GitHub/exynos8895/android_kernel_samsung_universal8895.git
19 years ago[PATCH] md: fail IO request to md that require a barrier.
NeilBrown [Fri, 9 Sep 2005 23:23:41 +0000 (16:23 -0700)]
[PATCH] md: fail IO request to md that require a barrier.

md does not yet support BIO_RW_BARRIER, so be honest about it and fail
(-EOPNOTSUPP) any such requests.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] md: fix minor error in raid10 read-balancing calculation.
NeilBrown [Fri, 9 Sep 2005 23:23:40 +0000 (16:23 -0700)]
[PATCH] md: fix minor error in raid10 read-balancing calculation.

'this_sector' is a virtual (array) address while 'head_position' is a physical
(device) address, so substraction doesn't make any sense.  devs[slot].addr
should be used instead of this_sector.

However, this patch doesn't make much practical different to the read
balancing due to the effects of later code.

Signed-off-by: Neil Brown <neilb@cse.unsw.edu.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/sam/kbuild
Linus Torvalds [Fri, 9 Sep 2005 22:46:49 +0000 (15:46 -0700)]
Merge /pub/scm/linux/kernel/git/sam/kbuild

19 years agoRemove "must_check" attributes in PCI-land
Linus Torvalds [Fri, 9 Sep 2005 22:43:46 +0000 (15:43 -0700)]
Remove "must_check" attributes in PCI-land

Don't just irritate all other kernel developers.  Fix the users first,
then you can re-introduce the must-check infrastructure to avoid new
cases creeping in.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years agoPreempt-safe RCU file usage
Linus Torvalds [Fri, 9 Sep 2005 22:42:34 +0000 (15:42 -0700)]
Preempt-safe RCU file usage

Fix up fs/compat.c fixes.

19 years ago[PATCH] Update PCI IOMEM allocation start
Daniel Ritz [Thu, 8 Sep 2005 22:57:14 +0000 (00:57 +0200)]
[PATCH] Update PCI IOMEM allocation start

This fixes the problem with "Averatec 6240 pcmcia_socket0: unable to
apply power", which was due to the CardBus IOMEM register region being
allocated at an address that was actually inside the RAM window that had
been reserved for video frame-buffers in an UMA setup.

The BIOS _should_ have marked that region reserved in the e820 memory
descriptor tables, but did not.

It is fixed by rounding up the default starting address of PCI memory
allocations, so that we leave a bigger gap after the final known memory
location.  The amount of rounding depends on how big the unused memory
gap is that we can allocate IOMEM from.

Based on example code by Linus.

Acked-by: Greg KH <greg@kroah.com>
Acked-by: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/gregkh/pci-2.6
Linus Torvalds [Fri, 9 Sep 2005 22:17:00 +0000 (15:17 -0700)]
Merge /pub/scm/linux/kernel/git/gregkh/pci-2.6

19 years agoFix up lost patch in compat_sys_select() for new RCU files world order
Linus Torvalds [Fri, 9 Sep 2005 22:10:52 +0000 (15:10 -0700)]
Fix up lost patch in compat_sys_select() for new RCU files world order

Andrew lost this in patch reject resolution, and never noticed, since
the compat code isn't in use on x86.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6
Linus Torvalds [Fri, 9 Sep 2005 21:25:22 +0000 (14:25 -0700)]
Merge /pub/scm/linux/kernel/git/davem/net-2.6

19 years ago[PATCH] Lost sockfd_put() in routing_ioctl()
Kirill Korotaev [Fri, 9 Sep 2005 09:59:48 +0000 (13:59 +0400)]
[PATCH] Lost sockfd_put() in routing_ioctl()

This patch adds lost sockfd_put() in 32bit compat rounting_ioctl() on
64bit platforms

Signed-Off-By: Kirill Korotaev <dev@sw.ru>
Signed-Off-By: Maxim Giryaev <gem@sw.ru>
Signed-off-By: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] lost fput in 32bit ioctl on x86-64
Kirill Korotaev [Fri, 9 Sep 2005 20:05:53 +0000 (13:05 -0700)]
[PATCH] lost fput in 32bit ioctl on x86-64

This adds a lost fput in 32bit tiocgdev ioctl on x86-64

[ chrisw: Updated to use fget_light/fput_light ]

Signed-Off-By: Kirill Korotaev <dev@sw.ru>
Signed-Off-By: Maxim Giryaev <gem@sw.ru>
Signed-off-by: Chris Wright <chrisw@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years agokbuild: um fix so it compile with generic asm-offsets.h support
Sam Ravnborg [Fri, 9 Sep 2005 21:10:54 +0000 (23:10 +0200)]
kbuild: um fix so it compile with generic asm-offsets.h support

um has it own set of files for asm-offsets. So for now the
gen-asm-offset macro is just duplicated in the um Makefile.

This may well be the final solution since um is a bit special compared
to other architectures - time will tell.

Also added a dummy arch/um/kernel/asm-offsets.h file to keep kbuild happy.

Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
19 years ago[PATCH] trivial iomem annotations in qla2xxx/qla_dbg.c
viro@ZenIV.linux.org.uk [Fri, 9 Sep 2005 20:42:25 +0000 (21:42 +0100)]
[PATCH] trivial iomem annotations in qla2xxx/qla_dbg.c

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] trivial __user annotations (md)
viro@ZenIV.linux.org.uk [Fri, 9 Sep 2005 19:36:43 +0000 (20:36 +0100)]
[PATCH] trivial __user annotations (md)

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] trivial __user annotations (evdev)
viro@ZenIV.linux.org.uk [Fri, 9 Sep 2005 19:29:12 +0000 (20:29 +0100)]
[PATCH] trivial __user annotations (evdev)

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] trivial __user cleanup (video1394)
viro@ZenIV.linux.org.uk [Fri, 9 Sep 2005 19:26:01 +0000 (20:26 +0100)]
[PATCH] trivial __user cleanup (video1394)

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] trivial __user annotations (ipmi)
viro@ZenIV.linux.org.uk [Fri, 9 Sep 2005 19:14:05 +0000 (20:14 +0100)]
[PATCH] trivial __user annotations (ipmi)

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] more SPIN_LOCK_UNLOCKED -> DEFINE_SPINLOCK conversions
Ingo Molnar [Fri, 9 Sep 2005 20:10:41 +0000 (13:10 -0700)]
[PATCH] more SPIN_LOCK_UNLOCKED -> DEFINE_SPINLOCK conversions

This converts the final 20 DEFINE_SPINLOCK holdouts.  (another 580 places
are already using DEFINE_SPINLOCK).  Build tested on x86.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] timer initialization cleanup: DEFINE_TIMER
Ingo Molnar [Fri, 9 Sep 2005 20:10:40 +0000 (13:10 -0700)]
[PATCH] timer initialization cleanup: DEFINE_TIMER

Clean up timer initialization by introducing DEFINE_TIMER a'la
DEFINE_SPINLOCK.  Build and boot-tested on x86.  A similar patch has been
been in the -RT tree for some time.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE: don't allow restarting of system calls
Miklos Szeredi [Fri, 9 Sep 2005 20:10:39 +0000 (13:10 -0700)]
[PATCH] FUSE: don't allow restarting of system calls

This patch removes ability to interrupt and restart operations while there
hasn't been any side-effect.

The reason: applications.  There are some apps it seems that generate
signals at a fast rate.  This means, that if the operation cannot make
enough progress between two signals, it will be restarted for ever.  This
bug actually manifested itself with 'krusader' trying to open a file for
writing under sshfs.  Thanks to Eduard Czimbalmos for the report.

The problem can be solved just by making open() uninterruptible, because in
this case it was the truncate operation that slowed down the progress.  But
it's better to solve this by simply not allowing interrupts at all (except
SIGKILL), because applications don't expect file operations to be
interruptible anyway.  As an added bonus the code is simplified somewhat.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE: add fsync operation for directories
Miklos Szeredi [Fri, 9 Sep 2005 20:10:38 +0000 (13:10 -0700)]
[PATCH] FUSE: add fsync operation for directories

This patch adds a new FSYNCDIR request, which is sent when fsync is called
on directories.  This operation is available in libfuse 2.3-pre1 or
greater.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fuse: don't update file times
Miklos Szeredi [Fri, 9 Sep 2005 20:10:38 +0000 (13:10 -0700)]
[PATCH] fuse: don't update file times

Don't change mtime/ctime/atime to local time on read/write.  Rather invalidate
file attributes, so next stat() will force a GETATTR call.  Bug reported by
Ben Grimm.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fuse: more flexible caching
Miklos Szeredi [Fri, 9 Sep 2005 20:10:37 +0000 (13:10 -0700)]
[PATCH] fuse: more flexible caching

Make data caching behavior selectable on a per-open basis instead of
per-mount.  Compatibility for the old mount options 'kernel_cache' and
'direct_io' is retained in the userspace library (version 2.4.0-pre1 or
later).

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fuse: transfer readdir data through device
Miklos Szeredi [Fri, 9 Sep 2005 20:10:36 +0000 (13:10 -0700)]
[PATCH] fuse: transfer readdir data through device

This patch removes a long lasting "hack" in FUSE, which used a separate
channel (a file descriptor refering to a disk-file) to transfer directory
contents from userspace to the kernel.

The patch adds three new operations (OPENDIR, READDIR, RELEASEDIR), which
have semantics and implementation exactly maching the respective file
operations (OPEN, READ, RELEASE).

This simplifies the directory reading code.  Also disk space is not
necessary, which can be important in embedded systems.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE - direct I/O
Miklos Szeredi [Fri, 9 Sep 2005 20:10:35 +0000 (13:10 -0700)]
[PATCH] FUSE - direct I/O

This patch adds support for the "direct_io" mount option of FUSE.

When this mount option is specified, the page cache is bypassed for
read and write operations.  This is useful for example, if the
filesystem doesn't know the size of files before reading them, or when
any kind of caching is harmful.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fuse: stricter mount option checking
Miklos Szeredi [Fri, 9 Sep 2005 20:10:34 +0000 (13:10 -0700)]
[PATCH] fuse: stricter mount option checking

Check for the presence of all mandatory mount options.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE: tighten check for processes allowed access
Miklos Szeredi [Fri, 9 Sep 2005 20:10:34 +0000 (13:10 -0700)]
[PATCH] FUSE: tighten check for processes allowed access

This patch tightens the check for allowing processes to access non-privileged
mounts.  The rational is that the filesystem implementation can control the
behavior or get otherwise unavailable information of the filesystem user.  If
the filesystem user process has the same uid, gid, and is not suid or sgid
application, then access is safe.  Otherwise access is not allowed unless the
"allow_other" mount option is given (for which policy is controlled by the
userspace mount utility).

Thanks to everyone linux-fsdevel, especially Martin Mares who helped uncover
problems with the previous approach.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE - readpages operation
Miklos Szeredi [Fri, 9 Sep 2005 20:10:33 +0000 (13:10 -0700)]
[PATCH] FUSE - readpages operation

This patch adds readpages support to FUSE.

With the help of the readpages() operation multiple reads are bundled
together and sent as a single request to userspace.  This can improve
reading performace.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE: add padding
Miklos Szeredi [Fri, 9 Sep 2005 20:10:32 +0000 (13:10 -0700)]
[PATCH] FUSE: add padding

Add padding to structures to make sizes the same on 32bit and 64bit archs.
Initial testing and test machine generously provided by Franco Broi.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE - extended attribute operations
Miklos Szeredi [Fri, 9 Sep 2005 20:10:31 +0000 (13:10 -0700)]
[PATCH] FUSE - extended attribute operations

This patch adds the extended attribute operations to FUSE.

The following operations are added:

 o getxattr
 o setxattr
 o listxattr
 o removexattr

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE - mount options
Miklos Szeredi [Fri, 9 Sep 2005 20:10:31 +0000 (13:10 -0700)]
[PATCH] FUSE - mount options

This patch adds miscellaneous mount options to the FUSE filesystem.

The following mount options are added:

 o default_permissions:  check permissions with generic_permission()
 o allow_other:          allow other users to access files
 o allow_root:           allow root to access files
 o kernel_cache:         don't invalidate page cache on open

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE - file operations
Miklos Szeredi [Fri, 9 Sep 2005 20:10:30 +0000 (13:10 -0700)]
[PATCH] FUSE - file operations

This patch adds the file operations of FUSE.

The following operations are added:

 o open
 o flush
 o release
 o fsync
 o readpage
 o commit_write

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE - read-write operations
Miklos Szeredi [Fri, 9 Sep 2005 20:10:29 +0000 (13:10 -0700)]
[PATCH] FUSE - read-write operations

This patch adds the write filesystem operations of FUSE.

The following operations are added:

 o setattr
 o symlink
 o mknod
 o mkdir
 o create
 o unlink
 o rmdir
 o rename
 o link

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE - read-only operations
Miklos Szeredi [Fri, 9 Sep 2005 20:10:28 +0000 (13:10 -0700)]
[PATCH] FUSE - read-only operations

This patch adds the read-only filesystem operations of FUSE.

This contains the following files:

 o dir.c
    - directory, symlink and file-inode operations

The following operations are added:

 o lookup
 o getattr
 o readlink
 o follow_link
 o directory open
 o readdir
 o directory release
 o permission
 o dentry revalidate
 o statfs

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE - device functions
Miklos Szeredi [Fri, 9 Sep 2005 20:10:27 +0000 (13:10 -0700)]
[PATCH] FUSE - device functions

This adds the FUSE device handling functions.

This contains the following files:

 o dev.c
    - fuse device operations (read, write, release, poll)
    - registers misc device
    - support for sending requests to userspace

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE - core
Miklos Szeredi [Fri, 9 Sep 2005 20:10:26 +0000 (13:10 -0700)]
[PATCH] FUSE - core

This patch adds FUSE core.

This contains the following files:

 o inode.c
    - superblock operations (alloc_inode, destroy_inode, read_inode,
      clear_inode, put_super, show_options)
    - registers FUSE filesystem

 o fuse_i.h
    - private header file

Requirements
============

 The most important difference between orinary filesystems and FUSE is
 the fact, that the filesystem data/metadata is provided by a userspace
 process run with the privileges of the mount "owner" instead of the
 kernel, or some remote entity usually running with elevated
 privileges.

 The security implication of this is that a non-privileged user must
 not be able to use this capability to compromise the system.  Obvious
 requirements arising from this are:

  - mount owner should not be able to get elevated privileges with the
    help of the mounted filesystem

  - mount owner should not be able to induce undesired behavior in
    other users' or the super user's processes

  - mount owner should not get illegitimate access to information from
    other users' and the super user's processes

 These are currently ensured with the following constraints:

  1) mount is only allowed to directory or file which the mount owner
    can modify without limitation (write access + no sticky bit for
    directories)

  2) nosuid,nodev mount options are forced

  3) any process running with fsuid different from the owner is denied
     all access to the filesystem

 1) and 2) are ensured by the "fusermount" mount utility which is a
    setuid root application doing the actual mount operation.

 3) is ensured by a check in the permission() method in kernel

 I started thinking about doing 3) in a different way because Christoph
 H. made a big deal out of it, saying that FUSE is unacceptable into
 mainline in this form.

 The suggested use of private namespaces would be OK, but in their
 current form have many limitations that make their use impractical (as
 discussed in this thread).

 Suggested improvements that would address these limitations:

   - implement shared subtrees

   - allow a process to join an existing namespace (make namespaces
     first-class objects)

   - implement the namespace creation/joining in a PAM module

 With all that in place the check of owner against current->fsuid may
 be removed from the FUSE kernel module, without compromising the
 security requirements.

 Suid programs still interesting questions, since they get access even
 to the private namespace causing some information leak (exact
 order/timing of filesystem operations performed), giving some
 ptrace-like capabilities to unprivileged users.  BTW this problem is
 not strictly limited to the namespace approach, since suid programs
 setting fsuid and accessing users' files will succeed with the current
 approach too.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] FUSE - MAINTAINERS, Kconfig and Makefile changes
Miklos Szeredi [Fri, 9 Sep 2005 20:10:22 +0000 (13:10 -0700)]
[PATCH] FUSE - MAINTAINERS, Kconfig and Makefile changes

This patch adds FUSE filesystem to MAINTAINERS, fs/Kconfig and
fs/Makefile.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] Documentation/sparse snapshot URL
Ben Dooks [Fri, 9 Sep 2005 20:10:20 +0000 (13:10 -0700)]
[PATCH] Documentation/sparse snapshot URL

The URL for Documentation/sparse is wrong now that it is in git.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] VFS: update documentation
Pekka J Enberg [Fri, 9 Sep 2005 20:10:19 +0000 (13:10 -0700)]
[PATCH] VFS: update documentation

This patch brings the now out-of-date Documentation/filesystems/vfs.txt
back to life.  Thanks to Carsten Otte, Trond Myklebust, and Anton
Altaparmakov for their help on updating this documentation.

Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] Kdump: Documentation Update
Vivek Goyal [Fri, 9 Sep 2005 20:10:19 +0000 (13:10 -0700)]
[PATCH] Kdump: Documentation Update

There are minor changes in command line options in kexec-tools for kdump.
This patch updates the documentation to reflect those changes.

Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] DocBook: fix kernel-api documentation generation
Martin Waitz [Fri, 9 Sep 2005 20:10:17 +0000 (13:10 -0700)]
[PATCH] DocBook: fix kernel-api documentation generation

This patch changes a macro definition so that kernel-doc can understand it.

Signed-off-by: Martin Waitz <tali@admingilde.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] update kfree, vfree, and vunmap kerneldoc
Pekka Enberg [Fri, 9 Sep 2005 20:10:16 +0000 (13:10 -0700)]
[PATCH] update kfree, vfree, and vunmap kerneldoc

This patch clarifies NULL handling of kfree() and vfree().  I addition,
wording of calling context restriction for vfree() and vunmap() are changed
from "may not" to "must not."

Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Acked-by: Manfred Spraul <manfred@colorfullife.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] docs: fix misinformation about overcommit_memory
Chuck Ebbert [Fri, 9 Sep 2005 20:10:15 +0000 (13:10 -0700)]
[PATCH] docs: fix misinformation about overcommit_memory

Someone complained about the docs for vm_overcommit_memory being wrong.
This patch copies the text from the vm documentation into procfs.

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] ISA DMA API documentation
Pierre Ossman [Fri, 9 Sep 2005 20:10:13 +0000 (13:10 -0700)]
[PATCH] ISA DMA API documentation

Documentation for how the ISA DMA controller is handled in the kernel.

Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] Documentation: how to apply patches for various trees
Jesper Juhl [Fri, 9 Sep 2005 20:10:12 +0000 (13:10 -0700)]
[PATCH] Documentation: how to apply patches for various trees

Add a new document describing the major kernel trees and how to apply their
patches.

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] Update Documentation/DocBook/kernel-hacking.tmpl
Rusty Russell [Fri, 9 Sep 2005 20:10:11 +0000 (13:10 -0700)]
[PATCH] Update Documentation/DocBook/kernel-hacking.tmpl

Update the hacking guide, before CONFIG_PREEMPT_RT goes in and it needs
rewriting again.

Changes include modernization of quotes, removal of most references to
bottom halves (some mention required because we still use bh in places to
mean softirq).

It would be nice to have a discussion of sparse and various annotations.
Please send patches straight to akpm.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (authored)
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] s3c2410fb: Platform support for ARM S3C2410 framebuffer driver
Arnaud Patard [Fri, 9 Sep 2005 20:10:10 +0000 (13:10 -0700)]
[PATCH] s3c2410fb: Platform support for ARM S3C2410 framebuffer driver

This patch add the plateform specific stuff needed to configure and use the
driver.

Signed-Off-By: Arnaud Patard <arnaud.patard@rtp-net.org>
Signed-off-by: Antonino Daplas <adaplas@pol.net>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Ben Dooks <ben@trinity.fluff.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] s3c2410fb: ARM S3C2410 framebuffer driver
Arnaud Patard [Fri, 9 Sep 2005 20:10:07 +0000 (13:10 -0700)]
[PATCH] s3c2410fb: ARM S3C2410 framebuffer driver

This set of two patches add support for the framebuffer of the Samsung S3C2410
ARM SoC.  This driver was started about one year ago and is now used on iPAQ
h1930/h1940, Acer n30 and probably other s3c2410-based machines I'm not aware
of.  I've also heard yesterday that it's working also on iPAQ rx3715/rx3115
(s3c2440-based machines).

Signed-Off-By: Arnaud Patard <arnaud.patard@rtp-net.org>
Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Ben Dooks <ben@trinity.fluff.org>
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] quiet non-x86 option ROM warnings
Olaf Hering [Fri, 9 Sep 2005 20:10:06 +0000 (13:10 -0700)]
[PATCH] quiet non-x86 option ROM warnings

Quiet an incorrect warning in aty128fb and radeonfb about the PCI ROM
content.  Macs work just find without that signature.

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] i810fb: Stop LCD displays from flickering
Antonino A. Daplas [Fri, 9 Sep 2005 20:10:05 +0000 (13:10 -0700)]
[PATCH] i810fb: Stop LCD displays from flickering

Stop LCD displays from flickering during high loads.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] i810fb: Add i2c/DDC support
Antonino A. Daplas [Fri, 9 Sep 2005 20:10:04 +0000 (13:10 -0700)]
[PATCH] i810fb: Add i2c/DDC support

Add ddc/i2c support for i810fb.  This will allow the driver to get display
information, especially for monitors with fickle timings.  The i2c support
depends on CONFIG_FB_I810_GTF.

Changed __init* to __devinit*

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Alexander Nyberg <alexn@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fbcon: Break up bit_putcs into its component functions
Antonino A. Daplas [Fri, 9 Sep 2005 20:10:04 +0000 (13:10 -0700)]
[PATCH] fbcon: Break up bit_putcs into its component functions

The function bit_putcs() in drivers/video/console/bitblit.c is becoming large.
 Break it up into its component functions (bit_putcs_unaligned and
bit_putcs_aligned).

Incorporated fb_pad_aligned_buffer() optimization by Roman Zippel.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] pxafb: Add hsync time reporting hook
Richard Purdie [Fri, 9 Sep 2005 20:10:03 +0000 (13:10 -0700)]
[PATCH] pxafb: Add hsync time reporting hook

To solve touchscreen interference problems devices like the Sharp Zaurus
SL-C3000 need to know the length of the horitzontal sync pulses.  This patch
adds a hook to pxafb so the touchscreen driver can function correctly.

Signed-Off-By: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fbdev: Initialize var structure in calc_mode_timings
Antonino A. Daplas [Fri, 9 Sep 2005 20:10:02 +0000 (13:10 -0700)]
[PATCH] fbdev: Initialize var structure in calc_mode_timings

The var structure in calc_mode_timings is not properly initialized (zero set)
which leads to undefined behavior when it is passed to fb_get_mode().

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] nvidiafb: Fixed mirrored characters in big endian machines
Antonino A. Daplas [Fri, 9 Sep 2005 20:10:01 +0000 (13:10 -0700)]
[PATCH] nvidiafb: Fixed mirrored characters in big endian machines

nvidiafb_imageblit converts the bitdata stream from big_endian to little
endian.  This produces mirrored characters when machine is big_endian.  Do not
endian convert on big endian machines.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fbcon: Stop cursor timer if console is inactive
Antonino A. Daplas [Fri, 9 Sep 2005 20:10:00 +0000 (13:10 -0700)]
[PATCH] fbcon: Stop cursor timer if console is inactive

If console is blanked or in KD_GRAPHICS mode, delete cursor timer.  No sense
flashing the cursor when there's nothing to be seen.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] savagefb: Make mode_option available when compiled as a module
Antonino A. Daplas [Fri, 9 Sep 2005 20:09:59 +0000 (13:09 -0700)]
[PATCH] savagefb: Make mode_option available when compiled as a module

Make "mode_option" available when compiled as a module.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] nvidiafb: Use CVT to get mode for digital displays
Antonino A. Daplas [Fri, 9 Sep 2005 20:09:59 +0000 (13:09 -0700)]
[PATCH] nvidiafb: Use CVT to get mode for digital displays

If no EDID block is probed, if the display is digital and if no mode option is
specified by the user, get the timings by CVT instead of using the global mode
database.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fbdev: Add VESA Coordinated Video Timings (CVT) support
Antonino A. Daplas [Fri, 9 Sep 2005 20:09:58 +0000 (13:09 -0700)]
[PATCH] fbdev: Add VESA Coordinated Video Timings (CVT) support

The Coordinated Video Timings (CVT) is the latest standard approved by VESA
concerning video timings generation.  It addresses the limitation of GTF which
is designed mainly for CRT displays.  CRT's have a high blanking requirement
(as much as 25% of the horizontal frame length) which artificially increases
the pixelclock.  Digital displays, on the other hand, needs to conserve the
pixelclock as much as possible.  The GTF also does not take into account the
different aspect ratios in its calculation.

The new function added is fb_find_mode_cvt().  It is called by fb_find_mode()
if it recognizes a mode option string formatted for CVT.  The format is:

<xres>x<yres>[M][R][-<bpp>][<at-sign><refresh>][i][m]

The 'M' tells the function to calculate using CVT.  On it's own, it will
compute a timing for CRT displays at 60Hz.  If the 'R' is specified, 'reduced
blanking' computation will be used, best for flatpanels.  The 'i' and the 'm'
is for 'interlaced mode' and 'with margins' respectively.

To determine if CVT was used, check for dmesg for something like this:

CVT Mode - <pix>M<n>[-R], ie: .480M3-R  (800x600 reduced blanking)

where: pix - product of xres and yres, in MB
    M   - is a CVT mode
    n   - the aspect ratio (3 - 4:3; 4 - 5:4; 9 - 16:9, 15:9; A - 16:10)
    -R   - reduced blanking

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] PCI: Small rearrangement of PCI probing code
Paul Mackerras [Mon, 5 Sep 2005 23:31:03 +0000 (09:31 +1000)]
[PATCH] PCI: Small rearrangement of PCI probing code

This patch makes some small rearrangements of the PCI probing code in
order to make it possible for arch code to set up the PCI tree
without needing to duplicate code from the PCI layer unnecessarily.
PPC64 will use this to set up the PCI tree from the Open Firmware
device tree, which we need to do on logically-partitioned pSeries
systems.

Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
19 years ago[PATCH] PCI: Unhide SMBus on Compaq Evo N620c
Rumen Ivanov Zarev [Tue, 6 Sep 2005 20:39:32 +0000 (13:39 -0700)]
[PATCH] PCI: Unhide SMBus on Compaq Evo N620c

Trivial patch against 2.6.13 to unhide SMBus on Compaq Evo N620c laptop using
Intel 82855PM chipset.

Signed-off-by: Rumen Zarev <rzarev@caltech.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
19 years ago[PATCH] PCI: Fix PCI bus mastering enable problem in pciehp
Rajesh Shah [Wed, 17 Aug 2005 00:32:04 +0000 (17:32 -0700)]
[PATCH] PCI: Fix PCI bus mastering enable problem in pciehp

Martin Franc reported that the pciehp driver was not enabling bus
master capability on his hot-plugged card. pciehprm_enable_card()
was updating the PCI command register only if _HPP indicated a
value for SERR or PERR that was different from the current setting.
I don't have hardware that reproduces this problem, but Martin
reports that this patch fixes the problem for him.

Signed-off-by: Rajesh Shah <rajesh.shah@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
From rzarev@its.caltech.edu Tue Sep  6 18:29:50 2005
Date: Tue, 6 Sep 2005 13:39:32 -0700 (PDT)
From: Rumen Ivanov Zarev <rzarev@its.caltech.edu>
Message-Id: <200509062039.j86KdWMr014934@inky.its.caltech.edu>
To: gregkh@suse.de
Subject: PCI: Unhide SMBus on Compaq Evo N620c
Cc: linux-kernel@vger.kernel.org
Trivial patch against 2.6.13 to unhide SMBus on Compaq Evo N620c laptop using
Intel 82855PM chipset.

Signed-off-by: Rumen Zarev <rzarev@caltech.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/pci/quirks.c |    6 ++++++
 1 file changed, 6 insertions(+)

--- gregkh-2.6.orig/drivers/pci/quirks.c 2005-09-09 10:28:55.000000000 -0700
+++ gregkh-2.6/drivers/pci/quirks.c 2005-09-09 13:51:44.000000000 -0700
@@ -876,6 +876,12 @@ static void __init asus_hides_smbus_host
                        case 0xC00C: /* Samsung P35 notebook */
                                asus_hides_smbus = 1;
                        }
+ } else if (unlikely(dev->subsystem_vendor == PCI_VENDOR_ID_COMPAQ)) {
+ if (dev->device == PCI_DEVICE_ID_INTEL_82855PM_HB)
+ switch(dev->subsystem_device) {
+ case 0x0058: /* Compaq Evo N620c */
+ asus_hides_smbus = 1;
+ }
  }
 }
 DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82845_HB, asus_hides_smbus_hostbridge );

19 years ago[PATCH] radeonfb: Only request resources we need
Daniel Burcaw [Fri, 9 Sep 2005 20:04:59 +0000 (13:04 -0700)]
[PATCH] radeonfb: Only request resources we need

This patch changes radeon to request only resources 0 and 2 instead of all
3.  This works around problems with some setups where BAR 1 (IO BAR) has
not been assigned by the firmware since it's not used on the machine and
the kernel fails to assign something to it due to the card being between a
P2P bridge that was configured without an IO range at all.

This typically fixes radeonfb on some Apple Xserve G5 machines

Signed-off-by: Daniel Burcaw <dburcaw@terrasoftsolutions.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] framebuffer: bit_putcs() optimization for 8x* fonts
Knut Petersen [Fri, 9 Sep 2005 20:04:59 +0000 (13:04 -0700)]
[PATCH] framebuffer: bit_putcs() optimization for 8x* fonts

This trivial patch gives a performance boost to the framebuffer console

Constructing the bitmaps that are given to the bitblit functions of the
framebuffer drivers is time consuming.  Here we avoide a call to the slow
fb_pad_aligned_buffer().  The patch replaces that call with a simple but
much more efficient bytewise copy.

The kernel spends a significant time at this place if you use 8x* fonts.
Every pixel displayed on your screen is prepared here.

Some benchmark results:

Displaying a file of 2000 lines with 160 characters each takes 889 ms
system time using cyblafb on my system (I´m using a 1280x1024 video mode,
resulting in a 160x64 character console)

Displaying the same file with the enclosed patch applied to 2.6.13 only
takes 760 ms system time, saving 129 ms or 14.5%.

Font widths other than 8 are not affected.

The advantage and correctness of this patch should be obvious.

Signed-off-by: Knut Petersen <Knut_Petersen@t-online.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] framebuffer: new driver for cyberblade/i1 graphics core
Knut Petersen [Fri, 9 Sep 2005 20:04:56 +0000 (13:04 -0700)]
[PATCH] framebuffer: new driver for cyberblade/i1 graphics core

This is a framebuffer driver for the Cyberblade/i1 graphics core.

Currently tridenfb claims to support the cyberblade/i1 graphics core.  This
is of very limited truth.  Even vesafb is faster and provides more working
modes and a much better quality of the video signal.  There is a great
number of bugs in tridentfb ...  but most often it is impossible to decide
if these bugs are real bugs or if fixing them for the cyberblade/i1 core
would break support for one of the other supported chips.

Tridentfb seems to be unmaintained,and documentation for most of the
supported chips is not available.  So "fixing" cyberblade/i1 support inside
of tridentfb was not an option, it would have caused numerous
if(CYBERBLADEi1) else ...  cases and would have rendered the code to be
almost unmaintainable.

A first version of this driver was published on 2005-07-31.  A fix for a
bug reported by Jochen Hein was integrated as well as some changes
requested by Antonino A.  Daplas.

A message has been added to tridentfb to inform current users of tridentfb
to switch to cyblafb if the cyberblade/i1 graphics core is detected.

This patch is one logical change, but because of the included documentation
it is bigger than 70kb.  Therefore it is not sent to lkml and
linux-fbdev-devel,

Signed-off-by: Knut Petersen <Knut_Petersen@t-online.de>
Cc: Muli Ben-Yehuda <mulix@mulix.org>
Acked-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] better error handing in savagefb_probe
Olaf Hering [Fri, 9 Sep 2005 20:04:55 +0000 (13:04 -0700)]
[PATCH] better error handing in savagefb_probe

err remains uninitialized of pci_request_regions fails.

Found by Thorsten Kukuk, I added a few more checks.

Signed-off-by: Olaf Hering <olh@suse.de>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] sisfb update
Thomas Winischhofer [Fri, 9 Sep 2005 20:04:45 +0000 (13:04 -0700)]
[PATCH] sisfb update

This lifts sisfb from version 1.7.17 to version 1.8.9. Changes include:

- Added support for XGI V3XT, V5, V8, Z7 chipsets, including POSTing of
  all of these chipsets.

- Added support for latest SiS chipsets (761).

- Added support for SiS76x memory "hybrid" mode.

- Added support for new LCD resolutions (eg 1280x854, 856x480).

- Fixed support for 320x240 STN panels (for embedded devices).

- Fixed many HDTV modes (525p, 750p, 1080i).

- Fixed PCI config register reading/writing to use proper kernel
  functions for this purpose.

- Fixed PCI ROM handling to use the kernel's proper functions.

- Removed lots of "typedef"s.

- Removed lots of code which was for X.org/XFree86 only.

- Fixed coding style in many places.

- Removed lots of 2.4 cruft.

- Reduced stack size by unifying two previously separate structs into
  one.

- Added new hooks for memory allocation (for DRM).  Now the driver can
  truly handle multiple cards, including memory management.

- Fixed numerous minor bugs.

Signed-off-by: Thomas Winischhofer <thomas@winischhofer.net>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] matroxfb: read MGA PInS data on PowerPC
Ian Romanick [Fri, 9 Sep 2005 20:04:42 +0000 (13:04 -0700)]
[PATCH] matroxfb: read MGA PInS data on PowerPC

This updates the matroxfb code so that it can find the PInS data embedded
in the BIOS on PowerPC cards.  The process for finding the data is
different on OpenFirmware cards than on x86 cards, and the code for doing
so was missing.

After patching, building, installing, and booting a kernel, you should grep
for "PInS" in /var/log/messages.  You should see two messages in the log:

PInS data found at offset XXXXX
PInS memtype = X

On the GXT135p card I get "31168" and "5".  The first value is irrelevant,
but it's presence lets me know that the PInS data was actually found.  On a
GXT130p, the second value should be 3.  Since I don't have access to that
hardware, if someone can verify that, I will submit a follow-on patch that
rips out all the memtype parameter stuff.

Signed-off-by: Ian Romanick <idr@us.ibm.com>
Signed-off-by: Petr Vandrovec <vandrove@vc.cvut.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] radeonfb_old: Fix broken link
Antonino A. Daplas [Fri, 9 Sep 2005 20:04:41 +0000 (13:04 -0700)]
[PATCH] radeonfb_old: Fix broken link

The link for ATI's product page in drivers/video/Kconfig for FB_RADEON is
broken.  Replace with a product comparison page.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] atyfb: Remove code that sets sync polarity unconditionally
Alexander Kern [Fri, 9 Sep 2005 20:04:40 +0000 (13:04 -0700)]
[PATCH] atyfb: Remove code that sets sync polarity unconditionally

Currently, atyfb has code that sets the hsync and vsync polarity based on the
current video mode, without letting the user override the settings.

Remove this particular code.  The sync polarities are set by the PROM, the
user or by the videomode.

Signed-off-by: Alexander Kern <alex.kern@gmx.de>
Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] console: Fix buffer copy on vc resize
Antonino A. Daplas [Fri, 9 Sep 2005 20:04:39 +0000 (13:04 -0700)]
[PATCH] console: Fix buffer copy on vc resize

On a vc resize, the contents of the old screen buffer are transferred to the
new screenbuffer.  If the new screenbuffer is smaller than the old one, only
the contents from the bottom are copied to new.  If the contents of the old
buffer are located at the top, then the contents will not be copied to the new
buffer resulting in a blank screen.

This bug will happen only if the vc in question is not in the foreground.
Doing an fbset -a or con2fbmap will trigger this bug.

To fix this problem, base the start of the copy from the location of the
current cursor.  If the cursor is near the top of the buffer, copy the
contents at the top, and if the cursor is near the bottom of the buffer, then
copy the contents at the bottom.  In the unlikely case where the new row size
is greater than 2x smaller than the old one, and the cursor is in the middle,
copy 1/2 screenful from the top and bottom of the cursor position.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fbcon: Saner 16-color to 4-color conversion
Antonino A. Daplas [Fri, 9 Sep 2005 20:04:38 +0000 (13:04 -0700)]
[PATCH] fbcon: Saner 16-color to 4-color conversion

Currently, the default linux 16-colors are converted to 4-colors by simply
dividing the values by 4.  However, this is not necessarily correct since the
first 4 colors are converted to black, rendering them invisible.

So, for black, no conversion; for light colors, convert to gray, for normal
text color, no conversion, and for bright colors, convert to intense white.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fbdev: Fix greater than 1 bit monochrome color handling
Antonino A. Daplas [Fri, 9 Sep 2005 20:04:37 +0000 (13:04 -0700)]
[PATCH] fbdev: Fix greater than 1 bit monochrome color handling

Currently, fbcon assumes that the visual FB_VISUAL_MONO* is always 1 bit.
According to Geert, there are old hardware where it's possible to have
monochrome at 8-bit, but has only 2 colors, black - 0x00 and white - 0xff.
Fix color handlers (fb_get_color_depth, and get_color) for this special case.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] nvidiafb: Fallback to firmware EDID
Antonino A. Daplas [Fri, 9 Sep 2005 20:04:36 +0000 (13:04 -0700)]
[PATCH] nvidiafb: Fallback to firmware EDID

If nvidiafb fails to probe the EDID block, get the EDID from the BIOS.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] savagefb: Driver updates
Antonino A. Daplas [Fri, 9 Sep 2005 20:04:35 +0000 (13:04 -0700)]
[PATCH] savagefb: Driver updates

- Fallback to firmware EDID if chipset has no DDC/I2C support or if I2C
  probing failed

- Add fb_blank hook

- Fix savagefb_suspend/resume to enable driver to successfully suspend and
  resume from S3, memory or disk

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fbdev: Resurrect hooks to get EDID from firmware
Antonino A. Daplas [Fri, 9 Sep 2005 20:04:34 +0000 (13:04 -0700)]
[PATCH] fbdev: Resurrect hooks to get EDID from firmware

For the i386, code is already present in video.S that gets the EDID from the
video BIOS.  Make this visible so drivers can also use this data as fallback
when i2c does not work.

To ensure that the EDID block is returned for the primary graphics adapter
only, by check if the IORESOURCE_ROM_SHADOW flag is set.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fbdev: geode updates
David Vrabel [Fri, 9 Sep 2005 20:04:32 +0000 (13:04 -0700)]
[PATCH] fbdev: geode updates

Geode framebuffer driver updates:

- Local mode list (taken from modedb.c) containing only relevant modes.
  This also makes the driver work as a module.

- Make it a PCI driver (from James Simmons <jsimmons@infradead.org>).

- A few other minor cosmetic bits and pieces.

Signed-off-by: David Vrabel <dvrabel@arcom.com>
Signed-off-by: James Simmons <jsimmons@infradead.org>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fbdev: prevent drivers that have hardware cursors from calling software curso...
James Simmons [Fri, 9 Sep 2005 20:04:31 +0000 (13:04 -0700)]
[PATCH] fbdev: prevent drivers that have hardware cursors from calling software cursor code

This patch removes drivers that have hardware cursors from calling the
software cursor code.  Also if the driver sets a no hardware cursor flag
then the driver reports a error it someone attempts to use the cursor.

Signed-off-by: James Simmons <jsimmons@infradead.org>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] vesafb: Add blanking support
Antonino A. Daplas [Fri, 9 Sep 2005 20:04:31 +0000 (13:04 -0700)]
[PATCH] vesafb: Add blanking support

Add rudimentary support by manipulating the VGA registers.  However, not
all vesa modes are VGA compatible, so VGA compatiblity is checked first.
Only 2 levels are supported, powerup and powerdown.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] fbdev: Add fbset -a support
Antonino A. Daplas [Fri, 9 Sep 2005 20:04:29 +0000 (13:04 -0700)]
[PATCH] fbdev: Add fbset -a support

Add capability to fbdev to listen to the FB_ACTIVATE_ALL flag.  If set, it
notifies fbcon that all consoles must be set to the current var.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: fix handling of malformed 9P messages
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:28 +0000 (13:04 -0700)]
[PATCH] v9fs: fix handling of malformed 9P messages

This patch attempts to do a better job of cleaning up after detecting errors
on the transport.  This should also improve error reporting on broken
connections to servers.

Signed-off-by: Latchesar Ionkov <lucho@ionkov.net>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: readlink extended mode check
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:27 +0000 (13:04 -0700)]
[PATCH] v9fs: readlink extended mode check

LANL reported some issues with random crashes during mount of legacy protocol
servers (9P2000 versus 9P2000.u) -- crash was always happening in readlink
(which should never happen in legacy mode).  Added some sanity conditionals to
the get_inode code which should prevent the errors LANL was seeing.  Code
tested benign through regression.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: Fix support for special files (devices, named pipes, etc.)
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:27 +0000 (13:04 -0700)]
[PATCH] v9fs: Fix support for special files (devices, named pipes, etc.)

Fix v9fs special files (block, char devices) support.

Signed-off-by: Latchesar Ionkov <lucho@ionkov.net>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: Clean-up vfs_inode and setattr functions
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:26 +0000 (13:04 -0700)]
[PATCH] v9fs: Clean-up vfs_inode and setattr functions

Cleanup code in v9fs vfs_inode as suggested by Alexey Dobriyan.  Did some
major revamping of the v9fs setattr code to remove unnecessary allocations and
clean up some dead-code.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: Change error magic numbers to defined constants
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:25 +0000 (13:04 -0700)]
[PATCH] v9fs: Change error magic numbers to defined constants

Change magic error numbers to system defined constants in v9fs error.h As
suggested by Jan-Benedict Glaw.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: debug and support routines
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:24 +0000 (13:04 -0700)]
[PATCH] v9fs: debug and support routines

This part of the patch contains debug and other misc routines.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: Support to force umount
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:23 +0000 (13:04 -0700)]
[PATCH] v9fs: Support to force umount

Support for force umount

Signed-off-by: Latchesar Ionkov <lucho@ionkov.net>
Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: transport modules
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:22 +0000 (13:04 -0700)]
[PATCH] v9fs: transport modules

This part of the patch contains transport routines.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Latchesar Ionkov <lucho@ionkov.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: 9P protocol implementation
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:21 +0000 (13:04 -0700)]
[PATCH] v9fs: 9P protocol implementation

This part of the patch contains the 9P protocol functions.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: VFS superblock operations and glue
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:20 +0000 (13:04 -0700)]
[PATCH] v9fs: VFS superblock operations and glue

This part of the patch contains VFS superblock and mapping code.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: VFS inode operations
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:19 +0000 (13:04 -0700)]
[PATCH] v9fs: VFS inode operations

This part of the patch contains the VFS inode interfaces.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: VFS file, dentry, and directory operations
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:18 +0000 (13:04 -0700)]
[PATCH] v9fs: VFS file, dentry, and directory operations

This part of the patch contains the VFS file, dentry & directory interfaces.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] v9fs: Documentation, Makefiles, Configuration
Eric Van Hensbergen [Fri, 9 Sep 2005 20:04:18 +0000 (13:04 -0700)]
[PATCH] v9fs: Documentation, Makefiles, Configuration

OVERVIEW

V9FS is a distributed file system for Linux which provides an
implementation of the Plan 9 resource sharing protocol 9P.  It can be
used to share all sorts of resources: static files, synthetic file servers
(such as /proc or /sys), devices, and application file servers (such as
FUSE).

BACKGROUND

Plan 9 (http://plan9.bell-labs.com/plan9) is a research operating
system and associated applications suite developed by the Computing
Science Research Center of AT&T Bell Laboratories (now a part of
Lucent Technologies), the same group that developed UNIX , C, and C++.
Plan 9 was initially released in 1993 to universities, and then made
generally available in 1995. Its core operating systems code laid the
foundation for the Inferno Operating System released as a product by
Lucent Bell-Labs in 1997. The Inferno venture was the only commercial
embodiment of Plan 9 and is currently maintained as a product by Vita
Nuova (http://www.vitanuova.com). After updated releases in 2000 and
2002, Plan 9 was open-sourced under the OSI approved Lucent Public
License in 2003.

The Plan 9 project was started by Ken Thompson and Rob Pike in 1985.
Their intent was to explore potential solutions to some of the
shortcomings of UNIX in the face of the widespread use of high-speed
networks to connect machines. In UNIX, networking was an afterthought
and UNIX clusters became little more than a network of stand-alone
systems. Plan 9 was designed from first principles as a seamless
distributed system with integrated secure network resource sharing.
Applications and services were architected in such a way as to allow
for implicit distribution across a cluster of systems. Configuring an
environment to use remote application components or services in place
of their local equivalent could be achieved with a few simple command
line instructions. For the most part, application implementations
operated independent of the location of their actual resources.

Commercial operating systems haven't changed much in the 20 years
since Plan 9 was conceived. Network and distributed systems support is
provided by a patchwork of middle-ware, with an endless number of
packages supplying pieces of the puzzle. Matters are complicated by
the use of different complicated protocols for individual services,
and separate implementations for kernel and application resources.
The V9FS project (http://v9fs.sourceforge.net) is an attempt to bring
Plan 9's unified approach to resource sharing to Linux and other
operating systems via support for the 9P2000 resource sharing
protocol.

V9FS HISTORY

V9FS was originally developed by Ron Minnich and Maya Gokhale at Los
Alamos National Labs (LANL) in 1997.  In November of 2001, Greg Watson
setup a SourceForge project as a public repository for the code which
supported the Linux 2.4 kernel.

About a year ago, I picked up the initial attempt Ron Minnich had
made to provide 2.6 support and got the code integrated into a 2.6.5
kernel.   I then went through a line-for-line re-write attempting to
clean-up the code while more closely following the Linux Kernel style
guidelines.  I co-authored a paper with Ron Minnich on the V9FS Linux
support including performance comparisons to NFSv3 using Bonnie and
PostMark - this paper appeared at the USENIX/FREENIX 2005
conference in April 2005:
( http://www.usenix.org/events/usenix05/tech/freenix/hensbergen.html ).

CALL FOR PARTICIPATION/REQUEST FOR COMMENTS

Our 2.6 kernel support is stabilizing and we'd like to begin pursuing
its integration into the official kernel tree.  We would appreciate any
review, comments, critiques, and additions from this community and are
actively seeking people to join our project and help us produce
something that would be acceptable and useful to the Linux community.

STATUS

The code is reasonably stable, although there are no doubt corner cases
our regression tests haven't discovered yet.  It is in regular use by several
of the developers and has been tested on x86 and PowerPC
(32-bit and 64-bit) in both small and large (LANL cluster) deployments.
Our current regression tests include fsx, bonnie, and postmark.

It was our intention to keep things as simple as possible for this
release -- trying to focus on correctness within the core of the
protocol support versus a rich set of features.  For example: a more
complete security model and cache layer are in the road map, but
excluded from this release.   Additionally, we have removed support for
mmap operations at Al Viro's request.

PERFORMANCE

Detailed performance numbers and analysis are included in the FREENIX
paper, but we show comparable performance to NFSv3 for large file
operations based on the Bonnie benchmark, and superior performance for
many small file operations based on the PostMark benchmark.   Somewhat
preliminary graphs (from the FREENIX paper) are available
(http://v9fs.sourceforge.net/perf/index.html).

RESOURCES

The source code is available in a few different forms:

tarballs: http://v9fs.sf.net
CVSweb: http://cvs.sourceforge.net/viewcvs.py/v9fs/linux-9p/
CVS: :pserver:anonymous@cvs.sourceforge.net:/cvsroot/v9fs/linux-9p
Git: rsync://v9fs.graverobber.org/v9fs (webgit: http://v9fs.graverobber.org)
9P: tcp!v9fs.graverobber.org!6564

The user-level server is available from either the Plan 9 distribution
or from http://v9fs.sf.net
Other support applications are still being developed, but preliminary
version can be downloaded from sourceforge.

Documentation on the protocol has historically been the Plan 9 Man
pages (http://plan9.bell-labs.com/sys/man/5/INDEX.html), but there is
an effort under way to write a more complete Internet-Draft style
specification (http://v9fs.sf.net/rfc).

There are a couple of mailing lists supporting v9fs, but the most used
is v9fs-developer@lists.sourceforge.net -- please direct/cc your
comments there so the other v9fs contibutors can participate in the
conversation.  There is also an IRC channel: irc://freenode.net/#v9fs

This part of the patch contains Documentation, Makefiles, and configuration
file changes.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] files: files locking doc
Dipankar Sarma [Fri, 9 Sep 2005 20:04:15 +0000 (13:04 -0700)]
[PATCH] files: files locking doc

Add documentation describing the new locking scheme for file descriptor table.

Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] files: lock-free fd look-up
Dipankar Sarma [Fri, 9 Sep 2005 20:04:14 +0000 (13:04 -0700)]
[PATCH] files: lock-free fd look-up

With the use of RCU in files structure, the look-up of files using fds can now
be lock-free.  The lookup is protected by rcu_read_lock()/rcu_read_unlock().
This patch changes the readers to use lock-free lookup.

Signed-off-by: Maneesh Soni <maneesh@in.ibm.com>
Signed-off-by: Ravikiran Thirumalai <kiran_th@gmail.com>
Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] files: files struct with RCU
Dipankar Sarma [Fri, 9 Sep 2005 20:04:13 +0000 (13:04 -0700)]
[PATCH] files: files struct with RCU

Patch to eliminate struct files_struct.file_lock spinlock on the reader side
and use rcu refcounting rcuref_xxx api for the f_count refcounter.  The
updates to the fdtable are done by allocating a new fdtable structure and
setting files->fdt to point to the new structure.  The fdtable structure is
protected by RCU thereby allowing lock-free lookup.  For fd arrays/sets that
are vmalloced, we use keventd to free them since RCU callbacks can't sleep.  A
global list of fdtable to be freed is not scalable, so we use a per-cpu list.
If keventd is already handling the current cpu's work, we use a timer to defer
queueing of that work.

Since the last publication, this patch has been re-written to avoid using
explicit memory barriers and use rcu_assign_pointer(), rcu_dereference()
premitives instead.  This required that the fd information is kept in a
separate structure (fdtable) and updated atomically.

Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] files-sparc64-fix 2
Dipankar Sarma [Fri, 9 Sep 2005 20:04:12 +0000 (13:04 -0700)]
[PATCH] files-sparc64-fix 2

Fix sparc64 timod to use the new files_fdtable() api to get the fd table.
This is necessary for RCUification.

Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] files: break up files struct
Dipankar Sarma [Fri, 9 Sep 2005 20:04:10 +0000 (13:04 -0700)]
[PATCH] files: break up files struct

In order for the RCU to work, the file table array, sets and their sizes must
be updated atomically.  Instead of ensuring this through too many memory
barriers, we put the arrays and their sizes in a separate structure.  This
patch takes the first step of putting the file table elements in a separate
structure fdtable that is embedded withing files_struct.  It also changes all
the users to refer to the file table using files_fdtable() macro.  Subsequent
applciation of RCU becomes easier after this.

Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] files: rcuref APIs
Dipankar Sarma [Fri, 9 Sep 2005 20:04:09 +0000 (13:04 -0700)]
[PATCH] files: rcuref APIs

Adds a set of primitives to do reference counting for objects that are looked
up without locks using RCU.

Signed-off-by: Ravikiran Thirumalai <kiran_th@gmail.com>
Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
19 years ago[PATCH] files: fix rcu initializers
Dipankar Sarma [Fri, 9 Sep 2005 20:04:07 +0000 (13:04 -0700)]
[PATCH] files: fix rcu initializers

First of a number of files_lock scaability patches.

 Here are the x86 numbers -

 tiobench on a 4(8)-way (HT) P4 system on ramdisk :

                                         (lockfree)
 Test            2.6.10-vanilla  Stdev   2.6.10-fd       Stdev
 -------------------------------------------------------------
 Seqread         1400.8          11.52   1465.4          34.27
 Randread        1594            8.86    2397.2          29.21
 Seqwrite        242.72          3.47    238.46          6.53
 Randwrite       445.74          9.15    446.4           9.75

 The performance improvement is very significant.
 We are getting killed by the cacheline bouncing of the files_struct
 lock here. Writes on ramdisk (ext2) seems to vary just too
 much to get any meaningful number.

 Also, With Tridge's thread_perf test on a 4(8)-way (HT) P4 xeon system :

 2.6.12-rc5-vanilla :

 Running test 'readwrite' with 8 tasks
 Threads     0.34 +/- 0.01 seconds
 Processes   0.16 +/- 0.00 seconds

 2.6.12-rc5-fd :

 Running test 'readwrite' with 8 tasks
 Threads     0.17 +/- 0.02 seconds
 Processes   0.17 +/- 0.02 seconds

 I repeated the measurements on ramfs (as opposed to ext2 on ramdisk in
 the earlier measurement) and I got more consistent results from tiobench :

 4(8) way xeon P4
 -----------------
                                         (lock-free)
 Test            2.6.12-rc5      Stdev   2.6.12-rc5-fd   Stdev
 -------------------------------------------------------------
 Seqread         1282            18.59   1343.6          26.37
 Randread        1517            7       2415            34.27
 Seqwrite        702.2           5.27    709.46           5.9
 Randwrite       846.86          15.15   919.68          21.4

 4-way ppc64
 ------------
                                         (lock-free)
 Test            2.6.12-rc5      Stdev   2.6.12-rc5-fd   Stdev
 -------------------------------------------------------------
 Seqread         1549            91.16   1569.6          47.2
 Randread        1473.6          25.11   1585.4          69.99
 Seqwrite        1096.8          20.03   1136            29.61
 Randwrite       1189.6           4.04   1275.2          32.96

 Also running Tridge's thread_perf test on ppc64 :

 2.6.12-rc5-vanilla
 --------------------
 Running test 'readwrite' with 4 tasks
 Threads     0.20 +/- 0.02 seconds
 Processes   0.16 +/- 0.01 seconds

 2.6.12-rc5-fd
 --------------------
 Running test 'readwrite' with 4 tasks
 Threads     0.18 +/- 0.04 seconds
 Processes   0.16 +/- 0.01 seconds

 The benefits are huge (upto ~60%) in some cases on x86 primarily
 due to the atomic operations during acquisition of ->file_lock
 and cache line bouncing in fast path. ppc64 benefits are modest
 due to LL/SC based locking, but still statistically significant.

This patch:

RCU head initilizer no longer needs the head varible name since we don't use
list.h lists anymore.

Signed-off-by: Dipankar Sarma <dipankar@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>