14 years ago[SCTP]: Fix sctp_assoc_seq_show() panics on big-endian systems.
Vlad Yasevich [Tue, 17 Jan 2006 19:54:06 +0000]
[SCTP]: Fix sctp_assoc_seq_show() panics on big-endian systems.

This patch corrects the panic by casting the argument to the
pointer of correct size.  On big-endian systems we ended up loading
only 32 bits of data because we are treating the pointer as an int*.
By treating this pointer as loff_t*, we'll load the full 64 bits
and then let regular integer demotion take place which will give us
the correct value.

Signed-off-by: Vlad Yaseivch <vladislav.yasevich@hp.com>
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>

14 years ago[SCTP]: sctp doesn't show all associations/endpoints in /proc
Vlad Yasevich [Tue, 17 Jan 2006 19:53:06 +0000]
[SCTP]: sctp doesn't show all associations/endpoints in /proc

When creating a very large number of associations (and endpoints),
/proc/assocs and /proc/eps will not show all of them.  As a result
netstat will not show all of the either.  This is particularly evident
when creating 1000+ associations (or endpoints).  As an example with
1500 tcp style associations over loopback, netstat showed 1420 on my
system instead of 3000.

The reason for this is that the seq_operations start method is invoked
multiple times bacause of the amount of data that is provided.  The
start method always increments the position parameter and since we use
the position as the hash bucket id, we end up skipping hash buckets.

This patch corrects this situation and get's rid of the silly hash-1
decrement.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>

14 years ago[SCTP]: Fix sctp_cookie alignment in the packet.
Vlad Yasevich [Tue, 17 Jan 2006 19:52:12 +0000]
[SCTP]: Fix sctp_cookie alignment in the packet.

On 64 bit architectures, sctp_cookie sent as part of INIT-ACK is not
aligned on a 64 bit boundry and thus causes unaligned access exceptions.

The layout of the cookie prameter is this:
|<----- Parameter Header --------------------|<--- Cookie DATA --------
-----------------------------------------------------------------------
| param type (16 bits) | param len (16 bits) | sig [32 bytes] | cookie..
-----------------------------------------------------------------------

The cookie data portion contains 64 bit values on 64 bit architechtures
(timeval) that fall on a 32 bit alignment boundry when used as part of
the on-wire format, but align correctly when used in internal
structures.  This patch explicitely pads the on-wire format so that
it is properly aligned.

Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>

14 years ago[SCTP]: Fix potential race condition between sctp_close() and sctp_rcv().
Sridhar Samudrala [Tue, 17 Jan 2006 19:51:28 +0000]
[SCTP]: Fix potential race condition between sctp_close() and sctp_rcv().

Do not release the reference to association/endpoint if an incoming skb is
added to backlog. Instead release it after the chunk is processed in
sctp_backlog_rcv().

Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>

14 years agoLinux v2.6.16-rc1
Linus Torvalds [Tue, 17 Jan 2006 07:44:47 +0000]
Linux v2.6.16-rc1

14 years agoMerge branch 'for-linus' of git://brick.kernel.dk/data/git/linux-2.6-block
Linus Torvalds [Tue, 17 Jan 2006 07:43:11 +0000]
Merge branch 'for-linus' of git://brick.kernel.dk/data/git/linux-2.6-block

14 years ago[PATCH] Remove unused code from rioctrl.c (Last for this batch of work)
Alan Cox [Mon, 16 Jan 2006 17:27:38 +0000]
[PATCH] Remove unused code from rioctrl.c (Last for this batch of work)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove rio_table.c unused code
Alan Cox [Mon, 16 Jan 2006 17:25:42 +0000]
[PATCH] Remove rio_table.c unused code

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove unused code from rio_linux.c
Alan Cox [Mon, 16 Jan 2006 17:24:47 +0000]
[PATCH] Remove unused code from rio_linux.c

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove unused CHECK code from riocmd.c
Alan Cox [Mon, 16 Jan 2006 17:21:25 +0000]
[PATCH] Remove unused CHECK code from riocmd.c

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove unused code from rioroute.h
Alan Cox [Mon, 16 Jan 2006 17:19:52 +0000]
[PATCH] Remove unused code from rioroute.h

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove unused code from rioboot.h
Alan Cox [Mon, 16 Jan 2006 17:18:05 +0000]
[PATCH] Remove unused code from rioboot.h

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove unused code from rioboot
Alan Cox [Mon, 16 Jan 2006 17:14:25 +0000]
[PATCH] Remove unused code from rioboot

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove #if 0 and other long dead code from rio_tty
Alan Cox [Mon, 16 Jan 2006 17:11:50 +0000]
[PATCH] Remove #if 0 and other long dead code from rio_tty

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove long dead #if 0 code from rio_param
Alan Cox [Mon, 16 Jan 2006 17:10:23 +0000]
[PATCH] Remove long dead #if 0 code from rio_param

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove old firmware headers from rio drivers
Alan Cox [Mon, 16 Jan 2006 17:07:54 +0000]
[PATCH] Remove old firmware headers from rio drivers

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove rtahw.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 17:05:55 +0000]
[PATCH] Remove rtahw.h from rio driver (unused file)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove file riscos.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 17:03:49 +0000]
[PATCH] Remove file riscos.h from rio driver (unused file)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove file riowinif.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 17:02:43 +0000]
[PATCH] Remove file riowinif.h from rio driver (unused file)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove riotime.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 17:01:41 +0000]
[PATCH] Remove riotime.h from rio driver (unused file)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove file riolocks.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 16:59:40 +0000]
[PATCH] Remove file riolocks.h from rio driver (unused file)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove proto.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 16:58:38 +0000]
[PATCH] Remove proto.h from rio driver (unused file)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove poll.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 16:57:18 +0000]
[PATCH] Remove poll.h from rio driver (unused file)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove mesg.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 16:56:16 +0000]
[PATCH] Remove mesg.h from rio driver (unused file)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove mca.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 16:54:30 +0000]
[PATCH] Remove mca.h from rio driver (unused file)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove internal firmware building files from rio
Alan Cox [Mon, 16 Jan 2006 16:53:38 +0000]
[PATCH] Remove internal firmware building files from rio

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove hosthw.h from rio (unused file)
Alan Cox [Mon, 16 Jan 2006 16:52:36 +0000]
[PATCH] Remove hosthw.h from rio (unused file)

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove formpkt.h from rio (unused file)
Alan Cox [Mon, 16 Jan 2006 16:51:37 +0000]
[PATCH] Remove formpkt.h from rio (unused file)

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove enable.h from rio (unused file)
Alan Cox [Mon, 16 Jan 2006 16:50:07 +0000]
[PATCH] Remove enable.h from rio (unused file)

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove enable.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 16:49:15 +0000]
[PATCH] Remove enable.h from rio driver (unused file)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove debug.h from rio.h (unused file)
Alan Cox [Mon, 16 Jan 2006 16:48:12 +0000]
[PATCH] Remove debug.h from rio.h (unused file)

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove data.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 16:47:14 +0000]
[PATCH] Remove data.h from rio driver (unused file)

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove chan.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 16:46:17 +0000]
[PATCH] Remove chan.h from rio driver (unused file)

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove cmd.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 16:45:18 +0000]
[PATCH] Remove cmd.h from rio driver (unused file)

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Remove brates.h from rio driver (unused file)
Alan Cox [Mon, 16 Jan 2006 16:44:05 +0000]
[PATCH] Remove brates.h from rio driver (unused file)

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] README updated
Xose Vazquez Perez [Sat, 14 Jan 2006 18:56:28 +0000]
[PATCH] README updated

Replace old information with newer from kernel.org

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] Fix compile warning in bt8xx module
Dmitry Torokhov [Mon, 16 Jan 2006 04:32:53 +0000]
[PATCH] Fix compile warning in bt8xx module

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] kernel/hrtimer.c sparse warning fix
Ingo Molnar [Mon, 16 Jan 2006 09:59:41 +0000]
[PATCH] kernel/hrtimer.c sparse warning fix

fix the following sparse warning:

 kernel/hrtimer.c:665:34: warning: incorrect type in argument 2 (different address spaces)
 kernel/hrtimer.c:665:34:    expected void const *from
 kernel/hrtimer.c:665:34:    got struct timespec [noderef] *<noident><asn:1>
 kernel/hrtimer.c:664:2: warning: dereference of noderef expression

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years agoMerge master.kernel.org:/pub/scm/linux/kernel/git/mchehab/v4l-dvb
Linus Torvalds [Tue, 17 Jan 2006 07:20:01 +0000]
Merge /pub/scm/linux/kernel/git/mchehab/v4l-dvb

14 years ago[PATCH] x86_64: Fix VSMP build
Ravikiran G Thirumalai [Tue, 17 Jan 2006 06:03:47 +0000]
[PATCH] x86_64: Fix VSMP build

Patch fixes a build problem with CONFIG_X86_VSMP.  The vSMP bits probably

gathered some fuzz on its way to mainline, and safe_halt() which was outside
the #endif (CONFIG_X86_VSMP) somehow got inside the !CONFIG_X86_VSMP condition,
hence being undefined and breaking CONFIG_X86_VSMP builds.  Patch takes
safe_halt() and halt() macros out of the #endif

Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: add __meminit for memory hotplug
Matt Tolentino [Tue, 17 Jan 2006 06:03:44 +0000]
[PATCH] x86_64: add __meminit for memory hotplug

Add __meminit to the __init lineup to ensure functions default
to __init when memory hotplug is not enabled.  Replace __devinit
with __meminit on functions that were changed when the memory
hotplug code was introduced.

Signed-off-by: Matt Tolentino <matthew.e.tolentino@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: add x86-64 support for memory hot-add
Matt Tolentino [Tue, 17 Jan 2006 06:03:41 +0000]
[PATCH] x86_64: add x86-64 support for memory hot-add

Add x86-64 specific memory hot-add functions, Kconfig options,
and runtime kernel page table update functions to make
hot-add usable on x86-64 machines.  Also, fixup the nefarious
conditional locking and exports pointed out by Andi.

Tested on Intel and IBM x86-64 memory hot-add capable systems.

Signed-off-by: Matt Tolentino <matthew.e.tolentino@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: Flexmap for 32bit and randomized mappings for 64bit
Andi Kleen [Tue, 17 Jan 2006 06:03:38 +0000]
[PATCH] x86_64: Flexmap for 32bit and randomized mappings for 64bit

Another try at this.

For 32bit follow the 32bit implementation from Ingo -
mappings are growing down from the end of stack now
and vary randomly by 1GB.

Randomized mappings for 64bit just vary the normal mmap break
by 1TB. I didn't bother implementing full flex mmap for 64bit
because it shouldn't be needed there.

Cc: mingo@elte.hu

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: Remove elf32_map in 32bit ELF loader
Andi Kleen [Tue, 17 Jan 2006 06:03:35 +0000]
[PATCH] x86_64: Remove elf32_map in 32bit ELF loader

It's identical to the standard elf_map.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: eliminate empty_bad_{page,{pte,pmd}_table}
Jan Beulich [Tue, 17 Jan 2006 06:03:32 +0000]
[PATCH] x86_64: eliminate empty_bad_{page,{pte,pmd}_table}

... as they are no longer needed. Since there were hard-coded numbers in the
file, the patch also adds a mechanism to avoid these (otherwise potential
future changes would again and again require adjusting these numbers).

Signed-Off-By: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: Update defconfig
Andi Kleen [Tue, 17 Jan 2006 06:03:29 +0000]
[PATCH] x86_64: Update defconfig

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] md: Clear clevel whenever level is set.
NeilBrown [Tue, 17 Jan 2006 06:14:57 +0000]
[PATCH] md: Clear clevel whenever level is set.

The 'level' of an md array can be set as either a number of a string.  When
one is set, the other must be marked 'undefined'.  This wasn't being done
in one place: where new arrays are created.

Result: if md1 is a raid1, it is stopped and a raid5 is created there, it
might still appear to be a raid1.

Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] elevator=as back-compatibility
Chuck Ebbert [Tue, 17 Jan 2006 06:14:55 +0000]
[PATCH] elevator=as back-compatibility

As of 2.6.15 you need to use "anticipatory" instead of "as".  Fix that up
so that `elevator=as' still works.

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Jens Axboe <axboe@suse.de>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] cs89x0: credit Dmitry Pervushin
Lennert Buytenhek [Tue, 17 Jan 2006 06:14:53 +0000]
[PATCH] cs89x0: credit Dmitry Pervushin

Credit Dmitry Pervushin for the PNX010X platform support.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: fix bitfield race
Miklos Szeredi [Tue, 17 Jan 2006 06:14:52 +0000]
[PATCH] fuse: fix bitfield race

Fix race in setting bitfields of fuse_conn.  Spotted by Andrew Morton.

The two fields ->connected and ->mounted were always changed with the
fuse_lock held.  But other bitfields in the same structure were changed
without the lock.  In theory this could lead to losing the assignment of
even the ones under lock.  The chosen solution is to change these two
fields to be a full unsigned type.  The other bitfields aren't "important"
enough to warrant the extra complexity of full locking or changing them to
bitops.

For all bitfields document why they are safe wrt. concurrent
assignments.

Also make the initialization of the 'num_waiting' atomic counter explicit.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: update documentation for sysfs
Miklos Szeredi [Tue, 17 Jan 2006 06:14:47 +0000]
[PATCH] fuse: update documentation for sysfs

Add documentation for new attributes in sysfs.  Also describe the filesystem.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: use asynchronous READ requests for readpages
Miklos Szeredi [Tue, 17 Jan 2006 06:14:46 +0000]
[PATCH] fuse: use asynchronous READ requests for readpages

This patch changes fuse_readpages() to send READ requests asynchronously.

This makes it possible for userspace filesystems to utilize the kernel
readahead logic instead of having to implement their own (resulting in double
caching).

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: READ request initialization
Miklos Szeredi [Tue, 17 Jan 2006 06:14:45 +0000]
[PATCH] fuse: READ request initialization

Add a separate function for filling in the READ request.  This will make it
possible to send asynchronous READ requests as well as synchronous ones.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: move INIT handling to inode.c
Miklos Szeredi [Tue, 17 Jan 2006 06:14:44 +0000]
[PATCH] fuse: move INIT handling to inode.c

Now the INIT requests can be completely handled in inode.c and the
fuse_send_init() function need not be global any more.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: add asynchronous request support
Miklos Szeredi [Tue, 17 Jan 2006 06:14:42 +0000]
[PATCH] fuse: add asynchronous request support

Add possibility for requests to run asynchronously and call an 'end' callback
when finished.

With this, the special handling of the INIT and RELEASE requests can be
cleaned up 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>

14 years ago[PATCH] fuse: add connection aborting
Miklos Szeredi [Tue, 17 Jan 2006 06:14:41 +0000]
[PATCH] fuse: add connection aborting

Add ability to abort a filesystem connection.

With the introduction of asynchronous reads, the ability to interrupt any
request is not enough to dissolve deadlocks, since now waiting for the request
completion (page unlocked) is independent of the actual request, so in a
deadlock all threads will be uninterruptible.

The solution is to make it possible to abort all requests, even those
currently undergoing I/O to/from userspace.  The natural interface for this is
'mount -f mountpoint', but that only works as long as the filesystem is
attached.  So also add an 'abort' attribute to the sysfs view of the
connection.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: add number of waiting requests attribute
Miklos Szeredi [Tue, 17 Jan 2006 06:14:38 +0000]
[PATCH] fuse: add number of waiting requests attribute

This patch adds the 'waiting' attribute which indicates how many filesystem
requests are currently waiting to be completed.  A non-zero value without any
filesystem activity indicates a hung or deadlocked filesystem.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: make fuse connection a kobject
Miklos Szeredi [Tue, 17 Jan 2006 06:14:35 +0000]
[PATCH] fuse: make fuse connection a kobject

Kobjectify fuse_conn, and make it visible under /sys/fs/fuse/connections.

Lacking any natural naming, connections are numbered.

This patch doesn't add any attributes, just the infrastructure.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: extend semantics of connected flag
Miklos Szeredi [Tue, 17 Jan 2006 06:14:34 +0000]
[PATCH] fuse: extend semantics of connected flag

The ->connected flag for a fuse_conn object previously only indicated whether
the device file for this connection is currently open or not.

Change it's meaning so that it indicates whether the connection is active or
not: now either umount or device release will clear the flag.

The separate ->mounted flag is still needed for handling background requests.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: introduce list for requests under I/O
Miklos Szeredi [Tue, 17 Jan 2006 06:14:31 +0000]
[PATCH] fuse: introduce list for requests under I/O

Create a new list for requests in the process of being transfered to/from
userspace.  This will be needed to be able to abort all requests even those
currently under I/O

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: introduce unified request state
Miklos Szeredi [Tue, 17 Jan 2006 06:14:31 +0000]
[PATCH] fuse: introduce unified request state

The state of request was made up of 2 bitfields (->sent and ->finished) and of
the fact that the request was on a list or not.

Unify this into a single state field.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: miscellaneous cleanup
Miklos Szeredi [Tue, 17 Jan 2006 06:14:29 +0000]
[PATCH] fuse: miscellaneous cleanup

 - remove some unneeded assignments

 - use kzalloc instead of kmalloc + memset

 - simplify setting sb->s_fs_info

 - in fuse_send_init() use fuse_get_request() instead of
   do_get_request() helper

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: uninline some functions
Miklos Szeredi [Tue, 17 Jan 2006 06:14:28 +0000]
[PATCH] fuse: uninline some functions

Inline keyword is unnecessary in most cases.  Clean them up.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: handle error INIT reply
Miklos Szeredi [Tue, 17 Jan 2006 06:14:27 +0000]
[PATCH] fuse: handle error INIT reply

Handle the case when the INIT request is answered with an error.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: fix request_end()
Miklos Szeredi [Tue, 17 Jan 2006 06:14:26 +0000]
[PATCH] fuse: fix request_end()

This function used the request object after decrementing its reference count
and releasing the lock.  This could in theory lead to all sorts of problems.

Fix and simplify at the same time.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fuse: fuse_copy_finish() order fix
Miklos Szeredi [Tue, 17 Jan 2006 06:14:25 +0000]
[PATCH] fuse: fuse_copy_finish() order fix

fuse_copy_finish() must be called before request_end(), since the later might
sleep, and no sleeping is allowed between fuse_copy_one() and
fuse_copy_finish() because of kmap_atomic()/kunmap_atomic() used in them.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] add /sys/fs
Miklos Szeredi [Tue, 17 Jan 2006 06:14:23 +0000]
[PATCH] add /sys/fs

This patch adds an empty /sys/fs, which filesystems can use.

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] partitions: Read Rio Karma partition table
Bob Copeland [Tue, 17 Jan 2006 06:14:20 +0000]
[PATCH] partitions: Read Rio Karma partition table

The Rio Karma portable MP3 player has its own proprietary partition table.

The partition layout is similar to a DOS boot sector but it begins at a
different offset and uses a different magic number (0xAB56 instead of
0xAA55).  Add support for it to enable mounting the device.

Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] sh: Move CPU subtype configuration to its own Kconfig
Paul Mundt [Tue, 17 Jan 2006 06:14:19 +0000]
[PATCH] sh: Move CPU subtype configuration to its own Kconfig

Currently the CPU subtype options are cluttering up arch/sh/Kconfig somewhat.

Given that, this moves all of that in to its own arch/sh/mm/Kconfig.  Things
like cache configuration are also moved to this new location.

This also adds support for strict CPU tuning on newer cores, which requires
the addition of as-option.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] sh: Simple timer framework
Paul Mundt [Tue, 17 Jan 2006 06:14:18 +0000]
[PATCH] sh: Simple timer framework

This builds on some of the clock framework code to support a simple system
timer interface.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] sh: Simplistic clock framework
Paul Mundt [Tue, 17 Jan 2006 06:14:17 +0000]
[PATCH] sh: Simplistic clock framework

This adds a relatively simplistic clock framework for sh.  The initial goal
behind this is to clean up the arch/sh/kernel/time.c mess and to get the CPU
subtype-specific frequency setting and calculation code moved somewhere more
sensible.

This only deals with the core clocks at the moment, though it's trivial for
other drivers to define their own clocks as desired.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] sh: I/O routine cleanups and ioremap() overhaul
Paul Mundt [Tue, 17 Jan 2006 06:14:15 +0000]
[PATCH] sh: I/O routine cleanups and ioremap() overhaul

This introduces a few changes in the way that the I/O routines are defined on
SH, specifically so that things like the iomap API properly wrap through the
machvec for board-specific quirks.

In addition to this, the old p3_ioremap() work is converted to a more generic
__ioremap() that will map through the PMB if it's available, or fall back on
page tables for everything else.

An alpha-like IO_CONCAT is also added so we can start to clean up the
board-specific io.h mess, which will be handled in board update patches..

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] sh: IRQ handler updates
Paul Mundt [Tue, 17 Jan 2006 06:14:14 +0000]
[PATCH] sh: IRQ handler updates

This moves the various IRQ controller drivers into a new subdirectory, and
also extends the INTC2 IRQ handler to also deal with SH7760 and SH7780
interrupts, rather than just ST-40.

The old CONFIG_SH_GENERIC has also been removed from the IRQ definitions, as
new ports are expected to be based off of CONFIG_SH_UNKNOWN.  Since there are
plenty of incompatible machvecs, CONFIG_SH_GENERIC doesn't make sense anymore.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] sh: kexec() support
kogiidena [Tue, 17 Jan 2006 06:14:10 +0000]
[PATCH] sh: kexec() support

This adds kexec() support for SH.

Signed-off-by: kogiidena <kogiidena@eggplant.ddo.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Cc: <fastboot@lists.osdl.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] sh: DMA updates
Paul Mundt [Tue, 17 Jan 2006 06:14:09 +0000]
[PATCH] sh: DMA updates

This extends the current SH DMA API somewhat to support a proper virtual
channel abstraction, and also works to represent this through the driver model
by giving each DMAC its own platform device.

There's also a few other minor changes to support a few new CPU subtypes, and
make TEI generation for the SH DMAC configurable.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] sh: consolidate hp620/hp680/hp690 targets into hp6xx
Paul Mundt [Tue, 17 Jan 2006 06:14:08 +0000]
[PATCH] sh: consolidate hp620/hp680/hp690 targets into hp6xx

Most of the reasons for keeping these separate before was due to hp690
discontig, and since we have a workaround for that now (abusing some shadow
space so everything is magically contiguous), there's no reason to keep the
targets separate.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fix up sbuslib for new fb_comapt_ioctl prototype
Christoph Hellwig [Tue, 17 Jan 2006 06:14:06 +0000]
[PATCH] fix up sbuslib for new fb_comapt_ioctl prototype

After ages my fb ioctl prototype cleanup finally got in.  Unfortunately the
patch are so old that the sbus compat_ioctl helper didn't exist back then,
so it's not covered.  This patch should fix that issue.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] arm26: s/task_threas_info/task_thread_info/
Alexey Dobriyan [Tue, 17 Jan 2006 06:14:05 +0000]
[PATCH] arm26: s/task_threas_info/task_thread_info/

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Ian Molton <spyro@f2s.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] mips: add pm_power_off
Yoichi Yuasa [Tue, 17 Jan 2006 06:14:01 +0000]
[PATCH] mips: add pm_power_off

Adds pm_power_off() to MIPS.

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] drivers/char/esp.c spinlock fix
Ingo Molnar [Tue, 17 Jan 2006 06:14:00 +0000]
[PATCH] drivers/char/esp.c spinlock fix

There's incorrect spinlock usage in espserial_init(): autoconfig() uses
info->lock before it's initialized.  The fix is to initialize the spinlock
earlier.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] build kernel/intermodule.c only when required
Adrian Bunk [Tue, 17 Jan 2006 06:13:59 +0000]
[PATCH] build kernel/intermodule.c only when required

Build kernel/intermodule.c only when required.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] no longer mark MTD_OBSOLETE_CHIPS as BROKEN
Adrian Bunk [Tue, 17 Jan 2006 06:13:58 +0000]
[PATCH] no longer mark MTD_OBSOLETE_CHIPS as BROKEN

This patch removes the wrong dependency of MTD_OBSOLETE_CHIPS on BROKEN and
marks the non-compiling MTD_AMDSTD and MTD_JEDEC drivers as BROKEN.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] arm26: kernel/irq.c: fix compilation
Alexey Dobriyan [Tue, 17 Jan 2006 06:13:57 +0000]
[PATCH] arm26: kernel/irq.c: fix compilation

It's trying to "continue;" in "if" statement.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: Ian Molton <spyro@f2s.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] arm26: add L1_CACHE_SHIFT
Alexey Dobriyan [Tue, 17 Jan 2006 06:13:56 +0000]
[PATCH] arm26: add L1_CACHE_SHIFT

Fix reiserfs compilation as a side effect =)

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: Ian Molton <spyro@f2s.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] fix arm26 THREAD_SIZE
Adrian Bunk [Tue, 17 Jan 2006 06:13:55 +0000]
[PATCH] fix arm26 THREAD_SIZE

arm26 currently has a 256 kB THREAD_SIZE (sic).

Looking at the comment in the code, this seems to be based on a
misunderstanding.

The comment says:
this needs attention (see kernel/fork.c which gets a nice div by zero if
this is lower than 8*32768

kernel/fork.c does:
  max_threads = mempages / (8 * THREAD_SIZE / PAGE_SIZE)

Therefore, a division by 0 is impossible for all reasonable cases with
THREAD_SIZE >= PAGE_SIZE.

Since the minimum PAGE_SIZE Linux uses on the arm26 architecture is 16k,
PAGE_SIZE should be sufficient for THREAD_SIZE.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Ian Molton <spyro@f2s.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] i386: remove gcc version check for CONFIG_REGPARM
Adrian Bunk [Tue, 17 Jan 2006 06:13:53 +0000]
[PATCH] i386: remove gcc version check for CONFIG_REGPARM

Since we do no longer support any gcc < 3.0, there's no need to check
for it..

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years agoMerge branch 'release' of git://git.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6
Linus Torvalds [Tue, 17 Jan 2006 04:56:49 +0000]
Merge branch 'release' of git://git./linux/kernel/git/aegl/linux-2.6

14 years ago[PATCH] hrtimer comment tweak
Jonathan Corbet [Mon, 16 Jan 2006 22:58:55 +0000]
[PATCH] hrtimer comment tweak

Fix a comment which missed an update cycle somewhere.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years agoMerge git://oss.sgi.com:8090/oss/git/xfs-2.6
Linus Torvalds [Tue, 17 Jan 2006 04:26:19 +0000]
Merge git://oss.sgi.com:8090/oss/git/xfs-2.6

14 years ago[PATCH] Fix drivers/block/ps2esdi.c compile
Jens Axboe [Mon, 16 Jan 2006 23:23:49 +0000]
[PATCH] Fix drivers/block/ps2esdi.c compile

Looks like fallout from the geo stuff.

14 years agoMerge branch 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/scjody/ieee1394
Linus Torvalds [Tue, 17 Jan 2006 04:23:21 +0000]
Merge branch 'for-linus' of /linux/kernel/git/scjody/ieee1394

14 years agoPull perfmon-montecito into release branch
Tony Luck [Tue, 17 Jan 2006 04:02:24 +0000]
Pull perfmon-montecito into release branch

14 years ago[IA64] Cleanup of arch/ia64/sn and include/asm-ia64/sn
Prarit Bhargava [Tue, 17 Jan 2006 03:54:40 +0000]
[IA64] Cleanup of arch/ia64/sn and include/asm-ia64/sn

Replace uintX_t declarations with uX declarations.
Replace intX_t declarations with sX declarations.

Signed-off-by: Prarit Bhargava <prarit@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>

14 years agoActually remove amdtp.[ch], cmp.[ch].
Jody McIntyre [Tue, 17 Jan 2006 00:21:57 +0000]
Actually remove amdtp.[ch], cmp.[ch].

The feature removal was done in 7301c8d3a05dc52d33598364da7c4eb6ab6357eb
but these files were not removed for some reason.

Signed-off-by: Jody McIntyre <scjody@modernduck.com>

14 years ago[IA64] pal cache flush patch
Xu, Anthony [Mon, 9 Jan 2006 02:36:35 +0000]
[IA64] pal cache flush patch

Because PAL spec has changed since 2002, you can goto
http://developer.intel.com/design/itanium/manuals/iiasdmanual.htm to
download new SDM, all PAL calls should be invoked with psr.ic=1, and
it's caller's responsibility to handle possible tlb miss.
Ia64_pal_cache_flush was written according to old spec, it is obsolete,
and this patch has ia64_pal_cache_flush conform to new spec.

Signed-off-by Anthony Xu <anthony.xu@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>

14 years ago[IA64] Stop multiple pci_claim_resource() call for the same resource
Kenji Kaneshige [Mon, 16 Jan 2006 04:45:23 +0000]
[IA64] Stop multiple pci_claim_resource() call for the same resource

This patch fixes the bug that pci_claim_resource() is called multiple
times for the same P2P bridge's resource structures if P2P bridges
require their own PCI I/O resources.

Signed-off-by: Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>

14 years agoMake alloc_page_buffers() initialise buffer_heads using init_buffer(),
Nathan Scott [Mon, 16 Jan 2006 22:02:07 +0000]
Make alloc_page_buffers() initialise buffer_heads using init_buffer(),
like other routines here, to ensure buffers are correctly initialised
with respect to b_private/b_end_io.  Fixes an odd interaction between
XFS and reiserfs.

Signed-off-by: Nathan Scott <nathans@sgi.com>

14 years ago[IA64] Simple memory hot-add for ia64.
Yasunori Goto [Sat, 7 Jan 2006 02:50:38 +0000]
[IA64] Simple memory hot-add for ia64.

First step to memory hotplug for ia64 (add only,
all new memory is added to node 0, does not use
ZONE_EASY_RECLAIM yet).

Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>

14 years agox86-64: fix initrd freeing
Linus Torvalds [Mon, 16 Jan 2006 19:33:09 +0000]
x86-64: fix initrd freeing

The comparison of the initrd start address against "&_end" is
unnecessary and incorrect.  Make it match the x86 code that just
compares the passed-in arguments.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>

14 years ago[PATCH] x86_64: Increase NR_IRQ_VECTORS to 32 * NR_CPUS
Andi Kleen [Mon, 16 Jan 2006 00:56:54 +0000]
[PATCH] x86_64: Increase NR_IRQ_VECTORS to 32 * NR_CPUS

This prevents running out of GSIs on large Unisys ES7000 machines.
Follows i386

Cc: "Protasevich, Natalie" <Natalie.Protasevich@unisys.com>

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>