vma->vm_pgoff is an pfn _offset_ relatif to the begining
of the memory start. The previous code was doing at first:

	vma->vm_pgoff << PAGE_SHIFT

which results into a wrong physical address since some
platforms have a physical mem start that can be different
from 0. After that the previous call __pa() on this
wrong physical address, however __pa() is used to convert
a _virtual_ address into a physical one.

This patch rewrites this convertion. It calculates the
pfn of PAGE_OFFSET which is the pfn of the mem start
then it adds the vma->vm_pgoff to it.

It also uses virt_to_phys() instead of __pa() since the
latter shouldn't be used by drivers.
Signed-off-by: Franck Bui-Huu <fbuihuu@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

* git://git.kernel.org/pub/scm/linux/kernel/git/steve/gfs2-2.6-fixes:
  [GFS2] Update git tree name/location
  [DLM] fix iovec length in recvmsg
  [GFS2] Pass the correct value to kunmap_atomic
  [GFS2] Fix bug where lock not held
  [DLM] Kconfig: don't show an empty DLM menu
  [GFS2] Fix uninitialised variable
  [GFS2] Fix a size calculation error

* master.kernel.org:/pub/scm/linux/kernel/git/jejb/voyager-2.6:
  [VOYAGER] fix up ptregs removal mess
  [VOYAGER] fix up attribute packed specifiers in voyager.h
  [VOYAGER] fix genirq mess

The file based core dump code was broken by pipe changes - a relative
llseek returns the absolute file position on success, not the relative
one, so dump_seek() always failed when invoked with non-zero current
position.

Only success/failure can be tested with relative lseek, we have to trust
kernel that on success we've got right file offset.  With this fix in
place I have finally real core files instead of 1KB fragments...
Signed-off-by: Petr Vandrovec <petr@vandrovec.name>
[ Cleaned it up a bit while here - use SEEK_CUR instead of hardcoding 1 ]
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

* git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6: (27 commits)
  [CIFS] Missing flags2 for DFS
  [CIFS] Workaround incomplete byte length returned by some
  [CIFS] cifs Kconfig: don't select CONNECTOR
  [CIFS] Level 1 QPathInfo needed for proper OS2 support
  [CIFS] fix typo in previous patch
  [CIFS] Fix old DOS time conversion to handle timezone
  [CIFS] Do not need to adjust for Jan/Feb for leap day
  [CIFS] Fix leaps year calculation for years after 2100
  [CIFS] readdir (ffirst) enablement of accurate timestamps from legacy servers
  [CIFS] Fix compiler warning with previous patch
  [CIFS] Fix typo
  [CIFS] Allow for 15 minute TZs (e.g. Nepal) and be more explicit about
  [CIFS] Fix readdir of large directories for backlevel servers
  [CIFS] Allow LANMAN21 support even in both POSIX non-POSIX path
  [CIFS] Make use of newer QFSInfo dependent on capability bit instead of
  [CIFS] Do not send newer QFSInfo to legacy servers which can not support it
  [CIFS] Fix typo in name of new cifs_show_stats
  [CIFS] Rename server time zone field
  [CIFS] Handle legacy servers which return undefined time zone
  [CIFS] CIFS support for /proc/<pid>/mountstats part 1
  ...

Manual conflict resolution in fs/cifs/connect.c

Apparently whoever converted voyager never actually checked that the
patch would compile ...

Remove as much of the pt_regs references as possible and move the
remaining ones into line with what's in x86 generic.
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>

The old style (attribute on each structure entry) never really worked.
Move it to an attribute per structure
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>

The implementation of genirq in x86 completely broke voyager (and
presumably visws).  Since it's plugged into so much of the x86
infrastructure, you can't expect it to work unconverted.

This patch introduces a voyager IRQ handler type and switches voyager
to the genirq infrastructure.
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>

Partly suggested by Igor Mammedov
Signed-off-by: Steve French <sfrench@us.ibm.com>

The plan is to have two trees. One for bug fixes to be sent on a
regular basis (-fixes) and another called -nmw which will contain items
queued for the next merge window (hence the name). The -nmw tree
will come & go according to need.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

The DLM always passes the iovec length as 1, this is wrong when the circular
buffer wraps round.
Signed-Off-By: Patrick Caulfield <pcaulfie@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

Pass kaddr rather than (incorrect) struct page to kunmap_atomic.
Signed-off-by: Russell Cattelan <cattelan@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

The log lock needs to be held when manipulating the counter
for the number of free journal blocks.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

Don't show an empty "Distributed Lock Manager" menu if IP_SCTP=n.

Reported by Dmytro Bagrii in kernel Bugzilla #7268.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: David Teigland <teigland@redhat.com>
Signed-off-by: Patrick Caulfield <pcaulfie@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

This fixes a bug where, in certain cases an uninitialised variable
could cause a dereference of a NULL pointer in gfs2_commit_write().
Also a typo in a comment is fixed at the same time.
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

Fix a size calculation error.
The size was incorrect being computed as a
negative length and then being passed to an
unsigned parameter.

This in turn would cause the allocator to
think it needed enough meta data to store
a gigabyte file for every file created.
Signed-off-by: Russell Cattelan <cattelan@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>

drivers/sbus/char stuff using kernel_execve() needs linux/syscalls.h
now; includes trimmed, while we are at it.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Kernel build breaks with CONFIG_X86_VSMP.  Probably due to some header
file cleanups in 2.6.19-rc1.
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

kernel_execve() fallout
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

servers on small SMB responses
Signed-off-by: Steve French <sfrench@us.ibm.com>

* master.kernel.org:/pub/scm/linux/kernel/git/lethal/sh-2.6:
  sh: SH-4A UBC support
  sh: interrupt exception handling rework
  sh: Default enable R7780RP IRQs.
  sh: Zero-out coherent buffer in consistent_alloc().
  sh: Convert IPR-IRQ to IRQ chip.
  sh: Convert INTC2 IRQ handler to irq_chip.
  sh: Fix pr_debug statements for sh4
  sh: Convert r7780rp IRQ handler to IRQ chip.
  sh: Updates for IRQ handler changes.
  sh: Kill off timer_ops get_frequency().
  sh: First step at generic timeofday support.

Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

`select' is a bit obnoxious: the option keeps on coming back
and it's hard to work out what to do to make it go away again.
The use of `depends on' is preferred (although it has
usability problems too..)
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Steve French <sfrench@us.ibm.com>

* 'for-linus' of git://brick.kernel.dk/data/git/linux-2.6-block:
  [PATCH] block layer: ioprio_best function fix
  [PATCH] ide-cd: fix breakage with internally queued commands
  [PATCH] block layer: elv_iosched_show should get elv_list_lock
  [PATCH] splice: fix pipe_to_file() ->prepare_write() error path
  [PATCH] block layer: elevator_find function cleanup
  [PATCH] elevator: elevator_type member not used

* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6:
  [PKT_SCHED] sch_htb: use rb_first() cleanup
  [RTNETLINK]: Fix use of wrong skb in do_getlink()
  [DECNET]: Fix sfuzz hanging on 2.6.18
  [NET]: Do not memcmp() over pad bytes of struct flowi.
  [NET]: Introduce protocol-specific destructor for time-wait sockets.
  [NET]: Use typesafe inet_twsk() inline function instead of cast.
  [NET]: Use hton{l,s}() for non-initializers.
  [TCP]: Use TCPOLEN_TSTAMP_ALIGNED macro instead of magic number.
  [IPV6]: Seperate sit driver to extra module (addrconf.c changes)
  [IPV6]: Seperate sit driver to extra module
  [NET]: File descriptor loss while receiving SCM_RIGHTS
  [SCTP]: Fix the RX queue size shown in /proc/net/sctp/assocs output.
  [SCTP]: Fix receive buffer accounting.
  SELinux: Bug fix in polidydb_destroy
  IPsec: fix handling of errors for socket policies
  IPsec: correct semantics for SELinux policy matching
  IPsec: propagate security module errors up from flow_cache_lookup
  NetLabel: use SECINITSID_UNLABELED for a base SID
  NetLabel: fix a cache race condition

* master.kernel.org:/pub/scm/linux/kernel/git/davem/sparc-2.6:
  [SPARC32]: Fix sparc32 modpost warnings.
  [SPARC32]: Fix sparc32 modpost warnings with sunzilog
  [SPARC32]: Mark srmmu_nocache_init as __init.
  [SPARC32]: pcic.c needs asm/irq_regs.h

This patch fixes my one line thinko where I was clearing
the vector_irq entries on the wrong cpus.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Currently ioprio_best function first checks wethere aioprio or bioprio equals
IOPRIO_CLASS_NONE (ioprio_valid() macros does that) and if it is so it returns
bioprio/aioprio appropriately. Thus the next four lines, that set aclass/bclass
to IOPRIO_CLASS_BE, if aclass/bclass == IOPRIO_CLASS_NONE, are never executed.

The second problem: if aioprio from class IOPRIO_CLASS_NONE and bioprio from
class IOPRIO_CLASS_IDLE are passed to ioprio_best function, it will return
IOPRIO_CLASS_IDLE. It means that during __make_request we can merge two
requests and set the priority of merged request to IDLE, while one of
the initial requests originates from a process with NONE (default) priority.
So we can get a situation when a process with default ioprio will experience
IO starvation, while there is no process from real-time class in the system.

Just removing ioprio_valid check should correct situation.
Signed-off-by: Vasily Tarasov <vtaras@openvz.org>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

We still need to maintain a private PC style command, since it
isn't completely unified with REQ_TYPE_BLOCK_PC yet.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

elv_iosched_show function iterates other elv_list, hence
elv_list_lock should be got.
Signed-off-by: Vasily Tarasov <vtaras@openvz.org>
Signed-off-by: Vasily Tarasov <jens.axboe@oracle.com>

Don't jump to the unlock+release path, we already did that.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

We can easily produce search through the elevator list
without introducing additional elevator_type variable.
Signed-off-by: Vasily Tarasov <vtaras@openvz.org>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

elevator_type field in elevator_type structure is useless:
it isn't used anywhere in kernel sources.
Signed-off-by: Vasily Tarasov <vtaras@openvz.org>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>

Use rb_first() to get first entry in rb tree.
Signed-off-by: Akinbou Mita <akinobu.mita@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

skb is the netlink query, nskb is the reply message.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

Dave Jones wrote:
> sfuzz         D 724EF62A  2828 28717  28691                     (NOTLB)
>        cd69fe98 00000082 0000012d 724ef62a 0001971a 00000010 00000007 df6d22b0 
>        dfd81080 725bbc5e 0001971a 000cc634 00000001 df6d23bc c140e260 00000202 
>        de1d5ba0 cd69fea0 de1d5ba0 00000000 00000000 de1d5b60 de1d5b8c de1d5ba0 
> Call Trace:
>  [<c05b1708>] lock_sock+0x75/0xa6
>  [<e0b0b604>] dn_getname+0x18/0x5f [decnet]
>  [<c05b083b>] sys_getsockname+0x5c/0xb0
>  [<c05b0b46>] sys_socketcall+0xef/0x261
>  [<c0403f97>] syscall_call+0x7/0xb
> DWARF2 unwinder stuck at syscall_call+0x7/0xb
> 
> I wonder if the plethora of lockdep related changes inadvertantly broke something?

Looks like unbalanced locking.
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>

They are not necessarily initialized to zero by the compiler,
for example when using run-time initializers of automatic
on-stack variables.

Noticed by Eric Dumazet and Patrick McHardy.
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>