From Andi Kleen:

Some fixes for CONFIG_X86_MCE_NONFATAL

Calling smp_call_function from interrupt context is forbidden

Unless I'm mistaken it would BUG on any box with more than two CPUs because
it would expect smp_call_function callback to run only on a single CPU??

Also handle the Hammer.

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

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

SuSv3 says: "The mprotect() function shall fail if:

  [EACCES]
    The prot argument specifies a protection that violates the access
    permission the process has to the underlying memory object.

  [EAGAIN]
    The prot argument specifies PROT_WRITE over a MAP_PRIVATE mapping and
    there are insufficient memory resources to reserve for locking the
    private page.

  [EINVAL]
    The addr argument is not a multiple of the page size as returned by
    sysconf().

  [ENOMEM]
    Addresses in the range [addr,addr+len) are invalid for the address
    space of a process, or specify one or more pages which are not mapped.

  [ENOMEM]
    The prot argument specifies PROT_WRITE on a MAP_PRIVATE mapping, and
    it would require more space than the system is able to supply for
    locking the private pages, if required.

  [ENOTSUP]
    The implementation does not support the combination of accesses
    requested in the prot argument."

This fixes error code of mprotect() of the ENOMEM case.

SuSv3 says: "The msync() function shall fail if:

[EBUSY]
    Some or all of the addresses in the range starting at addr and
    continuing for len bytes are locked, and MS_INVALIDATE is
    specified.

[EINVAL]
    The value of flags is invalid.

[EINVAL]
    The value of addr is not a multiple of the page size {PAGESIZE}.

[ENOMEM]
    The addresses in the range starting at addr and continuing for len
    bytes are outside the range allowed for the address space of a process
    or specify one or more pages that are not mapped."

This fixes error code of msync() of the EINVAL case.

A partitioned ppc64 machine can have a boot cpuid anywhere from 0 to 31.
With the non linear cpu changes in 2.5.24 we must set up the initial task
to start on the boot cpu. (since it isnt always 0 now)

With this patch I am able to boot on cpus other than 0. I also tested
discontiguous cpuids. So the non linear cpu stuff checks out OK.

I introduced warning about 3 releases ago. This prototypes are
actually unneccessary and killing them makes warning go away. Please
apply,

I'd like Florent credited -- he is maintaining 2.4.X version and
helping with development. Kill warnings by rearranging code / adding
prototypes. Enable using separate console (so user sees progress and X
suspend/resume works properly), forward-port of updates from Florent
and stop using own PRINTK stuff (mostly). Reading now primarily uses
block_device(), this should enable more cleanups. Fixed double free on
error path.

This fixes a problem introduced by the pageattr ioremap/unmap patches.
iounmap lost the ability to free non page aligned addresses, which
are e.g. used by the bootflag code.  This patch fixes this.

Also fix a potential off by one bug.

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

This patch for 2.5.25 is a rewrite of how smbfs builds requests. It allows
for more parallellism, better error handling and supporting oplocks with
further patches.

This was found by Dan Carpenter <error27@email.com>, using an smatch
script.  Looks to me like like an error caused during all the BKL
pushing.  1 more coming...

Not tested (Greg, please don't hurt me :p )

This was found by Dan Carpenter <error27@email.com>, using an smatch
script.  Looks to me like like an error caused during all the BKL
pushing.

Since I turned on the IO-APICs on secondary quads, we are receiving
timer interrupts on *all* quads, not just the first quad, each from their
local timer chip. This causes time to progress far too rapidly ;-)

The simple patch below turns off the timer interrupts for IO-APICs other
than interrupt 0, and has been tested to fix the problem. As it switches
on clustered_apic_mode, it should be safe from hurting anyone else.

This fix is already in 2.4 - I'm playing catchup with 2.5 - the same patch
applies with just a line offset

The current Intel-specific cache detection code is incomplete and hard to
read. This patch puts all the cache descriptors in a table, which then is
examined for matches in a loop.

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

Replace the global semaphore_lock with the spinlock embedded in
the wait_queue_head_t.  None of the data protected by semaphore_lock
is global and there is no need to restrict the system to only allow
one semaphore to be dealt with at a time.

This removes 2 lock round trips from __down() and __down_interruptible().
It also reduces the number of cache lines touched by 1 (the cache line
with seamphore_lock).

The patch below fixes a minor memory leak (forgotten kfree() on
initialization errors) in the msp3400 module.

This patch updates the bttv driver.  Changes:

 * moved much code to the generic video-buf.c helper module
   (bttv-driver.c, bttv-vbi.c, videobuf.c).
 * a number of changes in the card list and the card-specific code
   (bttv-cards.c).
 * misc small fixes here and there.

This patch updates the bttv driver documentation.

This patch fixes/improves the TDA8425 support in tvaudio.c.
It also makes the tvmixer and tvaudio modules look for i2c hardware on
riva cards (the rivatv project needs this).

This patch updates the btaudio driver.  btaudio can automatically
configures the sampling rate for autodetected cards (by PCI Subsystem
ID) now.

This patch adds support for a new philips tuner chip to the tuner
module.  It also makes the tuner module look for hardware on riva
cards.

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

kernel/softirq.c :: do_softirq() has a typo in the type of the `cpu'
value.  I incorrectly set it to `unsigned long' in a previous patch...
this is a CPU value not a bitmask and is thus an `int'.

The attached patch makes use of the new list_move_tail macro in 2.5 as
needed in sched.c, replacing a list_del and list_add_tail combo with the
optimized list_move_tail.

inode->i_sb->s_bdev is NULL when the inode refers to a blockdev.
Use the get_block() result instead.

Here's a patch which converts O_DIRECT to go direct-to-BIO, bypassing
the kiovec layer.  It's followed by a patch which converts the raw
driver to use the O_DIRECT engine.

CPU utilisation is about the same as the kiovec-based implementation.
Read and write bandwidth are the same too, for 128k chunks.   But with
one megabyte chunks, this implementation is 20% faster at writing.

I assume this is because the kiobuf-based implementation has to stop
and wait for each 128k chunk, whereas this code streams the entire
request, regardless of its size.

This is with a single (oldish) scsi disk on aic7xxx.  I'd expect the
margin to widen on higher-end hardware which likes to have more
requests in flight.

Question is: what do we want to do with this sucker?  These are the
remaining users of kiovecs:

	drivers/md/lvm-snap.c
	drivers/media/video/video-buf.c
	drivers/mtd/devices/blkmtd.c
	drivers/scsi/sg.c

the video and mtd drivers seems to be fairly easy to de-kiobufize.
I'm aware of one proprietary driver which uses kiobufs.  XFS uses
kiobufs a little bit - just to map the pages.

So with a bit of effort and maintainer-irritation, we can extract
the kiobuf layer from the kernel.

Synchronize with 2.5.25.

Incorporate IDE-94, as well as 95, 96, 97 and 98-pre as announced by Bartek and
unfortunately still not included in 2.5.25, which makes admittedly things
still fall appart:

Missing changelog for 98-pre by Bartlomiej Zolnierkiewicz (BTW.  Handling
Unicode should be essential at least to make proper crediting of many many
peoples possible!) follows here:

 - add missing channel->lock unlocking/locking and fix some comments
   in ide_timer_expiry()

 - allow PCI drivers to disable autodma in ->init_dma()
   (bug introduced in IDE 97, affects sl82c105.c only)

   noticed by Russell King

 - alim15x3.c, if revision is <= 0x20 disable autodma

 - remove unneeded checks (drive.dn > 3) from pdc202xx.c and sis5513.c

 - use block layer wrappers

And my additions follow:

 - Fix TCQ code. Patch based on work by Alexander Atanasov.

 - Use the FreeBSD derived request handler return values:

	ATA_OP_FINISHED
	ATA_OP_CONTINUES
	ATA_OP_RELEASED
	ATA_OP_READY	/* for status ready reporting during poll */

 - PMAC compilation fix by Paul Mackerras.

 - Simplify the ata_status_poll function significantly.

 - Fix logic used to prevent drive IRQ assertion from drive on channels sharing
   our interrupt.

NOTE: We will move it later to the time where a request is really finished
soon.

 - Don't use ata_busy_poll() use ata_status_poll() instead. This increases code
   unification.

NOTE: We should maybe invent some way to prevent the error recovery path to be
taken at all. In esp to prevent ata_error from trying to reissue commands.

"getpid()" returns "tgid" not "pid", so this is correct.

Fix from Saurabh Desai.

Eeek....

futex.c is seriously b0rken.

This is against 2.5.25. Lots of fixes. Brings things inline with 2.5.25.
Probably makes the subsystem actually work now (well, it will work, but
I bet without it, it doesn't). Merged in changes from the current 2.5.x
source.

about the "sendpage()" callback. Don't try to fall back on
a write with the page kmap'ed

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

- Move the directory index bitmap to use an attribute inode instead of
  having special fields for it inside the ntfs inode structure. This
  means that the index bitmaps now use the page cache for i/o, too,
  and also as a side effect we get support for non-resident index
  bitmaps for free.
- Simplify/cleanup error handling in fs/ntfs/dir.c::ntfs_readdir() and
  fix a page leak that manifested itself in some cases.
- Add fs/ntfs/inode.c::ntfs_put_inode(), which we need to release the
  index bitmap inode on the final iput().

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

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

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