With the recent irq migration fixes (post 2.6.32), Gary Hade has noticed
"No IRQ handler for vector" messages during the 2.6.33-rc1 kernel boot on IBM
AMD platforms and root caused the issue to this commit:

> commit 23359a88
> Author: Suresh Siddha <suresh.b.siddha@intel.com>
> Date:   Mon Oct 26 14:24:33 2009 -0800
>
>    x86: Remove move_cleanup_count from irq_cfg

As part of this patch, we have removed the move_cleanup_count check
in smp_irq_move_cleanup_interrupt(). With this change, we can run into a
situation where an irq cleanup interrupt on a cpu can cleanup the vector
mappings associated with multiple irqs, of which one of the irq's migration
might be still in progress. As such when that irq hits the old cpu, we get
the "No IRQ handler" messages.

Fix this by checking for the irq_cfg's move_in_progress and if the move
is still in progress delay the vector cleanup to another irq cleanup
interrupt request (which will happen when the irq starts arriving at the
new cpu destination).
Reported-and-tested-by: NGary Hade <garyhade@us.ibm.com>
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1262804191.2732.7.camel@sbs-t61.sc.intel.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

This reverts commit ae1b22f6.

As Linus said in 982d007a: "There was something really messy about
cmpxchg8b and clone CPU's, so if you enable it on other CPUs later, do it
carefully."

This breaks lguest for those configs, but we can fix that by emulating
if we have to.

Fixes: http://bugzilla.kernel.org/show_bug.cgi?id=14884Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Cc: stable@kernel.org
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Callers of copy_from_user() expect it to return the number of bytes
it could not copy. In no case it is supposed to return -EFAULT.

In case of a detected buffer overflow just return the requested
length. In addition one could think of a memset that would clear
the size of the target object.

[ hpa: code is not in .32 so not needed for -stable ]
Signed-off-by: NHeiko Carstens <heiko.carstens@de.ibm.com>
Acked-by: NArjan van de Ven <arjan@linux.intel.com>
LKML-Reference: <20100105131911.GC5480@osiris.boeblingen.de.ibm.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

This reverts commit ae1b22f6.

As Linus said in 982d007a: "There was something really messy about
cmpxchg8b and clone CPU's, so if you enable it on other CPUs later, do it
carefully."

This breaks lguest for those configs, but we can fix that by emulating
if we have to.

Fixes: http://bugzilla.kernel.org/show_bug.cgi?id=14884Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
LKML-Reference: <201001051248.49700.rusty@rustcorp.com.au>
Cc: stable@kernel.org
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

It is above 0x100 (PCI-Express extended register space), so if mmconf
is not enable, we can't access it.

[ hpa: changed the bound from 0x200 to 0x120, which is the tight
  bound. ]
Reported-by: NJens Axboe <jens.axboe@oracle.com>
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
LKML-Reference: <1261525263-13763-3-git-send-email-yinghai@kernel.org>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

fix for error that is introduced by
| x86: Use find_e820() instead of hard coded trampoline address

it should end with PAGE_SIZE + PAGE_SIZE
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
LKML-Reference: <1261525263-13763-2-git-send-email-yinghai@kernel.org>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Pass the frame pointer from the regs of the interrupted path
to dump_trace() while processing the stack trace.

Currently, dump_trace() takes the current bp and starts the
callchain from dump_trace() itself. This is wasteful because
we need to walk through the entire NMI/DEBUG stack before
retrieving the interrupted point.

We can fix that by just using the frame pointer from the
captured regs. It points exactly where we want to start.
Signed-off-by: NFrederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1262235183-5320-1-git-send-regression-fweisbec@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>

Introduce kernel parameter acpi_sleep=sci_force_enable

some laptop requires SCI_EN being set directly on resume,
or else they hung somewhere in the resume code path.

We already have a blacklist for these laptops but we still need
this option, especially when debugging some suspend/resume problems,
in case there are systems that need this workaround and are not yet
in the blacklist.
Signed-off-by: NZhang Rui <rui.zhang@intel.com>
Acked-by: NRafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: NLen Brown <len.brown@intel.com>

The wrong address was being used to write the SCIR led regs on remote
hubs.  Also, there was an inconsistency between how BIOS and the kernel
indexed these regs.  Standardize on using the lower 6 bits of the APIC
ID as the index.

This patch fixes the problem of writing to an errant address to a
cpu # >= 64.
Signed-off-by: NMike Travis <travis@sgi.com>
Reviewed-by: NJack Steiner <steiner@sgi.com>
Cc: stable@kernel.org
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The wrong address was being used to write the SCIR led regs on
remote hubs.  Also, there was an inconsistency between how BIOS
and the kernel indexed these regs.  Standardize on using the
lower 6 bits of the APIC ID as the index.

This patch fixes the problem of writing to an errant address to
a cpu # >= 64.
Signed-off-by: NMike Travis <travis@sgi.com>
Reviewed-by: NJack Steiner <steiner@sgi.com>
Cc: Robin Holt <holt@sgi.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: stable@kernel.org
LKML-Reference: <4B3922F9.3060905@sgi.com>
[ v2: fix a number of annoying checkpatch artifacts and whitespace noise ]
Signed-off-by: NIngo Molnar <mingo@elte.hu>

As suggested by Vegard Nossum, use KERN_WARNING for error
reporting to make sure kmemcheck reports end up in syslog.
Suggested-by: NVegard Nossum <vegard.nossum@gmail.com>
Signed-off-by: NPekka Enberg <penberg@cs.helsinki.fi>
Cc: Andrew Morton <akpm@linux-foundation.org>
LKML-Reference: <1261990935.4641.7.camel@penberg-laptop>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Andrew Morton reported a strange looking kmemcheck warning:

  WARNING: kmemcheck: Caught 32-bit read from uninitialized memory (ffff88004fba6c20)
  0000000000000000310000000000000000000000000000002413000000c9ffff
   u u u u u u u u u u u u u u u u i i i i i i i i u u u u u u u u

   [<ffffffff810af3aa>] kmemleak_scan+0x25a/0x540
   [<ffffffff810afbcb>] kmemleak_scan_thread+0x5b/0xe0
   [<ffffffff8104d0fe>] kthread+0x9e/0xb0
   [<ffffffff81003074>] kernel_thread_helper+0x4/0x10
   [<ffffffffffffffff>] 0xffffffffffffffff

The above printout is missing register dump completely. The
problem here is that the output comes from syslog which doesn't
show KERN_INFO log-level messages. We didn't see this before
because both of us were testing on 32-bit kernels which use the
_default_ log-level.

Fix that up by explicitly using KERN_DEFAULT log-level for
__show_regs() printks.
Signed-off-by: NPekka Enberg <penberg@cs.helsinki.fi>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Arjan van de Ven <arjan@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
LKML-Reference: <1261988819.4641.2.camel@penberg-laptop>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

User space may not want to overwrite asynchronously changing VCPU event
states on write-back. So allow to skip nmi.pending and sipi_vector by
setting corresponding bits in the flags field of kvm_vcpu_events.

[avi: advertise the bits in KVM_GET_VCPU_EVENTS]
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

The vcpus are initialized with irr_pending set to false, but
loading the LAPIC registers with pending IRR fails to reset
the irr_pending variable.

Cc: stable@kernel.org
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

The invlpg prefault optimization breaks Windows 2008 R2 occasionally.

The visible effect is that the invlpg handler instantiates a pte which
is, microseconds later, written with a different gfn by another vcpu.

The OS could have other mechanisms to prevent a present translation from
being used, which the hypervisor is unaware of.

While the documentation states that the cpu is at liberty to prefetch tlb
entries, it looks like this is not heeded, so remove tlb prefetch from
invlpg.

Cc: stable@kernel.org
Signed-off-by: NMarcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

Recently, some distros have started shipping versions of gcc which
default to -march=i686.  This breaks building kernels for pre-i686
machines, even if they have been selected in Kconfig, due to the
generation of CMOV instructions.

There isn't enough benefit to try to preserve the generation of these
instructions even when selected, so simply force -march=i386 for the
decompressor when building a 32-bit kernel.
Reported-and-tested-by: NChris Rankin <rankincj@yahoo.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>
LKML-Reference: <219280.97558.qm@web52907.mail.re2.yahoo.com>

This reverts commit 9f15226e.  It's just
wrong, and broke resume for Rafael even on a non-AMD CPU.

As Rafael says:
 "... it causes microcode_init_cpu() to be called during resume even for
  CPUs for which there's no microcode to apply.  That, in turn, results
  in executing request_firmware() (on Intel CPUs at least) which doesn't
  work at this stage of resume (we have device interrupts disabled, I/O
  devices are still suspended and so on).

  If I'm not mistaken, the "if (uci->valid)" logic means "if that CPU is
  known to us" , so before commit 9f15226e microcode_resume_cpu() was
  called for all CPUs already in the system during suspend, which was
  the right thing to do.  The commit changed it so that the CPUs without
  microcode to apply are now treated as "unknown", which is not quite
  right.

  The problem this commit attempted to solve has to be handled
  differently."

Bisected-and -requested-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c: avoid cross-CPU interrupts by using smp_call_function_any()

Presently acpi-cpufreq will perform the MSR read on the first CPU in the
mask.  That's inefficient if that CPU differs from the current CPU.
Because we have to perform a cross-CPU call, but we could have run the
rdmsr on the current CPU.

So switch to using the new smp_call_function_any(), which will perform the
call on the current CPU if that CPU is present in the mask (it is).

Cc: "Zhang, Yanmin" <yanmin_zhang@linux.intel.com>
Cc: Dave Jones <davej@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jaswinder Singh Rajput <jaswinder@kernel.org>
Cc: Len Brown <len.brown@intel.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLen Brown <len.brown@intel.com>

The x86 and ia64 implementations of the function in $subject are
exactly the same.

Also, since the arch-specific implementations of setting _PDC have
been completely hollowed out, remove the empty shells.

Cc: Tony Luck <tony.luck@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAlex Chiang <achiang@hp.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

The only thing arch-specific about calling _PDC is what bits get
set in the input obj_list buffer.

There's no need for several levels of indirection to twiddle those
bits. Additionally, since we're just messing around with a buffer,
we can simplify the interface; no need to pass around the entire
struct acpi_processor * just to get at the buffer.

Cc: Tony Luck <tony.luck@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAlex Chiang <achiang@hp.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

Both x86 and ia64 initialize _PDC with mostly common bit settings.

Factor out the common settings and leave the arch-specific ones alone.

Cc: Tony Luck <tony.luck@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAlex Chiang <achiang@hp.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

The x86 and ia64 implementations of arch_acpi_processor_init_pdc()
are almost exactly the same. The only difference is in what bits
they set in obj_list buffer.

Combine the boilerplate memory management code, and leave the
arch-specific bit twiddling in separate implementations.

Cc: Tony Luck <tony.luck@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAlex Chiang <achiang@hp.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

arch dependent helper function that tells us if we should attempt to
evaluate _PDC on this machine or not.

The x86 implementation assumes that the CPUs in the machine must be
homogeneous, and that you cannot mix CPUs of different vendors.

Cc: Tony Luck <tony.luck@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NAlex Chiang <achiang@hp.com>
Signed-off-by: NLen Brown <len.brown@intel.com>

The assumption that acpi_table_parse passes the return value
of the hanlder function to the caller proved wrong
recently. The return value of the handler function is
totally ignored. This makes the initialization code for AMD
IOMMU buggy in a way that could cause a kernel panic on
initialization. This patch fixes the issue in the AMD IOMMU
driver.

Cc: stable@kernel.org
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>

Different version of objdump says its version in different way;

GNU objdump 2.16.1

or

GNU objdump version 2.19.51.0.14-1.fc11 20090722

This patch uses the first argument which starts with a number
as version string.

Changes in v2:
 - Remove unneeded increment.
Signed-off-by: NMasami Hiramatsu <mhiramat@redhat.com>
LKML-Reference: <20091218154012.16960.5113.stgit@dhcp-100-2-132.bos.redhat.com>
Suggested-by: NH. Peter Anvin <hpa@zytor.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

The kbuild's select command doesn't propagate through the config
dependencies.

Hence the current rules of hardware breakpoint's config can't
ensure perf can never be disabled under us.

We have:

config X86
	selects HAVE_HW_BREAKPOINTS

config HAVE_HW_BREAKPOINTS
	select PERF_EVENTS

config PERF_EVENTS
	[...]

x86 will select the breakpoints but that won't propagate to perf
events. The user can still disable the latter, but it is
necessary for the breakpoints.

What we need is:

 - x86 selects HAVE_HW_BREAKPOINTS and PERF_EVENTS
 - HAVE_HW_BREAKPOINTS depends on PERF_EVENTS

so that we ensure PERF_EVENTS is enabled and frozen for x86.

This fixes the following kind of build errors:

 In file included from arch/x86/kernel/hw_breakpoint.c:31:
 include/linux/hw_breakpoint.h: In function 'hw_breakpoint_addr':
 include/linux/hw_breakpoint.h:39: error: 'struct perf_event' has no member named 'attr'

v2: Select also ANON_INODES from x86, required for perf
Reported-by: NCyrill Gorcunov <gorcunov@gmail.com>
Reported-by: NMichal Marek <mmarek@suse.cz>
Reported-by: NAndrew Randrianasulu <randrik_a@yahoo.com>
Signed-off-by: NFrederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Cc: K.Prasad <prasad@linux.vnet.ibm.com>
LKML-Reference: <1261010034-7786-1-git-send-regression-fweisbec@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

John Blackwood reported:
> on an older Dell PowerEdge 6650 system with 8 cpus (4 are hyper-threaded),
> and  32 bit (x86) kernel, once you change the irq smp_affinity of an irq
> to be less than all cpus in the system, you can never change really the
> irq smp_affinity back to be all cpus in the system (0xff) again,
> even though no error status is returned on the "/bin/echo ff >
> /proc/irq/[n]/smp_affinity" operation.
>
> This is due to that fact that BAD_APICID has the same value as
> all cpus (0xff) on 32bit kernels, and thus the value returned from
> set_desc_affinity() via the cpu_mask_to_apicid_and() function is treated
> as a failure in set_ioapic_affinity_irq_desc(), and no affinity changes
> are made.

set_desc_affinity() is already checking if the incoming cpu mask
intersects with the cpu online mask or not. So there is no need
for the apic op cpu_mask_to_apicid_and() to check again
and return BAD_APICID.

Remove the BAD_APICID return value from cpu_mask_to_apicid_and()
and also fix set_desc_affinity() to return -1 instead of using BAD_APICID
to represent error conditions (as cpu_mask_to_apicid_and() can return
logical or physical apicid values and BAD_APICID is really to represent
bad physical apic id).
Reported-by: NJohn Blackwood <john.blackwood@ccur.com>
Root-caused-by: NJohn Blackwood <john.blackwood@ccur.com>
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1261103386.2535.409.camel@sbs-t61>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

It says

Warning: objdump version  is older than 2.19
Warning: Skipping posttest.

because it used the wrong field from `objdump -v':

akpm:/usr/src/25> /opt/crosstool/gcc-4.0.2-glibc-2.3.6/x86_64-unknown-linux-gnu/bin/x86_64-unknown-linux-gnu-objdump -v
GNU objdump 2.16.1
Copyright 2005 Free Software Foundation, Inc.
This program is free software; you may redistribute it under the terms of
the GNU General Public License.  This program has absolutely no warranty.

Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
LKML-Reference: <200912172326.nBHNQaQl024796@imap1.linux-foundation.org>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>
Cc: Masami Hiramatsu <mhiramat@redhat.com>

Commit 83ce4009 did the following change
If the TSC is constant and non-stop, also set it reliable.

But, there seems to be few systems that will end up with TSC warp across
sockets, depending on how the cpus come out of reset. Skipping TSC sync
test on such systems may result in time inconsistency later.

So, reenable TSC sync test even on constant and non-stop TSC systems.
Set, sched_clock_stable to 1 by default and reset it in
mark_tsc_unstable, if TSC sync fails.

This change still gives perf benefit mentioned in 83ce4009 for systems
where TSC is reliable.
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Acked-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20091217202702.GA18015@linux-os.sc.intel.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

The loop condition is fragile: we compare an unsigned value to zero, and
then decrement it by something larger than one in the loop.  All the
callers should be passing in appropriately aligned buffer lengths, but
it's better to just not rely on it, and have some appropriate defensive
loop limits.
Acked-by: NRoland McGrath <roland@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Not all awk implementations (including the default awk in Ubuntu 9.10)
support POSIX character classes.  Since x86-opcode-map.txt is plain
ASCII, we can just use explicit ranges for lower case, alphabetic, and
alphanumeric characters instead.
Signed-off-by: NRoland Dreier <rolandd@cisco.com>
Acked-by: NMasami Hiramatsu <mhiramat@redhat.com>
LKML-Reference: <adabphy750b.fsf@roland-alpha.cisco.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

perf events, x86/stacktrace: Fix performance/softlockup by providing a special frame pointer-only stack walker

It's just wasteful for stacktrace users like perf to walk
through every entries on the stack whereas these only accept
reliable ones, ie: that the frame pointer validates.

Since perf requires pure reliable stacktraces, it needs a stack
walker based on frame pointers-only to optimize the stacktrace
processing.

This might solve some near-lockup scenarios that can be triggered
by call-graph tracing timer events.
Signed-off-by: NFrederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1261024834-5336-2-git-send-regression-fweisbec@gmail.com>
[ v2: fix for modular builds and small detail tidyup ]
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The current print_context_stack helper that does the stack
walking job is good for usual stacktraces as it walks through
all the stack and reports even addresses that look unreliable,
which is nice when we don't have frame pointers for example.

But we have users like perf that only require reliable
stacktraces, and those may want a more adapted stack walker, so
lets make this function a callback in stacktrace_ops that users
can tune for their needs.
Signed-off-by: NFrederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <1261024834-5336-1-git-send-regression-fweisbec@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Noone uses this wrapper yet, and Ingo asked that it be kept consistent
with current task_struct usage.

(One user crept in via linux-next: fixed)

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au.
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Tejun Heo <tj@kernel.org>

Due to recent changes wakeup and mptable, we run out of early
reservations on 32-bit NUMA.  Thus, adjust the available number.
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
LKML-Reference: <4B22D754.2020706@kernel.org>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Found one system that boot from socket1 instead of socket0, SRAT get rejected...

[    0.000000] SRAT: Node 1 PXM 0 0-a0000
[    0.000000] SRAT: Node 1 PXM 0 100000-80000000
[    0.000000] SRAT: Node 1 PXM 0 100000000-2080000000
[    0.000000] SRAT: Node 0 PXM 1 2080000000-4080000000
[    0.000000] SRAT: Node 2 PXM 2 4080000000-6080000000
[    0.000000] SRAT: Node 3 PXM 3 6080000000-8080000000
[    0.000000] SRAT: Node 4 PXM 4 8080000000-a080000000
[    0.000000] SRAT: Node 5 PXM 5 a080000000-c080000000
[    0.000000] SRAT: Node 6 PXM 6 c080000000-e080000000
[    0.000000] SRAT: Node 7 PXM 7 e080000000-10080000000
...
[    0.000000] NUMA: Allocated memnodemap from 500000 - 701040
[    0.000000] NUMA: Using 20 for the hash shift.
[    0.000000] Adding active range (0, 0x2080000, 0x4080000) 0 entries of 3200 used
[    0.000000] Adding active range (1, 0x0, 0x96) 1 entries of 3200 used
[    0.000000] Adding active range (1, 0x100, 0x7f750) 2 entries of 3200 used
[    0.000000] Adding active range (1, 0x100000, 0x2080000) 3 entries of 3200 used
[    0.000000] Adding active range (2, 0x4080000, 0x6080000) 4 entries of 3200 used
[    0.000000] Adding active range (3, 0x6080000, 0x8080000) 5 entries of 3200 used
[    0.000000] Adding active range (4, 0x8080000, 0xa080000) 6 entries of 3200 used
[    0.000000] Adding active range (5, 0xa080000, 0xc080000) 7 entries of 3200 used
[    0.000000] Adding active range (6, 0xc080000, 0xe080000) 8 entries of 3200 used
[    0.000000] Adding active range (7, 0xe080000, 0x10080000) 9 entries of 3200 used
[    0.000000] SRAT: PXMs only cover 917504MB of your 1048566MB e820 RAM. Not used.
[    0.000000] SRAT: SRAT not used.

the early_node_map is not sorted because node0 with non zero start come first.

so try to sort it right away after all regions are registered.

also fixs refression by 8716273c (x86: Export srat physical topology)

-v2: make it more solid to handle cross node case like node0 [0,4g), [8,12g) and node1 [4g, 8g), [12g, 16g)
-v3: update comments.
Reported-and-tested-by: NJens Axboe <jens.axboe@oracle.com>
Signed-off-by: NYinghai Lu <yinghai@kernel.org>
LKML-Reference: <4B2579D2.3010201@kernel.org>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

xsave_cntxt_init() does something like:

	cpuid(0xd, ..);	// find out what features FP/SSE/.. etc are supported

	xsetbv();	// enable the features known to OS

	cpuid(0xd, ..);	// find out the size of the context for features enabled

Depending on what features get enabled in xsetbv(), value of the
cpuid.eax=0xd.ecx=0.ebx changes correspondingly (representing the
size of the context that is enabled).

As we don't have volatile keyword for native_cpuid(), gcc 4.1.2
optimizes away the second cpuid and the kernel continues to use
the cpuid information obtained before xsetbv(), ultimately leading to kernel
crash on processors supporting more state than the legacy FP/SSE.

Add "volatile" for native_cpuid().
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1261009542.2745.55.camel@sbs-t61.sc.intel.com>
Cc: stable@kernel.org
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Randy Dunlap reported the following build error:

"When CONFIG_SMP=n, CONFIG_X86_MSR=m:

ERROR: "msrs_free" [drivers/edac/amd64_edac_mod.ko] undefined!
ERROR: "msrs_alloc" [drivers/edac/amd64_edac_mod.ko] undefined!"

This is due to the fact that <arch/x86/lib/msr.c> is conditioned on
CONFIG_SMP and in the UP case we have only the stubs in the header.
Fork off SMP functionality into a new file (msr-smp.c) and build
msrs_{alloc,free} unconditionally.
Reported-by: NRandy Dunlap <randy.dunlap@oracle.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: NBorislav Petkov <petkovbb@gmail.com>
LKML-Reference: <20091216231625.GD27228@liondog.tnic>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Use NodeId MSR to get NodeId and number of nodes per processor.
Signed-off-by: NAndreas Herrmann <andreas.herrmann3@amd.com>
LKML-Reference: <20091216144355.GB28798@alberich.amd.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Remark update_res from __init to __devinit as it is called also
from __devinit functions.

This patch removes the following warning message:

  WARNING: vmlinux.o(.devinit.text+0x774a): Section mismatch
  in reference from the function pci_root_bus_res() to the
  function .init.text:update_res()
  The function __devinit pci_root_bus_res() references
  a function __init update_res().
  If update_res is only used by pci_root_bus_res then
  annotate update_res with a matching annotation.
Signed-off-by: NJiri Slaby <jslaby@suse.cz>
Cc: Aristeu Sergio <arozansk@redhat.com>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: linux-pci@vger.kernel.org
Cc: x86@kernel.org
Signed-off-by: NHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>