THe CFI driver in 2.6.24 kernel is broken.  Not so intensive read/write
operations cause incomplete writes which lead to kernel panics in JFFS2.

We investigated the issue - it is caused by bug in FL_SHUTDOWN parsing code.
Sometimes chip returns -EIO as if it is in FL_SHUTDOWN state when it should
wait in FL_PONT (error in order of conditions).

The following patch fixes the bug in state parsing code of CFI.  Also I've
added comments to notify developers if they want to add new case in future.
Signed-off-by: Alexey Korolev <akorolev@infradead.org>
Reviewed-by: Joern Engel <joern@logfs.org>
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

A boot option for the memory controller was discussed on lkml.  It is a good
idea to add it, since it saves memory for people who want to turn off the
memory controller.

By default the option is on for the following two reasons:

1. It provides compatibility with the current scheme where the memory
   controller turns on if the config option is enabled
2. It allows for wider testing of the memory controller, once the config
   option is enabled

We still allow the create, destroy callbacks to succeed, since they are not
aware of boot options.  We do not populate the directory will memory resource
controller specific files.
Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Paul Menage <menage@google.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Sudhir Kumar <skumar@linux.vnet.ibm.com>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

The effects of cgroup_disable=foo are:

- foo isn't auto-mounted if you mount all cgroups in a single hierarchy
- foo isn't visible as an individually mountable subsystem

As a result there will only ever be one call to foo->create(), at init time;
all processes will stay in this group, and the group will never be mounted on
a visible hierarchy.  Any additional effects (e.g.  not allocating metadata)
are up to the foo subsystem.

This doesn't handle early_init subsystems (their "disabled" bit isn't set be,
but it could easily be extended to do so if any of the early_init systems
wanted it - I think it would just involve some nastier parameter processing
since it would occur before the command-line argument parser had been run.

Hugh said:

  Ballpark figures, I'm trying to get this question out rather than
  processing the exact numbers: CONFIG_CGROUP_MEM_RES_CTLR adds 15% overhead
  to the affected paths, booting with cgroup_disable=memory cuts that back to
  1% overhead (due to slightly bigger struct page).

  I'm no expert on distros, they may have no interest whatever in
  CONFIG_CGROUP_MEM_RES_CTLR=y; and the rest of us can easily build with or
  without it, or apply the cgroup_disable=memory patches.

Unix bench's execl test result on x86_64 was

== just after boot without mounting any cgroup fs.==
mem_cgorup=off : Execl Throughput       43.0     3150.1      732.6
mem_cgroup=on  : Execl Throughput       43.0     2932.6      682.0
==

[lizf@cn.fujitsu.com: fix boot option parsing]
Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Paul Menage <menage@google.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Sudhir Kumar <skumar@linux.vnet.ibm.com>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86:
  x86: revert assign IRQs to hpet timer
  x86: tsc prevent time going backwards
  xen: Clear PG_pinned in release_{pt,pd}()
  xen: Do not pin/unpin PMD pages
  xen: refactor xen_{alloc,release}_{pt,pd}()
  x86, agpgart: scary messages are fortunately obsolete
  xen: fix grant table bug
  x86: fix breakage of vSMP irq operations
  x86: print message if nmi_watchdog=2 cannot be enabled
  x86: fix nmi_watchdog=2 on Pentium-D CPUs

Long overdue update of the m68k defconfigs
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

The default defconfig should be one from arch/m68k/configs/

arch/m68k/defconfig was not exactly identical to amiga_defconfig but
also considering how long they have been without any update that doesn't
seem to have been on purpose.
Signed-off-by: Adrian Bunk <adrian.bunk@movial.fi>
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

* 'upstream-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev:
  pata_ali: disable ATAPI DMA
  libata: ATA_12/16 doesn't fall into ATAPI_MISC
  libata: uninline atapi_cmd_type()
  libata: fix IDENTIFY order in ata_bus_probe()

Mikulas Patocka noted that the optimization where we check if a buffer
was already dirty (and we avoid re-dirtying it) was not really SMP-safe.

Since the read of the old status was not synchronized with anything, an
aggressive CPU re-ordering of memory accesses might have moved that read
up to before the data was even written to the buffer, and another CPU
that cleaned it again, causing the newly dirty state to never actually
hit the disk.

Admittedly this would probably never trigger in practice, but it's still
wrong.

Mikulas sent a patch that fixed the problem, but I dislike the subtlety
of the whole optimization, so this is an alternate fix that is more
explicit about the particular SMP ordering for the optimization, and
separates out the speculative reads of the buffer state into its own
conditional (and makes the memory barrier only happen if we are likely
to actually hit the optimized case in the first place).

I considered removing the optimization entirely, but Andrew argued for
it's continued existence. I'm a push-over.

Cc: Mikulas Patocka <mikulas@artax.karlin.mff.cuni.cz>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Commit f63fd7e2 ("parport_pc: detection
for SuperIO IT87XX POST") only released the IO port region on success,
not when the probe for the IT87XX chip failed.

That caused not only a reserved region to leak, but also caused an oops
when the driver module was unloaded and somebody tried to cat
/proc/ioports - because the string that was assigned to the IO port
region was a static string in the module virtual address area.
Reported-by: Lubos Lunak <l.lunak@suse.cz>
Cc: Jan Kara <jack@suse.cz>
Cc: Petr Cvek <petr.cvek@tul.cz>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

The commits:

commit 37a47db8
Author: Balaji Rao <balajirrao@gmail.com>
Date:   Wed Jan 30 13:30:03 2008 +0100

    x86: assign IRQs to HPET timers, fix

and

commit e3f37a54
Author: Balaji Rao <balajirrao@gmail.com>
Date:   Wed Jan 30 13:30:03 2008 +0100

    x86: assign IRQs to HPET timers

have been identified to cause a regression on some platforms due to
the assignement of legacy IRQs which makes the legacy devices
connected to those IRQs disfunctional.

Revert them.

This fixes http://bugzilla.kernel.org/show_bug.cgi?id=10382Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

We already catch most of the TSC problems by sanity checks, but there
is a subtle bug which has been in the code for ever. This can cause
time jumps in the range of hours.

This was reported in:
     http://lkml.org/lkml/2007/8/23/96
and
     http://lkml.org/lkml/2008/3/31/23

I was able to reproduce the problem with a gettimeofday loop test on a
dual core and a quad core machine which both have sychronized
TSCs. The TSCs seems not to be perfectly in sync though, but the
kernel is not able to detect the slight delta in the sync check. Still
there exists an extremly small window where this delta can be observed
with a real big time jump. So far I was only able to reproduce this
with the vsyscall gettimeofday implementation, but in theory this
might be observable with the syscall based version as well.

CPU 0 updates the clock source variables under xtime/vyscall lock and
CPU1, where the TSC is slighty behind CPU0, is reading the time right
after the seqlock was unlocked.

The clocksource reference data was updated with the TSC from CPU0 and
the value which is read from TSC on CPU1 is less than the reference
data. This results in a huge delta value due to the unsigned
subtraction of the TSC value and the reference value. This algorithm
can not be changed due to the support of wrapping clock sources like
pm timer.

The huge delta is converted to nanoseconds and added to xtime, which
is then observable by the caller. The next gettimeofday call on CPU1
will show the correct time again as now the TSC has advanced above the
reference value.

To prevent this TSC specific wreckage we need to compare the TSC value
against the reference value and return the latter when it is larger
than the actual TSC value.

I pondered to mark the TSC unstable when the readout is smaller than
the reference value, but this would render an otherwise good and fast
clocksource unusable without a real good reason.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Cc: xen-devel@lists.xensource.com
Cc: Mark McLoughlin <markmc@redhat.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

i.e. with this simple test case:

    int fd = open("/dev/zero", O_RDONLY);
    munmap(mmap((void *)0x40000000, 0x1000_LEN, PROT_READ, MAP_PRIVATE, fd, 0), 0x1000);
    close(fd);

we currently get:

   kernel BUG at arch/x86/xen/enlighten.c:678!
   ...
   EIP is at xen_release_pt+0x79/0xa9
   ...
   Call Trace:
    [<c041da25>] ? __pmd_free_tlb+0x1a/0x75
    [<c047a192>] ? free_pgd_range+0x1d2/0x2b5
    [<c047a2f3>] ? free_pgtables+0x7e/0x93
    [<c047b272>] ? unmap_region+0xb9/0xf5
    [<c047c1bd>] ? do_munmap+0x193/0x1f5
    [<c047c24f>] ? sys_munmap+0x30/0x3f
    [<c0408cce>] ? syscall_call+0x7/0xb
    =======================

and xen complains:

  (XEN) mm.c:2241:d4 Mfn 1cc37 not pinned

Further details at:

  https://bugzilla.redhat.com/436453Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Cc: xen-devel@lists.xensource.com
Cc: Mark McLoughlin <markmc@redhat.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Cc: xen-devel@lists.xensource.com
Cc: Mark McLoughlin <markmc@redhat.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Fix obsolete printks in aperture-64. We used not to handle missing
agpgart, but we handle it okay now.
Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

fix memory corruption and crash due to mis-sized grant table.

A PV OS has two grant table data structures: the grant table itself
and a free list.  The free list is composed of an array of pages,
which grow dynamically as the guest OS requires more grants.  While
the grant table contains 8-byte entries, the free list contains 4-byte
entries.  So we have half as many pages in the free list than in the
grant table.

There was a bug in the free list allocation code. The free list was
indexed as if it was the same size as the grant table.  But it's only
half as large.  So memory got corrupted, and I was seeing crashes in
the slab allocator later on.

Taken from:

  http://xenbits.xensource.com/linux-2.6.18-xen.hg?rev/4018c0da3360Signed-off-by: Michael Abd-El-Malek <mabdelmalek@cmu.edu>
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

25-rc* stopped working with CONFIG_X86_VSMP on vSMP machines.

Looks like the vsmp irq ops got accidentally removed during merge of x86_64
pvops in 2.6.25. -- commit 6abcd98f removed
vsmp irq ops.

Tested with both CONFIG_X86_VSMP and without CONFIG_X86_VSMP, on vSMP and non
vSMP x86_64 machines.

Please apply.
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

right now if there's no CPU support for nmi_watchdog=2 we'll just
refuse it silently.

print a useful warning.
Signed-off-by: Ingo Molnar <mingo@elte.hu>

implement nmi_watchdog=2 on this class of CPUs:

  cpu family      : 15
  model           : 6
  model name      : Intel(R) Pentium(R) D CPU 3.00GHz

the watchdog's ->setup() method is safe anyway, so if the CPU
cannot support it we'll bail out safely.
Signed-off-by: Ingo Molnar <mingo@elte.hu>

ATAPI DMA just doesn't work reliably on pata_ali.  The IDE driver can
do it but for some mysterious reason, pata_ali can't.  This patch
disables it by default and makes the driver whine during
initialization.  "pata_ali.atapi_dma" parameter is added so that user
can bypass the workaround.
Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

SAT passthrus don't really fit into ATAPI_MISC class.  SAT passthru
commands always transfer multiple of 512 bytes and variable length
response is not allowed.  This patch creates a separate category -
ATAPI_PASS_THRU - for these.

This fixes HSM violation on "hdparm -I".
Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

Uninline atapi_cmd_type().  It doesn't really have to be inline and
more case will be added which need to access unexported libata
variable.
Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

Commit f58229f8 accidentally made
ata_bus_probe() not use reverse order probing.  Fix it.

There currently isn't any PATA driver which uses obsolete
ata_bus_probe() path, so this patch is mainly for correctness.
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/selinux-2.6:
  selinux: prevent rentry into the FS

This avoids using wrmsr on MSR_IA32_DEBUGCTLMSR when it's not needed.
No wrmsr ever needs to be done if noone has ever used block stepping.

Without this change, using ptrace on 2.6.25 on an x86 KVM guest
will tickle KVM's missing support for the MSR and crash the guest
kernel.  Though host KVM is the buggy one, this makes for a regression
in the guest behavior from 2.6.24->2.6.25 that we can easily avoid.

I also corrected some bad whitespace.
Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
  Input: appletouch - add product IDs for the 4th generation MacBooks

* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/paulus/powerpc:
  [POWERPC] Fix MPC5200 (not B!) device tree so FEC ethernet works
  [POWERPC] mpc5200: Amalgamated DTS fixes and updates
  [POWERPC] Fix rtas_flash procfs interface
  [POWERPC] Fix deadlock with mmu_hash_lock in hash_page_sync
  [POWERPC] Fix iSeries hard irq enabling regression
  [POWERPC] Fix CPM2 SCC1 clock initialization.
  [POWERPC] Fix defconfigs so we dont set both GENRTC and RTCLIB
  [POWERPC] fsldma: Use compatiable binding as spec
  [POWERPC] sata_fsl: reduce compatibility to fsl,pq-sata
  [POWERPC] 83xx: enable usb in 837x rdb and 83xx defconfigs
  [POWERPC] 83xx: Fix wrong USB phy type in mpc837xrdb dts

Signed-off-by: Sven Schnelle <svens@stackframe.org>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: Sven Schnelle <svens@stackframe.org>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

The loop block driver is careful to mask __GFP_IO|__GFP_FS out of its
mapping_gfp_mask, to avoid hangs under memory pressure.  But nowadays
it uses splice, usually going through __generic_file_splice_read.  That
must use mapping_gfp_mask instead of GFP_KERNEL to avoid those hangs.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

BUG fix.  Keep us from re-entering the fs when we aren't supposed to.

See discussion at
http://marc.info/?t=120716967100004&r=1&w=2Signed-off-by: Josef Bacik <jbacik@redhat.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: James Morris <jmorris@namei.org>

This gets the FEC ethernet driver working again on the lite5200
platform.

The FEC driver is also compatible with the MPC5200, not only with the
MPC5200B, so this adds a suitable entry to the driver's match list.
Furthermore this adds the settings for the PHY in the dts file for the
Lite5200.  Note, that this is not exactly the same as in the
Lite5200B, because the PHY is located at f0003000:01 for the 5200, and
at :00 for the 5200B.  This was tested on a Lite5200 and a Lite5200B,
both booted a kernel via tftp and mounted the root via nfs
successfully.
Signed-off-by: René Bürgel <r.buergel@unicontrol.de>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: Paul Mackerras <paulus@samba.org>

DTS updates that fix booting problems on mpc5200-based boards:
- change to ethernet reg property
- addition of mdio and phy nodes
- removal of pci node (Motion-Pro board)

Other DTS updates:
- update i2c device tree nodes
- add lpb bus node and flash device (without partitions defined)
- add rtc i2c nodes
Signed-off-by: Marian Balakowicz <m8@semihalf.com>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: Paul Mackerras <paulus@samba.org>

Handling of the proc_dir_entry->count was changed in 2.6.24-rc5.
After this change, the default value for pde->count is 1 and not 0 as
before.  Therefore, if we want to check whether our procfs file is
already opened (already in use), we have to check if pde->count is
greater than 2 rather than 1.
Signed-off-by: Maxim Shchetynin <maxim@de.ibm.com>
Signed-off-by: Jens Osterkamp <jens@de.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>

hash_page_sync() takes and releases the low level mmu hash
lock in order to sync with other processors disposing of page
tables.  Because that lock can be needed to service hash misses
triggered by interrupt handlers, taking it must be done with
interrupts off.  However, hash_page_sync() appears to be called
with interrupts enabled, thus causing occasional deadlocks.

We fix it by making sure hash_page_sync() masks interrupts while
holding the lock.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>

A subtle bug sneaked into iSeries recently.  On this platform, we must
not normally clear MSR:EE (the hardware external interrupt enable)
except for short periods of time.  Taking an interrupt while
soft-disabled doesn't cause us to clear it for example.

The iSeries kernel expects to mostly run with MSR:EE enabled at all
times except in a few exception entry/exit code paths.  Thus
local_irq_enable() doesn't check if it needs to hard-enable as it
expects this to be unnecessary on iSeries.

However, hard_irq_disable() _does_ cause MSR:EE to be cleared,
including on iSeries.  A call to it was recently added to the
context switch code, thus causing interrupts to become disabled
for a long periods of time, causing the iSeries watchdog to kick
in under some circumstances and other nasty things.

This patch fixes it by making local_irq_enable() properly re-enable
MSR:EE on iSeries.  It basically removes a return statement here
to make iSeries use the same code path as everybody else.  That does
mean that we might occasionally get spurious decrementer interrupts
but I don't think that matters.

Another option would have been to make hard_irq_disable() a nop
on iSeries but I didn't like it much, in case we have good reasons
to hard-disable.

Part of the patch is fixes to make sure the hard_enabled PACA field
is properly set on iSeries as it used not to be before, since it
was mostly unused.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>

A missing break statement in a switch caused cpm2_clk_setup() to initialize
SCC2 instead of SCC1.
Signed-off-by: Laurent Pinchart <laurentp@cse-semaphore.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>

* git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb-2.6:
  USB: ohci: fix 2 timers to fire at jiffies + 1s
  USB: Allow initialization of broken keyspan serial adapters.
  USB: fix bug in sg initialization in usbtest
  USB: serial: fix regression in Visor/Palm OS module for kernels >= 2.6.24
  USB: cp2101: Add identifiers for the Telegesys ETRX2USB
  USB: serial: ti_usb_3410_5052: Correct TUSB3410 endpoint requirements.
  USB: another ehci_iaa_watchdog fix

A nasty compile error:

In file included from security/keys/internal.h:16,
                 from security/keys/sysctl.c:14:
include/linux/key-ui.h: In function 'key_permission':
include/linux/key-ui.h:51: error: invalid use of undefined type 'struct task_struct'

apparently the compiler has decided that it needs to know sizeof(task_struct)
so that it can add zero to a task_struct* (which is rather dumb of it).

Getting task_struct in scope in these deeply-nested headers is scary-looking,
so let's just remove the "+ 0".

Cc: David Howells <dhowells@redhat.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Markers do not mix well with CONFIG_PREEMPT_RCU because it uses
preempt_disable/enable() and not rcu_read_lock/unlock for minimal
intrusiveness.  We would need call_sched and sched_barrier primitives.

Currently, the modification (connection and disconnection) of probes
from markers requires changes to the data structure done in RCU-style :
a new data structure is created, the pointer is changed atomically, a
quiescent state is reached and then the old data structure is freed.

The quiescent state is reached once all the currently running
preempt_disable regions are done running.  We use the call_rcu mechanism
to execute kfree() after such quiescent state has been reached.
However, the new CONFIG_PREEMPT_RCU version of call_rcu and rcu_barrier
does not guarantee that all preempt_disable code regions have finished,
hence the race.

The "proper" way to do this is to use rcu_read_lock/unlock, but we don't
want to use it to minimize intrusiveness on the traced system.  (we do
not want the marker code to call into much of the OS code, because it
would quickly restrict what can and cannot be instrumented, such as the
scheduler).

The temporary fix, until we get call_rcu_sched and rcu_barrier_sched in
mainline, is to use synchronize_sched before each call_rcu calls, so we
wait for the quiescent state in the system call code path.  It will slow
down batch marker enable/disable, but will make sure the race is gone.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>