changelog.jfs

jfs-2.4-1.1.7.tar.gz jfs 2.4－1.1.7 源码

IBM's Journaled File System (JFS) for Linux

Release July 22, 2004 (version 1.1.7)

This is our seventieth release of IBM's Enterprise JFS technology port to
Linux.  Version 1.0.0 was released 6/28/2001.

Function and Fixes in drop 70 (1.1.7)
   - Fix null-pointer dereferences
   - Fix memory leaks
   - get rid of buggy txAbortCommit
   - Avoid race invalidating metadata page
   - Reduce stack usage
   - module unload was not removing /proc/fs/jfs
   - More robust error recovery in add_index
   - ea_get was calling __get_metapage with invalid size
   - Make sizes of tid_t & lid_t consistent
   - Don't return EPERM for system xattrs
   - Don't call d_instantiate if txCommit fails
   - Handle out of space errors more gracefully
   - Fix stack overrun in dtSearch
   - Don't allow reading beyond inode map's EOF
   - zero new log pages, etc.
   - Updated field isn't always written to disk during truncate
   - Protect active_ag with spinlock
   - Prevent concurrent calls to txCommit on the imap inode
   - Check for dmap corruption before using leafidx

Function and Fixes in drop 69 (1.1.6)
   - zero uninitialized data
   - Add lots of missing statics and remove dead code (Christoph Hellwig)
   - Prevent hang in __lock_metapage
   - Fix race in jfs_sync

Function and Fixes in drop 68 (1.1.5)
   - Protect i_nlink with commit_sem
   - Avoid segfault when dirty inodes are written on readonly mount
   - Fix setting of ea_buf->flag to avoid assert
   - rename should update mtime on source and target directories

Function and Fixes in drop 67 (1.1.4)
   - Make sure journal buffer gets flushed to disk
   - Improved error handling
   - Remove racy, redundant call to block_flushpage
   - Fix race between link() and unlink()

Function and Fixes in drop 66 (1.1.3)
   - Fix compat.h so that compile does not break on unlikely() and dump_stack()
   - jfs_lookup should check for bad inode returned by iget
   - Prevent rare deadlock
   - Fix compile problem with recent Redhat kernels
   - Code cleanup suggested by static analysis tool
   - Peformance improvement
   - Use block device inode/mapping instead of direc_inode/direct_mapping
   - Support for kernels older than 2.4.18 has been removed
   - Fix resize errors
   - Fix possible trap/data loss when fixing directory index table
   - If unicode conversion fails, operation should fail
   - Code cleanup: make error return codes negative
   - add nointegrity mount option
   - Merge xattr code.  This has been in the official kernel since 2.4.20
   - jfs_write_super_lockfs should mark superblock clean

Function and Fixes in drop 65 (1.1.2)
   - Clean up code flushing outstanding transactions to the journal
   - Replace ugly debug macros with simpler ones
   - Add get_index_page to eliminate unneeded I/O
   - Fix hang while flushing outstanding transactions under heavy load
   - Avoid deadlock under very heavy load
   - Don't zero s_op during failed mount cleanup

Function and Fixes in drop 64 (1.1.1)
   - calling brelse() with wrong bh during writing out of secondary superblock 
     while resizing the FS
   - Remove COMMIT_Holdlock
   - jfs_clear_inode was assuming that active_ag should never be set comming into
     this routine. Since it's possible for a file to be extended after the file
     descriptor has been closed, we need to allow the possibility. (Dirty pages
     of memory mapped files can be written after the file has been closed.)
   - Avoid writing partial log pages for lazy transactions.
   - Move index table out of directory inode's address space.
   - jfs_truncate needs to call block_truncate_page


Function and Fixes in drop 63 (1.1.0)
   - Fix off-by-one error when symbolic link length == 256 (bug 1513)
   - jfs_clear_inode should skip bad inodes instead of choking on them (bug 1445)
   - Make txForce actually force the metadata to disk
   - Fix hang on umount after stress test(#21357) ACL problem
   - Fix byte-swapping problem in getting ealist->size (bug 21085) EA problem
  
Please send bugs, comments, cards and letters to [e-mail].

The JFS mailing lists can be subscribed to by using any of the links
containing "subscribe" along the left side of our web page at
http://oss.software.ibm.com/jfs/.