Add Joseph Chan's explanation of the problem and workaround to the
VT6421 magic fix.
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Joseph Chan <JosephChan@via.com.tw>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

This reverts commit 962400e8, which was
entirely bogus.

The code used to multiply the character offset by "vc->vc_cols", and
that's actually correct, because 'd' itself is an 'unsigned short'.  So
the pointer arithmetic already takes the size of a VGA character into
account.  Changing it to use vc_size_row (which is just "vc_cols"
shifted up to take the size of the character into account) ends up
multiplying with the VGA character size twice.

This got reported as bugs for various other subsystems, because what it
actually results in is writing the 16-bit vc_video_erase_char pattern
(usually 0x0720: 0x07 is the default attribute, 0x20 is ASCII space)
into some random other allocation.

So Markus ended up reporting this as a ext4 bug, while to Torsten Kaiser
it looked like a problem with KMS or libata.  Jeff Chua saw it in
different places.

And finally - Justin Mattock had slab poisoning enabled, and saw it as a
slab poison overwritten.  And bisected and reverted this to verify the
buggy commit.
Reported-by: NMarkus Trippelsdorf <markus@trippelsdorf.de>
Reported-by: NTorsten Kaiser <just.for.lkml@googlemail.com>
Reported-by: NJeff Chua <jeff.chua.linux@gmail.com>
Reported-by: NJustin P. Mattock <justinmattock@gmail.com>
Reported-bisected-and-tested-by: NJustin P. Mattock <justinmattock@gmail.com>
Acked-by: NDave Airlie <airlied@redhat.com>
Cc: Frank Pan <frankpzh@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Cursors need to be in the GTT domain when being accessed by the GPU.
Previously this was a fortuitous byproduct of userspace using pwrite()
to upload the image data into the cursor. The redundant clflush was
removed in commit 9b8c4a and so the image was no longer being flushed
out of the caches into main memory. One could also devise a scenario
where the cursor was rendered by the GPU, prior to being attached as the
cursor, resulting in similar corruption due to the missing MI_FLUSH.

Fixes:

  Bug 28335 - Cursor corruption caused by commit 9b8c4a0b
  https://bugs.freedesktop.org/show_bug.cgi?id=28335Signed-off-by: NChris Wilson <chris@chris-wilson.co.uk>
Reported-and-tested-by: NJeff Chua <jeff.chua.linux@gmail.com>
Tested-by: NLinus Torvalds <torvalds@linux-foundation.org>
Reported-by: NAndy Isaacson <adi@hexapodia.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4:
  ext4: Fix remaining racy updates of EXT4_I(inode)->i_flags
  ext4: Make sure the MOVE_EXT ioctl can't overwrite append-only files

A few functions were still modifying i_flags in a racy manner.
Signed-off-by: NDmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>

* 'for-linus' of git://oss.sgi.com/xfs/xfs:
  xfs: improve xfs_isilocked
  xfs: skip writeback from reclaim context
  xfs: remove done roadmap item from xfs-delayed-logging-design.txt
  xfs: fix race in inode cluster freeing failing to stale inodes
  xfs: fix access to upper inodes without inode64
  xfs: fix might_sleep() warning when initialising per-ag tree
  fs/xfs/quota: Add missing mutex_unlock
  xfs: remove duplicated #include
  xfs: convert more trace events to DEFINE_EVENT
  xfs: xfs_trace.c: remove duplicated #include
  xfs: Check new inode size is OK before preallocating
  xfs: clean up xlog_align
  xfs: cleanup log reservation calculactions
  xfs: be more explicit if RT mount fails due to config
  xfs: replace E2BIG with EFBIG where appropriate

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (30 commits)
  X25: remove duplicated #include
  tcp: use correct net ns in cookie_v4_check()
  rps: tcp: fix rps_sock_flow_table table updates
  ppp_generic: fix multilink fragment sizes
  syncookies: remove Kconfig text line about disabled-by-default
  ixgbe: only check pfc bits in hang logic if pfc is enabled
  net: check for refcount if pop a stacked dst_entry
  ixgbe: return IXGBE_ERR_RAR_INDEX when out of range
  act_pedit: access skb->data safely
  sfc: Store port number in net_device::dev_id
  epic100: Test __BIG_ENDIAN instead of (non-existent) CONFIG_BIG_ENDIAN
  tehuti: return -EFAULT on copy_to_user errors
  isdn/kcapi: return -EFAULT on copy_from_user errors
  e1000e: change logical negate to bitwise
  sfc: Get port number from CS_PORT_NUM, not PCI function number
  cls_u32: use skb_header_pointer() to dereference data safely
  TCP: tcp_hybla: Fix integer overflow in slow start increment
  act_nat: fix the wrong checksum when addr isn't in old_addr/mask
  net/fec: fix pm to survive to suspend/resume
  korina: count RX DMA OVR as rx_fifo_error
  ...

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ryusuke/nilfs2:
  nilfs2: remove obsolete declarations of cache constructor and destructor
  nilfs2: fix style issue in nilfs_destroy_cachep

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs-2.6:
  Minix: Clean up left over label
  fix truncate inode time modification breakage
  fix setattr error handling in sysfs, configfs
  fcntl: return -EFAULT if copy_to_user fails
  wrong type for 'magic' argument in simple_fill_super()
  fix the deadlock in qib_fs
  mqueue doesn't need make_bad_inode()

* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
  module: fix bne2 "gave up waiting for init of module libcrc32c"
  module: verify_export_symbols under the lock
  module: move find_module check to end
  module: make locking more fine-grained.
  module: Make module sysfs functions private.
  module: move sysfs exposure to end of load_module
  module: fix kdb's illicit use of struct module_use.
  module: Make the 'usage' lists be two-way

Problem: it's hard to avoid an init routine stumbling over a
request_module these days.  And it's not clear it's always a bad idea:
for example, a module like kvm with dynamic dependencies on kvm-intel
or kvm-amd would be neater if it could simply request_module the right
one.

In this particular case, it's libcrc32c:

	libcrc32c_mod_init
	 crypto_alloc_shash
	  crypto_alloc_tfm
	   crypto_find_alg
	    crypto_alg_mod_lookup
	     crypto_larval_lookup
	      request_module

If another module is waiting inside resolve_symbol() for libcrc32c to
finish initializing (ie. bne2 depends on libcrc32c) then it does so
holding the module lock, and our request_module() can't make progress
until that is released.

Waiting inside resolve_symbol() without the lock isn't all that hard:
we just need to pass the -EBUSY up the call chain so we can sleep
where we don't hold the lock.  Error reporting is a bit trickier: we
need to copy the name of the unfinished module before releasing the
lock.

Other notes:
1) This also fixes a theoretical issue where a weak dependency would allow
   symbol version mismatches to be ignored.
2) We rename use_module to ref_module to make life easier for the only
   external user (the out-of-tree ksplice patches).
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Tim Abbot <tabbott@ksplice.com>
Tested-by: NBrandon Philips <bphilips@suse.de>

It disabled preempt so it was "safe", but nothing stops another module
slipping in before this module is added to the global list now we don't
hold the lock the whole time.

So we check this just after we check for duplicate modules, and just
before we put the module in the global list.

(find_symbol finds symbols in coming and going modules, too).
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>

I think Rusty may have made the lock a bit _too_ finegrained there, and
didn't add it to some places that needed it. It looks, for example, like
PATCH 1/2 actually drops the lock in places where it's needed
("find_module()" is documented to need it, but now load_module() didn't
hold it at all when it did the find_module()).

Rather than adding a new "module_loading" list, I think we should be able
to just use the existing "modules" list, and just fix up the locking a
bit.

In fact, maybe we could just move the "look up existing module" a bit
later - optimistically assuming that the module doesn't exist, and then
just undoing the work if it turns out that we were wrong, just before
adding ourselves to the list.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>

Kay Sievers <kay.sievers@vrfy.org> reports that we still have some
contention over module loading which is slowing boot.

Linus also disliked a previous "drop lock and regrab" patch to fix the
bne2 "gave up waiting for init of module libcrc32c" message.

This is more ambitious: we only grab the lock where we need it.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Cc: Brandon Philips <brandon@ifup.org>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>

These were placed in the header in ef665c1a to get the various
SYSFS/MODULE config combintations to compile.

That may have been necessary then, but it's not now.  These functions
are all local to module.c.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Cc: Randy Dunlap <randy.dunlap@oracle.com>

This means a little extra work, but is more logical: we don't put
anything in sysfs until we're about to put the module into the
global list an parse its parameters.

This also gives us a logical place to put duplicate module detection
in the next patch.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>

Linus changed the structure, and luckily this didn't compile any more.
Reported-by: NStephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Cc: Jason Wessel <jason.wessel@windriver.com>
Cc: Martin Hicks <mort@sgi.com>

When adding a module that depends on another one, we used to create a
one-way list of "modules_which_use_me", so that module unloading could
see who needs a module.

It's actually quite simple to make that list go both ways: so that we
not only can see "who uses me", but also see a list of modules that are
"used by me".

In fact, we always wanted that list in "module_unload_free()": when we
unload a module, we want to also release all the other modules that are
used by that module.  But because we didn't have that list, we used to
first iterate over all modules, and then iterate over each "used by me"
list of that module.

By making the list two-way, we simplify module_unload_free(), and it
allows for some trivial fixes later too.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (cleaned & rebased)

Remove duplicated #include('s) in drivers/net/wan/x25_asy.c
Signed-off-by: NHuang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Its better to make a route lookup in appropriate namespace.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

I believe a moderate SYN flood attack can corrupt RFS flow table
(rps_sock_flow_table), making RPS/RFS much less effective.

Even in a normal situation, server handling short lived sessions suffer
from bad steering for the first data packet of a session, if another SYN
packet is received for another session.

We do following action in tcp_v4_rcv() :

	sock_rps_save_rxhash(sk, skb->rxhash);

We should _not_ do this if sk is a LISTEN socket, as about each
packet received on a LISTEN socket has a different rxhash than
previous one.
 -> RPS_NO_CPU markers are spread all over rps_sock_flow_table.

Also, it makes sense to protect sk->rxhash field changes with socket
lock (We currently can change it even if user thread owns the lock
and might use rxhash)

This patch moves sock_rps_save_rxhash() to a sock locked section,
and only for non LISTEN sockets.
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fix bug in multilink fragment size calculation introduced by
commit 9c705260
"ppp: ppp_mp_explode() redesign"
Signed-off-by: NBen McKeegan <ben@netservers.co.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

syncookies default to on since
e994b7c9
(tcp: Don't make syn cookies initial setting depend on CONFIG_SYSCTL).
Signed-off-by: NFlorian Westphal <fw@strlen.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Only check pfc bits in hang logic if PFC is enabled.  Previously,
if DCB was enabled but PFC was disabled the incorrect pause
bits would be checked.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Acked-by: NDon Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

xfrm triggers a warning if dst_pop() drops a refcount
on a noref dst. This patch changes dst_pop() to
skb_dst_pop(). skb_dst_pop() drops the refcnt only
on a refcounted dst. Also we don't clone the child
dst_entry, so it is not refcounted and we can use
skb_dst_set_noref() in xfrm_output_one().
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>
Signed-off-by: NEric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
  Input: wacom - add Cintiq 21UX2 and Intuos4 WL
  Input: ads7846 - fix compiler warning in ads7846_probe()
  Input: tps6507x-ts - a couple work queue cleanups
  Input: s3c2410_ts - tone down logging
  Input: s3c2410_ts - fix build error due to ADC Kconfig rename

* git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6: (23 commits)
  sh: Make intc messages consistent via pr_fmt.
  sh: make sure static declaration on ms7724se
  sh: make sure static declaration on mach-migor
  sh: make sure static declaration on mach-ecovec24
  sh: make sure static declaration on mach-ap325rxa
  clocksource: sh_cmt: compute mult and shift before registration
  clocksource: sh_tmu: compute mult and shift before registration
  sh: PIO disabling for x3proto and urquell.
  sh: mach-sdk7786: conditionally disable PIO support.
  sh: support for platforms without PIO.
  usb: r8a66597-hcd pio to mmio accessor conversion.
  usb: gadget: r8a66597-udc pio to mmio accessor conversion.
  usb: gadget: m66592-udc pio to mmio accessor conversion.
  sh: add romImage MMCIF boot for sh7724 and Ecovec V2
  sh: add boot code to MMCIF driver header
  sh: prepare MMCIF driver header file
  sh: allow romImage data between head.S and the zero page
  sh: Add support MMCIF for ecovec
  sh: remove duplicated #include
  input: serio: disable i8042 for non-cayman sh platforms.
  ...

* 'linux_next' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/i7core: (83 commits)
  i7core_edac: Better describe the supported devices
  Add support for Westmere to i7core_edac driver
  i7core_edac: don't free on success
  i7core_edac: Add support for X5670
  Always call i7core_[ur]dimm_check_mc_ecc_err
  i7core_edac: fix memory leak of i7core_dev
  EDAC: add __init to i7core_xeon_pci_fixup
  i7core_edac: Fix wrong device id for channel 1 devices
  i7core: add support for Lynnfield alternate address
  i7core_edac: Add initial support for Lynnfield
  i7core_edac: do not export static functions
  edac: fix i7core build
  edac: i7core_edac produces undefined behaviour on 32bit
  i7core_edac: Use a more generic approach for probing PCI devices
  i7core_edac: PCI device is called NONCORE, instead of NOCORE
  i7core_edac: Fix ringbuffer maxsize
  i7core_edac: First store, then increment
  i7core_edac: Better parse "any" addrmask
  i7core_edac: Use a lockless ringbuffer
  edac: Create an unique instance for each kobj
  ...

* 'v4l_for_2.6.35' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-2.6: (87 commits)
  V4L/DVB: ivtv: Timing tweaks and code re-order to try and improve stability
  V4L/DVB: ivtv: Avoid accidental video standard change
  V4L/DVB: ivtvfb : Module load / unload fixes
  V4L/DVB: cx2341x: Report correct temporal setting for log-status
  V4L/DVB: cx18, cx23885, v4l2 doc, MAINTAINERS: Update Andy Walls' email address
  V4L/DVB: drivers/media: Eliminate a NULL pointer dereference
  V4L/DVB: dvb-core: Fix ULE decapsulation bug
  V4L/DVB: Bug fix: make IR work again for dm1105
  V4L/DVB: media/IR: nec-decoder needs to select BITREV
  V4L/DVB: video/saa7134: change dprintk() to i2cdprintk()
  V4L/DVB: video/saa7134: remove duplicate break
  V4L/DVB: IR/imon: add auto-config for 0xffdc rf device
  V4L/DVB: IR/imon: clean up usage of bools
  V4L/DVB: em28xx: remove unneeded null checks
  V4L/DVB: ngene: remove unused #include <linux/version.h>
  V4L/DVB: ak881x needs slab.h
  V4L/DVB: FusionHDTV: Use quick reads for I2C IR device probing
  V4L/DVB: Technotrend S2-3200 ships with a TT 1500 remote
  V4L/DVB: drivers/media: Use kzalloc
  V4L/DVB: m920x: Select simple tuner
  ...

* 'for-linus' of git://git.kernel.dk/linux-2.6-block: (27 commits)
  block: make blk_init_free_list and elevator_init idempotent
  block: avoid unconditionally freeing previously allocated request_queue
  pipe: change /proc/sys/fs/pipe-max-pages to byte sized interface
  pipe: change the privilege required for growing a pipe beyond system max
  pipe: adjust minimum pipe size to 1 page
  block: disable preemption before using sched_clock()
  cciss: call BUG() earlier
  Preparing 8.3.8rc2
  drbd: Reduce verbosity
  drbd: use drbd specific ratelimit instead of global printk_ratelimit
  drbd: fix hang on local read errors while disconnected
  drbd: Removed the now empty w_io_error() function
  drbd: removed duplicated #includes
  drbd: improve usage of MSG_MORE
  drbd: need to set socket bufsize early to take effect
  drbd: improve network latency, TCP_QUICKACK
  drbd: Revert "drbd: Create new current UUID as late as possible"
  brd: support discard
  Revert "writeback: fix WB_SYNC_NONE writeback from umount"
  Revert "writeback: ensure that WB_SYNC_NONE writeback with sb pinned is sync"
  ...

* 'rc-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild-2.6:
  gconfig: fix build failure on fedora 13

* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging-2.6: (27 commits)
  Staging: sep: return -EFAULT on copy_to_user errors
  Staging: rc2860: return -EFAULT on copy_to_user errors
  Staging: Eliminate a NULL pointer dereference
  staging: Use GFP_ATOMIC when a lock is held
  Staging: comedi - correct parameter gainlkup for DAQCard-6024E in driver ni_mio_cs.c
  Staging: comedi: fixing ni_labpc to mite dependancy
  Staging: wlags49_h2, wlags49_h25: fixed Kconfig dependencies
  Staging: phison: depends on ATA_BMDMA
  Staging: iio-utils: fix memory overflow for dynamically allocateded memory to hold filename
  Staging: adis16255: add proper section markings to hotplug funcs
  Staging: adis16255: fix typo in Kconfig
  Staging: batman-adv: Don't allocate icmp packet with GFP_KERNEL
  Staging: batman-adv: Don't call free_netdev twice
  Staging: batman-adv: Call unregister_netdev on failures to get rtnl lock
  Staging: batman-adv: fix rogue packets on shutdown
  Staging: add MSM framebuffer driver
  Staging: comedi: fixing ni_tio to mite PCI dependancy
  Staging: comedi: fix 8255 and DAS08 Kconfig dependancies.
  Staging: comedi: For COMEDI_BUFINFO, check access to command
  Staging: comedi: COMEDI_BUFINFO with no async - report no bytes read or written
  ...

* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core-2.6:
  fix setattr error handling in sysfs, configfs
  kobject: free memory if netlink_kernel_create() fails
  lib/kobject_uevent.c: fix CONIG_NET=n warning

* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty-2.6:
  serial: add support for various Titan PCI cards
  vt_ioctl: return -EFAULT on copy_from_user errors
  serial: altera_uart: Proper section for altera_uart_remove
  tty: fix a little bug in scrup, vt.c
  altera_uart: Simplify altera_uart_console_putc
  altera_uart: Don't take spinlock in already protected functions
  TTY/n_gsm: potential double lock
  serial: bfin_5xx: fix typo in IER check
  serial: bfin_5xx: IRDA is not affected by anomaly 05000230
  serial_cs: add and sort IDs for serial and modem cards
  msm_serial: fix serial on trout

* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6:
  USB: unbind all interfaces before rebinding them
  USB: serial: digi_acceleport: Eliminate a NULL pointer dereference
  usb: fix ehci_hcd build failure when both generic-OF and xilinx is selected
  USB: cdc-acm: fix resource reclaim in error path of acm_probe
  USB: ftdi_sio: fix DTR/RTS line modes
  USB: s3c-hsotg: Ensure FIFOs are fully flushed after layout
  USB: s3c-hsotg: SoftDisconnect minimum 3ms
  USB: s3c-hsotg: Ensure TX FIFO addresses setup when initialising FIFOs
  USB: s3c_hsotg: define USB_GADGET_DUALSPEED in Kconfig
  USB: s3c: Enable soft disconnect during initialization
  USB: xhci: Print NEC firmware version.
  USB: xhci: Wait for host to start running.
  USB: xhci: Wait for controller to be ready after reset.
  USB: isp1362: fix inw warning on Blackfin systems
  USB: mos7840: fix null-pointer dereference

Remove a BUG_ON for when interrupts are disabled during an MMC request.

During boot, interrupts can be disabled when a request is made, causing
this bug to be triggered.  In reality, there's no reason this should halt
the kernel, as the driver has proved reliable in spite of disabled
interrupts, and additionally, there's nothing in this code that would
require interrupts to be enabled.

The only setup I've managed to make it trigger on is on the HTC Herald
during bootup when the driver is built into the kernel (mostly because
that's all I have).  I believe it's related to the fact that on bootup I
get many timeout errors on "CMD5" while initializing the card.  Each CMD5
timeout triggers that bug (I changed it to a WARN_ON to get it to boot in)
due to the fact that part of the timeout code involves sending the request
again.  With interrupts turned off, that BUG would be triggered.
Signed-off-by: NCory Maccarrone <darkstar6262@gmail.com>
Acked-by: NTony Lindgren <tony@atomide.com>
Cc: <linux-mmc@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Greg Thelen reported recent Johannes's stack diet patch makes kernel hang.
 His test is following.

  mount -t cgroup none /cgroups -o memory
  mkdir /cgroups/cg1
  echo $$ > /cgroups/cg1/tasks
  dd bs=1024 count=1024 if=/dev/null of=/data/foo
  echo $$ > /cgroups/tasks
  echo 1 > /cgroups/cg1/memory.force_empty

Actually, This OOM hard to try logic have been corrupted since following
two years old patch.

	commit a41f24ea
	Author: Nishanth Aravamudan <nacc@us.ibm.com>
	Date:   Tue Apr 29 00:58:25 2008 -0700

	    page allocator: smarter retry of costly-order allocations

Original intention was "return success if the system have shrinkable zones
though priority==0 reclaim was failure".  But the above patch changed to
"return nr_reclaimed if .....".  Oh, That forgot nr_reclaimed may be 0 if
priority==0 reclaim failure.

And Johannes's patch 0aeb2339 ("vmscan: remove all_unreclaimable scan
control") made it more corrupt.  Originally, priority==0 reclaim failure
on memcg return 0, but this patch changed to return 1.  It totally
confused memcg.

This patch fixes it completely.
Reported-by: NGreg Thelen <gthelen@google.com>
Signed-off-by: NKOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Acked-by: NJohannes Weiner <hannes@cmpxchg.org>
Acked-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Tested-by: NGreg Thelen <gthelen@google.com>
Acked-by: NBalbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The commit 80b5184c ("kernel/: convert cpu
notifier to return encapsulate errno value") changed the return value of
cpu notifier callbacks.

Those callbacks don't return NOTIFY_BAD on failures anymore.  But there
are a few callbacks which are called directly at init time and checking
the return value.

I forgot to change BUG_ON checking by the direct callers in the commit.
Signed-off-by: NAkinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Child groups should have a greater depth than their parents.  Prior to
this change, the parent would incorrectly report zero memory usage for
child cgroups when use_hierarchy is enabled.

test script:
  mount -t cgroup none /cgroups -o memory
  cd /cgroups
  mkdir cg1

  echo 1 > cg1/memory.use_hierarchy
  mkdir cg1/cg11

  echo $$ > cg1/cg11/tasks
  dd if=/dev/zero of=/tmp/foo bs=1M count=1

  echo
  echo CHILD
  grep cache cg1/cg11/memory.stat

  echo
  echo PARENT
  grep cache cg1/memory.stat

  echo $$ > tasks
  rmdir cg1/cg11 cg1
  cd /
  umount /cgroups

Using fae9c791, a recent patch that changed alloc_css_id() depth computation,
the parent incorrectly reports zero usage:
  root@ubuntu:~# ./test
  1+0 records in
  1+0 records out
  1048576 bytes (1.0 MB) copied, 0.0151844 s, 69.1 MB/s

  CHILD
  cache 1048576
  total_cache 1048576

  PARENT
  cache 0
  total_cache 0

With this patch, the parent correctly includes child usage:
  root@ubuntu:~# ./test
  1+0 records in
  1+0 records out
  1048576 bytes (1.0 MB) copied, 0.0136827 s, 76.6 MB/s

  CHILD
  cache 1052672
  total_cache 1052672

  PARENT
  cache 0
  total_cache 1052672
Signed-off-by: NGreg Thelen <gthelen@google.com>
Acked-by: NPaul Menage <menage@google.com>
Acked-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: NLi Zefan <lizf@cn.fujitsu.com>
Cc: <stable@kernel.org>		[2.6.34.x]
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>