Marcelo released 2.4.20-pre4 today. Included in the bug fixes and driver updates was the merge of JFS. JFS is IBM's journaled filesystem from OS/2. JFS had previously been merged into the -ac tree (2.4.18pre9-ac4) and was merged into the 2.5 tree early on (2.5.6). JFS joins ext3 and reiserfs in the 2.4 tree. SGI's XFS is still awaiting inclusion into the stable tree.
So here goes -pre4, with JFS merged.
Also, if you got bootup lockups or some unexpected weird error try
-pre4 ;)
Summary of changes from v2.4.20-pre3 to v2.4.20-pre4
============================================
:
o setup_arch() cleanups
o (2/4) discontigmem support for i386 against 2.4.20pre3
:
o Fix a problem that when doing online resizing, resizer code forgot to update bitmap usage counters
o Fix a problem where bitmap usage counters were possibly incorrectly updated on bigendian and 64 bit boxes
:
o VM docs from -ac
o fix current BK tree compilation with devfs enabled
:
o 686-notsc_A0
o [PATCH] notsc-warning_A0
:
o USB: hpusbscsi driver updates
:
o USB storage: get rid of DMA to stack
Adrian Bunk :
o Fix ftape build problems
Christoph Hellwig :
o JFS: Initial import of version 1.0.18 for Linux 2.4
Dave Kleikamp :
o JFS: Fix structure alignment problem on 64-bit machines
o JFS: Add hch's copyright
o JFS: sanitize ->clear_inode, remove ->put-inode
o Fix races in JFS threads
o JFS: Yet another truncation fix
o JFS does not need to set i_version. It is never used
o JFS: fix fsync
o procfs entries should be created when CONFIG_JFS_STATISTICS is set
o JFS: set s_maxbytes to 1 byte lower
o Rework JFS's inode locking
o JFS: Dynamically allocate metapage structures
o Remove d_delete calls from jfs_rmdir & jfs_unlink
o JFS: Fix handling of commit_sem
o Add resize function to JFS
o fix typo in fs/jfs/resize.c
o JFS: Replace depreciated initializer syntax with C99 style
o JFS: Trivial fixes
Geert Uytterhoeven :
o Fix compile warning in init/do_mounts.c
Greg Kroah-Hartman :
o USB: serial Config.in cleanups
o USB: ftdi_sio driver update
o USB: ipaq driver updates
o USB: pl2303 driver update
o USB: serial driver minor fixes
o USB: ir-usb driver minor fixes
o USB: add usb-storage sddr-55 driver
o USB: bluetooth driver fixes
o USB: scanner driver update and maintainer change
Marcelo Tosatti :
o Changed EXTRAVERSION to -pre4
o Added arch/i386/kernel/time.o to exportobj list
Niels Kristian Bech Jensen :
o Avoiding implicit declaration in net/netsyms.c
o Fixing a compiler warning in drivers/block/genhd.c
Paul Mackerras :
o fix bug in yield()
Richard Gooch :
o Switched to ISO C structure field initialisers
o base.c
Simon Evans :
o 2.4.19 - add support for f5u011 to catc.c
Steven Cole :
o 2.4.20-pre2 add module text for 58 options
Trond Myklebust :
o Add round trip timing to RPC over UDP client [1/3]
o Add round trip timing to RPC over UDP client [2/3]
o Add round trip timing to RPC over UDP client [3/3]
o Improve RPC request ordering
o Improve network congestion code [1/3]
o Improve network congestion code [2/3]
o Improve network congestion code [3/3]
o Fix RPC write_space() code
o Increase UDP socket buffer size
V. Ganesh :
o typo in usb/serial/ipaq.h
Inclusion of JFS
Hmm, I must be getting confused with the different systems I have, running different kernels (Vanilla, -ac*, etc). For some reason I thought JFS was included several releases ago, since it always appears in menuconfig->filesystems :) (and no, I'm not confusing it with jffs)
Re: Inclusion
It was in the -ac tree a while ago, that's probably what you're thinking of. For the original post, the JFS Linux port is actually a port of JFS from OS/2, not from AIX. There are differences between them, but probably not many the average user would notice.
AIX, not OS/2
JFS is from AIX, not from OS/2.
Kris
OS/2
from this story on IBM.com anouncing the JFS project:
IBM is donating its Journaled File System (JFS) technology to the Linux open source community with the goal of completing a port to Linux. The JFS technology, which is currently used in the IBM OS/2 Warp Server, is now an open source project on developerWorks (see Resources). The code is available under the GPL license.
[emphasis added]
I was wrong.
Well waddya know
Spoke too soon! It does come from AIX by way of the OS/2 port. What releases of OS/2 come with JFS?
Kris
surprising
I'm a little surprised that JFS made the merge before XFS. I have been under the impression that XFS for linux is more stable than JFS. I guess I have bad information. I'm glad to see JFS make the merge. I guess I should try it one of these days. ;o)>
re: surprising
i guess it's because jfs is less intrusive than xfs.
re: re: surprising
This is correct - due to this, XFS is not yet even in 2.5, though they are trying to get it in before the freeze. It's not likely that XFS will make it into 2.4 at all.
It is in 2.4 kernels
In the distro's kernels (e.g. Mandrake)
Distro Kernel is not the Linux kernel
The subject under consideration is the Linux Kernel, not Manadrake, Red Hat, SuSe or any other Distro Kernel. There are a great many features that appear in many Distro Kernels but not the Linux Kernel, although we might it to be otherwise. Saying something is in a Distro 2.4 Kernel is not similar to saying ,"It is in the Linux 2.4 Kernel."
I'm also surprised
If you read things: Sun sell linux workstations with XFS, and XFS is in remarked on the title.., Gentoo linux has a XFS version.., The only journalized file system that _really_ works seems to work fine is XFS.. WHY it is not on linux tree???
I hope it will be on linux distributions before a 2.5 stable branch. Things as XFS, Java2jdk,.. are good things to be in linux.
As already stated...
XFS is too intrusive. In its initial form, the port made too many changes to the VFS layer. That's being worked on for inclusion. I also don't agree with your statement that it's the only one that really works fine, I consider ext3 to work great.
ReiserFS WFM
Been running ReiserFS for a network of dozens of diskless boxes for over 2 years now without a hiccup. Running over a Linux-RAID1 array and serving through NFS always shows me that it's not in bad shape, and both performance and stability have been tops. I like it.
Re: ReiserFS WFM
I too have been running ReiserFS for a while, but I had numerous problems in the early 2.4 series with both reliability and kNFSd. These have since sorted themselves out, but I didn't mention it in the "it just works" group for that reason.