The problems are that, with the ACPI vs timer overring issue _fixed_,
after using the box for some time (between several seconds and 1 hour, at
random) processes get very high CPU loads (once I've got X using 107% of
the CPU, for example) and the system becomes unresponsive, as though there
were interrupts lost or something similar.

Andreas Herrman reproduced similar problems:

> Ok, now I've reproduced the stability problem.
> - Using tip/master,
> - reverting e38502eb8aa82314d5ab0eba45f50e6790dadd88 and
> - applying your patch from this posting
>   http://marc.info/?l=linux-kernel&m=121539354224562&w=4
>
> Starting X, firefox, gimp, tuxpaint and doing some drawing in tuxpaint
> results in a slow system. Drawing is almost not possible anymore --
> Selections of new colors, cursors etc. is performed with huge delay
> if it's performed at all.
>
> BTW, the code sets up timer IRQ as Virtual Wire IRQ:
>
> Jul  8 14:57:58 kodscha IO-APIC (apicid-pin) 2-22, 2-23 not connected.
> Jul  8 14:57:58 kodscha ..TIMER: vector=0x30 apic1=0 pin1=2 apic2=-1 pin2=-1
> Jul  8 14:57:58 kodscha ...trying to set up timer as Virtual Wire IRQ... works.
>
> and both INT0 and INT2 of IOAPIC are masked:
>
> Jul  8 14:57:58 kodscha NR Dst Mask Trig IRR Pol Stat Dmod Deli Vect:
> Jul  8 14:57:58 kodscha 00 000 1    0    0   0   0    0    0    00
> Jul  8 14:57:58 kodscha 01 003 0    0    0   0   0    1    1    31
> Jul  8 14:57:58 kodscha 02 003 1    0    0   0   0    0    0    30
>
> I've also seen strange CPU utilization -- with syslog-ng:
>
> top - 15:33:06 up 35 min,  4 users,  load average: 1.70, 0.68, 0.37
> Tasks:  64 total,   4 running,  60 sleeping,   0 stopped,   0 zombie
> Cpu0  :  0.0%us,100.0%sy,  0.0%ni,  0.0%id,  0.0%wa,  0.0%hi,  0.0%si,  0.0%st
> Cpu1  :  6.4%us, 87.2%sy,  0.0%ni,  5.8%id,  0.0%wa,  0.6%hi,  0.0%si,  0.0%st
> Mem:    895384k total,   283568k used,   611816k free,    35492k buffers
> Swap:  1959920k total,        0k used,  1959920k free,   163044k cached
>
>   PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
>  4632 root      20   0 17216  800  580 S  104  0.1   0:34.22 syslog-ng
> 28505 root      20   0  205m  11m 4024 S    6  1.3   0:21.16 X
> 28518 root      20   0 56292 5652 4492 S    1  0.6   0:01.80 fluxbox
>     1 root      20   0  3724  608  508 S    0  0.1   0:00.36 init
>
> So far I have no clue why C1E-idle in conjunction with virtual wire
> mode causes this strange behaviour.
>
> ... and I start to think about the root cause of all this.
>
> I've performed similar tests under X with the IRQ0/INT0 configuration and
> I did not see above symptoms.

So lets fall back to the IRQ0/INT0 configuration on this box.

This basically restores the dont-use-the-lapic-timer exception mechanism
that was unconditional on this box prior commit 8750bf5 ("x86: add C1E
aware idle function").
Signed-off-by: NIngo Molnar <mingo@elte.hu>

When system have 4g less ram installed, and acpi table sit
near end of ram, make max_pfn cover them too,
so 64bit kernel don't need to mess up fixmap.
Signed-off-by: NYinghai Lu <yhlu.kernel@gmail.com>
Cc: "Suresh Siddha" <suresh.b.siddha@intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

fix:

 In file included from arch/x86/kernel/tlb_uv.c:14:
 include/asm/uv/uv_mmrs.h:986: error: redefinition of ‘union uvh_rh_gam_cfg_overlay_config_mmr_u’
 include/asm/uv/uv_mmrs.h:988: error: redefinition of ‘struct uvh_rh_gam_cfg_overlay_config_mmr_s’
 include/asm/uv/uv_mmrs.h:1064: error: redefinition of ‘union uvh_rh_gam_mmioh_overlay_config_mmr_u’
 include/asm/uv/uv_mmrs.h:1066: error: redefinition of ‘struct uvh_rh_gam_mmioh_overlay_config_mmr_s’

caused by another duplicate section (cut & paste error) in commit
5d061e39 "x86, uv: update x86 mmr list for SGI uv".
Signed-off-by: NIngo Molnar <mingo@elte.hu>

fix:

In file included from arch/x86/kernel/genx2apic_uv_x.c:25:
include/asm/uv/uv_mmrs.h:986: error: redefinition of ‘union uvh_rh_gam_cfg_overlay_config_mmr_u’
include/asm/uv/uv_mmrs.h:988: error: redefinition of ‘struct uvh_rh_gam_cfg_overlay_config_mmr_s’
include/asm/uv/uv_mmrs.h:1064: error: redefinition of ‘union uvh_rh_gam_mmioh_overlay_config_mmr_u’
include/asm/uv/uv_mmrs.h:1066: error: redefinition of ‘struct uvh_rh_gam_mmioh_overlay_config_mmr_s’

caused by duplicate section (cut & paste error) in commit
5d061e39 "x86, uv: update x86 mmr list for SGI uv".
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Rename the paravirtualized calculate_cpu_khz to calibrate_tsc.
In all cases, we actually calibrate_tsc and use that as the cpu_khz value.
Signed-off-by: NAlok N Kataria <akataria@vmware.com>
Signed-off-by: NDan Hecht <dhecht@vmware.com>
Cc: Dan Hecht <dhecht@vmware.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Unify the clocksource code.
Unify the tsc_init code.
Signed-off-by: NAlok N Kataria <akataria@vmware.com>
Signed-off-by: NDan Hecht <dhecht@vmware.com>
Cc: Dan Hecht <dhecht@vmware.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Merge the tsc calibration code for the 32bit and 64bit kernel.
The paravirtualized calculate_cpu_khz for 64bit now points to the correct
tsc_calibrate code as in 32bit.
Original native_calculate_cpu_khz for 64 bit is now called as calibrate_cpu.

Also moved the recalibrate_cpu_khz function in the common file.
Note that this function is called only from powernow K7 cpu freq driver.
Signed-off-by: NAlok N Kataria <akataria@vmware.com>
Signed-off-by: NDan Hecht <dhecht@vmware.com>
Cc: Dan Hecht <dhecht@vmware.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This patch updates the X86 mmr list for SGI uv.
Signed-off-by: NDimitri Sivanich <sivanich@sgi.com>
Cc: Jack Steiner <steiner@sgi.com>
Cc: Russ Anderson <rja@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Create page table entries to map the SGI UV chipset GRU. local MMR &
global MMR ranges.
Signed-off-by: NJack Steiner <steiner@sgi.com>
Cc: linux-mm@kvack.org
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Add boot-time function for creating additional 2MB page table entries for
mapping chipset specific cached/uncached ranges.
Signed-off-by: NJack Steiner <steiner@sgi.com>
Cc: linux-mm@kvack.org
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This patch uses the /sys/firmware/memmap interface provided in the last patch
on the x86 architecture when E820 is used. The patch copies the E820
memory map very early, and registers the E820 map afterwards via
firmware_map_add_early().
Signed-off-by: NBernhard Walle <bwalle@suse.de>
Acked-by: NGreg KH <gregkh@suse.de>
Acked-by: NVivek Goyal <vgoyal@redhat.com>
Cc: kexec@lists.infradead.org
Cc: yhlu.kernel@gmail.com
Signed-off-by: NIngo Molnar <mingo@elte.hu>

use ACPI_NUMA directly

and move srat_32.c to mm/
Signed-off-by: NYinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

From: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>

Older x86-32 processors do not support global mappings (PGD), so must
only use it if the processor supports it.

The _PAGE_KERNEL* flags always have _PAGE_KERNEL set, since logically
we always want it set.

This is OK even on processors which do not support PGD, since all
_PAGE flags are masked with __supported_pte_mask before being turned
into a real in-pagetable pte.  On 32-bit systems, __supported_pte_mask
is initialized to not contain _PAGE_GLOBAL, and it is then added if
the CPU is found to support it.

The x86-32 code used to use __PAGE_KERNEL/__PAGE_KERNEL_EXEC for this
purpose, but they're now redundant and can be removed.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Consistently set _PAGE_GLOBAL in _PAGE_KERNEL flags.  This makes 32-
and 64-bit code consistent, and removes some special cases where
__PAGE_KERNEL* did not have _PAGE_GLOBAL set, causing confusion as a
result of the inconsistencies.

This patch only affects x86-64, which generally always supports PGD.
The x86-32 patch is next.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

fix:

arch/x86/kernel/built-in.o: In function `setup_arch':
: undefined reference to `prefill_possible_map'
Signed-off-by: NIngo Molnar <mingo@elte.hu>

call it right after we are done with MADT/mptable handling, instead of
doing that in setup_per_cpu_areas() later on...

this way for_possible_cpu() can be used early.
Signed-off-by: NYinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

move out e820_register_active_regions from non numa zones_sizes_init()
and remove numa version zones_sizes_init().

and let 32 bit call remove_all_active_ranges() in setup_arch() directly
like 64-bit
Signed-off-by: NYinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

so it has a more meaningful name.
also change it to static.
Signed-off-by: NYinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

e820_search_gap also take a end_addr parameter to limit search from
start_addr to end_addr.
Signed-off-by: NAloK N Kataria <akataria@vmware.com>
Acked-by: NYinghai Lu <yhlu.kernel@gmail.com>
Cc: "lenb@kernel.org" <lenb@kernel.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

  * When CONFIG_DEBUG_PER_CPU_MAPS is set, the node passed to
    node_to_cpumask and node_to_cpumask_ptr should be validated.
    If invalid, then a dump_stack is performed and a zero cpumask
    is returned.

v2: Slightly different version to remove a compiler warning.
v3: Redone to reflect moving setup.c -> setup_percpu.c
Signed-off-by: NMike Travis <travis@sgi.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
Cc: "akpm@linux-foundation.org" <akpm@linux-foundation.org>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Ying Huang would like setup_data to be reserved, but not included in the
no save range.

Here we try to modify the e820 table to reserve that range early.
also add that in early_res in case bootloader messes up with the ramdisk.

other solution would be
1. add early_res_to_highmem...
2. early_res_to_e820...
but they could reserve another type memory wrongly, if early_res has some
resource reserved early, and not needed later, but it is not removed from
early_res in time. Like the RAMDISK (already handled).
Signed-off-by: NYinghai Lu <yhlu.kernel@gmail.com>
Cc: andi@firstfloor.org
Tested-by: NHuang, Ying <ying.huang@intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

now that the early-ioremap code is unified, move the prototypes too from
io_32.h to io.h.

this fixes:

arch/x86/kernel/setup.c:531: error: implicit declaration of function ‘early_ioremap_init'

on 64-bit.
Signed-off-by: NIngo Molnar <mingo@elte.hu>

change the enable_local_apic to static force_enable_local_apic for 32bit
Signed-off-by: NYinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Make sure SWAPGS and PARAVIRT_ADJUST_EXCEPTION_FRAME are properly
defined when CONFIG_PARAVIRT is off.

Fixes Ingo's build failure:
arch/x86/kernel/entry_64.S: Assembler messages:
arch/x86/kernel/entry_64.S:1201: Error: invalid character '_' in mnemonic
arch/x86/kernel/entry_64.S:1205: Error: invalid character '_' in mnemonic
arch/x86/kernel/entry_64.S:1209: Error: invalid character '_' in mnemonic
arch/x86/kernel/entry_64.S:1213: Error: invalid character '_' in mnemonic
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Mark McLoughlin <markmc@redhat.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Stephen Tweedie <sct@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

instead of calling it from trap_init()

also move init ioapic mapping out of apic_32.c

so 32 bit do same as 64 bit
Signed-off-by: NYinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

asm-x86/paravirt.h already have protection with CONFIG_PARAVIRT inside
Signed-off-by: NYinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

64-bit Xen pushes a couple of extra words onto an exception frame.
Add a hook to deal with them.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

It's never safe to call a swapgs pvop when the user stack is current -
it must be inline replaced.  Rather than making a call, the
SWAPGS_UNSAFE_STACK pvop always just puts "swapgs" as a placeholder,
which must either be replaced inline or trap'n'emulated (somehow).
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

In a 64-bit system, we need separate sysret/sysexit operations to
return to a 32-bit userspace.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citirx.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

There's no need to combine restoring the user rsp within the sysret
pvop, so split it out.  This makes the pvop's semantics closer to the
machine instruction.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citirx.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Don't conflate sysret and sysexit; they're different instructions with
different semantics, and may be in use at the same time (at least
within the same kernel, depending on whether its an Intel or AMD
system).

sysexit - just return to userspace, does no register restoration of
    any kind; must explicitly atomically enable interrupts.

sysret - reloads flags from r11, so no need to explicitly enable
    interrupts on 64-bit, responsible for restoring usermode %gs
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citirx.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

We will need to set a pte on l3_user_pgt. Extract set_pte_vaddr_pud()
from set_pte_vaddr(), that will accept the l3 page table as parameter.

This change should be a no-op for existing code.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NMark McLoughlin <markmc@redhat.com>
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Because Xen doesn't support PSE mappings in guests, all code which
assumed the presence of PSE has been changed to fall back to smaller
mappings if necessary.  As a result, PSE is optional rather than
required (though still used whereever possible).
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Split x86_64_start_kernel() into two pieces:

   The first essentially cleans up after head_64.S.  It clears the
   bss, zaps low identity mappings, sets up some early exception
   handlers.

   The second part preserves the boot data, reserves the kernel's
   text/data/bss, pagetables and ramdisk, and then starts the kernel
   proper.

This split is so that Xen can call the second part to do the set up it
needs done.  It doesn't need any of the first part setups, because it
doesn't boot via head_64.S, and its redundant or actively damaging.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This matches 32 bit.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Set __PAGE_OFFSET to the most negative possible address +
16*PGDIR_SIZE.  The gap is to allow a space for a hypervisor to fit.
The gap is more or less arbitrary, but it's what Xen needs.

When booting native, kernel/head_64.S has a set of compile-time
generated pagetables used at boot time.  This patch removes their
absolutely hard-coded layout, and makes it parameterised on
__PAGE_OFFSET (and __START_KERNEL_map).
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

On 32-bit it's best to use a %cs: prefix to access memory where the
other segments may not bet set up properly yet.  On 64-bit it's best
to use a rip-relative addressing mode.  Define PARA_INDIRECT() to
abstract this and generate the proper addressing mode in each case.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Rather than just jumping to 0 when there's a missing operation, raise a BUG.
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Add hooks which are called at pgd_alloc/free time.  The pgd_alloc hook
may return an error code, which if non-zero, causes the pgd allocation
to be failed.  The hooks may be used to allocate/free auxillary
per-pgd information.

also fix:

> * Ingo Molnar <mingo@elte.hu> wrote:
>
>  include/asm/pgalloc.h: In function ‘paravirt_pgd_free':
>  include/asm/pgalloc.h:14: error: parameter name omitted
>  arch/x86/kernel/entry_64.S: In file included from
>  arch/x86/kernel/traps_64.c:51:include/asm/pgalloc.h: In function ‘paravirt_pgd_free':
>  include/asm/pgalloc.h:14: error: parameter name omitted
Signed-off-by: NJeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: xen-devel <xen-devel@lists.xensource.com>
Cc: Stephen Tweedie <sct@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>