Remove the clocks from the list of regulators to correct the value of
MAX77686_REG_MAX which is used in the regulator driver to represent the no.
of regulators present in max77686.
Signed-off-by: NYadwinder Singh Brar <yadi.brar@samsung.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

in hw design, 800 is mainly for pmic control, while 805 for audio.
but there are 3 registers which controls class D speaker property,
and they are defined in 800 i2c client domain. so 805 codec driver
needs to use 800 i2c client to access class D speaker reg for
audio path management. so add this workaround for the purpose to
let 805 access 800 i2c in some scenario.
Signed-off-by: NQiao Zhou <zhouqiao@marvell.com>
Reviewed-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

88PM800 and 88PM805 are two discrete chips used for power management.
Hardware designer can use them together or only one of them according
to requirement.

88pm80x.c provides common i2c driver handling for both 800 and
805, such as i2c_driver init, regmap init, read/write api etc.

88pm800.c handles specifically for 800, such as chip init, irq
init/handle, mfd device register, including rtc, onkey, regulator(
to be add later) etc. besides that, 800 has three i2c device, one
regular i2c client, two other i2c dummy for gpadc and power purpose.

88pm805.c handles specifically for 805, such as chip init, irq
init/handle, mfd device register, including codec, headset/mic detect
etc.

the i2c operation of both 800 and 805 are via regmap, and 88pm80x-i2c
exported a group of r/w bulk r/w and bits set API for facility.
Signed-off-by: NQiao Zhou <zhouqiao@marvell.com>
Reviewed-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

These registers will be used in future devices.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Rather than open coding the enable GPIO control in the MFD core use the
API to push the management on to the regulator driver. The immediate
advantage is slight for most systems but this will in future allow device
configurations where an external regulator is used for DCVDD.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Now this driver is using regmap API, the iolock mutex is not used and
can be removed.
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Pre-regulator of 88pm8606 is mainly for support charging based on vbus,
it needs to be enabled for charging battery, and will be disabled in
some exception condition like over-temp.
Add the pre-regulator device init data and resource for mfd subdev.
Signed-off-by: NJett.Zhou <jtzhou@marvell.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Add flag to platform data to enable external 32-kHz crystal oscillator
(or square wave) input.

The tps6591x can use either an internal 32-kHz RC oscillator or an
external crystal (or square wave) to generate the 32-kHz clock.

The default setting depends on the selected boot mode. In boot mode 00
the internal RC oscillator is used at power-on, but the external crystal
oscillator (or square wave) can be enabled by clearing the ck32k_ctrl
flag in the device control register.

Note that there is no way to switch from the external crystal oscillator
to the internal RC oscillator.
Signed-off-by: NJohan Hovold <jhovold@gmail.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Add irq domain support for max8997 interrupts. The reverse mapping method
used is linear mapping since the sub-drivers of max8997 such as regulator
and charger drivers can use the max8997 irq_domain to get the linux irq
number for max8997 interrupts. All uses of irq_base in platform data and
max8997 driver private data are removed.
Reviwed-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: NMyungJoo Ham <myungjoo.ham@samsung.com>
Acked-by: NGrant Likely <grant.likely@secretlab.ca>
Signed-off-by: NThomas Abraham <thomas.abraham@linaro.org>
Signed-off-by: NChanwoo Choi <cw00.choi@samsung.com>
Signed-off-by: NKyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Applying a succinct description to the of_compatible variable recently
added to the mfd_cell struct. Also link to the documentation page where
more information can be found about compatible properties.
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Force the Modem wakeup by asserting the CaWakeReq signal before the
hostaccess_req/ack ping-pong sequence. The Awake_req signal is de-asserted
asserted at the same time than the hostaccess_req. Return error on failure
case so that the client using this can take appropiate steps.
Signed-off-by: NArun Murthy <arun.murthy@stericsson.com>
Acked-by: NLinus Walleij <linus.walleij@stericsson.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Since none of the users now reference the cache directly we can happily
remove the custom cache code and rely on the regmap cache.

For simplicity we don't bother with the register defaults tables but
instead read the defaults from the device - regmap is capable of doing
this, unlike our old cache infrastructure. This saves a lot of code and
allows us to cache the device revision information too.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Use the most simple possible transformation on the existing code so keep
the table sitting around, further patches in this series will delete the
existing cache code - the main purpose of this patch is to ensure that
we always have a cache for bisection.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Signed-off-by: NChris Rattray <crattray@opensource.wolfsonmicro.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

During Device Tree enablement of the ab8500 and db8500-prcmu drivers,
a decision was made to omit registration through the MFD API and use
Device Tree directly. However, because MFD devices have a different
address space and the ab8500 and db8500 both use I2C to communicate,
this causes issues with address translation during execution of
of_platform_populate(). So the solution is to make the MFD core aware
of Device Tree and have it assign the correct node pointers instead.

To make this work the MFD core also needs to be awere of IRQ domains,
as Device Tree insists on IRQ domain compatibility. So, instead of
providing an irq-base via platform code, in the DT case we simply
look up the IRQ domain and map to the correct virtual IRQ.
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

As the AB8500 is an IRQ controller in its own right, here we provide
the AB8500 driver with IRQ domain support. This is required if we wish
to reference any of its IRQs from a platform's Device Tree.

Cc: Naga Radheshy <naga.radheshy@stericsson.com>
Cc: Mattias Wallin <mattias.wallin@stericsson.com>
Cc: Daniel Willerud <daniel.willerud@stericsson.com>
Signed-off-by: NLee Jones <lee.jones@linaro.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Now this driver is using regmap APIs, the iolock mutex is not used and can be
removed.
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

This patch is device driver for MAX77686 chip.
MAX77686 is PMIC and includes regulator and rtc on it.
This driver is core of MAX77686 chip, so provides common support for
accessing on-chip devices. It uses irq_domain to manage irq and regmap
to read/write data to its register with i2c bus.
Signed-off-by: NChiwoong Byun <woong.byun@samsung.com>
Signed-off-by: NJonghwa Lee <jonghwa3.lee@samsung.com>
Signed-off-by: NMyungjoo Ham <myungjoo.ham@samsung.com>
Signed-off-by: NKyungmin Park <kyungmin.park@samsung.com>
Reviewed-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

The documentation didn't actually mention how to enable no_new_privs.
This also adds a note about possible interactions between
no_new_privs and LSMs (i.e. why teaching systemd to set no_new_privs
is not necessarily a good idea), and it references the new docs
from include/linux/prctl.h.
Suggested-by: NRob Landley <rob@landley.net>
Signed-off-by: NAndy Lutomirski <luto@amacapital.net>
Acked-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NJames Morris <james.l.morris@oracle.com>

Ocfs2 uses kiocb.*private as a flag of unsigned long size. In
commit a11f7e63 ocfs2: serialize unaligned aio, the unaligned
io flag is involved in it to serialize the unaligned aio. As
*private is not initialized in init_sync_kiocb() of do_sync_write(),
this unaligned io flag may be unexpectly set in an aligned dio.
And this will cause OCFS2_I(inode)->ip_unaligned_aio decreased
to -1 in ocfs2_dio_end_io(), thus the following unaligned dio
will hang forever at ocfs2_aiodio_wait() in ocfs2_file_aio_write().
Signed-off-by: NJunxiao Bi <junxiao.bi@oracle.com>
Cc: stable@vger.kernel.org
Acked-by: NJeff Moyer <jmoyer@redhat.com>
Signed-off-by: NJoel Becker <jlbec@evilplan.org>

Prune this down to just the struct kvm_irqfd so we can avoid
changing function definition for every flag or field we use.
Signed-off-by: NAlex Williamson <alex.williamson@redhat.com>
Acked-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>

It was noticed recently that when we send data on a transport, its possible that
we might bundle a sack that arrived on a different transport.  While this isn't
a major problem, it does go against the SHOULD requirement in section 6.4 of RFC
2960:

 An endpoint SHOULD transmit reply chunks (e.g., SACK, HEARTBEAT ACK,
   etc.) to the same destination transport address from which it
   received the DATA or control chunk to which it is replying.  This
   rule should also be followed if the endpoint is bundling DATA chunks
   together with the reply chunk.

This patch seeks to correct that.  It restricts the bundling of sack operations
to only those transports which have moved the ctsn of the association forward
since the last sack.  By doing this we guarantee that we only bundle outbound
saks on a transport that has received a chunk since the last sack.  This brings
us into stricter compliance with the RFC.

Vlad had initially suggested that we strictly allow only sack bundling on the
transport that last moved the ctsn forward.  While this makes sense, I was
concerned that doing so prevented us from bundling in the case where we had
received chunks that moved the ctsn on multiple transports.  In those cases, the
RFC allows us to select any of the transports having received chunks to bundle
the sack on.  so I've modified the approach to allow for that, by adding a state
variable to each transport that tracks weather it has moved the ctsn since the
last sack.  This I think keeps our behavior (and performance), close enough to
our current profile that I think we can do this without a sysctl knob to
enable/disable it.
Signed-off-by: NNeil Horman <nhorman@tuxdriver.com>
CC: Vlad Yaseivch <vyasevich@gmail.com>
CC: David S. Miller <davem@davemloft.net>
CC: linux-sctp@vger.kernel.org
Reported-by: NMichele Baldessari <michele@redhat.com>
Reported-by: Nsorin serban <sserban@redhat.com>
Acked-by: NVlad Yasevich <vyasevich@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fix kernel-doc warning.  This struct member was removed in commit
87568264 ("irq: Remove irq_chip->release()") so remove its
associated kernel-doc entry also.

  Warning(include/linux/irq.h:338): Excess struct/union/enum/typedef member 'release' description in 'irq_chip'
Signed-off-by: NRandy Dunlap <rdunlap@xenotime.net>
Cc: Richard Weinberger <richard@nod.at>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit 2603efa3 ("bug.h: Fix up powerpc build regression") corrected
the powerpc build case and extended the __ASSEMBLY__ guards, but it also
got caught in pre-processor hell accidentally matching the else case of
CONFIG_BUG resulting in the BUG disabled case tripping up on
-Werror=implicit-function-declaration.

It's not possible to __ASSEMBLY__ guard the entire file as architecture
code needs to get at the BUGFLAG_WARNING definition in the GENERIC_BUG
case, but the rest of the CONFIG_BUG=y/n case needs to be guarded.

Rather than littering endless __ASSEMBLY__ checks in each of the if/else
cases we just move the BUGFLAG definitions up under their own
GENERIC_BUG test and then shove everything else under one big
__ASSEMBLY__ guard.

Build tested on all of x86 CONFIG_BUG=y, CONFIG_BUG=n, powerpc (due to
it's dependence on BUGFLAG definitions in assembly code), and sh (due to
not bringing in linux/kernel.h to satisfy the taint flag definitions used
by the generic bug code).

Hopefully that's the end of the corner cases and I can abstain from ever
having to touch this infernal header ever again.
Reported-by: NFengguang Wu <fengguang.wu@intel.com>
Tested-by: NFengguang Wu <wfg@linux.intel.com>
Acked-by: NRandy Dunlap <rdunlap@xenotime.net>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Several forthcoming Wolfson devices are based on a common platform
known as Arizona allowing a great deal of reuse of driver code. This
patch adds core support for these devices.

In order to handle systems which do not use the generic clock API a
simple wrapper for the 32kHz clock domain in the devices is provided.
Once the generic clock API is widely available this code will be moved
over to use that.

For simplicity some WM5102 specific code is included in the core driver,
the effort involved in splitting the device out isn't worth it.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Several forthcoming Wolfson devices are based on a common platform
known as Arizona allowing a great deal of reuse of driver code. This
patch adds register definitions for these devices.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Several bug reports have been received recently for USB mass-storage
devices that don't handle READ CAPACITY(16) commands properly.  They
report bogus sizes, in some cases becoming unusable as a result.

The bugs were triggered by commit
09b6b51b (SCSI & usb-storage: add
flags for VPD pages and REPORT LUNS), which caused usb-storage to stop
overriding the SCSI level reported by devices.  By default, the sd
driver will try READ CAPACITY(16) first for any device whose level is
above SCSI_SPC_2.

It seems likely that any device large enough to require the use of
READ CAPACITY(16) (i.e., 2 TB or more) would be able to handle READ
CAPACITY(10) commands properly.  Indeed, I don't know of any devices
that don't handle READ CAPACITY(10) properly.

Therefore this patch (as1559) adds a new flag telling the sd driver
to try READ CAPACITY(10) before READ CAPACITY(16), and sets this flag
for every USB mass-storage device.  If a device really is larger than
2 TB, sd will fall back to READ CAPACITY(16) just as it used to.

This fixes Bugzilla #43391.
Signed-off-by: NAlan Stern <stern@rowland.harvard.edu>
Acked-by: NHans de Goede <hdegoede@redhat.com>
CC: "James E.J. Bottomley" <JBottomley@parallels.com>
CC: Matthew Dharm <mdharm-usb@one-eyed-alien.net>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This comment is well out of date.
Signed-off-by: NDave Airlie <airlied@redhat.com>

Guard vga_switcheroo.h against multiple inclusion.
Signed-off-by: NOzan Çağlayan <ozancag@gmail.com>
Signed-off-by: NDave Airlie <airlied@redhat.com>

viresh.kumar@st.com email-id doesn't exist anymore as I have left the
company.  Replace ST's id with viresh.linux@gmail.com.

It also updates .mailmap file to fix address for 'git shortlog'
Signed-off-by: NViresh Kumar <viresh.linux@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In the x86 32bit PAE CONFIG_TRANSPARENT_HUGEPAGE=y case while holding the
mmap_sem for reading, cmpxchg8b cannot be used to read pmd contents under
Xen.

So instead of dealing only with "consistent" pmdvals in
pmd_none_or_trans_huge_or_clear_bad() (which would be conceptually
simpler) we let pmd_none_or_trans_huge_or_clear_bad() deal with pmdvals
where the low 32bit and high 32bit could be inconsistent (to avoid having
to use cmpxchg8b).

The only guarantee we get from pmd_read_atomic is that if the low part of
the pmd was found null, the high part will be null too (so the pmd will be
considered unstable).  And if the low part of the pmd is found "stable"
later, then it means the whole pmd was read atomically (because after a
pmd is stable, neither MADV_DONTNEED nor page faults can alter it anymore,
and we read the high part after the low part).

In the 32bit PAE x86 case, it is enough to read the low part of the pmdval
atomically to declare the pmd as "stable" and that's true for THP and no
THP, furthermore in the THP case we also have a barrier() that will
prevent any inconsistent pmdvals to be cached by a later re-read of the
*pmd.
Signed-off-by: NAndrea Arcangeli <aarcange@redhat.com>
Cc: Jonathan Nieder <jrnieder@gmail.com>
Cc: Ulrich Obergfell <uobergfe@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Hugh Dickins <hughd@google.com>
Cc: Larry Woodman <lwoodman@redhat.com>
Cc: Petr Matousek <pmatouse@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Jan Beulich <jbeulich@suse.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@gmail.com>
Tested-by: NAndrew Jones <drjones@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

On arches that do not support this_cpu_cmpxchg_double() slab_lock is used
to do atomic cmpxchg() on double word which contains page->_count.  The
page count can be changed from get_page() or put_page() without taking
slab_lock.  That corrupts page counter.

Fix it by moving page->_count out of cmpxchg_double data.  So that slub
does no change it while updating slub meta-data in struct page.

[akpm@linux-foundation.org: use standard comment layout, tweak comment text]
Reported-by: NAmey Bhide <abhide@nicira.com>
Signed-off-by: NPravin B Shelar <pshelar@nicira.com>
Acked-by: NChristoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NKay Sievers <kay@vrfy.org>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Reported-by: NRandy Dunlap <rdunlap@xenotime.net>
Reported-by: NFengguang Wu <wfg@linux.intel.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The asm-generic/bug.h __ASSEMBLY__ guarding is completely bogus, which
tripped up the powerpc build when the kernel.h include was added:

	In file included from include/asm-generic/bug.h:5:0,
			 from arch/powerpc/include/asm/bug.h:127,
			 from arch/powerpc/kernel/head_64.S:31:
	include/linux/kernel.h:44:0: warning: "ALIGN" redefined [enabled by default]
	include/linux/linkage.h:57:0: note: this is the location of the previous definition
	include/linux/sysinfo.h: Assembler messages:
	include/linux/sysinfo.h:7: Error: Unrecognized opcode: `struct'
	include/linux/sysinfo.h:8: Error: Unrecognized opcode: `__kernel_long_t'

Moving the __ASSEMBLY__ guard up and stashing the kernel.h include under
it fixes this up, as well as covering the case the original fix was
attempting to handle.
Tested-by: NStephen Rothwell <sfr@canb.auug.org.au>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit 5963e317 ("ftrace/x86: Do not change stacks in DEBUG when
calling lockdep") prevented lockdep calls from the int3 breakpoint handler
from reseting the stack if a function that was called was in the process
of being converted for tracing and had a breakpoint on it. The idea is,
before calling the lockdep code, do a load_idt() to the special IDT that
kept the breakpoint stack from reseting. This worked well as a quick fix
for this kernel release, until a certain config caused a lockup in the
function tracer start up tests.

Investigating it, I found that the load_idt that was used to prevent
the int3 from changing stacks was itself being traced!

Even though the config had CONFIG_OPTIMIZE_INLINING disabled, and
all 'inline' tags were set to always inline, there were still cases that
it did not inline! This was caused by CONFIG_PARAVIRT_GUEST, where it
would add a pointer to the native_load_idt() which made that function
to be traced.

Commit 45959ee7 ("ftrace: Do not function trace inlined functions")
only touched the 'inline' tags when CONFIG_OPMITIZE_INLINING was enabled.
PARAVIRT_GUEST shows that this was not enough and we need to also
mark always_inline with notrace as well.
Reported-by: NFengguang Wu <wfg@linux.intel.com>
Tested-by: NFengguang Wu <wfg@linux.intel.com>
Signed-off-by: NSteven Rostedt <rostedt@goodmis.org>

Currently there is a 'chicken and egg' issue when the DS is also the mounted
MDS. The nfs_match_client() reference from nfs4_set_ds_client bumps the
cl_count, the nfs_client is not freed at umount, and nfs4_deviceid_purge_client
is not called to dereference the MDS usage of a deviceid which holds a
reference to the DS nfs_client.  The result is the umount program returns,
but the nfs_client is not freed, and the cl_session hearbeat continues.

The MDS (and all other nfs mounts) lose their last nfs_client reference in
nfs_free_server when the last nfs_server (fsid) is umounted.
The file layout DS lose their last nfs_client reference in destroy_ds
when the last deviceid referencing the data server is put and destroy_ds is
called. This is triggered by a call to nfs4_deviceid_purge_client which
removes references to a pNFS deviceid used by an MDS mount.

The fix is to track how many pnfs enabled filesystems are mounted from
this server, and then to purge the device id cache once that count reaches
zero.
Reported-by: NJorge Mora <Jorge.Mora@netapp.com>
Reported-by: NAndy Adamson <andros@netapp.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Signed-off-by: NRémi Denis-Courmont <remi@remlab.net>
Cc: Sakari Ailus <sakari.ailus@nokia.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fix warnings on some architectures/configs (not on x86):

include/linux/vga_switcheroo.h:28:30: warning: 'struct pci_dev' declared inside parameter list [enabled by default]
include/linux/vga_switcheroo.h:28:30: warning: its scope is only this definition or declaration, which is probably not what you want [enabled by default]
Signed-off-by: NRandy Dunlap <rdunlap@xenotime.net>
Cc:	Takashi Iwai <tiwai@suse.de>
Reported-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NDave Airlie <airlied@redhat.com>

This reverts commit 2a0c451a.

It causes crashes, because now ip6_null_entry is used before
it is initialized.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Minchan Kim reports that when a system has many swap areas, and tmpfs
swaps out to the ninth or more, shmem_getpage_gfp()'s attempts to read
back the page cannot locate it, and the read fails with -ENOMEM.

Whoops.  Yes, I blindly followed read_swap_header()'s pte_to_swp_entry(
swp_entry_to_pte()) technique for determining maximum usable swap
offset, without stopping to realize that that actually depends upon the
pte swap encoding shifting swap offset to the higher bits and truncating
it there.  Whereas our radix_tree swap encoding leaves offset in the
lower bits: it's swap "type" (that is, index of swap area) that was
truncated.

Fix it by reducing the SWP_TYPE_SHIFT() in swapops.h, and removing the
broken radix_to_swp_entry(swp_to_radix_entry()) from read_swap_header().

This does not reduce the usable size of a swap area any further, it
leaves it as claimed when making the original commit: no change from 3.0
on x86_64, nor on i386 without PAE; but 3.0's 512GB is reduced to 128GB
per swapfile on i386 with PAE.  It's not a change I would have risked
five years ago, but with x86_64 supported for ten years, I believe it's
appropriate now.

Hmm, and what if some architecture implements its swap pte with offset
encoded below type? That would equally break the maximum usable swap
offset check.  Happily, they all follow the same tradition of encoding
offset above type, but I'll prepare a check on that for next.
Reported-and-Reviewed-and-Tested-by: NMinchan Kim <minchan@kernel.org>
Signed-off-by: NHugh Dickins <hughd@google.com>
Cc: stable@vger.kernel.org [3.1, 3.2, 3.3, 3.4]
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>