Instead of sending ->older_than_this to queue_io() and
move_expired_inodes(), send the entire wb_writeback_work
structure.  There are other fields of a work item that are
useful in these routines and in tracepoints.
Acked-by: NJan Kara <jack@suse.cz>
Signed-off-by: NCurt Wohlgemuth <curtw@google.com>
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

Useful for analyzing the dynamics of the throttling algorithms and
debugging user reported problems.
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

It helps understand how various throttle bandwidths are updated.
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

Fix powerpc compile warnings

mm/page-writeback.c: In function 'bdi_position_ratio':
mm/page-writeback.c:622:3: warning: comparison of distinct pointer types lacks a cast [enabled by default]
page-writeback.c:635:4: warning: comparison of distinct pointer types lacks a cast [enabled by default]

Also fix gcc "uninitialized var" warnings.
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

One thing puzzled me is that in JBOD case, the per-disk writeout
performance is smaller than the corresponding single-disk case even
when they have comparable bdi_thresh. Tracing shows find that in single
disk case, bdi_writeback is always kept high while in JBOD case, it
could drop low from time to time and correspondingly bdi_reclaimable
could sometimes rush high.

The fix is to watch bdi_reclaimable and kick background writeback as
soon as it goes high. This resembles the global background threshold
but in per-bdi manner. The trick is, as long as bdi_reclaimable does
not go high, bdi_writeback naturally won't go low because
bdi_reclaimable+bdi_writeback ~= bdi_thresh.

With less fluctuated writeback pages, JBOD performance is observed to
increase noticeably in various cases.

vmstat:nr_written values before/after patch:

  3.1.0-rc4-wo-underrun+      3.1.0-rc4-bgthresh3+  
------------------------  ------------------------  
               125596480       +25.9%    158179363  JBOD-10HDD-16G/ext4-100dd-1M-24p-16384M-20:10-X
                61790815      +110.4%    130032231  JBOD-10HDD-16G/ext4-10dd-1M-24p-16384M-20:10-X
                58853546        -0.1%     58823828  JBOD-10HDD-16G/ext4-1dd-1M-24p-16384M-20:10-X
               110159811       +24.7%    137355377  JBOD-10HDD-16G/xfs-100dd-1M-24p-16384M-20:10-X
                69544762       +10.8%     77080047  JBOD-10HDD-16G/xfs-10dd-1M-24p-16384M-20:10-X
                50644862        +0.5%     50890006  JBOD-10HDD-16G/xfs-1dd-1M-24p-16384M-20:10-X
                42677090       +28.0%     54643527  JBOD-10HDD-thresh=100M/ext4-100dd-1M-24p-16384M-100M:10-X
                47491324       +13.3%     53785605  JBOD-10HDD-thresh=100M/ext4-10dd-1M-24p-16384M-100M:10-X
                52548986        +0.9%     53001031  JBOD-10HDD-thresh=100M/ext4-1dd-1M-24p-16384M-100M:10-X
                26783091       +36.8%     36650248  JBOD-10HDD-thresh=100M/xfs-100dd-1M-24p-16384M-100M:10-X
                35526347       +14.0%     40492312  JBOD-10HDD-thresh=100M/xfs-10dd-1M-24p-16384M-100M:10-X
                44670723        -1.1%     44177606  JBOD-10HDD-thresh=100M/xfs-1dd-1M-24p-16384M-100M:10-X
               127996037       +22.4%    156719990  JBOD-10HDD-thresh=2G/ext4-100dd-1M-24p-16384M-2048M:10-X
                57518856        +3.8%     59677625  JBOD-10HDD-thresh=2G/ext4-10dd-1M-24p-16384M-2048M:10-X
                51919909       +12.2%     58269894  JBOD-10HDD-thresh=2G/ext4-1dd-1M-24p-16384M-2048M:10-X
                86410514       +79.0%    154660433  JBOD-10HDD-thresh=2G/xfs-100dd-1M-24p-16384M-2048M:10-X
                40132519       +38.6%     55617893  JBOD-10HDD-thresh=2G/xfs-10dd-1M-24p-16384M-2048M:10-X
                48423248        +7.5%     52042927  JBOD-10HDD-thresh=2G/xfs-1dd-1M-24p-16384M-2048M:10-X
               206041046       +44.1%    296846536  JBOD-10HDD-thresh=4G/xfs-100dd-1M-24p-16384M-4096M:10-X
                72312903       -19.4%     58272885  JBOD-10HDD-thresh=4G/xfs-10dd-1M-24p-16384M-4096M:10-X
                50635672        -0.5%     50384787  JBOD-10HDD-thresh=4G/xfs-1dd-1M-24p-16384M-4096M:10-X
                68308534      +115.7%    147324758  JBOD-10HDD-thresh=800M/ext4-100dd-1M-24p-16384M-800M:10-X
                57882933       +14.5%     66269621  JBOD-10HDD-thresh=800M/ext4-10dd-1M-24p-16384M-800M:10-X
                52183472       +12.8%     58855181  JBOD-10HDD-thresh=800M/ext4-1dd-1M-24p-16384M-800M:10-X
                53788956       +94.2%    104460352  JBOD-10HDD-thresh=800M/xfs-100dd-1M-24p-16384M-800M:10-X
                44493342       +35.5%     60298210  JBOD-10HDD-thresh=800M/xfs-10dd-1M-24p-16384M-800M:10-X
                42641209       +18.9%     50681038  JBOD-10HDD-thresh=800M/xfs-1dd-1M-24p-16384M-800M:10-X
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

Keep a minimal pool of dirty pages for each bdi, so that the disk IO
queues won't underrun. Also gently increase a small bdi_thresh to avoid
it stuck in 0 for some light dirtied bdi.

It's particularly useful for JBOD and small memory system.

It may result in (pos_ratio > 1) at the setpoint and push the dirty
pages high. This is more or less intended because the bdi is in the
danger of IO queue underflow.
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

The dirty pause time shall ultimately be controlled by adjusting
nr_dirtied_pause, since there is relationship

	pause = pages_dirtied / task_ratelimit

Assuming

	pages_dirtied ~= nr_dirtied_pause
	task_ratelimit ~= dirty_ratelimit

We get

	nr_dirtied_pause ~= dirty_ratelimit * desired_pause

Here dirty_ratelimit is preferred over task_ratelimit because it's
more stable.

It's also important to limit possible large transitional errors:

- bw is changing quickly
- pages_dirtied << nr_dirtied_pause on entering dirty exceeded area
- pages_dirtied >> nr_dirtied_pause on btrfs (to be improved by a
  separate fix, but still expect non-trivial errors)

So we end up using the above formula inside clamp_val().

The best test case for this code is to run 100 "dd bs=4M" tasks on
btrfs and check its pause time distribution.
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

Apply two policies to scale down the max pause time for

1) small number of concurrent dirtiers
2) small memory system (comparing to storage bandwidth)

MAX_PAUSE=200ms may only be suitable for high end servers with lots of
concurrent dirtiers, where the large pause time can reduce much overheads.

Otherwise, smaller pause time is desirable whenever possible, so as to
get good responsiveness and smooth user experiences. It's actually
required for good disk utilization in the case when all the dirty pages
can be synced to disk within MAX_PAUSE=200ms.
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

As proposed by Chris, Dave and Jan, don't start foreground writeback IO
inside balance_dirty_pages(). Instead, simply let it idle sleep for some
time to throttle the dirtying task. In the mean while, kick off the
per-bdi flusher thread to do background writeback IO.

RATIONALS
=========

- disk seeks on concurrent writeback of multiple inodes (Dave Chinner)

  If every thread doing writes and being throttled start foreground
  writeback, it leads to N IO submitters from at least N different
  inodes at the same time, end up with N different sets of IO being
  issued with potentially zero locality to each other, resulting in
  much lower elevator sort/merge efficiency and hence we seek the disk
  all over the place to service the different sets of IO.
  OTOH, if there is only one submission thread, it doesn't jump between
  inodes in the same way when congestion clears - it keeps writing to
  the same inode, resulting in large related chunks of sequential IOs
  being issued to the disk. This is more efficient than the above
  foreground writeback because the elevator works better and the disk
  seeks less.

- lock contention and cache bouncing on concurrent IO submitters (Dave Chinner)

  With this patchset, the fs_mark benchmark on a 12-drive software RAID0 goes
  from CPU bound to IO bound, freeing "3-4 CPUs worth of spinlock contention".

  * "CPU usage has dropped by ~55%", "it certainly appears that most of
    the CPU time saving comes from the removal of contention on the
    inode_wb_list_lock" (IMHO at least 10% comes from the reduction of
    cacheline bouncing, because the new code is able to call much less
    frequently into balance_dirty_pages() and hence access the global
    page states)

  * the user space "App overhead" is reduced by 20%, by avoiding the
    cacheline pollution by the complex writeback code path

  * "for a ~5% throughput reduction", "the number of write IOs have
    dropped by ~25%", and the elapsed time reduced from 41:42.17 to
    40:53.23.

  * On a simple test of 100 dd, it reduces the CPU %system time from 30% to 3%,
    and improves IO throughput from 38MB/s to 42MB/s.

- IO size too small for fast arrays and too large for slow USB sticks

  The write_chunk used by current balance_dirty_pages() cannot be
  directly set to some large value (eg. 128MB) for better IO efficiency.
  Because it could lead to more than 1 second user perceivable stalls.
  Even the current 4MB write size may be too large for slow USB sticks.
  The fact that balance_dirty_pages() starts IO on itself couples the
  IO size to wait time, which makes it hard to do suitable IO size while
  keeping the wait time under control.

  Now it's possible to increase writeback chunk size proportional to the
  disk bandwidth. In a simple test of 50 dd's on XFS, 1-HDD, 3GB ram,
  the larger writeback size dramatically reduces the seek count to 1/10
  (far beyond my expectation) and improves the write throughput by 24%.

- long block time in balance_dirty_pages() hurts desktop responsiveness

  Many of us may have the experience: it often takes a couple of seconds
  or even long time to stop a heavy writing dd/cp/tar command with
  Ctrl-C or "kill -9".

- IO pipeline broken by bumpy write() progress

  There are a broad class of "loop {read(buf); write(buf);}" applications
  whose read() pipeline will be under-utilized or even come to a stop if
  the write()s have long latencies _or_ don't progress in a constant rate.
  The current threshold based throttling inherently transfers the large
  low level IO completion fluctuations to bumpy application write()s,
  and further deteriorates with increasing number of dirtiers and/or bdi's.

  For example, when doing 50 dd's + 1 remote rsync to an XFS partition,
  the rsync progresses very bumpy in legacy kernel, and throughput is
  improved by 67% by this patchset. (plus the larger write chunk size,
  it will be 93% speedup).

  The new rate based throttling can support 1000+ dd's with excellent
  smoothness, low latency and low overheads.

For the above reasons, it's much better to do IO-less and low latency
pauses in balance_dirty_pages().

Jan Kara, Dave Chinner and me explored the scheme to let
balance_dirty_pages() wait for enough writeback IO completions to
safeguard the dirty limit. However it's found to have two problems:

- in large NUMA systems, the per-cpu counters may have big accounting
  errors, leading to big throttle wait time and jitters.

- NFS may kill large amount of unstable pages with one single COMMIT.
  Because NFS server serves COMMIT with expensive fsync() IOs, it is
  desirable to delay and reduce the number of COMMITs. So it's not
  likely to optimize away such kind of bursty IO completions, and the
  resulted large (and tiny) stall times in IO completion based throttling.

So here is a pause time oriented approach, which tries to control the
pause time in each balance_dirty_pages() invocations, by controlling
the number of pages dirtied before calling balance_dirty_pages(), for
smooth and efficient dirty throttling:

- avoid useless (eg. zero pause time) balance_dirty_pages() calls
- avoid too small pause time (less than   4ms, which burns CPU power)
- avoid too large pause time (more than 200ms, which hurts responsiveness)
- avoid big fluctuations of pause times

It can control pause times at will. The default policy (in a followup
patch) will be to do ~10ms pauses in 1-dd case, and increase to ~100ms
in 1000-dd case.

BEHAVIOR CHANGE
===============

(1) dirty threshold

Users will notice that the applications will get throttled once crossing
the global (background + dirty)/2=15% threshold, and then balanced around
17.5%. Before patch, the behavior is to just throttle it at 20% dirtyable
memory in 1-dd case.

Since the task will be soft throttled earlier than before, it may be
perceived by end users as performance "slow down" if his application
happens to dirty more than 15% dirtyable memory.

(2) smoothness/responsiveness

Users will notice a more responsive system during heavy writeback.
"killall dd" will take effect instantly.
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

Add two fields to task_struct.

1) account dirtied pages in the individual tasks, for accuracy
2) per-task balance_dirty_pages() call intervals, for flexibility

The balance_dirty_pages() call interval (ie. nr_dirtied_pause) will
scale near-sqrt to the safety gap between dirty pages and threshold.

The main problem of per-task nr_dirtied is, if 1k+ tasks start dirtying
pages at exactly the same time, each task will be assigned a large
initial nr_dirtied_pause, so that the dirty threshold will be exceeded
long before each task reached its nr_dirtied_pause and hence call
balance_dirty_pages().

The solution is to watch for the number of pages dirtied on each CPU in
between the calls into balance_dirty_pages(). If it exceeds ratelimit_pages
(3% dirty threshold), force call balance_dirty_pages() for a chance to
set bdi->dirty_exceeded. In normal situations, this safeguarding
condition is not expected to trigger at all.

On the sqrt in dirty_poll_interval():

It will serve as an initial guess when dirty pages are still in the
freerun area.

When dirty pages are floating inside the dirty control scope [freerun,
limit], a followup patch will use some refined dirty poll interval to
get the desired pause time.

   thresh-dirty (MB)    sqrt
		   1      16
		   2      22
		   4      32
		   8      45
		  16      64
		  32      90
		  64     128
		 128     181
		 256     256
		 512     362
		1024     512

The above table means, given 1MB (or 1GB) gap and the dd tasks polling
balance_dirty_pages() on every 16 (or 512) pages, the dirty limit won't
be exceeded as long as there are less than 16 (or 512) concurrent dd's.

So sqrt naturally leads to less overheads and more safe concurrent tasks
for large memory servers, which have large (thresh-freerun) gaps.

peter: keep the per-CPU ratelimit for safeguarding the 1k+ tasks case

CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
Reviewed-by: NAndrea Righi <andrea@betterlinux.com>
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

There are some imperfections in balanced_dirty_ratelimit.

1) large fluctuations

The dirty_rate used for computing balanced_dirty_ratelimit is merely
averaged in the past 200ms (very small comparing to the 3s estimation
period for write_bw), which makes rather dispersed distribution of
balanced_dirty_ratelimit.

It's pretty hard to average out the singular points by increasing the
estimation period. Considering that the averaging technique will
introduce very undesirable time lags, I give it up totally. (btw, the 3s
write_bw averaging time lag is much more acceptable because its impact
is one-way and therefore won't lead to oscillations.)

The more practical way is filtering -- most singular
balanced_dirty_ratelimit points can be filtered out by remembering some
prev_balanced_rate and prev_prev_balanced_rate. However the more
reliable way is to guard balanced_dirty_ratelimit with task_ratelimit.

2) due to truncates and fs redirties, the (write_bw <=> dirty_rate)
match could become unbalanced, which may lead to large systematical
errors in balanced_dirty_ratelimit. The truncates, due to its possibly
bumpy nature, can hardly be compensated smoothly. So let's face it. When
some over-estimated balanced_dirty_ratelimit brings dirty_ratelimit
high, dirty pages will go higher than the setpoint. task_ratelimit will
in turn become lower than dirty_ratelimit.  So if we consider both
balanced_dirty_ratelimit and task_ratelimit and update dirty_ratelimit
only when they are on the same side of dirty_ratelimit, the systematical
errors in balanced_dirty_ratelimit won't be able to bring
dirty_ratelimit far away.

The balanced_dirty_ratelimit estimation may also be inaccurate near
@limit or @freerun, however is less an issue.

3) since we ultimately want to

- keep the fluctuations of task ratelimit as small as possible
- keep the dirty pages around the setpoint as long time as possible

the update policy used for (2) also serves the above goals nicely:
if for some reason the dirty pages are high (task_ratelimit < dirty_ratelimit),
and dirty_ratelimit is low (dirty_ratelimit < balanced_dirty_ratelimit),
there is no point to bring up dirty_ratelimit in a hurry only to hurt
both the above two goals.

So, we make use of task_ratelimit to limit the update of dirty_ratelimit
in two ways:

1) avoid changing dirty rate when it's against the position control target
   (the adjusted rate will slow down the progress of dirty pages going
   back to setpoint).

2) limit the step size. task_ratelimit is changing values step by step,
   leaving a consistent trace comparing to the randomly jumping
   balanced_dirty_ratelimit. task_ratelimit also has the nice smaller
   errors in stable state and typically larger errors when there are big
   errors in rate.  So it's a pretty good limiting factor for the step
   size of dirty_ratelimit.

Note that bdi->dirty_ratelimit is always tracking balanced_dirty_ratelimit.
task_ratelimit is merely used as a limiting factor.
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

It's all about bdi->dirty_ratelimit, which aims to be (write_bw / N)
when there are N dd tasks.

On write() syscall, use bdi->dirty_ratelimit
============================================

    balance_dirty_pages(pages_dirtied)
    {
        task_ratelimit = bdi->dirty_ratelimit * bdi_position_ratio();
        pause = pages_dirtied / task_ratelimit;
        sleep(pause);
    }

On every 200ms, update bdi->dirty_ratelimit
===========================================

    bdi_update_dirty_ratelimit()
    {
        task_ratelimit = bdi->dirty_ratelimit * bdi_position_ratio();
        balanced_dirty_ratelimit = task_ratelimit * write_bw / dirty_rate;
        bdi->dirty_ratelimit = balanced_dirty_ratelimit
    }

Estimation of balanced bdi->dirty_ratelimit
===========================================

balanced task_ratelimit
-----------------------

balance_dirty_pages() needs to throttle tasks dirtying pages such that
the total amount of dirty pages stays below the specified dirty limit in
order to avoid memory deadlocks. Furthermore we desire fairness in that
tasks get throttled proportionally to the amount of pages they dirty.

IOW we want to throttle tasks such that we match the dirty rate to the
writeout bandwidth, this yields a stable amount of dirty pages:

        dirty_rate == write_bw                                          (1)

The fairness requirement gives us:

        task_ratelimit = balanced_dirty_ratelimit
                       == write_bw / N                                  (2)

where N is the number of dd tasks.  We don't know N beforehand, but
still can estimate balanced_dirty_ratelimit within 200ms.

Start by throttling each dd task at rate

        task_ratelimit = task_ratelimit_0                               (3)
                         (any non-zero initial value is OK)

After 200ms, we measured

        dirty_rate = # of pages dirtied by all dd's / 200ms
        write_bw   = # of pages written to the disk / 200ms

For the aggressive dd dirtiers, the equality holds

        dirty_rate == N * task_rate
                   == N * task_ratelimit_0                              (4)
Or
        task_ratelimit_0 == dirty_rate / N                              (5)

Now we conclude that the balanced task ratelimit can be estimated by

                                                      write_bw
        balanced_dirty_ratelimit = task_ratelimit_0 * ----------        (6)
                                                      dirty_rate

Because with (4) and (5) we can get the desired equality (1):

                                                       write_bw
        balanced_dirty_ratelimit == (dirty_rate / N) * ----------
                                                       dirty_rate
                                 == write_bw / N

Then using the balanced task ratelimit we can compute task pause times like:

        task_pause = task->nr_dirtied / task_ratelimit

task_ratelimit with position control
------------------------------------

However, while the above gives us means of matching the dirty rate to
the writeout bandwidth, it at best provides us with a stable dirty page
count (assuming a static system). In order to control the dirty page
count such that it is high enough to provide performance, but does not
exceed the specified limit we need another control.

The dirty position control works by extending (2) to

        task_ratelimit = balanced_dirty_ratelimit * pos_ratio           (7)

where pos_ratio is a negative feedback function that subjects to

1) f(setpoint) = 1.0
2) df/dx < 0

That is, if the dirty pages are ABOVE the setpoint, we throttle each
task a bit more HEAVY than balanced_dirty_ratelimit, so that the dirty
pages are created less fast than they are cleaned, thus DROP to the
setpoints (and the reverse).

Based on (7) and the assumption that both dirty_ratelimit and pos_ratio
remains CONSTANT for the past 200ms, we get

        task_ratelimit_0 = balanced_dirty_ratelimit * pos_ratio         (8)

Putting (8) into (6), we get the formula used in
bdi_update_dirty_ratelimit():

                                                write_bw
        balanced_dirty_ratelimit *= pos_ratio * ----------              (9)
                                                dirty_rate
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

No behavior change.
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

bdi_position_ratio() provides a scale factor to bdi->dirty_ratelimit, so
that the resulted task rate limit can drive the dirty pages back to the
global/bdi setpoints.

Old scheme is,
                                          |
                           free run area  |  throttle area
  ----------------------------------------+---------------------------->
                                    thresh^                  dirty pages

New scheme is,

  ^ task rate limit
  |
  |            *
  |             *
  |              *
  |[free run]      *      [smooth throttled]
  |                  *
  |                     *
  |                         *
  ..bdi->dirty_ratelimit..........*
  |                               .     *
  |                               .          *
  |                               .              *
  |                               .                 *
  |                               .                    *
  +-------------------------------.-----------------------*------------>
                          setpoint^                  limit^  dirty pages

The slope of the bdi control line should be

1) large enough to pull the dirty pages to setpoint reasonably fast

2) small enough to avoid big fluctuations in the resulted pos_ratio and
   hence task ratelimit

Since the fluctuation range of the bdi dirty pages is typically observed
to be within 1-second worth of data, the bdi control line's slope is
selected to be a linear function of bdi write bandwidth, so that it can
adapt to slow/fast storage devices well.

Assume the bdi control line

	pos_ratio = 1.0 + k * (dirty - bdi_setpoint)

where k is the negative slope.

If targeting for 12.5% fluctuation range in pos_ratio when dirty pages
are fluctuating in range

	[bdi_setpoint - write_bw/2, bdi_setpoint + write_bw/2],

we get slope

	k = - 1 / (8 * write_bw)

Let pos_ratio(x_intercept) = 0, we get the parameter used in code:

	x_intercept = bdi_setpoint + 8 * write_bw

The global/bdi slopes are nicely complementing each other when the
system has only one major bdi (indicated by bdi_thresh ~= thresh):

1) slope of global control line    => scaling to the control scope size
2) slope of main bdi control line  => scaling to the writeout bandwidth

so that

- in memory tight systems, (1) becomes strong enough to squeeze dirty
  pages inside the control scope

- in large memory systems where the "gravity" of (1) for pulling the
  dirty pages to setpoint is too weak, (2) can back (1) up and drive
  dirty pages to bdi_setpoint ~= setpoint reasonably fast.

Unfortunately in JBOD setups, the fluctuation range of bdi threshold
is related to memory size due to the interferences between disks.  In
this case, the bdi slope will be weighted sum of write_bw and bdi_thresh.

Given equations

        span = x_intercept - bdi_setpoint
        k = df/dx = - 1 / span

and the extremum values

        span = bdi_thresh
        dx = bdi_thresh

we get

        df = - dx / span = - 1.0

That means, when bdi_dirty deviates bdi_thresh up, pos_ratio and hence
task ratelimit will fluctuate by -100%.

peter: use 3rd order polynomial for the global control line

CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: NJan Kara <jack@suse.cz>
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

Introduce the BDI_DIRTIED counter. It will be used for estimating the
bdi's dirty bandwidth.

CC: Jan Kara <jack@suse.cz>
CC: Michael Rubin <mrubin@google.com>
CC: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>

* 'for-linus' of git://git.infradead.org/users/sameo/mfd-2.6:
  mfd: Fix generic irq chip ack function name for jz4740-adc

* 'for-linus' of git://github.com/tiwai/sound:
  ALSA: hda - Fix a regression of the position-buffer check

* 'perf-urgent-for-linus' of git://tesla.tglx.de/git/linux-2.6-tip:
  perf tools: Fix raw sample reading

Merge branches 'irq-urgent-for-linus', 'x86-urgent-for-linus' and 'sched-urgent-for-linus' of git://tesla.tglx.de/git/linux-2.6-tip

* 'irq-urgent-for-linus' of git://tesla.tglx.de/git/linux-2.6-tip:
  irq: Fix check for already initialized irq_domain in irq_domain_add
  irq: Add declaration of irq_domain_simple_ops to irqdomain.h

* 'x86-urgent-for-linus' of git://tesla.tglx.de/git/linux-2.6-tip:
  x86/rtc: Don't recursively acquire rtc_lock

* 'sched-urgent-for-linus' of git://tesla.tglx.de/git/linux-2.6-tip:
  posix-cpu-timers: Cure SMP wobbles
  sched: Fix up wchan borkage
  sched/rt: Migrate equal priority tasks to available CPUs

David reported:

  Attached below is a watered-down version of rt/tst-cpuclock2.c from
  GLIBC.  Just build it with "gcc -o test test.c -lpthread -lrt" or
  similar.

  Run it several times, and you will see cases where the main thread
  will measure a process clock difference before and after the nanosleep
  which is smaller than the cpu-burner thread's individual thread clock
  difference.  This doesn't make any sense since the cpu-burner thread
  is part of the top-level process's thread group.

  I've reproduced this on both x86-64 and sparc64 (using both 32-bit and
  64-bit binaries).

  For example:

  [davem@boricha build-x86_64-linux]$ ./test
  process: before(0.001221967) after(0.498624371) diff(497402404)
  thread:  before(0.000081692) after(0.498316431) diff(498234739)
  self:    before(0.001223521) after(0.001240219) diff(16698)
  [davem@boricha build-x86_64-linux]$ 

  The diff of 'process' should always be >= the diff of 'thread'.

  I make sure to wrap the 'thread' clock measurements the most tightly
  around the nanosleep() call, and that the 'process' clock measurements
  are the outer-most ones.

  ---
  #include <unistd.h>
  #include <stdio.h>
  #include <stdlib.h>
  #include <time.h>
  #include <fcntl.h>
  #include <string.h>
  #include <errno.h>
  #include <pthread.h>

  static pthread_barrier_t barrier;

  static void *chew_cpu(void *arg)
  {
	  pthread_barrier_wait(&barrier);
	  while (1)
		  __asm__ __volatile__("" : : : "memory");
	  return NULL;
  }

  int main(void)
  {
	  clockid_t process_clock, my_thread_clock, th_clock;
	  struct timespec process_before, process_after;
	  struct timespec me_before, me_after;
	  struct timespec th_before, th_after;
	  struct timespec sleeptime;
	  unsigned long diff;
	  pthread_t th;
	  int err;

	  err = clock_getcpuclockid(0, &process_clock);
	  if (err)
		  return 1;

	  err = pthread_getcpuclockid(pthread_self(), &my_thread_clock);
	  if (err)
		  return 1;

	  pthread_barrier_init(&barrier, NULL, 2);
	  err = pthread_create(&th, NULL, chew_cpu, NULL);
	  if (err)
		  return 1;

	  err = pthread_getcpuclockid(th, &th_clock);
	  if (err)
		  return 1;

	  pthread_barrier_wait(&barrier);

	  err = clock_gettime(process_clock, &process_before);
	  if (err)
		  return 1;

	  err = clock_gettime(my_thread_clock, &me_before);
	  if (err)
		  return 1;

	  err = clock_gettime(th_clock, &th_before);
	  if (err)
		  return 1;

	  sleeptime.tv_sec = 0;
	  sleeptime.tv_nsec = 500000000;
	  nanosleep(&sleeptime, NULL);

	  err = clock_gettime(th_clock, &th_after);
	  if (err)
		  return 1;

	  err = clock_gettime(my_thread_clock, &me_after);
	  if (err)
		  return 1;

	  err = clock_gettime(process_clock, &process_after);
	  if (err)
		  return 1;

	  diff = process_after.tv_nsec - process_before.tv_nsec;
	  printf("process: before(%lu.%.9lu) after(%lu.%.9lu) diff(%lu)\n",
		 process_before.tv_sec, process_before.tv_nsec,
		 process_after.tv_sec, process_after.tv_nsec, diff);
	  diff = th_after.tv_nsec - th_before.tv_nsec;
	  printf("thread:  before(%lu.%.9lu) after(%lu.%.9lu) diff(%lu)\n",
		 th_before.tv_sec, th_before.tv_nsec,
		 th_after.tv_sec, th_after.tv_nsec, diff);
	  diff = me_after.tv_nsec - me_before.tv_nsec;
	  printf("self:    before(%lu.%.9lu) after(%lu.%.9lu) diff(%lu)\n",
		 me_before.tv_sec, me_before.tv_nsec,
		 me_after.tv_sec, me_after.tv_nsec, diff);

	  return 0;
  }

This is due to us using p->se.sum_exec_runtime in
thread_group_cputime() where we iterate the thread group and sum all
data. This does not take time since the last schedule operation (tick
or otherwise) into account. We can cure this by using
task_sched_runtime() at the cost of having to take locks.

This also means we can (and must) do away with
thread_group_sched_runtime() since the modified thread_group_cputime()
is now more accurate and would deadlock when called from
thread_group_sched_runtime().

Aside of that it makes the function safe on 32 bit systems. The old
code added t->se.sum_exec_runtime unprotected. sum_exec_runtime is a
64bit value and could be changed on another cpu at the same time.
Reported-by: NDavid Miller <davem@davemloft.net>
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Cc: stable@kernel.org
Link: http://lkml.kernel.org/r/1314874459.7945.22.camel@twinsTested-by: NDavid Miller <davem@davemloft.net>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

The commit a810364a
    ALSA: hda - Handle -1 as invalid position, too
caused a regression on some machines that require the position-buffer
instead of LPIB, e.g. resulting in noises with mic recording with
PulseAudio.

This patch fixes the detection by delaying the test at the timing as
same as 3.0, i.e. doing the position check only when requested in
azx_position_ok().
Reported-and-tested-by: NRocko Requin <rockorequin@hotmail.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

__find_resource() incorrectly returns a resource window which overlaps
an existing allocated window.  This happens when the parent's
resource-window spans 0x00000000 to 0xffffffff and is entirely allocated
to all its children resource-windows.

__find_resource() looks for gaps in resource allocation among the
children resource windows.  When it encounters the last child window it
blindly tries the range next to one allocated to the last child.  Since
the last child's window ends at 0xffffffff the calculation overflows,
leading the algorithm to believe that any window in the range 0x0000000
to 0xfffffff is available for allocation.  This leads to a conflicting
window allocation.

Michal Ludvig reported this issue seen on his platform.  The following
patch fixes the problem and has been verified by Michal.  I believe this
bug has been there for ages.  It got exposed by git commit 2bbc6942
("PCI : ability to relocate assigned pci-resources")
Signed-off-by: NRam Pai <linuxram@us.ibm.com>
Tested-by: NMichal Ludvig <mludvig@logix.net.nz>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* 'for-linus' of git://github.com/NewDreamNetwork/ceph-client:
  libceph: fix pg_temp mapping update
  libceph: fix pg_temp mapping calculation
  libceph: fix linger request requeuing
  libceph: fix parse options memory leak
  libceph: initialize ack_stamp to avoid unnecessary connection reset

* 'v4l_for_linus' of git://linuxtv.org/mchehab/for_linus:
  [media] omap3isp: Fix build error in ispccdc.c
  [media] uvcvideo: Fix crash when linking entities
  [media] v4l: Make sure we hold a reference to the v4l2_device before using it
  [media] v4l: Fix use-after-free case in v4l2_device_release
  [media] uvcvideo: Set alternate setting 0 on resume if the bus has been reset
  [media] OMAP_VOUT: Fix build break caused by update_mode removal in DSS2

* 'for-linus' of git://git390.marist.edu/pub/scm/linux-2.6:
  [S390] cio: fix cio_tpi ignoring adapter interrupts
  [S390] gmap: always up mmap_sem properly
  [S390] Do not clobber personality flags on exec

* git://github.com/davem330/sparc:
  sparc64: Force the execute bit in OpenFirmware's translation entries.
  sparc: Make '-p' boot option meaningful again.
  sparc, exec: remove redundant addr_limit assignment
  sparc64: Future proof Niagara cpu detection.

* 'drm-intel-fixes' of git://people.freedesktop.org/~keithp/linux:
  drm/i915: FBC off for ironlake and older, otherwise on by default
  drm/i915: Enable SDVO hotplug interrupts for HDMI and DVI
  drm/i915: Enable dither whenever display bpc < frame buffer bpc

Apple Quad G5 has some oddity in it's device-tree which causes the new
generic matching code to fail to relate nodes for PCI-E devices below U4
with their respective struct pci_dev.  This breaks graphics on those
machines among others.

This fixes it using a quirk which copies the node pointer from the host
bridge for the root complex, which makes the generic code work for the
children afterward.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit d5767c53 ("bootup: move 'usermodehelper_enable()' to the end
of do_basic_setup()") moved 'usermodehelper_enable()' to end of
do_basic_setup() to after the initcalls.  But then I get failed to let
uvesafb work on my computer, and lose the splash boot.

So maybe we could start usermodehelper_enable a little early to make
some task work that need eary init with the help of user mode.

[ I would *really* prefer that initcalls not call into user space - even
  the real 'init' hasn't been execve'd yet, after all! But for uvesafb
  it really does look like we don't have much choice.

  I considered doing this when we mount the root filesystem, but
  depending on config options that is in multiple places.  We could do
  the usermode helper enable as a rootfs_initcall()..

  So I'm just using wang yanqing's trivial patch.  It's not wonderful,
  but it's simple and should work.  We should revisit this some day,
  though.      - Linus ]
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Wrong pointer is being passed for raw data sanity checking, when parsing
sample event.

This ends up with invalid event and perf record being stuck in
__perf_session__process_events function during processing build IDs
(process_buildids function).

Following command hangs up in my setup:
	./perf record -e raw_syscalls:sys_enter ls

The fix is to use proper pointer to the raw data instead of the 'u'
union.
Reviewed-by: NDavid Ahern <dsahern@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/1317308709-9474-2-git-send-email-jolsa@redhat.comSigned-off-by: NJiri Olsa <jolsa@redhat.com>
Signed-off-by: NArnaldo Carvalho de Melo <acme@redhat.com>

In the OF 'translations' property, the template TTEs in the mappings
never specify the executable bit.  This is the case even though some
of these mappings are for OF's code segment.

Therefore, we need to force the execute bit on in every mapping.

This problem can only really trigger on Niagara/sun4v machines and the
history behind this is a little complicated.

Previous to sun4v, the sun4u TTE entries lacked a hardware execute
permission bit.  So OF didn't have to ever worry about setting
anything to handle executable pages.  Any valid TTE loaded into the
I-TLB would be respected by the chip.

But sun4v Niagara chips have a real hardware enforced executable bit
in their TTEs.  So it has to be set or else the I-TLB throws an
instruction access exception with type code 6 (protection violation).

We've been extremely fortunate to not get bitten by this in the past.

The best I can tell is that the OF's mappings for it's executable code
were mapped using permanent locked mappings on sun4v in the past.
Therefore, the fact that we didn't have the exec bit set in the OF
translations we would use did not matter in practice.

Thanks to Greg Onufer for helping me track this down.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Doing it just before starting to call into cpu_idle() made a sick kind
of sense only because the original bug we fixed (see commit
288d5abe: "Boot up with usermodehelper disabled") was about problems
with some scheduler data structures not being initialized, and they had
better be initialized at that point.

But it really didn't make any other conceptual sense, and doing it after
the initial "schedule()" call for the idle thread actually opened up a
race: what if the main initialization thread did everything without
needing to sleep, and got all the way into user land too? Without
actually having scheduled back to the idle thread?

Now, in normal circumstances that doesn't ever happen, but it looks like
Richard Cochran triggered exactly that on his ARM IXP4xx machines:

  "I have some ARM IXP4xx based machines that use the two on chip MAC
   ports (aka NPEs).  The NPE needs a firmware in order to function.
   Ever since the following commit [that 288d5abe one], it is no
   longer possible to bring up the interfaces during the init scripts."

with a call trace showing an ioctl coming from user space. Richard says:

  "The init is busybox, and the startup script does mount, syslogd, and
   then ifup, so that all can go by quickly."

The fix is to move the usermodehelper_enable() into the main 'init'
thread, and just put it after we've done all our initcalls.  By then,
everything really should be up, but we've obviously not actually started
the user-mode portion of init yet.
Reported-and-tested-by: NRichard Cochran <richardcochran@gmail.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The incremental map updates have a record for each pg_temp mapping that is
to be add/updated (len > 0) or removed (len == 0).  The old code was
written as if the updates were a complete enumeration; that was just wrong.
Update the code to remove 0-length entries and drop the rbtree traversal.

This avoids misdirected (and hung) requests that manifest as server
errors like

[WRN] client4104 10.0.1.219:0/275025290 misdirected client4104.1:129 0.1 to osd0 not [1,0] in e11/11
Signed-off-by: NSage Weil <sage@newdream.net>

We need to apply the modulo pg_num calculation before looking up a pgid in
the pg_temp mapping rbtree.  This fixes pg_temp mappings, and fixes
(some) misdirected requests that result in messages like

[WRN] client4104 10.0.1.219:0/275025290 misdirected client4104.1:129 0.1 to osd0 not [1,0] in e11/11

on the server and stall make the client block without getting a reply (at
least until the pg_temp mapping goes way, but that can take a long long
time).

Reorder calc_pg_raw() a bit to make more sense.
Signed-off-by: NSage Weil <sage@newdream.net>

* git://github.com/davem330/net:
  ipv6-multicast: Fix memory leak in IPv6 multicast.
  ipv6: check return value for dst_alloc
  net: check return value for dst_alloc
  ipv6-multicast: Fix memory leak in input path.
  bnx2x: add missing break in bnx2x_dcbnl_get_cap
  bnx2x: fix WOL by enablement PME in config space
  bnx2x: fix hw attention handling
  net: fix a typo in Documentation/networking/scaling.txt
  ath9k: Fix a dma warning/memory leak
  rtlwifi: rtl8192cu: Fix unitialized struct
  iwlagn: fix dangling scan request
  batman-adv: do_bcast has to be true for broadcast packets only
  cfg80211: Fix validation of AKM suites
  iwlegacy: do not use interruptible waits
  iwlegacy: fix command queue timeout
  ath9k_hw: Fix Rx DMA stuck for AR9003 chips

* git://bedivere.hansenpartnership.com/git/scsi-rc-fixes-2.6:
  [SCSI] 3w-9xxx: fix iommu_iova leak
  [SCSI] cxgb3i: convert cdev->l2opt to use rcu to prevent NULL dereference
  [SCSI] scsi: qla4xxx needs libiscsi.o
  [SCSI] libsas: fix failure to revalidate domain for anything but the first expander child.
  [SCSI] aacraid: reset should disable MSI interrupt

* 'for-linus' of git://git.kernel.dk/linux-block:
  block: Free queue resources at blk_release_queue()

* 'writeback-for-linus' of git://github.com/fengguang/linux:
  writeback: show raw dirtied_when in trace writeback_single_inode

A kernel crash is observed when a mounted ext3/ext4 filesystem is
physically removed. The problem is that blk_cleanup_queue() frees up
some resources eg by calling elevator_exit(), which are not checked for
in normal operation. So we should rather move these calls to the
destructor function blk_release_queue() as at that point all remaining
references are gone. However, in doing so we have to ensure that any
externally supplied queue_lock is disconnected as the driver might free
up the lock after the call of blk_cleanup_queue(),
Signed-off-by: NHannes Reinecke <hare@suse.de>
Signed-off-by: NJens Axboe <axboe@kernel.dk>