They look similar enough, and are merged. Only difference
(zap_low_mapping for i386) is inside ifdef
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

it is practically the same between arches now, so it is
moved to smpboot.c. Minor differences (gdt initialization)
live inside an ifdef
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

It now looks the same between architectures, so we
merge it in smpboot.c. Minor differences goes inside
an ifdef
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This is a very large patch, because it depends on a lot
of auxiliary static functions. But they all have been modified
to the point that they're sufficiently close now. So they're just
merged in smpboot.c
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This is to match i386. The former name was cuter,
but the current is more meaningful and more general,
since cpu_id can be a logical id.
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

wakeup_secondary_via_INIT => wakeup_secondary_cpu.

This is to match i386, where init is not always used.
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

After the inclusion, a lot of files needs fixing for conflicts,
some of them in the headers themselves, to accomodate for both
i386 and x86_64 versions.

[ mingo@elte.hu: build fix ]
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This patch provides minor adjustments for do_boot_cpus
in both architectures to allow for integration
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

We introduce empty macros just to make them look like the same
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

We do it and also fix conflicts, which makes x86_64 automatically
closer to i386
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Do it and also fix conflicts, which automatically makes
x86_64 look closer to i386
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

the cpu count is changed accordingly: now, what matters is
online cpus.
Also, we add those functions for x86_64
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

goal is to have i386 and x86_64 closer, so we
add barriers to match
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This patch does not change the behaviour of x86_64, since APIC_INTEGRATED
is always defined as (1). But the code now matches exactly i386 version
(well, this part of the code, at least)
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

it does not need to be inside lock. Do the way i386 does.
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

now that it is the same between arches, put it into smpboot.c
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Call it conditionally for secondary cpus. This behaviour
matches i386
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This will ease integration with i386
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

It is used to match i386. The definition for the non-paravirt
case is moved to smp.h instead of smp_32.h
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This patch replaces apic_read() for apic_read_around()
and apic_write for apic_write_around() in smpboot_64.c
We do it to have a common usage between x86_64 and i386.
In the former, it will always simply expand to apic_write
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

wraps the busy loop for wait_for_init_deasserted() in a function,
so smp_callin in x86_64 looks like more i386
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

setup_trampoline() looks very similar between architectures, and this
patch unifies them. The i386 version allocates bootmem memory, while
the x86_64 version uses a fixed address.

In this patch, we initialize the global trampoline_base to the x86_64 version,
and i386 allocation can later override it.
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

In here, they can serve both architectures
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

They are now equal, and are moved to a common file
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

it is already used in x86_64. In i386, it only
removes from cpu_online_map
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This lock does not protect cpu_online_map, so its
length can be shortened, and in some cases, removed.
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

it is equal between architectures
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

set_cpu_sibling_map() and remove_sibling_info() are
equal between architectures, and are now moved to common
file
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

this is done to match i386
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

And move its extern definition to smp.h, the common header
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

move definitions that are now equal in type from
smpboot_{32,64}.c to smpboot.c

cpu_callin_map is put temporarily in smp_64.h (already
exists in smp_32.h), and will soon be merged.
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

this patches moves prefill_possible_map() to smpboot.c
Right now it is x86_64-specific, but nothing intrinsically
prevents it to be used by i386
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

definition is moved to common header. x86_64 version is now called
native_smp_cpus_done
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

definition is moved to common header. x86_64 version is now called
native_smp_prepare_cpus
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

definition is moved to common header. x86_64 version is now called
native_prepare_boot_cpu
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

function definition is moved to common header. x86_64 version
is now called native_cpu_up
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Currently, c_idle is declared in the stack, and thus, have no static address.

Peter Zijlstra points out this simple solution, in which c_idle.work
is initializated separatedly. Note that the INIT_WORK macro has a static
declaration of a key inside.
Signed-off-by: NGlauber Costa <gcosta@redhat.com>
Acked-by: NPeter Zijlstra <pzijlstr@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Silence the following warning:
WARNING: o-x86_64/arch/x86/kernel/built-in.o(.text+0x17cd3): Section mismatch in reference from the function remove_cpu_from_maps() to the variable .cpuinit.data:cpu_initialized

remove_cpu:maps() had a single user: __cpu_disable() so
mark it static and annotate it with __ref to silence the
warning from modpost.

_cpu_disable() has a single user in kernel/cpu.c:
 => take_cpu_down()
    which again has a single user in the following call:
    => __stop_machine_run(take_cpu_down, &tcd_param, cpu);
Here a kthread is created.

So maybe the warning is correct and the right fix is to
remove the __cpuinitdata annotation of cpu_initialized?

Note: The analysis were disturbed by the fact that we had a variable
      with the same name in cpu/common.c - but this is 32 bit only]
Note: Should smpboot_64 use cpu_clear()?
Signed-off-by: NSam Ravnborg <sam@ravnborg.org>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

in init/main.c boot_cpu_init() does that before.
Signed-off-by: NYinghai Lu <yinghai.lu@sun.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

Change the following static arrays sized by NR_CPUS to
per_cpu data variables:

	char cpu_to_node_map[NR_CPUS];

fixup:

  - Split cpu_to_node function into "early" and "late" versions
    so that x86_cpu_to_node_map_early_ptr is not EXPORT'ed and
    the cpu_to_node inline function is more streamlined.

  - This also involves setting up the percpu maps as early as possible.

  - Fix X86_32 NUMA build errors that previous version of this
    patch caused.

V2->V3:
    - add early_cpu_to_node function to keep cpu_to_node efficient
    - move and rename smp_set_apicids() to setup_percpu_maps()
    - call setup_percpu_maps() as early as possible

V1->V2:
    - Removed extraneous casts
    - Fix !NUMA builds with '#ifdef CONFIG_NUMA"
Signed-off-by: NMike Travis <travis@sgi.com>
Reviewed-by: NChristoph Lameter <clameter@sgi.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>