Mark Langsdorf reported:

> One of my co-workers noticed that the powernow-k8
> driver no longer restarts when a CPU core is
> hot-disabled and then hot-enabled on AMD quad-core
> systems.
>
> The following comands work fine on 2.6.26 and fail
> on 2.6.27-rc1:
>
> echo 0 > /sys/devices/system/cpu/cpu3/online
> echo 1 > /sys/devices/system/cpu/cpu3/online
> find /sys -name cpufreq
>
> For 2.6.26, the find will return a cpufreq
> directory for each processor.  In 2.6.27-rc1,
> the cpu3 directory is missing.
>
> After digging through the code, the following
> logic is failing when the core is hot-enabled
> at runtime.  The code works during the boot
> sequence.
>
>       cpumask_t = current->cpus_allowed;
>       set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu));
>       if (smp_processor_id() != cpu)
>               return -ENODEV;

So set the CPU active before calling the CPU_ONLINE notifier chain,
there are a handful of notifiers that use set_cpus_allowed().

This fix also solves the problem with x86-microcode. I've sent
alternative patches for microcode, but as this "rely on
set_cpus_allowed_ptr() being workable in cpu-hotplug(CPU_ONLINE, ...)"
assumption seems to be more broad than what we thought, perhaps this fix
should be applied.

With this patch we define that by the moment CPU_ONLINE is being sent,
a 'cpu' is online and ready for tasks to be migrated onto it.
Signed-off-by: Dmitry Adamushko <dmitry.adamushko@gmail.com>
Reported-by: Mark Langsdorf <mark.langsdorf@amd.com>
Tested-by: Mark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Defer commit 6d299f1b to the next release.

Testing of the tip/sched/clock tree revealed a mysql+oltp regression
which bisection eventually traced back to this commit in mainline.

Pertinent test results:  Three run sysbench averages, throughput units
in read/write requests/sec.

clients         1     2     4     8    16    32    64
6e0534f2      9646 17876 34774 33868 32230 30767 29441
2.6.26.1     9112 17936 34652 33383 31929 30665 29232
6d299f1b      9112 14637 28370 33339 32038 30762 29204

Note: subsequent commits hide the majority of this regression until you
apply the clock fixes, at which time it reemerges at full magnitude.

We cannot see anything bad about the change itself so we defer it to the
next release until this problem is fully analysed.
Signed-off-by: Mike Galbraith <efault@gmx.de>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Gregory Haskins <ghaskins@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

Some arch's can't handle sched_clock() being called too early - delay
this until sched_clock_init() has been called.
Reported-by: Bill Gatliff <bgat@billgatliff.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Tested-by: Nishanth Aravamudan <nacc@us.ibm.com>
CC: Russell King - ARM Linux <linux@arm.linux.org.uk>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/davej/cpufreq:
  [CPUFREQ][2/2] preregister support for powernow-k8
  [CPUFREQ][1/2] whitespace fix for powernow-k8
  [CPUFREQ] Update MAINTAINERS to reflect new mailing list.
  [CPUFREQ] Fix warning in elanfreq
  [CPUFREQ] Fix -Wshadow warning in conservative governor.
  [CPUFREQ] Remove EXPERIMENTAL annotation from VIA C7 powersaver kconfig.

* 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/sfrench/cifs-2.6:
  [CIFS] list entry can not return null
  turn cifs_setattr into a multiplexor that calls the correct function
  move file time and dos attribute setting logic into new function
  spin off cifs_setattr with unix extensions to its own function
  [CIFS] Code cleanup in old sessionsetup code
  [CIFS] cifs_mkdir and cifs_create should respect the setgid bit on parent dir
  Rename CIFSSMBSetFileTimes to CIFSSMBSetFileInfo and add PID arg
  change CIFSSMBSetTimes to CIFSSMBSetPathInfo
  [CIFS] fix trailing whitespace
  bundle up Unix SET_PATH_INFO args into a struct and change name
  Fix missing braces in cifs_revalidate()
  remove locking around tcpSesAllocCount atomic variable
  [CIFS] properly account for new user= field in SPNEGO upcall string allocation
  [CIFS] remove level of indentation from decode_negTokenInit
  [CIFS] cifs send2 not retrying enough in some cases on full socket
  [CIFS] oid should also be checked against class in cifs asn

Fix fatal multi-line kernel-doc error in list.h:
function short description must be on one line.

Error(linux-2.6.27-rc2-git3//include/linux/list.h:318): duplicate section name 'Description'
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: Steve French <sfrench@us.ibm.com>

This patch provides support for the _PSD ACPI object in the Powernow-k8
driver.  Although it looks like an invasive patch, most of it is
simply the consequence of turning the static acpi_performance_data
structure into a pointer.

AMD has tested it on several machines over the past few days without issue.

[trivial checkpatch warnings fixed up by davej]
[X86_POWERNOW_K8_ACPI=n buildfix from Randy Dunlap]
Signed-off-by: Mark Langsdorf <mark.langsdorf@amd.com>
Tested-by: Frank Arnold <frank.arnold@amd.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Dave Jones <davej@redhat.com>

Trivial whitespace fix for powernow-k8.
Signed-off-by: Mark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: Dave Jones <davej@redhat.com>

Mailing list moved to vger.
Signed-off-by: Dave Jones <davej@redhat.com>

arch/x86/kernel/cpu/cpufreq/elanfreq.c:47:26: warning: symbol 'elan_multiplier' was not declared. Should it be static?

Yes, yes it should.
Signed-off-by: Dave Jones <davej@redhat.com>

drivers/cpufreq/cpufreq_conservative.c:336:15: warning: symbol 'freq_step' shadows an earlier one

Just rename the local variable.
Signed-off-by: Dave Jones <davej@redhat.com>

This has been pretty solid, and doesn't see much change at all.

Noticed by Harald Welte.
Signed-off-by: Dave Jones <davej@redhat.com>

* 'for-linus-merged' of master.kernel.org:/home/rmk/linux-2.6-arm:
  [ARM] 5177/1: arm/mach-sa1100/Makefile: remove CONFIG_SA1100_USB
  [ARM] 5166/1: magician: add MAINTAINERS entry
  [ARM] fix pnx4008 build errors
  [ARM] Fix SMP booting with non-zero PHYS_OFFSET
  [ARM] 5185/1: Fix spi num_chipselect for lubbock
  [ARM] Move include/asm-arm/arch-* to arch/arm/*/include/mach
  [ARM] Add support for arch/arm/mach-*/include and arch/arm/plat-*/include
  [ARM] Remove asm/hardware.h, use asm/arch/hardware.h instead
  [ARM] Eliminate useless includes of asm/mach-types.h
  [ARM] Fix circular include dependency with IRQ headers
  avr32: Use <mach/foo.h> instead of <asm/arch/foo.h>
  avr32: Introduce arch/avr32/mach-*/include/mach
  avr32: Move include/asm-avr32 to arch/avr32/include/asm
  [ARM] sa1100_wdt: use reset_status to remember watchdog reset status
  [ARM] pxa: introduce reset_status and clear_reset_status for driver's usage
  [ARM] pxa: introduce reset.h for reset specific header information

There doesn't seem to be a compelling reason why nfsd4_op_name() is
marked as "inline":

It's only used in a dprintk(), and as long as it has only one caller
non-ancient gcc versions anyway inline it automatically.

This patch fixes the following compile error with gcc 3.4:

  ...
    CC      fs/nfsd/nfs4proc.o
  nfs4proc.c: In function `nfsd4_proc_compound':
  nfs4proc.c:854: sorry, unimplemented: inlining failed in call to
  nfs4proc.c:897: sorry, unimplemented: called from here
  make[3]: *** [fs/nfsd/nfs4proc.o] Error 1
Reported-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
[ Also made it "const char *"  - Linus]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

Conflicts:

	drivers/watchdog/at91rm9200_wdt.c
	drivers/watchdog/davinci_wdt.c
	drivers/watchdog/ep93xx_wdt.c
	drivers/watchdog/ixp2000_wdt.c
	drivers/watchdog/ixp4xx_wdt.c
	drivers/watchdog/ks8695_wdt.c
	drivers/watchdog/omap_wdt.c
	drivers/watchdog/pnx4008_wdt.c
	drivers/watchdog/sa1100_wdt.c
	drivers/watchdog/wdt285.c

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (99 commits)
  pkt_sched: Fix actions referencing
  bnx2x: fix logical op
  tcp: (whitespace only) fix confusing indentation
  pkt_sched: Fix qdisc config when link is down.
  [Bluetooth] Add full quirk implementation for btusb driver
  [Bluetooth] Removal of unnecessary ignore module parameter
  [Bluetooth] Add parameters to control BNEP header compression
  ath9k: Revamp wireless mode usage
  ath9k: More unused macros
  ath9k: Remove a few unused macros and fix indentation
  ath9k: Use mac80211's band macros and remove enum hal_freq_band
  ath9k: Remove redundant data structure ath9k_txq_info
  ath9k: Cleanup data structures related to HW capabilities
  ath9k: work around gcc ICEs
  ath9k: Add new Atheros IEEE 802.11n driver
  ath5k: remove Atheros 11n devices from supported list
  list.h: add list_cut_position()
  list.h: Add list_splice_tail() and list_splice_tail_init()
  p54: swap short slot time dcf values
  rt2x00: Block all unsupported modes
  ...

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6:
  sparc64: Fix end-of-stack checking in save_stack_trace().
  sparc: don't use asm/of_device.h
  sparc64: Use kernel/uid16.c helpers instead of own copy.
  of/sparc: remove include of linux/of_platform.h from asm/of_platform.h
  sparc: i8042-sparcio.h: fix warning

Back when .gitignore file was added to arch/powerpc/ in 06f2138e ([POWERPC]
Add files build to .gitignore, 2006-11-26), there indeed was nothing
tracked in the ignored hierarchy and ignoring everything made sense.  But
we have very many tracked files there these days, and having a higher
level .gitignore that ignores everything is asking for future troubles..

This should have been part of b8b572e1 (powerpc: Move include files to
arch/powerpc/include/asm, 2008-08-01).
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

This patch removes the commented out line for the not available
CONFIG_SA1100_USB option.
Reported-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

Signed-off-by: Philipp Zabel <philipp.zabel@gmail.com>
Acked-by: Eric Miao <eric.miao@marvell.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

include/linux/i2c-pnx.h was missed when moving the include files.
Fix it now; it doesn't really need to include mach/i2c.h at all.
Successfully build tested with pnx4008_defconfig, which had
failed in linux-next.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>

Bug reported by Alexander Beregalov.

Before we dereference the stack frame or try to peek at the
pt_regs magic value, make sure the entire object is within
the kernel stack bounds.
Signed-off-by: David S. Miller <davem@davemloft.net>

When an action is added several times with the same exact index
it gets deleted on every even-numbered attempt.
This fixes that issue.
Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>

Fix dubious logical operation that was found by sparse:
linux-next-20080807/drivers/net/bnx2x_main.c:7205:27: warning: dubious: !x & y
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The indentation in part of tcp_minisocks makes it look like one of the if
statements is much more important than it actually is.
Signed-off-by: Adam Langley <agl@imperialviolet.org>
Signed-off-by: David S. Miller <davem@davemloft.net>

Bug reported by Stephen Hemminger.

We need to fetch the root from ->qdisc_sleeping not ->qdisc.
Signed-off-by: David S. Miller <davem@davemloft.net>

* 'tracehook' of git://git.kernel.org/pub/scm/linux/kernel/git/frob/linux-2.6-utrace:
  tracehook: fix CLONE_PTRACE

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband:
  IB/mad: Test ib_create_send_mad() return with IS_ERR(), not == NULL
  IB/mlx4: Allow 4K messages for UD QPs
  mlx4_core: Add ethernet fields to CQE struct
  IB/ipath: Fix printk format warnings
  RDMA/cxgb3: Fix deadlock initializing iw_cxgb3 device
  RDMA/cxgb3: Fix up MW access rights
  RDMA/cxgb3: Fix QP capabilities
  RDMA/cma: Remove padding arrays by using struct sockaddr_storage
  IB/ipath: Use unsigned long for irq flags
  IPoIB/cm: Set correct SG list in ipoib_cm_init_rx_wr()

* 'for-linus' of git://git.kernel.dk/linux-2.6-block:
  xen-blkfront.c: make blkif_ioctl() static
  bio: make use of bvec_nr_vecs
  cciss: fix bug if scsi tape support is disabled
  cciss: add support for multi lun tape devices
  cciss: change the way we notify scsi midlayer of tape drives
  cciss: fix negative logical drive count in procfs
  cciss: remove redundant code
  cciss: make rebuild_lun_table behave better
  cciss: return -EFAULT if copy_from_user() fails

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound-2.6:
  ALSA: ASoC: fix SNDCTL_DSP_SYNC support in Freescale 8610 sound drivers

Looks like somebody didn't get enough sleep.
Noticed-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gerg/m68knommu:
  Move all of include/asm-m68knommu to arch/m68knommu/include/asm.

This driver was declared obsolete over 2 years ago, the alternative
console driver for legacy iSeries (hvc_iseries) was made the default
over 1 year ago and this driver has been build broken for over 3
months, so remove it.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/v4l-dvb: (23 commits)
  V4L/DVB (8617): uvcvideo: don't use stack-based buffers for USB transfers.
  V4L/DVB (8616): uvcvideo: Add support for two Bison Electronics webcams
  V4L/DVB (8611): Add suspend/resume to pxa_camera driver
  V4L/DVB (8610): Add suspend/resume capabilities to soc_camera.
  V4L/DVB (8609): media: Clean up platform_driver_unregister() bogosity.
  V4L/DVB (8607): cxusb: fix OOPS and broken tuning regression on FusionHDTV Dual Digital 4
  V4L/DVB (8605): gspca: Fix of gspca_zc3xx oops - 2.6.27-rc1
  V4L/DVB (8604): gspca: Fix of "scheduling while atomic" crash.
  V4L/DVB (8602): gspca: Fix small bugs, simplify and cleanup ov519.
  V4L/DVB (8582): set mts_firmware for em2882 based Pinnacle Hybrid Pro
  V4L/DVB (8574): gspca: Bad bytesperlines of pixelformat in spca505/506/508 and vc023x.
  V4L/DVB (8573): gspca: Bad scan of frame in spca505/506/508.
  V4L/DVB (8572): gspca: Webcam 0c45:6143 in documentation.
  V4L/DVB (8571): gspca: Don't use CONFIG_VIDEO_ADV_DEBUG as a compile option.
  V4L/DVB (8569): gspca: Set back the old values of Sonix sn9c120 and cleanup source.
  V4L/DVB (8567): gspca: hflip and vflip controls added for ov519 - ov7670 plus init cleanup.
  V4L/DVB (8564): fix vino driver build error
  V4L/DVB (8563): fix drivers/media/video/arv.c compilation
  V4L/DVB (8562): DVB_DRX397XD: remove FW_LOADER select
  V4L/DVB (8558): media/video/Kconfig: fix a typo
  ...