This patch changes the syscall entry path to store the
current-frame-mask (CFM) in pt_regs->cr_ifs.  This just takes one
extra instruction (a "dep" to clear the bits other than 0-37) and is
free in terms of cycles.

The advantage of doing this is that it lets ptrace() avoid having to
walk the stack to determine the end of the user-level backing-store of
a process which is in the middle of a system-call.  Since this is what
strace does all the time, this speeds up strace quite a bit (by ~50%).
More importantly, it makes the syscall vs. non-syscall case much more
symmetric, which is always something I wanted.

Note that the change to ivt.S looks big but this is just a rippling
effect of instruction-scheduling to keep syscall latency the same.
All that's really going on there is that instead of storing 0 into
cr_ifs member we store the low 38 bits of ar.pfs.
Signed-off-by: David Mosberger <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>

into intel.com:/data/home/aegl/BK/linux-ia64-release-2.6.10

Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Because reference of fls becomes error.
Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Eliminate useless clobber.
Signed-off-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Kernel startup code specific to the Coldfire 5272 based CANcam board.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Move the ColdFire 5282 specific register definitions into a common
528x definition include. They can be used for all members of the
528x CPU family.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The ColdFire 527x and 528x use the same FEC ethernet register offsets.
Conditionally use the correct structure.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Remove unused include of asm/delay.h from ColdFire 5407 configuration
code.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Move the M5282EVB kernel startup code from the 5282 specific platform
directory to a common 528x platform directory. All ColdFire 582x
CPU cores are the same, just minor peripheral differences. Renaming
the sub-directory to match the proper board name (M5282EVB), not
use the generic MOTOROLA.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Move the ColdFire 5282 specific platform Makefile into the common
528x CPU platform directory.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Move the ColdFire 5282 specific CPU configuration code into the common
528x CPU platform directory. It can be used for all members of the
528x CPU family.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Add kernel startup code for the Freescale 5275EVB board.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Create basic hardware configuration and setup code for the newly
added Freescale 527x family of CPU's (5270/5271/5274/5275).
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Determine the DRAM size dynamically for the MOTOROLA 5206e platform.
We can just read back the SDRAM register configuration bits to 
determine what the RAM size really is.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Removed unused include of asm/delay.h.
Add support for getting boot args from dedicated flash segment
on the SCALES and CANcam boards.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Include linux/delay.h instead of asm/delay.h in ColdFire vector
handling code.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Build the PIT timer code in for 528x and 527x CPU targets.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Makefile to support building the Freescale 527x platform support.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Updated default configuration for the m68knommu architectures.
Adds newly created CPU and clock configurations.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Kernel startup code specific to the Freescale M5271EVB board.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

A big update to the ColdFire FEC ethernet driver.
This has been in the works for a while (and thus tested for
a while too). Fundamentally through there is 3 important
changes:

. support 528x and 527x device families
. allow for multiple FEC devices to be present (for example 2 on the 5274/5)
. support the Kendein 8721 PHY
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Move the PIT timer code from the 5282 specific platform directory
to the common ColdFire platform code directory (5307). Freescale
are using the PIT timer on other devices now (at least the 523x and
527x families), so put it near the other common timer code so they
can all use it easily.
Signed-off-by: Greg Ungerer <gerg@snapgear.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

Sun tells me:

  For many years both Solaris x86 and Linux have used the same fdisk
  partition id of 0x82.  Solaris uses that identifier to allocate disk
  space which is then further subdivided by the Solaris VTOC (Volume
  Table of Contents) and Linux uses that id as the identifier for its
  swap partition.

  The new release of Solaris should be out early next year. Solaris
  will begin using a partition id of 0xbf.  We will continue to support
  the older partition id value of 0x82 for compatibility reasons.

  Solaris's fdisk program has been enhanced to allow system administrators
  to switch back and forth between the new and old identifier without any
  loss of data.

So, this may be useful to some people a few months from now.

Not surprisingly, the ext3 code crashes in the same way
the ext2 code does when dividing by zero.

I have directories full of typo corrections - don't know
whether you want them. Below some wierds.

Added the hp and 5xbsd flavors of ufstype.

Removed barrier - it is not an ext2 option.

Corrected the distinction between kernel-selected defaults and values
read from the filesystem.

Fixed some typos.  Shortened line length to 80.

This implements support for doing a HW synchronization of the CPU
timebases on SMP G5 machines.

 When the proper clock chips are found on i2c, they are used to stop the
timebase clock source during the synchronization process.  This replaces
the software sync algorithm we used so far and provides slightly more
precise results.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

The code in pmac_low_i2c.c is a low level synchronous version of the i2c keywest
driver for use by platform code early during boot or during sleep/wakeup cycles
to communicate with some motherboard chips, typically clock chips.
It wasn't used on g5 until now, which is good because it wasn't 64 bits clean :)

This patch fixes it, and also remove the use of udelay() since it can be used for
synchronizing the HW timebase, and so must operate when it's frozen (and our
implementation of udelay uses that timebase).
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

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

Thanks to Kay Sievers for reporting this. 

Was caused by a change from Venkatesh Pallipadi as seen at:
  http://linux.bkbits.net:8080/linux-2.5/cset@41810e4aGZ0E5bn_hMb4JgIY5u90zASigned-off-by: Greg Kroah-Hartman <greg@kroah.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

EFI partitioning scheme was reading the last reported sector of the
block device to look for the alternate GPT header, before it had
confirmed that it should.  This causes problems for devices with the
following problems:  a) those who misreport their size (typically
off-by-one), and b) those who fail when asked to read a block
outside their range.

This patch moves the test for the Protective Master Boot Record (PMBR)
ahead of the tests for the Primary and Alternate GPT headers.  If the
PMBR is not valid, the disk is assumed to not be a GPT disk.  This can
be overridden with the 'gpt' kernel command line option.  If the
Primary GPT header is not valid, the Alternate GPT header is not
probed automatically unless the 'gpt' kernel command line option is
passed.  If the both the PMBR and Primary GPT header are valid, then
the Alternate GPT header at the end of the disk is probed.

Also re-enables CONFIG_EFI_PARTITION for all architectures.
Signed-off-by: Matt Domsch <Matt_Domsch@dell.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

This fixes a problem introduced in the previous set of driver model
changes that has been seen by a lot of people (most notibly the greater
than 256 pty users, but others might also be hitting this without
realizing it.)

Also add a comment so we don't try to "fix" this again.
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

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

into intel.com:/home/lenb/src/26-latest-dev