- Create struct cpu to generically describe cpus (it simply contains
  a struct sys_device in it).

- Define an array of size NR_CPUS in arch/i386/kernel/cpu/common.c 
  and register each on bootup. This gives us something like:

# tree -d /sys/root/sys/
/sys/root/sys/
|-- cpu0
|-- pic0
`-- rtc0

and:

# tree -d /sys/bus/system/devices/
/sys/bus/system/devices/
|-- cpu0 -> ../../../root/sys/cpu0

- Define arch-specific CPU driver that's also registered on boot.
  That gives us: 

# tree -d /sys/bus/system/drivers/
/sys/bus/system/drivers/
|-- cpu

- Create a CPU device class that's registered very early. That 
  gives us all the CPUs in the system in one place:

# tree -d /sys/class/cpu/
/sys/class/cpu/
|-- devices
|   `-- 0 -> ../../../root/sys/cpu0
`-- drivers

Other archs are encouraged to do the same.

The generic driver is used by the virtual USB bridge device. This makes sure that
the driver is registered before we try to use it (and it gets the bus type right).

We also check for equality when matching devices to drivers, because we don't 
want to match any device to it. 

In some cases, especially when dealing with system and platform devices, a 
device's driver is known when the device is registered. We still want to add
the device to the driver's list and add it to the class.

This makes splits driver binding into probe() and attach(). If the device already
has a driver, we simply call attach(). Otherwise, we try to match it on the bus
and still call found_match().

This requires that all drivers that are referenced are registered beforehand.

  
- Create struct sys_device to describe system-level devices (CPUs, PICs, etc.). This 
  structure includes a 'name' and 'id' field for drivers to fill in with a simple
  canonical name (like 'pic' or 'floppy') and the id of the device relative to its 
  discovery in the system (it's enumerated value).

  The core then constructs the bus_id for the device from these, giving them meaningful
  names when exporting them to userspace:

# tree -d /sys/root/sys/
/sys/root/sys/
|-- pic0
`-- rtc0

- Replace 
	int register_sys_device(struct device * dev);
	with 
	int sys_device_register(struct sys_device * sysdev);

- Fixup the users of the API.

- Add a system_bus_type for devices to associate themselves with. This provides a 
  bus/system/ directory in driverfs that looks like:

# tree -d /sys/bus/system/
/sys/bus/system/
|-- devices
|   |-- pic0 -> ../../../root/sys/pic0
|   `-- rtc0 -> ../../../root/sys/rtc0
`-- drivers
    `-- pic

This fixes all xchg()'s and a preemption bug.

This does the following things:

 - removes the ->thread_group list and uses a new PIDTYPE_TGID pid class
   to handle thread groups. This cleans up lots of code in signal.c and
   elsewhere.

 - fixes sys_execve() if a non-leader thread calls it. (2.5.38 crashed in
   this case.)

 - renames list_for_each_noprefetch to __list_for_each.

 - cleans up delayed-leader parent notification.

 - introduces link_pid() to optimize PIDTYPE_TGID installation in the
   thread-group case.

I've tested the patch with a number of threaded and non-threaded
workloads, and it works just fine. Compiles & boots on UP and SMP x86.

The session/pgrp bugs reported to lkml are probably still open, they are
the next on my todo - now that we have a clean pidhash architecture they
should be easier to fix.

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

T30_s * is part of a union, so the typedef needs to exist even when
CONFIG_ISDN_TTY_FAX is not set.

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

When converting all L_TARGETs to lib.a, I missed these instances.

Update alpha port to work with new nanosecond xtime, and the in_atomic()
requirements.

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

The problem is that the local APIC code references stuff in
mpparse, but 2.5.37 changed arch/i386/kernel/Makefile to only
compile mpparse for SMP.

This patch works around this by enforcing CONFIG_X86_MPPARSE
for all LOCAL_APIC-enabled configs.

Add bio_get_nr_vecs(). It returns an approximate number of pages that
can be added to a block device. It's just a ballpark number, but I think
this is quite fine for the type of thing it is needed for: mpage etc
need to know an approx size of a bio that they need to allocate. It
would be silly to continously allocate 64-page sized bio_vec entries, if
the target cannot do more than 8, for example.

make pdc4030 work

Bad merge from 2.4.20-pre-ac, ide_build_dmatable() does not need data
direction argument in 2.5 (it's implicit in the request)

I'm terribly sorry - I've sent you the wrong diff, it was
some intermediate variant. Actually it added extra breakage to
ide_hwif_configure().

Desired behavior was:

if ctl == base == 0, the device is in "true legacy" mode (as per PCI
spec); use values from the base address registers otherwise.

cleanup end_that_request_first() end_io handling, and fix bug where
partial completes didn't get accounted right wrt blk_recalc_rq_sectors()

into tp1.ruhr-uni-bochum.de:/home/kai/kernel/v2.5/linux-2.5.isdn

It was (only partially) protected by cli() before, which we want to
get rid of.

Simplifies the code which was previously using an open coded
singly linked list.

Also, deleting a phone number during dial-out could easily oops
the kernel before this patch.

ISDN_GLOBAL_STOPPED is a way to globally stop the system
from dialing out / accepting incoming calls. Instead of
spreading checks all over the place, just catch dial
commands / incoming call indications in one place.

Also, kill isdn_net_phone typedef and clean up affected code.

It's not used for the timeout controlled hangup anymore, only to
hangup depending on the dialmode, which we handle directly now.

o PPP_IPX is defined in a header these days
o isdn_net_hangup takes an isdn_net_local *, simplifying
  code a bit.

into tp1.ruhr-uni-bochum.de:/home/kai/kernel/v2.5/linux-2.5.make

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

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

Convert the VM to not wait on other people's dirty data.

 - If we find a dirty page and its queue is not congested, do some writeback.

 - If we find a dirty page and its queue _is_ congested then just
   refile the page.

 - If we find a PageWriteback page then just refile the page.

 - There is additional throttling for write(2) callers.  Within
   generic_file_write(), record their backing queue in ->current.
   Within page reclaim, if this tasks encounters a page which is dirty
   or under writeback onthis queue, block on it.  This gives some more
   writer throttling and reduces the page refiling frequency.

It's somewhat CPU expensive - under really heavy load we only get a 50%
reclaim rate in pages coming off the tail of the LRU.  This can be
fixed by splitting the inactive list into reclaimable and
non-reclaimable lists.  But the CPU load isn't too bad, and latency is
much, much more important in these situations.

Example: with `mem=512m', running 4 instances of `dbench 100', 2.5.34
took 35 minutes to compile a kernel.  With this patch, it took three
minutes, 45 seconds.

I haven't done swapcache or MAP_SHARED pages yet.  If there's tons of
dirty swapcache or mmap data around we still stall heavily in page
reclaim.  That's less important.

This patch also has a tweak for swapless machines: don't even bother
bringing anon pages onto the inactive list if there is no swap online.

The key concept here is that pdflush does not block on request queues
any more.  Instead, it circulates across the queues, keeping any
non-congested queues full of write data.  When all queues are full,
pdflush takes a nap, to be woken when *any* queue exits write
congestion.

This code can keep sixty spindles saturated - we've never been able to
do that before.

 - Add the `nonblocking' flag to struct writeback_control, and teach
   the writeback paths to honour it.

 - Add the `encountered_congestion' flag to struct writeback_control
   and teach the writeback paths to set it.

So as soon as a mapping's backing_dev_info indicates that it is getting
congested, bale out of writeback.  And don't even start writeback
against filesystems whose queues are congested.

 - Convert pdflush's background_writeback() function to use
   nonblocking writeback.

This way, a single pdflush thread will circulate around all the
dirty queues, keeping them filled.

 - Convert the pdlfush `kupdate' function to do the same thing.

This solves the problem of pdflush thread pool exhaustion.

It solves the problem of pdflush startup latency.

It solves the (minor) problem wherein `kupdate' writeback only writes
back a single disk at a time (it was getting blocked on each queue in
turn).

It probably means that we only ever need a single pdflush thread.

Use the new queue congestion detector in ext2_preread_inode().  Don't
try the speculative read if the read queue is congested.

Also, don't try it if the disk is write-congested.  Presumably it is
more important to get the dirty memory cleaned out.

The patch provides a means for the VM to be able to determine whether a
request queue is in a "congested" state.  If it is congested, then a
write to (or read from) the queue may cause blockage in
get_request_wait().

So the VM can do:

	if (!bdi_write_congested(page->mapping->backing_dev_info))
		writepage(page);

This is not exact.  The code assumes that if the request queue still
has 1/4 of its capacity (queue_nr_requests) available then a request
will be non-blocking.  There is a small chance that another CPU could
zoom in and consume those requests.  But on the rare occasions where
that may happen the result will mereley be some unexpected latency -
it's not worth doing anything elaborate to prevent this.

The patch decreases the size of `batch_requests'.  batch_requests is
positively harmful - when a "heavy" writer and a "light" writer are
both writing to the same queue, batch_requests provides a means for the
heavy writer to massively stall the light writer.  Instead of waiting
for one or two requests to come free, the light writer has to wait for
32 requests to complete.

Plus batch_requests generally makes things harder to tune, understand
and predict.  I wanted to kill it altogether, but Jens says that it is
important for some hardware - it allows decent size requests to be
submitted.

The VM changes which go along with this code cause batch_requests to be
not so painful anyway - the only processes which sleep in
get_request_wait() are the ones which we elect, by design, to wait in
there - typically heavy writers.


The patch changes the meaning of `queue_nr_requests'.  It used to mean
"total number of requests per queue".  Half of these are for reads, and
half are for writes.  This always confused the heck out of me, and the
code needs to divide queue_nr_requests by two all over the place.

So queue_nr_requests now means "the number of write requests per queue"
and "the number of read requests per queue".  ie: I halved it.

Also, queue_nr_requests was converted to static scope.  Nothing else
uses it.


The accuracy of bdi_read_congested() and bdi_write_congested() depends
upon the accuracy of mapping->backing_dev_info.  With complex block
stacking arrangements it is possible that ->backing_dev_info is
pointing at the wrong queue.  I don't know.

But the cost of getting this wrong is merely latency, and if it is a
problem we can fix it up in the block layer, by getting stacking
devices to communicate their congestion state upwards in some manner.

__set_page_dirty_buffers() is calling __mark_inode_dirty under
mapping->private_lock.

We don't need to hold ->private_lock across that call.  It's only there
to pin page->buffers.

This simplifies the VM locking heirarchy.

When I converted ext3 to use to use direct-to-BIO writeback for
data=writeback mode I forgot that we need to hold a transaction open on
behalf of MAP_SHARED pages.  The fileystem is BUGging in get_block()
because there is no transaction open.

So let's forget that idea for now and send data=writeback mode back to
ext3_writepage.

into nuts.ninka.net:/home/davem/src/BK/sparc-2.5

into nuts.ninka.net:/home/davem/src/BK/net-2.5

into nuts.ninka.net:/home/davem/src/BK/net-2.5

Slowly killing the ugly struct forest.