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.

With this llc_ui_sockets is almost not needed anymore, next
changesets will deal with the dataunit/xid/test primitives, that
are still using it.

It is the same bug fixed some months ago in tcp_v6_get_port,
i.e. we can't touch ipv6 private areas without checking if
the socket is AF_INET6.

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

it is an ex-parrot

assorted compile fixes

mtdblock switched to use of gendisks + compile fixes

z2ram.c switched to use of gendisks

ataflop.c switched to use of gendisks

amiflop.c switched to use of gendisks

macroectomy a-la pf.c and pcd.c ones, ditto for passing pointers to
structures instead of minors.

pd.c fed through Lindent

dumb expansion of macro - it had #define CURRENT current_req

tapeblock never assignes anything to its elements of blk_size[][]; we could
not bother allocating it in the first place.

More trivial fixes: typos in partitions/check.c, block/floppy.c and
acorn/block/fd1772.c + replacement of #define with inline in block/floppy.c
(fd_eject()).

Some trivial fixes for some typos introduced by Al's gendisk changes..

 - missing comma in cdu31a
 - missing semicolon in cdu31a
 - comma instead of colon in gscd
 - semicolon instead of comma in mcd
 - missing closing bracket in sonycd535

From Andries.

- Fix module usage counting for ip6_tables.o
- make ipt_ULOG compile on SPARC
- fix experimental ipt_unclean match, do not consider udp w/o csum unclean

From davem: replace `unsigned' with size_t.

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