Patch from Erich Focht

This adds a hook to rebalance globally across nodes every NODE_BALANCE_RATE
iterations of the rebalancer. This allows us to easily tune on an architecture
specific basis how often we wish to rebalance - machines with higher NUMA
ratios (more expensive off-node access) will want to do this less often.
It's currently set to 100 for NUMA-Q and 10 for other machines. If the
imbalance between nodes is > 125%, we'll rebalance them. The hook for this
is added to the NUMA definition of cpus_to_balance, so again, no impact
on non-NUMA machines.

Patch from Michael Hohnbaum

This adds a hook, sched_balance_exec(), to the exec code, to make it
place the exec'ed task on the least loaded queue. We have less state
to move at exec time than fork time, so this is the cheapest point
to cross-node migrate. Experience in Dynix/PTX and testing on Linux
has confirmed that this is the cheapest time to move tasks between nodes.

It also macro-wraps changes to nr_running, to allow us to keep track of
per-node nr_running as well. Again, no impact on non-NUMA machines.

Patch from Martin J. Bligh

This adds a small hook to the find_busiest_queue routine to allow us to
specify a mask of which CPUs to search over. In the NUMA case, it will
only balance inside the node (much cheaper to search, and stops tasks
from bouncing across nodes, which is very costly). The cpus_to_balance
routine is conditionally defined to ensure no impact to non-NUMA machines.

This is a tiny NUMA scheduler, but it needs the assistance of the second
and third patches in order to spread tasks across nodes.

Another left-over from ancient module code, it was supposed to return
non-zero if the module has a use count, but currently it always
evaluates to 0.

There are a few users of different types:
(1)  ioctl that perform a while(MOD_IN_USE) MOD_DEC_USE_COUNT loop.
     Just rip them out, we now have forced module unloading.
(2)  printk's that moan if the use-count in not zero in the exitfunc.
     Just rip them out, this can't happen.
(3)  if(MOD_IN_USE) MOD_DEC_USE_COUNT constructs in ->close of a few
     serial drivers.  Just remove the conditional, we did a
     MOD_INC_USE_COUNT in ->open.
(4)  This one is interesting:  drivers/sbus/char/display7seg.c uses
     the module use count to track openers.  Replace this with an
     atomic_t.

In addition remove tons of stale comments in network driver that aren't
understandable for anyone who doesn't know ancient Linux module semantics.

x86-64 updates for 2.5.58. Changes only x86-64 specific files.

 - Rewrote module allocation. Lots of bugs fixed. Module loading
   should work now again.
 - Kconfig help fixes from Randy Dunlap
 - Makefile cleanups from Pavel Machek and Sam Ravnborg
 - Assembly cleanups from Pavel
 - defconfig update
 - Better strlen_user/strnlen_user
 - Merge with i386: new ptrace commands, 32bit vsyscall signal trampolines
		new deactivate_mm, add asm/bug.h
 - Make sure initramfs is freed after booting (thanks to Kai for the hint)
 - User per cpu data for profile counters (Ravikiran Thirumalai)
 - 32bit compat_* updates from Stephen Rothwell
 - Fix race in context switch. The exception handler for bogus segment
   loads in __switch_to needs to keep interrupts disabled, otherwise an
   interrupt can deadlock on scheduler locks.  Also make sure they don't
   printk or set oops_in_progress during printk because printk does a
   wake_up too.
 - Disable 64bit GS base changes for processes.  I cannot get it to work
   reliably.
 - Clear IOPL on kernel entry

into home.transmeta.com:/home/torvalds/v2.5/linux

Without it, unloading a module
leads to some unpleasant oops...

Please note that the naming DB is now completely optional. If there is
no eisa.ids in the drivers/eisa/ directory, build will behave as if
CONFIG_EISA_NAMES is disabled. So this patch can be left out if there
is any objection.

Base patch adding sysfs support for the EISA bus

This is a left-over from the old modules code, Rusty stubbed it out
to always return 0.  Three scsi pcmcia driver check it for beeing non-NULL,
trying to work around their unload races.  I've added #warnings there
and stubbed out the GET_USE_COUNT so we can remove it from the core.

The Linux code for SGI partitions uses an int instead of an unsigned int
in the ondisk structure in two places, which breaks > TB partitions.

While porting the code over from an internal 2.4-based tree I've also
switched it to use the explicit uXX/sXX types everywhere and moved the
struct defintions above sgi_partition().

  Move ALSA before OSS

  The input system in 2.5 is able to see the volume keys on inspiron
  notebooks w/o help from i8k.c.  This patch therefore removes the
  new code from i8kutils-1.17 for feeding those keypresses to the
  keyboard driver.

  This leaves only MODULE_PARM(restricted, "i") as the useful addition
  to what was in 2.5.58's i8k.c.  This module parm restricts control of
  the system fans to processes with CAP_SYS_ADMIN set.

into home.transmeta.com:/home/torvalds/v2.5/linux

There's no need to alias the kallsyms-related symbols to a dummy
variable, we can as well just do the sanity check against NULL.

  Allow for different LMA vs VMA (logical/virtual memory address).
  IA-64 uses the LMA to tell the bootloader the physical location
  of the image, whereas the VMA as always represents the address the
  image gets mapped to.
  
  The default (used for non IA-64) is LMA == VMA, which is what
  the linker previously assumed anyway.
  
Also:
o remove duplicate .rodata1 section
o __vermagic doesn't need its own section in vmlinux, it can
  just go into .rodata
o .kstrtab hasn't been used since the introduction of the new
  module loader, so it should be deleted from the linker scripts
  as well (except for arch/um, which does not seem up to date
  w.r.t the new module loader yet)
o The kallsyms mechanism has changed to not need its own section,
  so again the references in the linker scripts can go away

The current code reserves 60 bytes for the symbol string of every
exported symbol, unnecessarily wasting kernel memory since most symbols
are much shorter. We revert to the 2.4 solution where the actual strings
are saved out of line and only the pointers are kept.

The latest module-init-tools already handle this case, people who are
using older versions need to update to make sure depmod works
properly.

Saves 80 KB in vmlinux with my .config.

From: James Bottomley <James.Bottomley@steeleye.com>
  
kallsyms is broken in parisc on 2.5.56 again because of assembler syntax
subtleties.  This is the offending line:
  
printf("\t.byte 0x%02x ; .asciz\t\"%s\"\n"
  
Note the `;' separating the two statements.  On some platforms `;' is a
comment in assembly code, and thus the following .asciz is ignored.

Mac/m68k NCR5380 SCSI updates (forward port of Ray Knight's changes in 2.4.x):
  - Forward port of pseudo-DMA from 2.2.20
  - Move SCSI host template definition from mac_scsi.h to mac_scsi.c

Generic RTC driver: fix spelling in documentation (from Geoffrey Lee
<glee@gnupilgrims.org>)

Q40/Q60 keyboard fixes:
  - IRQ definitions were prepended with Q40_
  - <asm/keyboard.h> no longer exists
  - Let q40kbd_init() fails if not running on a Q40/Q60
  - q40kbd_init() must return an error code
  - Make q40kbd_{init,exit}() static

Jeff removed the swap_device member from struct swap_info_struct
(http://www.uwsg.iu.edu/hypermail/linux/kernel/0202.1/0258.html)
but it is still used in the m68k arch for the ST-RAM. The below
should remove it.

Frankly, I didn't try compiling... My original intent was to move the
swap_list definition from swap.h to mm/swapfile.c, but m68k still
uses it here :( so perhaps this isn't possible. And I just happened to
stumble upon this.

(from Marcus Alanen <maalanen@ra.abo.fi> through Rusty Trivial Russell)

M68k generic RTC driver updates:
  - Revive help text for CONFIG_GEN_RTC
  - Re-add lost config option for CONFIG_GEN_RTC_X
  - Re-add lost mach_get_ss()
  - Export mach_[gs]et_rtc_pll()
  - Add implementation of mach_get_ss() and mach_[gs]et_rtc_pll() for Q40/Q60
  - Add missing include
  - Add implementation of get_rtc_ss()

Sun-3: Add missing deactivate_mm() (yes, there should be two of them in
include/asm-m68k/mmu_context.h: one for Motorola MMUs and one for Sun-3 MMUs)

M68k exception table updates to compensate for changes in 2.5.55

Q40/Q60 IRQ updates from 2.4.x

Amiga keyboard: the release bit indicates a key release, not a key press.

into home.transmeta.com:/home/torvalds/v2.5/linux

vm_enough_memory seems to call si_meminfo just to get the total
RAM, which seems far too expensive. This replaces the comment
saying "this is crap" with some code that's less crap.

Here is the s390x patch to use the new generic compatibility functions.

This creates compat_sys_sigpending and compat_sys_sigprocmask and
patches sent to maintainers remove all the arch specific versions.

With Martin's continuing approval, here is the s390x part of the patch.

This creates compat_sigset_t and compat_old_sigset_t i.e.  just the
types.  This is just the generic part, the architecture specific parts
will be sent to respective maintainers.

into home.transmeta.com:/home/torvalds/v2.5/linux

into samba.org:/scratch/anton/for-alan