I've got a new i/o scheduler in testing, some changes where needed in
the block layer to accomodate it. Basically because right now
assumptions are made about q->queue_head being the sort list. The
changes in detail:

o elevator_merge_requests_fn takes queue argument as well

o __make_request() inits insert_here to NULL instead of
  q->queue_head.prev, which means that the i/o schedulers must
  explicitly check for this condition now.

o incorporate elv_queue_empty(), it was just a place holder before

o add elv_get_sort_head(). it returns the sort head of the elevator for
  a given request. attempt_{back,front}_merge uses it to determine
  whether a request is valid or not. Maybe attempt_{back,front}_merge
  should just be killed, I doubt they have much relevance with the wake
  up batching.

o call the merge_cleanup functions of the elevator _after_ the merge has
  been done, not before. This way the elevator functions get the new
  state of the request, which is the most interesting.

o Kill extra nr_sectors check in ll_merge_requests_fn()

o bi->bi_bdev is always set in __make_request(), so kill check.

 - Change over queuedata to carry the device instead of the channel
   information.

 - cs5530 patches by Adam J. Richter. Small indent style adjustments.

 - qd65xx cli()/sti() adjustments.

 - Fix bogous command in ide.c pointed out by Peter Vendroviec.

 - Eliminate ide_stall_queue(). For those worried: we didn't sleep at
   all.

 - Eliminate support for "sector remapping". loop devices can handle
   stuff like that. All the custom DOS high system memmory loaded
   BIOS workaround tricks are obsolete right now. If anywhere it should
   be the FAT filesystem code which should be clever enough to deal with
   it by adjusting it's read/write methods.

 - PCI "scather gather" allocation handling revamp by Adam J. Richter.

 - Simplify do_ide_request after ->sleep removal.

 - Make do_ide_request prefer to handle the device matching the request
   queue it was called for first. RQ-queues are unique for devices.
   In a next step queuedata will be changed to point to the device
   not the channel.

Changeset 1.403.142.29 introduces a pretty nasty race into the RPC
code. Once we've decoded the RPC reply, it needs to be protected
against being overwritten by any resends.

The following patch achieves this by ensuring that the request is
removed from the list xprt->recv in xprt_complete_rqst(). This again
ensures that xprt_lookup_rqst() will fail to find that request until
we put it back on the list.

bogus dentry pointer to the VFS layer (which would oops on it
inside the dcache lock).

the d_drop is appropriate.

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

Linus, can we please not have that BUG() in 2.5.30?

One of the flakey things I see with 2.5.29 is that many
devices give trouble enumerating.  uhci still oopses khubd
enumerating pegasus (many control message errors first).
But that's not the only troublesome combo.

This patch fixes a failure that would hang the ohci driver
It's basically avoiding a few less IRQs (most noticable
with ISO), in this case so that a wedged SET_ADDRESS completes.
(Instead of hanging khubd in 'D' state.)  And it also updates
a diagnostic to provide more troubleshooting help for similar
issues in the future.

Linus, please apply. This fixes cpu_possible() on x86.

The following patches to nfs2xdr.c/nfs3xdr.c might be relevant: *entry
points into the page, and so the assignment should be made before we
kunmap().

Also includes a patch that converts references to inode->i_data into
inode->i_mapping.  Not likely to be the source of your hang, but should
still be done for consistency with the rest of the kernel.

One more lookup bug caught. Fix missing unlock_kernel() when doing
readdirplus cached lookups in nfs_lookup().

copy_process() just copies the process, it doesn't actually start it.

This is in preparation for doing a "atomically start process on CPU X"
or other cases where we want to change the state of the process before
we actually start running it.

 - Include first cut by Adam J. Richter on proper lock protection for
   tuning functions.

 - Rename ide_register_subdriver() to ata_register_device() and
   ide_unregister_subdriver() accordingly to ata_unregister_device(),
   since this is reflecting better what those functions are about.

 - Remove tons of "curicum vite" style useless documentation here and
   there.

 - typedef unsigned char byte; is finally gone. Everything using it should
   be just ported to u8 if I missed some place where it still gets used.

 - Next round of parameter parsing cleanups by Gerald Champagne.
   Adjusted a bit to harmonize with  hd.c.

 - Move IDE register bitfields declarations over from hdparm.h to
   ide.h.

 - Declare constants needed by hd.c directly there. Those are standard
   values not subject to change and we prefer a bit of code duplication
   in favour of making the two drivers independant from each other.

 - Move everything not ioctl related away from hdreg.h to ide.h.
   This header is in effect not private to the ATA code and should
   therefore not contain stuff only usefull there.

The CS5530 one seems unneeded looking at the databook.  Try the patch
below instead, which removes the irq lock and uses the proper kernel
functions to enable MWI and master.

This updates the 2.5.x PC Card drivers of the Bluetooth subsystem.

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

using a system call from kernel space. 

This avoids one level of hidden code, and makes what happens
much more explicit (and speeds it up too, fwiw)

This "warning fix" bug report is actually an OOPS bugfix.

locks_unlock_delete is buggy in a couple of different ways (previously
reported by Brian Dixon). Rather than fix it, this patch simply deletes
it and uses the normal posix file locking mechanisms to remove all locks
in locks_remove_posix instead.

Make cpu_relax() on all architectures a gcc barrier to match x86.

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

o Corrected dependencies for parportbook
o Introduced do_cmd, thus adhering to KBUILD_VERBOSE and make -s

do_cmd is a nice shorthand when creating rules that in one line
shall adhere to KBUILD_VERBOSE and make -s.
So far the only user is (will be) the docbook makefile

file locking LSM update

An acct flag was added to do_munmap, true everywhere but in mremap's
move_vma: instead of updating the arch and driver sources, revert that
that change and temporarily mask VM_ACCOUNT around that one do_munmap.

Also, noticed that do_mremap fails needlessly if both shrinking _and_
moving a mapping: update old_len to pass vm area boundaries test.

If we support mmap MAP_NORESERVE, we should support it on shared
anonymous objects: too bad that needs a few changes.  do_mmap_pgoff pass
VM_ACCOUNT (or not) down to shmem_file_setup, flag stored into shmem
info, for use by shmem_delete_inode later.  Also removed a harmless but
pointless call to shmem_truncate.

Update Doc and remove FIXME comment from fork.c now accounting right.

do_mmap_pgoff's (file == NULL) check was incorrect: it caused shared
MAP_ANONYMOUS objects to be counted twice (again in shmem_file_setup),
and again on fork(); whereas the equivalent shared /dev/zero objects
were correctly counted.  Conversely, a private readonly file mapping
was (correctly) not counted, but still not counted when mprotected to
writable: mprotect_fixup had pointless "charged = 0" changes, now it
does vm_enough_memory checking when private is first made writable
(but later we may want to refine behaviour on a noreserve mapping).

Also changed correct (flags & MAP_SHARED) test in do_mmap_pgoff to
equivalent (vm_flags & VM_SHARED) test: because do_mmap_pgoff is
dealing with vm_flags rather than the input flags by that stage.

Remove vm_unacct_vma function: it's only used in one place,
which can do it better by using vm_unacct_memory directly.

do_mmap_pgoff clears MAP_NORESERVE from vm_flags when VM accounts
strictly: but it's not in vm_flags, it's in flags (and tested there).

There is no point in do_mremap clearing MAP_NORESERVE from its flags:
it has already validated that only the MREMAP_ flags can be set,
and it has no use for MAP_NORESERVE in the code that follows anyway.

shmem_notify_change and shmem_file_write be careful about overflowingly
large loff_t before shifting it into unsigned long for vm_enough_memory.
Rename SHMEM_MAX_BLOCKS to SHMEM_MAX_INDEX (to avoid confusion with
512-byte blocks), define SHMEM_MAX_BYTES from it.

But 2.5 vmtruncate lacked the s_maxbytes error handling which
shmem_notify_change now expects: bring it in from the -dj tree.
shmem_file_write error handling needs a closer look later on.

Repeated overnight kernel builds in tmpfs showed insane Committed_AS
by morning.  The main bug was that shmem_file_write was passing
(newsize-oldsize)>>PAGE_SHIFT to vm_enough_memory, but it has to be
((newsize>>PAGE_SHIFT)-(oldsize>>PAGE_SHIFT)) - imagine 1k writes.

But actually, if we're going to do strict accounting, then we should
round up to next page not down - use VM_ACCT macro throughout (needs
unusual mix of PAGE_CACHE_SIZE with PAGE_SHIFT); and must count one
page for a long symlink.

Currently there is no way to find out the effective object size of a slab
cache.  XFS has lots of IRIX-derived code that want to do zalloc() style
allocations on zones (which are implemented as slab caches in XFS/Linux)
and thus needs to know about it.  There are three ways do implement it:

a) implement kmem_cache_zalloc
b) make the xfs zone a struct of kmem_cache_t and a size variable
c) implement kmem_cache_size

The current XFS tree does a) but I absolutely don't like it as encourages
people to use kmem_cache_zalloc for new code instead of thinking about how
to utilize slab object reuse.  b) would be easy, but I guess kmem_cache_size
is usefull enough to get into the kernel.  Here's the patch: