This patch fixes up a lot of problems in the m68knommu architecture
CO-MEMlite PCI interface code. The code was seriously out of date,
not implementing many of the requried PCI bios support functions.
Also fixes many wrong function return types and arguments.

Fix m68knommu 68VZ328 Makefile to only traverse its sub-directories.
There is no local code files for it to compile.

Include errno.h in the m68knommu 68360 specific interrupt setup code.
It needs a few E* types as return values.

Fix wrong argument prototype in m68knommu commproc.c.

This patch fixes the m68knommu de2 platform config to use the generic
68328 irq setup. All 68*328 family can use the same setup.

This patch fixes a couple of spelling errors in the m68knommy Kconfig
help entries. Also makes CONFIG_HIGHPROFILE dependant on CONFIG_COLDFIRE,
since that feature is only implemented on ColdFIre based platforms.

This patch fixes the m68knommu ucdimm platform config to use the generic
68328 irq setup. All 68*328 family can use the same setup.

Include errno.h in the m68knommu 68328 specific interrupt setup code.
It needs a few E* types as return values.

Include asm/unistd.h to get the NR_syscalls definition in the
m68knommu 68328 interrupt entry code.

Include asm/unistd.h to get the NR_syscalls definition in the
m68knommu 68360 interrupt entry code.

Here is the latest round of EISA/sysfs update.

 - Add documentation,
 - Add support for per EISA-id driver data,
 - Move virtual_root device to a platform device,
 - Update CREDITS.

Here is the parisc part of the patch with Willy's blessing.

Here is the s390x part of the patch with Martin's blessing.

Here is the x86_64 part of the patch with Andi's blessing.

This patch creates compat_sys_fcntl{,64}.  The diffstat for the whole
patch set is below, but this is just the generic part, the architecture
specific parts will follow.

This patch also removes struct flock64 from all the 64 bit architectures
(except parisc).

into tetrachloride.(none):/mnt/raid/src/kernel/2.5/cpufreq

Patch from Dave Miller.  Fixes a very long-standing bug.

If a process has an fd open against a now-removed directory, lookups on that
fd will end up calling ext2_find_entry() against a zero-length directory.

When this happens ext2_find_entry() will, on the first pass through the loop,
set `kaddr' to page_address(page) - 20.  Things get confused and the "zero
length directory entry" warning triggers.

This only happens on 64-bit machines, because ext2_last_byte() is returning
an unsigned (32-bit) value, and the arithmetic works out OK for 32-bit
machines.

So we change ext2_find_entry() to bale out immediately if the directory is
zero-length.  All other directory-walking functions do this, but
ext2_find_entry() forgot to, due to the search-from-the-last-place
optimisation.

Fix a couple of ext3 error handling routines to not assume that the
superblock has valid journal and buffer_head pointers.  These functions are
called during mount and unmount and that may not be true.

This should fix the oops which Zwane saw when mounting a corrupt filesystem.

I enabled the advanced use-after-free detector for large slab objects and the
kernel oopsed.  This is because that debug code adds things at the head of
the slab objects, and the kernel will die if task_structs are not
well-aligned.

The way to tell the slab allocator that it is not allowed to misalign objects
from this slab is SLAB_MUST_HWCACHE_ALIGN.

Patch from Oleg Drokin <green@linuxhacker.ru>

There seems to be a memleak on error exit path.

Patch from Oleg Drokin <green@linuxhacker.ru>

   Not freeing allocated memory on error exit path.

Patch from Petr Vandrovec <vandrove@vc.cvut.cz>
Reported by Oleg Drokin.

    In NCP_IOC_SETOBJECTNAME handler, we allocated space (newname pointer),
    copy stuff from userspace to there and then assign userspace
    pointer to our internal structure, whoops!

Force inlining even when gcc-3.x is too confused to do it for us.

The conversion of the console registration to an initcall-style thing has
broken lots of people's setups.  It is now dependent upon linkage order and
if you have both CONFIG_VT_CONSOLE and CONFIG_SERIAL_CONSOLE, no boot
messages come out on the screen because the kernel is selecting the serial
console first.

It can be fixed by specifying console=tty0, but nobody is doing that.

We can fix it up by placing drivers/char/ in front of drivers/serial/ in
linkage order.

The pnp_request_card_device() stub should return NULL, not -ENODEV.

Patch from Andreas Mohr <andi@rhlx01.fht-esslingen.de>

Need to retake the spinlock in __pdflush() before continuing.

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

Make x86-64 boot again after the INITIAL_JIFFIES changes
and some bug fixes. Also some work for NPTL.

 - Merge with i386/2.5.64-bk3
 - Fix memory leak in copy_thread
 - arch_prctl uses GDT for base if possible. Cleanup.
 - clone supports bases >32bit for SETTLS etc. %fs hardcoded now.
 - new ptrace support for 64bit TLS
 - Disable (set|get)_thread_* for 64bit processes.
 - Audit arch/x86_64 for jiffies wrap issues.
 - Fix initial jiffies problem (that caused hanging kernels)
 - FIx a few 32bit emulation bugs (sigaltstack, sigqueue)
 - Some cleanup from Pavel
 - Should compile again as UP
 - Shrink size a bit by not putting exception tables into object files.
 - Fix compilation with gcc 3.3 :- force inlining when needed
 - Work around 2.5.64-bk3 console init bug.
 - Fix some alignments in assembly code

Rationale:  devfs_only does nothing but disabling {un,}register_blkdev
and {un,}register_chrdev.  {un,}register_blkdev already do nothing but
adding it's name argument to a lookup table for the __bdevname and
/proc/device output so this use is already bogus.  The disabling of
the character device per-major arrays can work in practice but is
useless as any driver relying on it can't be used on non-devfs systems.

Cleanup the i2c procfs code a bit (less ifdef mess), partially based
on the lm_sensors CVS code.

This fixes a bunch of "src" versus "obj" confusions that apparently
existed in my bootsect removal patch.  How embarrassing...

This recent patch caused Nick Piggin's 2xPIII VIA686B chipset machine into an
interrupts-off lockup during IDE probing.

We don't really know why - it might be because an interrupt is delivered to a
secondary which doesn't expect it.

I have a second patch from Zwane which solves the same problem in a different
way, but until that's had some wider testing I suggest we just back off the
original.

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

 - Fix my email address to one which reaches me ;-)
 - Remove timer.h include as we don't use timers.
 - Add module.h.
 - Sort includes alphabetically.
 - Move EXPORT_SYMBOL from ksyms.c to locks.c.
 - Simplify locks_conflict().

entry routine load the real ESP0 off that per-cpu stack. Make this
even faster by putting the sysenter stack in the per-CPU TSS, so
that we can use the tss->esp0 value directly (which we have to
update on task switches anyway).

CAREFUL! This needs very subtle code for debug and NMI exceptions,
to make sure we don't run with the sysenter stack in any real kernel
code!