extents on directories)
174 Details : disable preallocation for non-regular files.
177 Frequency : blocks per group < blocksize*8 and uninit_groups is enabled
179 Description: ldiskfs error: XXX blocks in bitmap, YYY in gd
180 Details : If blocks per group is less than blocksize*8, set rest of the
184 Frequency : only for filesystems larger than 8TB
186 Description: ldiskfs BUG ldiskfs_mb_use_best_found()
187 Details : The ldiskfs mballoc3 code was using a __u16 to store the group
188 number, but with 8TB+ filesystems there are more than 65536
189 groups, causing an oops.
191 Severity : enhancement
193 Description: Add a FIEMAP(FIle Extent MAP) ioctl for ldiskfs
194 Details : FIEMAP ioctl will allow an application to efficiently fetch the
195 extent information of a file. It can be used to map logical blocks
196 in a file to physical blocks in the block device.
200 Description: mkfs.lustre: Unable to mount /dev/cciss/c0d1:Cannot allocate memory
201 Details : correctly handle device paths using a subdirectory in /dev when
202 creating the per-device procfs directory under /proc/fs/ldiskfs.
204 -------------------------------------------------------------------------------
206 2008-04-26 Sun Microsystems, Inc.
211 Description: Kernel BUG at ... ldiskfs/ldiskfs/mballoc.c:3853
212 Details : Removing the underlying device (e.g. via a SCSI rescan) can cause a
213 kernel BUG in ldiskfs_mb_discard_inode_preallocations().
217 Description: deadlock in ext3_expand_extra_isize
218 Details : Inode expansion during ext3_mark_inode_dirty may lead to deadlock.
219 This is solved by making sure that ext3_expand_extra_isize isn't
220 called from ext3_xattr_set_handle.
224 Description: soft lockups on 1.6.2 MDS (is_subdir)
225 Details : don't add dentries with ".." to dcache and ignore such dentries in
229 Frequency : very rare, if additional xattrs are used on kernels >= 2.6.12
231 Description: files may lose file attributes in some cases
232 Details : If there are multiple extended attributes stored on the inode,
233 in particular ACLs, SELinux, or user attributes (if user_xattr
234 is specified for the client mount options) then there is a risk
235 of attribute loss. If an additional attribute is be stored
236 initially in the inode and then increase in size enough to be
237 moved to the external attribute block (e.g. ACL growing in size)
238 for the attribute to be lost.
242 Description: inode version not being initialized on new inodes
243 Details : The inode i_version field was not being initialized on disk.
244 This field is currently unused but will be needed for VBR.
246 --------------------------------------------------------------------------------
248 2008-01-11 Sun Microsystems, Inc.
253 Description: Add support for vanilla-2.6.22 kernel.
255 --------------------------------------------------------------------------------
257 2007-12-07 Cluster File Systems, Inc. <info@clusterfs.com>
262 Description: statfs speedup patches.
263 Details : The patches improve statfs performance for very large
267 Frequency : if the uninit_groups feature is enabled on ldiskfs
269 Description: e2fsck reports "invalid unused inodes count"
270 Details : If a new ldiskfs filesystem is created with the "uninit_groups"
271 feature and only a single inode is created in a group then the
272 "bg_unused_inodes" count is incorrectly updated. Creating a
273 second inode in that group would update it correctly.
275 --------------------------------------------------------------------------------
277 2007-09-27 Cluster File Systems, Inc. <info@clusterfs.com>
280 Severity : enhancement
282 Description: Add a FIEMAP(FIle Extent MAP) ioctl for ldiskfs
283 Details : FIEMAP ioctl will allow an application to efficiently fetch the
284 extent information of a file. It can be used to map logical blocks
285 in a file to physical blocks in the block device.
287 --------------------------------------------------------------------------------
289 2007-08-29 Cluster File Systems, Inc. <info@clusterfs.com>
295 Description: Multiple mount protection(MMP) support.
296 Details : This feature will protect the filesystem from being mounted more
297 than once simultaneously. It will also protect changes by e2fsprogs
298 to the filesystem if it is mounted. This assumes high importance
299 in a shared storage environment where multiple mounts can severely
300 corrupt the filesystem.
302 --------------------------------------------------------------------------------
304 2007-08-10 Cluster File Systems, Inc. <info@clusterfs.com>
306 * Initial release of ldiskfs as a separate package.
311 Description: Updated patchess for new RHEL4 kernel
312 Details : Add patch ext3-unlink-race.patch
313 Updated series file ldiskfs-2.6-rhel4.series