I screwed up this conversion - we should be iterating across online CPUs, not
possible ones.

Spotted by Joe Perches <joe@perches.com>

Cc: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

When we stop allocating percpu memory for not-possible CPUs we must not touch
the percpu data for not-possible CPUs at all.  The correct way of doing this
is to test cpu_possible() or to use for_each_cpu().

This patch is a kernel-wide sweep of all instances of NR_CPUS.  I found very
few instances of this bug, if any.  But the patch converts lots of open-coded
test to use the preferred helper macros.

Cc: Mikael Starvik <starvik@axis.com>
Cc: David Howells <dhowells@redhat.com>
Acked-by: NKyle McMartin <kyle@parisc-linux.org>
Cc: Anton Blanchard <anton@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Andi Kleen <ak@muc.de>
Cc: Christian Zankel <chris@zankel.net>
Cc: Philippe Elie <phil.el@wanadoo.fr>
Cc: Nathan Scott <nathans@sgi.com>
Cc: Jens Axboe <axboe@suse.de>
Cc: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Ignore clock frequencies below 2Ghz for CPU's detected with N60 errata bug.
Signed-off-by: NBen Collins <bcollins@ubuntu.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NDave Jones <davej@redhat.com>

This patch fixes the following compile error:

...
  CC      arch/i386/kernel/cpu/cpufreq/gx-suspmod.o
arch/i386/kernel/cpu/cpufreq/gx-suspmod.c: In function 'gx_detect_chipset':
arch/i386/kernel/cpu/cpufreq/gx-suspmod.c:193: error: implicit declaration of function 'pci_match_id'
arch/i386/kernel/cpu/cpufreq/gx-suspmod.c:193: warning: comparison between pointer and integer
make[3]: *** [arch/i386/kernel/cpu/cpufreq/gx-suspmod.o] Error 1

<--  snip  -->
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Signed-off-by: NDave Jones <davej@redhat.com>

cpufreq init can be called when a CPU is set online.
Need to make powernow-k8's initialisation functions __cpuinit to
prevents oopses when a CPU is off/onlined on a AMD system

Cc: trenn@suse.de
Cc: mark.langsdorf@amd.com
Cc: davej@redhat.com
Signed-off-by: NAndi Kleen <ak@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

They previously tried to figure this out on their own.

Suggested by Venkatesh.

Cc: venkatesh.pallipadi@intel.com
Cc: davej@redhat.com
Signed-off-by: NAndi Kleen <ak@suse.de>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NBenoit Boissinot <benoit.boissinot@ens-lyon.org>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLen Brown <len.brown@intel.com>

This patch moves away PMBASE reading and only performs it at
cpufreq_register_driver time by exiting with -ENODEV if unable to read
the value.
Signed-off-by: NMattia Dongili <malattia@linux.it>
Acked-by: NDominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: NDave Jones <davej@redhat.com>

The attached patch introduces runtime latency measurement for ICH[234]
based chipsets instead of using CPUFREQ_ETERNAL. It includes
some sanity checks in case the measured value is out of range and
assigns a safe value of 500uSec that should still be enough on
problematics chipsets (current testing report values ~200uSec). The
measurement is currently done in speedstep_get_freqs in order to avoid
further unnecessary transitions and in the hope it'll come handy for SMI
also.
Signed-off-by: NMattia Dongili <malattia@linux.it>
Acked-by: NDominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: NDave Jones <davej@redhat.com>

 speedstep-ich.c |    4 ++--
 speedstep-lib.c |   32 +++++++++++++++++++++++++++++++-
 speedstep-lib.h |    1 +
 speedstep-smi.c |    1 +
 4 files changed, 35 insertions(+), 3 deletions(-)

If a user has booted with 'quiet', some important messages don't
get displayed which really should. We've seen at least one case
where powernow-k8 stopped working, and the user needed a BIOS update
that they didn't know about.
Signed-off-by: NDave Jones <davej@redhat.com>

Thanks to LinuxICC (http://linuxicc.sf.net), a comparison of a u32 less
than 0 was found, this patch changes the variable to a signed int so that
comparison is meaningful.
Signed-off-by: NGabriel A. Devenyi <ace@staticwave.ca>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NDave Jones <davej@redhat.com>

Linux invokes the AML _PDC method (Processor Driver Capabilities)
to tell the BIOS what features it can handle.  While the ACPI
spec says nothing about the OS invoking _PDC multiple times,
doing so with changing bits seems to hopelessly confuse the BIOS
on multiple platforms up to and including crashing the system.

Factor out the _PDC invocation so Linux invokes it only once.

http://bugzilla.kernel.org/show_bug.cgi?id=5483Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

Signed-off-by: NDave Jones <davej@redhat.com>

Future versions of the Opteron processor may support
frequency transitions of 100 MHz, instead of the=20
current 200 MHz.  This patch enables the powernow-k8
driver to transition to an odd FID code, indicating
a multiple of 100 MHz frequency.
Signed-off-by: NJacob Shin <jacob.shin@amd.com>
Signed-off-by: NMark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: NDave Jones <davej@redhat.com>

This patch cleans up some error messages in the
powernow-k8 driver and makes them more understandable.
Signed-off-by: NJacob Shin <jacob.shin@amd.com>
Signed-off-by: NMark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: NDave Jones <davej@redhat.com>

Use ARRAY_SIZE macro instead of sizeof(x)/sizeof(x[0])
Signed-off-by: NTobias Klauser <tklauser@nuerscht.ch>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

I recently picked up my older work to remove unnecessary #includes of
sched.h, starting from a patch by Dave Jones to not include sched.h
from module.h. This reduces the number of indirect includes of sched.h
by ~300. Another ~400 pointless direct includes can be removed after
this disentangling (patch to follow later).
However, quite a few indirect includes need to be fixed up for this.

In order to feed the patches through -mm with as little disturbance as
possible, I've split out the fixes I accumulated up to now (complete for
i386 and x86_64, more archs to follow later) and post them before the real
patch.  This way this large part of the patch is kept simple with only
adding #includes, and all hunks are independent of each other.  So if any
hunk rejects or gets in the way of other patches, just drop it.  My scripts
will pick it up again in the next round.
Signed-off-by: NTim Schmielau <tim@physik3.uni-rostock.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Not sure how it slipped by, but here's a trivial typo fix for powernow.
Signed-off-by: NChris Wright <chrisw@osdl.org>
[ It's "nurter" backwards.. Maybe we have a hillbilly The Shining fan? ]
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

AMD recently discovered that on some hardware, there is a race condition
possible when a C-state change request goes onto the bus at the same
time as a P-state change request.

Both requests happen, but the southbridge hardware only acknowledges the
C-state change.  The PowerNow! driver is then stuck in a loop, waiting
for the P-state change acknowledgement.  The driver eventually times
out, but can no longer perform P-state changes.

It turns out the solution is to resend the P-state change, which the
southbridge will acknowledge normally.

Thanks to Johannes Winkelmann for reporting this and testing the fix.
Signed-off-by: NMark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: NDave Jones <davej@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NDave Jones <davej@redhat.com>

Via reading the code, my understanding is that powernow-k8 uses
preempt_disable to ensure that driver->target doesn't migrate across cpus
whilst it's accessing per processor registers, however set_cpus_allowed
will provide this for us.  Additionally, remove schedule() calls from
set_cpus_allowed as set_cpus_allowed ensures that you're executing on the
target processor on return.
Signed-off-by: NZwane Mwaikambo <zwane@arm.linux.org.uk>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NDave Jones <davej@redhat.com>

i386 Inline asm cleanup.  Use cr/dr accessor functions.

Also, a potential bugfix.  Also, some CR accessors really should be volatile.
Reads from CR0 (numeric state may change in an exception handler), writes to
CR4 (flipping CR4.TSD) and reads from CR2 (page fault) prevent instruction
re-ordering.  I did not add memory clobber to CR3 / CR4 / CR0 updates, as it
was not there to begin with, and in no case should kernel memory be clobbered,
except when doing a TLB flush, which already has memory clobber.

I noticed that page invalidation does not have a memory clobber.  I can't find
a bug as a result, but there is definitely a potential for a bug here:

#define __flush_tlb_single(addr) \
	__asm__ __volatile__("invlpg %0": :"m" (*(char *) addr))
Signed-off-by: NZachary Amsden <zach@vmware.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

From: Denis Vlasenko <vda@ilport.com.ua>
Signed-off-by: NDave Jones <davej@redhat.com>

Minor fallout from my upcoming __attribute__((format(printf,x,y)))
patches. The variable 'result' is untouched, so this patch just removes
it.
Signed-off-by: NMika Kukkonen <mikukkon@gmail.com>
Signed-off-by: NDave Jones <davej@redhat.com>

Ho-hum, did not notice there was more printf fixes for cpufreq (you
should see the amount I have for isdn and reiser ...). Sorry for noise.
Signed-off-by: NMika Kukkonen <mikukkon@gmail.com>
Signed-off-by: NDave Jones <davej@redhat.com>

speedstep_centrino.c:extract_clock() assumes the bus speed of 100MHz, which is
not true with latest laptops. Due to this assumption and due to the encoded
frequency check during initialization, speedstep-centrino driver fails even
on systems that has proper ACPI information to do the P-state transition.

The change below moves the centrino-speedstep detection to be used only
when table based P-state transition is done. For ACPI based P-state
transition, we skip the centrino_cpu identification, and as a result we
don't use the bus speed assumption in extract_clock. This change makes
speedstep-centrino work on Pentium-M based systems, which have more than 100MHz
bus speed.
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NDave Jones <davej@redhat.com>

The acpi-cpufreq driver does a P-state get after a P-state set
to verify whether set went through successfully. This test
is kind of redundant as set goes throught most of the times,
and the test is also expensive as a get of P-states can
take a lot of time (same as a set operation) as it goes
through SMM mode. Effectively, we are doubling the P-state
latency due to this get opertion.

momdule parameter "acpi_pstate_strict" restores orginal paranoia.

http://bugzilla.kernel.org/show_bug.cgi?id=5129Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

Otherwise a platform that supports ACPI based cpufreq
and boots up at lowest possible speed could stay there
forever.  This because the governor may request max speed,
but the code doesn't update if there is no change in
speed, and it assumed the initial state of max speed.

http://bugzilla.kernel.org/show_bug.cgi?id=4634Signed-off-by: NDominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

arch/i386/kernel/cpu/cpufreq/powernow-k8.c:740: warning: unused variable `vid'
arch/i386/kernel/cpu/cpufreq/powernow-k8.c:739: warning: unused variable `fid'
arch/i386/kernel/cpu/cpufreq/powernow-k8.c:743: warning: unused variable `vid'
arch/i386/kernel/cpu/cpufreq/powernow-k8.c:742: warning: unused variable `fid'
arch/i386/kernel/cpu/cpufreq/powernow-k8.c:746: `fid' undeclared (first use in this function)
arch/i386/kernel/cpu/cpufreq/powernow-k8.c:746: `vid' undeclared (first use in this function)
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NDave Jones <davej@redhat.com>

From: Mark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: NDave Jones <davej@redhat.com>

powernow-k8.c:110: warning: `hi' may be used uninitialized in this function
Signed-off-by: NBrian Gerst <bgerst@didntduck.org>
Signed-off-by: NDave Jones <davej@redhat.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>

can be encoded in the current driver's 4 bit frequency
field.  This patch updates the driver to support Rev F
including 6 bit FIDs and processor ID updates.

This should apply cleanly whether or not the dual-core
bugfix I sent out last week is applied.  I'd prefer
that both get applied, of course.
Signed-off-by: NDavid Keck <david.keck@amd.com>
Signed-off-by: NMark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: NDave Jones <davej@redhat.com>

each core be created in the _cpu_init function
call.  The cpufreq infrastructure doesn't call
_cpu_init for the second core in each processor.
Some systems crashed when _get was called with
an odd-numbered core because it tried to
dereference a NULL pointer since the data
structure had not been created.

The attached patch solves the problem by
initializing data structures for all shared
cores in the _cpu_init function.  It should
apply to 2.6.12-rc6 and has been tested by
AMD and Sun.
Signed-off-by: NMark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: NDave Jones <davej@redhat.com>

powernow-k8.c: In function `query_current_values_with_pending_wait':
powernow-k8.c:110: warning: `hi' may be used uninitialized in this function
Signed-off-by: NBrian Gerst <bgerst@didntduck.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

http://bugzilla.kernel.org/show_bug.cgi?id=4401Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

The dynamic pci id logic has been bothering me for a while, and now that
I started to look into how to move some of this to the driver core, I
thought it was time to clean it all up.

It ends up making the code smaller, and easier to follow, and fixes a
few bugs at the same time (dynamic ids were not being matched
everywhere, and so could be missed on some call paths for new devices,
semaphore not needed to be grabbed when adding a new id and calling the
driver core, etc.)

I also renamed the function pci_match_device() to pci_match_id() as
that's what it really does.
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Signed-off-by: NJesper Juhl <juhl-lkml@dif.dk>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

cpfureq developers cant spel.
Signed-off-by: NDave Jones <davej@redhat.com>

From patch by: Ken Staton <ken_staton@agilent.com>
Signed-off-by: NDave Jones <davej@redhat.com>

As mandated by the spec, disable timer around transitions.

From code by : Ken Staton <ken_staton@agilent.com
Signed-off-by: NDave Jones <davej@redhat.com>