into nuts.davemloft.net:/disk1/BK/sparc-2.6

s/smb_/smp_/

Noticed by Werner Almesberger.
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

	* insb et.al. first argument is unsigned long, not void * (just look
at their callers).  There _is_ a variant that takes void __iomem *, but that's
ioread8_rep() and its friends.
	* exported the rest of these suckers (insl/outsl already had been).
Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: David S. Miller <davem@davemloft.net>

Noticed by Bob Breuer.
Signed-off-by: David S. Miller <davem@davemloft.net>

into ppc970.osdl.org:/home/torvalds/v2.6/linux

atomic32.c assumes that arbitrary stuff can be passed into
spin_lock() on non-SMP builds, which is true except for when
spinlock debugging is enabled.
Signed-off-by: David S. Miller <davem@davemloft.net>

into ppc970.osdl.org:/home/torvalds/v2.6/linux

Signed-off-by: David S. Miller <davem@davemloft.net>

smp_mb(), smp_rmp() and smp_wmb() definitions have a trailing semicolon 
and cause compilation errors in single statement context, like 
if-then-else on line 358 in include/linux/skbuff.h. This patch removes 
all three offending semicolons to make it compile.
Signed-off-by: Meelis Roos <mroos@linux.ee>
Acked-by: William Irwin <wli@holomorphy.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Helps with crazy Mac OS-X TCP implementations which delay the
recvmsg() wakeup of the user until push is seen.

Based upon ideas from Alexey Kuznetsov, and a preliminary
patch by Stephen Hemminger.
Signed-off-by: David S. Miller <davem@davemloft.net>

into ppc970.osdl.org:/home/torvalds/v2.6/linux

The nls_cp936.c is not synchronized with MS's translation table, there are
some characters have different code from the code in updated table of MS.
For example, the unicode for character 0xB8A3 (GBK) should be U+798F, but
the code is U+FA1B in nls_cp936.c.

Regenerated the tables.  The number of differences is rather large, so I
did not check every change, but looking at a random sample the changes seem
OK.  Mostly the CJK Compatibility Ideographs have been replaced by their
ordinary equivalents.  Some code points that were unassigned now got a
meaning.
Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

local variable is base, not vbase.
Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Richard Purdie provided a patch to fix support for XScale1 processors (this
is the PMU version i never had access to initially), we weren't clearing
the overflow flags after an overflow interrupt had triggered resulting in
no additional interrupts occuring.  Additionally i've added basic power
management support.
Signed-off-by: Zwane Mwaikambo <zwane@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This effectively enables executable stack and executable heap for all 32bit
programs on x86-64, except if noexec32=on is specified.

This does not support changing this with personality right now, this would
need more intrusive changes.  A 64bit process will always turn it off and a
32bit process turn it on.  

Also readd the noexec32=on option to turn this off and fix a minor bug in
noexec=...  (would be reported as unknown option)
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The linker doesn't complain, but i got this error on ARM which has similar
code.

oprofile_arch_exit: discarded in section `.exit.text' from arch/arm/oprofile/built-in.o
arch/arm/oprofile/built-in.o(.init.text+0x4c): In function `oprofile_init':
: relocation truncated to fit: R_ARM_PC24 oprofile_arch_exit

oprofile_arch_init()
	<error path>
	oprofile_arch_exit()
		__exit nmi_exit()
			__exit exit_driverfs()
Signed-off-by: Zwane Mwaikambo <zwane@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

So we discover that Borland's Kylix application builder emits weird elf
files which describe a non-writeable bss segment.

So remove the clear_user() check at the place where we zero out the bss.  I
don't _think_ there are any security implications here (plus we've never
checked that clear_user() return value, so whoops if it is a problem).
Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The second parameter of the sys_ipc system wrapper on ppc64, s390-64 and
sparc64 is an "int".  sys_shmget gets called with this 32 bit value as the
size parameter.  This limits the maximum shared memory segment on these
three architectures to 2GB.  To fix this the second parameter is declared
as an "unsigned long" and is then casted to the type required by the The
same int vs.  unsigned long bug is fixed for sys_msgsnd and sys_msgrcv as
well.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This fixes an incorrect sign extension in the compat layer that breaks
32bit shmget that are >2GB.  sys_shmget has a signed size_t size argument,
and the int size argument coming from 32bit user space would get sign
extended to 64bit, which is wrong.

I fixed it on all compat architectures, except PPC64 which was already ok. 

It was originally debugged and fixed by Karl Rister @ IBM for SLES9 on x86-64.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Attached patch defines dma_mapping_error on alpha.  Without this
libata-core.c won't compile.
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Another fix to the put_chip() concurrency logic.

Problem was occurring when:

1) one thread was erasing a block in partition x;
2) another thread suspended the erase in order to write to
   partition y;
3) a third thread came along to read a different block from
   partition x and, when it called put_chip(), chip->oldstate was
   FL_ERASING and the erase (mistakenly) resumed;
4) the write in partition y obviously failed at that point.

Incidentally, the fix for this problem also fixed the case where
suspending writes for MTD XIP usage was not working properly.
Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

into ppc970.osdl.org:/home/torvalds/v2.6/linux

into ppc970.osdl.org:/home/torvalds/v2.6/linux

Signed-off-by: Mike Miller <mike.miller@hp.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>

From: Andrew Vasquez <andrew.vasquez@qlogic.com>

replace them with _smp_processor_id()
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>

1. FAN Status/Control: you can now get the fan status (running or not) and
   also set the fan speed (for <5 seconds only). The problem is that there
   is an auto regulator that kicks in within about 5 seconds after that to
   restart the fan if it is above a threshold temperature (39 Degree C in
   my Vaio). It is useful just to get the fan status (primarily). It also
   appears that you can change the speed by increasing the values (much
   like the LCD control) - there are effectively only about 6 speeds (it
   seems - not sure, but from what I've played with on my Vaio).

2. Temperature: you can get the current temperature (same as reported by
   ACPI). This is primarily useful for APM users (since ACPI already gives
   this). I have used this to detect when the fan comes on in my Vaio (39
   Degree C).

From: Narayanan R S <nars@kadamba.org>
Signed-off-by: Stelian Pop <stelian@popies.net>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Fold the contents of sonypi.h into sonypi.c making some structures static.

Partly-from: Adrian Bunk <bunk@stusta.de>
Suggested-by: Dmitry Torokhov <dtor_core@ameritech.net>
Signed-off-by: Stelian Pop <stelian@popies.net>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Use MISC_DYNAMIC_MINOR in miscdevice.minor assignment.

Patch-from: Olaf Hering <olh@suse.de>
Signed-off-by: Stelian Pop <stelian@popies.net>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Add another HELP button event.
Increment the version number.
Signed-off-by: Stelian Pop <stelian@popies.net>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>