Create a "taskstats" interface based on generic netlink (NETLINK_GENERIC
family), for getting statistics of tasks and thread groups during their
lifetime and when they exit.  The interface is intended for use by multiple
accounting packages though it is being created in the context of delay
accounting.

This patch creates the interface without populating the fields of the data
that is sent to the user in response to a command or upon the exit of a task.
Each accounting package interested in using taskstats has to provide an
additional patch to add its stats to the common structure.

[akpm@osdl.org: cleanups, Kconfig fix]
Signed-off-by: NShailabh Nagar <nagar@us.ibm.com>
Signed-off-by: NBalbir Singh <balbir@in.ibm.com>
Cc: Jes Sorensen <jes@sgi.com>
Cc: Peter Chubb <peterc@gelato.unsw.edu.au>
Cc: Erich Focht <efocht@ess.nec.de>
Cc: Levent Serinol <lserinol@gmail.com>
Cc: Jay Lan <jlan@engr.sgi.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Initialization code related to collection of per-task "delay" statistics which
measure how long it had to wait for cpu, sync block io, swapping etc.  The
collection of statistics and the interface are in other patches.  This patch
sets up the data structures and allows the statistics collection to be
disabled through a kernel boot parameter.
Signed-off-by: NShailabh Nagar <nagar@watson.ibm.com>
Signed-off-by: NBalbir Singh <balbir@in.ibm.com>
Cc: Jes Sorensen <jes@sgi.com>
Cc: Peter Chubb <peterc@gelato.unsw.edu.au>
Cc: Erich Focht <efocht@ess.nec.de>
Cc: Levent Serinol <lserinol@gmail.com>
Cc: Jay Lan <jlan@engr.sgi.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Teach special (recursive) locking code to the lock validator.  Has no effect
on non-lockdep kernels.
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Do 'make oldconfig' and accept all the defaults for new config options -
reboot into the kernel and if everything goes well it should boot up fine and
you should have /proc/lockdep and /proc/lockdep_stats files.

Typically if the lock validator finds some problem it will print out
voluminous debug output that begins with "BUG: ..." and which syslog output
can be used by kernel developers to figure out the precise locking scenario.

What does the lock validator do?  It "observes" and maps all locking rules as
they occur dynamically (as triggered by the kernel's natural use of spinlocks,
rwlocks, mutexes and rwsems).  Whenever the lock validator subsystem detects a
new locking scenario, it validates this new rule against the existing set of
rules.  If this new rule is consistent with the existing set of rules then the
new rule is added transparently and the kernel continues as normal.  If the
new rule could create a deadlock scenario then this condition is printed out.

When determining validity of locking, all possible "deadlock scenarios" are
considered: assuming arbitrary number of CPUs, arbitrary irq context and task
context constellations, running arbitrary combinations of all the existing
locking scenarios.  In a typical system this means millions of separate
scenarios.  This is why we call it a "locking correctness" validator - for all
rules that are observed the lock validator proves it with mathematical
certainty that a deadlock could not occur (assuming that the lock validator
implementation itself is correct and its internal data structures are not
corrupted by some other kernel subsystem).  [see more details and conditionals
of this statement in include/linux/lockdep.h and
Documentation/lockdep-design.txt]

Furthermore, this "all possible scenarios" property of the validator also
enables the finding of complex, highly unlikely multi-CPU multi-context races
via single single-context rules, increasing the likelyhood of finding bugs
drastically.  In practical terms: the lock validator already found a bug in
the upstream kernel that could only occur on systems with 3 or more CPUs, and
which needed 3 very unlikely code sequences to occur at once on the 3 CPUs.
That bug was found and reported on a single-CPU system (!).  So in essence a
race will be found "piecemail-wise", triggering all the necessary components
for the race, without having to reproduce the race scenario itself!  In its
short existence the lock validator found and reported many bugs before they
actually caused a real deadlock.

To further increase the efficiency of the validator, the mapping is not per
"lock instance", but per "lock-class".  For example, all struct inode objects
in the kernel have inode->inotify_mutex.  If there are 10,000 inodes cached,
then there are 10,000 lock objects.  But ->inotify_mutex is a single "lock
type", and all locking activities that occur against ->inotify_mutex are
"unified" into this single lock-class.  The advantage of the lock-class
approach is that all historical ->inotify_mutex uses are mapped into a single
(and as narrow as possible) set of locking rules - regardless of how many
different tasks or inode structures it took to build this set of rules.  The
set of rules persist during the lifetime of the kernel.

To see the rough magnitude of checking that the lock validator does, here's a
portion of /proc/lockdep_stats, fresh after bootup:

 lock-classes:                            694 [max: 2048]
 direct dependencies:                  1598 [max: 8192]
 indirect dependencies:               17896
 all direct dependencies:             16206
 dependency chains:                    1910 [max: 8192]
 in-hardirq chains:                      17
 in-softirq chains:                     105
 in-process chains:                    1065
 stack-trace entries:                 38761 [max: 131072]
 combined max dependencies:         2033928
 hardirq-safe locks:                     24
 hardirq-unsafe locks:                  176
 softirq-safe locks:                     53
 softirq-unsafe locks:                  137
 irq-safe locks:                         59
 irq-unsafe locks:                      176

The lock validator has observed 1598 actual single-thread locking patterns,
and has validated all possible 2033928 distinct locking scenarios.

More details about the design of the lock validator can be found in
Documentation/lockdep-design.txt, which can also found at:

   http://redhat.com/~mingo/lockdep-patches/lockdep-design.txt

[bunk@stusta.de: cleanups]
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Generic lock debugging:

 - generalized lock debugging framework. For example, a bug in one lock
   subsystem turns off debugging in all lock subsystems.

 - got rid of the caller address passing (__IP__/__IP_DECL__/etc.) from
   the mutex/rtmutex debugging code: it caused way too much prototype
   hackery, and lockdep will give the same information anyway.

 - ability to do silent tests

 - check lock freeing in vfree too.

 - more finegrained debugging options, to allow distributions to
   turn off more expensive debugging features.

There's no separate 'held mutexes' list anymore - but there's a 'held locks'
stack within lockdep, which unifies deadlock detection across all lock
classes.  (this is independent of the lockdep validation stuff - lockdep first
checks whether we are holding a lock already)

Here are the current debugging options:

CONFIG_DEBUG_MUTEXES=y
CONFIG_DEBUG_LOCK_ALLOC=y

which do:

 config DEBUG_MUTEXES
          bool "Mutex debugging, basic checks"

 config DEBUG_LOCK_ALLOC
         bool "Detect incorrect freeing of live mutexes"
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

s390's console_init must enable interrupts, but early_boot_irqs_on() gets
called later.  To avoid problems move console_init() after local_irq_enable().
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: NIngo Molnar <mingo@elte.hu>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

We're not reay to take a timer interrupt until timekeeping_init() has run.
But time_init() will start the time interrupt and if it is called with
local interrupts enabled we'll immediately take an interrupt and die.

Fix that by running timekeeping_init() prior to time_init().

We don't know _why_ local interrupts got enabled on Jesse Brandeburg's
machine.  That's a separate as-yet-unsolved problem.  THe patch adds a little
bit of debugging to detect that.

This whole requirement that local interrupts be held off during early boot
keeps on biting us.
Signed-off-by: NJohn Stultz <johnstul@us.ibm.com>
Cc: Jesse Brandeburg <jesse.brandeburg@gmail.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

include/linux/version.h contained both actual KERNEL version
and UTS_RELEASE that contains a subset from git SHA1 for when
kernel was compiled as part of a git repository.
This had the unfortunate side-effect that all files including version.h
would be recompiled when some git changes was made due to changes SHA1.
Split it out so we keep independent parts in separate files.

Also update checkversion.pl script to no longer check for UTS_RELEASE.
Signed-off-by: NSam Ravnborg <sam@ravnborg.org>

Fix the INIT_ENV_ARG_LIMIT dependencies to what seems to have been
intended.

Spotted by Jean-Luc Leger.
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Acked-by: NPaolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Presently, smp_processor_id() isn't necessarily set up until setup_arch().
But it's used in boot_cpu_init() and printk() and perhaps in other places,
prior to setup_arch() being called.

So provide a new smp_setup_processor_id() which is called before anything
else, wire it up for Voyager (which boots on a CPU other than #0, and broke).

Cc: James Bottomley <James.Bottomley@steeleye.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The remaining counters in page_state after the zoned VM counter patches
have been applied are all just for show in /proc/vmstat.  They have no
essential function for the VM.

We use a simple increment of per cpu variables.  In order to avoid the most
severe races we disable preempt.  Preempt does not prevent the race between
an increment and an interrupt handler incrementing the same statistics
counter.  However, that race is exceedingly rare, we may only loose one
increment or so and there is no requirement (at least not in kernel) that
the vm event counters have to be accurate.

In the non preempt case this results in a simple increment for each
counter.  For many architectures this will be reduced by the compiler to a
single instruction.  This single instruction is atomic for i386 and x86_64.
 And therefore even the rare race condition in an interrupt is avoided for
both architectures in most cases.

The patchset also adds an off switch for embedded systems that allows a
building of linux kernels without these counters.

The implementation of these counters is through inline code that hopefully
results in only a single instruction increment instruction being emitted
(i386, x86_64) or in the increment being hidden though instruction
concurrency (EPIC architectures such as ia64 can get that done).

Benefits:
- VM event counter operations usually reduce to a single inline instruction
  on i386 and x86_64.
- No interrupt disable, only preempt disable for the preempt case.
  Preempt disable can also be avoided by moving the counter into a spinlock.
- Handling is similar to zoned VM counters.
- Simple and easily extendable.
- Can be omitted to reduce memory use for embedded use.

References:

RFC http://marc.theaimsgroup.com/?l=linux-kernel&m=113512330605497&w=2
RFC http://marc.theaimsgroup.com/?l=linux-kernel&m=114988082814934&w=2
local_t http://marc.theaimsgroup.com/?l=linux-kernel&m=114991748606690&w=2
V2 http://marc.theaimsgroup.com/?t=115014808400007&r=1&w=2
V3 http://marc.theaimsgroup.com/?l=linux-kernel&m=115024767022346&w=2
V4 http://marc.theaimsgroup.com/?l=linux-kernel&m=115047968808926&w=2Signed-off-by: NChristoph Lameter <clameter@sgi.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NJörn Engel <joern@wohnheim.fh-wedel.de>
Signed-off-by: NAdrian Bunk <bunk@stusta.de>

Core functions for the rt-mutex subsystem.
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

- add a proper prototype for the following global function:
  - buffer_init()

- make the following needlessly global function static:
  - end_buffer_async_write()
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Also fixes up all files that #include it.
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

This patch removes the devfs code from the init/ directory.
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Architecture specific configs like this have no business at all
in init/Kconfig. This prevents it from being set on x86-64

Pointed out by H.Peter Anvin
Signed-off-by: NAndi Kleen <ak@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

These are the generic bits needed to enable reliable stack traces based
on Dwarf2-like (.eh_frame) unwind information. Subsequent patches will
enable x86-64 and i386 to make use of this.

Thanks to Andi Kleen and Ingo Molnar, who pointed out several possibilities
for improvement.
Signed-off-by: NJan Beulich <jbeulich@novell.com>
Signed-off-by: NAndi Kleen <ak@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This patch ensures that initramfs overwrites work correctly, even when dealing
with device nodes of different types.  Furthermore, when replacing a file
which already exists, we must make very certain that we truncate the existing
file.
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>
Cc: Michael Neuling <mikey@neuling.org>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Modify the update_wall_time function so it increments time using the
clocksource abstraction instead of jiffies.  Since the only clocksource driver
currently provided is the jiffies clocksource, this should result in no
functional change.  Additionally, a timekeeping_init and timekeeping_resume
function has been added to initialize and maintain some of the new timekeping
state.

[hirofumi@mail.parknet.co.jp: fixlet]
Signed-off-by: NJohn Stultz <johnstul@us.ibm.com>
Signed-off-by: NOGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Make makes sysctl non-optional unless EMBEDDED is set.  There are a number
of interfaces exposed via sysctl, enough that it has to be considered core
kernel functionality at this point.
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

In this implementation, audit registers inotify watches on the parent
directories of paths specified in audit rules.  When audit's inotify
event handler is called, it updates any affected rules based on the
filesystem event.  If the parent directory is renamed, removed, or its
filesystem is unmounted, audit removes all rules referencing that
inotify watch.

To keep things simple, this implementation limits location-based
auditing to the directory entries in an existing directory.  Given
a path-based rule for /foo/bar/passwd, the following table applies:

    passwd modified -- audit event logged
    passwd replaced -- audit event logged, rules list updated
    bar renamed     -- rule removed
    foo renamed     -- untracked, meaning that the rule now applies to
		       the new location

Audit users typically want to have many rules referencing filesystem
objects, which can significantly impact filtering performance.  This
patch also adds an inode-number-based rule hash to mitigate this
situation.

The patch is relative to the audit git tree:
http://kernel.org/git/?p=linux/kernel/git/viro/audit-current.git;a=summary
and uses the inotify kernel API:
http://lkml.org/lkml/2006/6/1/145Signed-off-by: NAmy Griffis <amy.griffis@hp.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

This makes it possible to change two options which were hardcoded sofar.
1. Any symbol can now take the role of CONFIG_MODULES
2. The more useful option is to change the list of default file names,
   which kconfig uses to load the base configuration if .config isn't
   available.
Signed-off-by: NRoman Zippel <zippel@linux-m68k.org>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NSam Ravnborg <sam@ravnborg.org>

With this patch, "root=mtd3" and "root=mtd:foo" work for a JFFS2 rootfs.
Signed-off-by: NJoern Engel <joern@wh.fh-wedel.de>

When we fail to mount from a valid root device list out the filesystems we
have tried to mount it with.  This gives the user vital diagnostics as to
what is missing from their kernel.

For example in the fragment below the kernel does not have CRAMFS compiled
into the kernel and yet appears to recognise it at the RAMDISK detect
stage.  Later the mount fails as we don't have the filesystem.

  RAMDISK: cramfs filesystem found at block 0
  RAMDISK: Loading 1604KiB [1 disk] into ram disk... done.
  XFS: bad magic number
  XFS: SB validate failed
  No filesystem could mount root, tried: reiserfs ext3 ext2 msdos vfat
    iso9660 jfs xfs
  Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(8,1)
Signed-off-by: NAndy Whitcroft <apw@shadowen.org>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Copy the filenames of hardlinks when inserting them into the hash, since
the "name" pointer may point to scratch space (name_buf).  Not doing so
results in corruption if the scratch space is later overwritten: the wrong
file may be hardlinked, or, if the scratch space contains garbage, the link
will fail and a 0-byte file will be created instead.
Signed-off-by: NMark Huang <mlhuang@cs.princeton.edu>
Acked-by: NAl Viro <viro@zeniv.linux.org.uk>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This was already a bad plan when I argued against adding it in the first
place. Good riddance.
Signed-off-by: NDavid Woodhouse <dwmw2@infradead.org>

Suppress the initcall-return-value warnings unless initcall_debug was
specified.

They do find bugs, but they're extremely small ones and as Andi points out,
people get distressed.

Cc: Andi Kleen <ak@muc.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NAndi Kleen <ak@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Move the DOUBLEFAULT option from the top-level menu to the EMBEDDED menu.
Only applicable to X86_32.
Signed-off-by: NRandy Dunlap <rdunlap@xenotime.net>
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

replaces for_each_cpu with for_each_possible_cpu().
Signed-off-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

I noticed that after boot with an initrd in 2.6.16 the rootfs had:

--w-r-xr-T    1 root     root      6241141 Jan  1  1970 initrd.image

Which is caused by a small typo:
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Since the addition of boot_cpu_init(), fixup_cpu_present_map() has been a
no-op.  That's because fixup_cpu_present_map() won't touch cpu_present_map if
it has any bits set, and boot_cpu_init() sets a bit.

So remove fixup_cpu_present_map().

A consequence of this (actually of the boot_cpu_init() change) is that the
architecture _must_ populate cpu_present_map itself (probably in
smp_prepare_cpus()).  fixup_cpu_present_map() won't do it any more.

If the architecture doesn't do this, it'll only bring up a single CPU.

The other side effect (though less serious) is that smp_prepare_boot_cpu() no
longer needs to mark the boot cpu in the online and present maps -
boot_cpu_init() does that for everyone (to make early printks work).
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The ROOT_DEV comment is no longer accurate, it now seems to be
initialized in init/do_mounts.c.
Signed-off-by: NFlorin Malita <fmalita@gmail.com>
Signed-off-by: NAdrian Bunk <bunk@stusta.de>

Initramfs initrd images do not need a ramdisk device, so remove this
restriction in Kconfig.  BLK_DEV_RAM=n saves about 13k on i386.  Also
without ramdisk device there's no need for "dry run", so initramfs unpacks
much faster.

People using cramfs, squashfs, or gzipped ext2/minix initrd images are
probably smart enough not to turn off ramdisk support by accident.

Cc: Al Viro <viro@ftp.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This patch adds a proper prototype for setup_arch() in init.h.

This patch is based on a patch by Ben Dooks <ben-linux@fluff.org>.
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

We presently ignore the return values from initcalls.  But that can carry
useful debugging information.  So print it out if it's non-zero.

It turns out the -ENODEV happens quite a lot, due to built-in drivers which
have no hardware to drive.  So suppress that unless initcall_debug was
specified.

Also make the warning message more friendly by printing the name of the
initcall function.

Also drop the KERN_DEBUG from the initcall_debug message.  If we specified
inticall_debug then we obviously want to see the messages.
Acked-by: NPaul Jackson <pj@sgi.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

MODULE_PARM was actually breaking: recent gcc version optimize them out as
unused.  It's time to replace the last users, which are generally in the
most unloved drivers anyway.
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The meaning of MS_VERBOSE is backwards; if the bit is set, it really means,
"don't be verbose".  This is confusing and counter-intuitive.

In addition, there is also no way to set the MS_VERBOSE flag in the
mount(8) program in util-linux, but interesting, it does define options
which would do the right thing if MS_SILENT were defined, which
unfortunately we do not:

#ifdef MS_SILENT
  { "quiet",    0, 0, MS_SILENT    },   /* be quiet  */
  { "loud",     0, 1, MS_SILENT    },   /* print out messages. */
#endif

So the obvious fix is to deprecate the use of MS_VERBOSE and replace it
with MS_SILENT.
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Original patch from Paul Mundt, sysfs parts removed by me since they
were broken.
Signed-off-by: NJens Axboe <axboe@suse.de>