Commit Graph

88560 Commits

Author SHA1 Message Date
Glauber Costa
89b08200ad x86: make x86_64 accept the max_cpus parameter
The parameter passing parsing is done in the common smpboot.c

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
69c18c15d3 x86: merge __cpu_disable and cpu_die
They are now equal, and are moved to a common file

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
e9a6cb96fa x86: do not clear cpu_online_map
it was already cleared two lines above, and so, this removal
is bogus

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
045f9d2202 x86: use remove_from_maps in cpu_disable
it is already used in x86_64. In i386, it only
removes from cpu_online_map

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
fc25da9ec6 x86: remove vector_lock around cpu_online_map
This lock does not protect cpu_online_map, so its
length can be shortened, and in some cases, removed.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
70708a18e8 x86: move cpu_coregroup_map to common file
it is equal between architectures

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
768d95051b x86: move sibling functions to common file
set_cpu_sibling_map() and remove_sibling_info() are
equal between architectures, and are now moved to common
file

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
1dbb4726fa x86: move hotplug related extern definitions to smp.h
definitions that are inside CONFIG_HOTPLUG_CPU in
the arch-specific smp*.h files are moved to common
header

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
61d5989973 x86: make remove_siblinginfo non-static
this is done to match i386

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
1452207689 x86: make set_cpu_sibling_map nonstatic
And move its extern definition to smp.h, the common header

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
a355352b97 x86: move equal types to common file
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: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
5382e89670 x86: adjust types in smpcommon_32.c
so they can have the same type as x86_64

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
fe6762030c x86: remove cpu_llc_id from processor.h
it is already defined in smp.h

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
0941ecb55f x86: get rid of smp_32.c and smp_64.c
This patch merges the copyright notices, and valuable
comments that were left back on smp_{32,64}.c. With that,
files are empty, and are deleted

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
c048fdfe61 x86: create tlb files
this patch creates tlb_32.c and tlb_64.c, with
tlb-related functions that used to live in smp*.c files.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
8202350367 x86: create ipi.c
This patch moves all ipi and apic related functions
from smp_32.c to ipi.c

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
f9e47a126b x86: create smp.c
this patch moves all the functions and data structures that look
like exactly the same from smp_{32,64}.c to smp.c

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
377d698426 x86: unify smp_send_stop
function definition is moved to common header.
x86_64 version is now called native_smp_send_stop

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:56 +02:00
Glauber Costa
321183c145 x86: add reboot_force test to native_smp_send_stop
This can be safely added to i386. After that,
functions look exactly the same for both arches

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:55 +02:00
Glauber Costa
3be5b49e8f x86: make stop_this_cpu looks exactly equal in both arches
with the hlt_works change, it is possible to have i386
and x86_64 stop_this_cpu() looking exactly the same. They
can, after that, be merged.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:55 +02:00
Glauber Costa
3d3f487c58 x86: provide hlt_works function.
In x86_64, hlt always work. in i386, we'll query the cpuinfo associated
with this cpu

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:55 +02:00
Glauber Costa
2513926c28 x86: change x86_64 smp_call_function_mask to look alike i386
the two versions (the inner version, and the outer version, that takes
the locks) of smp_call_function_mask are made into one. With the changes,
i386 and x86_64 versions look exactly the same.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:55 +02:00
Glauber Costa
3a36d1e435 x86: provide __smp_call_function
This function is used in smp_send_stop(). It's like
smp_call_function_mask, but always go to all online cpus,
and does not take any locks.

It is added to x86_64, but will soon be unified in a common file

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:55 +02:00
Glauber Costa
e32640a2cd x86: create smpcommon.c
This patch creates smpcommon.c with functions that are
equal between architectures. The i386-only init_gdt
is ifdef'd.

Note that smpcommon.o figures twice in the Makefile:
this is because sub-architectures like voyager that does
not use the normal smp_$(BITS) files also have to access them

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:55 +02:00
Glauber Costa
3428f3d6ca x86: remove irqs disabled warning.
there's already a warning in the topmost function

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:55 +02:00
Glauber Costa
7930e53422 x86: remove export for smp_call_function_mask.
with this removal, exports for both i386 and x86_64,
regarding the "smp_call_function" series are now the same.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:54 +02:00
Glauber Costa
68a1c3f8cd x86: move prefill_possible_map to common file
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: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:53 +02:00
Glauber Costa
7b1292e237 x86: use disabled_cpus in i386
this patch allows a cpu to be marked as present but disabled in i386,
just as x86_64 currently does.

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:53 +02:00
Glauber Costa
93b016f8f3 x86: move disabled_cpus to common header
disabled_cpus is (up to now) a x86_64-only contruction.
But it's extern declaration can be moved to common header anyway

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:53 +02:00
Glauber Costa
c559764923 x86: unify smp_cpus_done
definition is moved to common header. x86_64 version is now called
native_smp_cpus_done

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:53 +02:00
Glauber Costa
7557da6720 x86: unify smp_prepare_cpus
definition is moved to common header. x86_64 version is now called
native_smp_prepare_cpus

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:53 +02:00
Glauber Costa
1e3fac83da x86: unify prepare_boot_cpu
definition is moved to common header. x86_64 version is now called
native_prepare_boot_cpu

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:53 +02:00
Glauber Costa
71d195492a x86: unify __cpu_up.
function definition is moved to common header. x86_64 version
is now called native_cpu_up

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:53 +02:00
Glauber Costa
64b1a21e09 x86: unify smp_call_function_mask
definition is moved to common header, x86_64 function name
now is native_smp_call_function_mask

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:53 +02:00
Glauber Costa
8678969e60 x86: merge smp_send_reschedule
function definition is moved to common header, x86_64 version is now called
native_smp_send_reschedule

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:53 +02:00
Glauber Costa
c76cb36846 x86: move smp_ops extern declaration to common header
the smp_ops symbol is temporarily defined in smp_64.c, but it will soon
be unified

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:53 +02:00
Glauber Costa
16694024d6 x86: define smp_ops in common header
x86_64 will benefit from it
Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Glauber Costa
53ebef4961 x86: merge extern variables definitions
move extern definitions that are the same between smp_{32,64}.h
to smp.h

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Glauber Costa
639acb16e6 x86: merge extern function definitions
move extern function definitions that are the same between smp_{32,64}.h
to smp.h

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Glauber Costa
c27cfeffad x86: commonize smp.h
this is the first step of integrating smp.h between x86_64
and i386

Signed-off-by: Glauber Costa <gcosta@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Ingo Molnar
8b6451fe5c x86: fix switch_to() clobbers
Liu Pingfan noticed that switch_to() clobbers more registers than its
asm constraints specify.

We get away with this due to luck mostly - schedule()
by its nature only has 'local' state which gets reloaded
automatically. Fix it nevertheless, we could hit this anytime.

it turns out that with the extra constraints gcc manages to make
schedule() even more compact:

   text	   data	    bss	    dec	    hex	filename
  28626	    684	   2640	  31950	   7cce	sched.o.before
  28613	    684	   2640	  31937	   7cc1	sched.o.after

Reported-by: Liu Pingfan <kernelfans@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Ingo Molnar
23b55bd9f3 x86: clean up switch_to()
Make the code more readable and more hackable:

 - use symbolic asm parameters
 - use readable indentation
 - add comments that explains the details

No code changed:

kernel/sched.o:

   text	   data	    bss	    dec	    hex	filename
  28626	    684	   2640	  31950	   7cce	sched.o.before
  28626	    684	   2640	  31950	   7cce	sched.o.after

md5:
   2823d406c18b781975cdb2e7cfea0059  sched.o.before.asm
   2823d406c18b781975cdb2e7cfea0059  sched.o.after.asm

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Alexander van Heukelum
ecd94c0809 x86: reserve end-of-conventional-memory to 1MB, 64-bit, use paravirt_enabled
Jeremy Fitzhardinge pointed out that looking at the boot_params
struct to determine if the system is running in a paravirtual
environment is not reliable for the Xen case, currently. He also
points out that there already exists a function to determine if
the system is running in a paravirtual environment. So let's use
that instead. This gets rid of the preprocessor test too.

Signed-off-by: Alexander van Heukelum <heukelum@fastmail.fm>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Alexander van Heukelum
2fde61fdb0 x86: reserve end-of-conventional-memory to 1MB, 32-bit, use paravirt_enabled
Jeremy Fitzhardinge pointed out that looking at the boot_params
struct to determine if the system is running in a paravirtual
environment is not reliable for the Xen case, currently. He also
points out that there already exists a function to determine if
the system is running in a paravirtual environment. So let's use
that instead. This gets rid of the preprocessor test too.

Signed-off-by: Alexander van Heukelum <heukelum@fastmail.fm>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Pavel Machek
0d7a1819e9 x86: wmb() confusion in system.h
Comment says wmb is a nop, but it is implemented as lock addl
below... Should it be compiled to nop if we know we are running on
"good" Intel cpu?

At least remove confusing comment for now.

Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Rusty Russell
3c2047cd32 x86: if we cannot calibrate the TSC, we panic.
The current tsc_init() clears the TSC feature bit if the TSC khz
cannot be calculated, causing us to panic in
arch/x86/kernel/cpu/bugs.c check_config().  We should simply mark it
unstable.

Frankly, someone should take an axe to this code.  mark_tsc_unstable()
not only marks it unstable, but sets tsc_enabled to 0, which seems
redundant but is actually important here because means it won't be
used by sched_clock() either.  Perhaps a tristate enum "UNUSABLE,
UNSTABLE, OK" would be clearer, and separate mark_tsc_unstable() and
mark_tsc_broken() functions?

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Ingo Molnar
9fc34113f6 x86: debug pmd_bad()
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Ingo Molnar
40869cd038 x86: redo cded932b75
redo commit cded932b75.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Ingo Molnar
ba748d221e x86: warn about RAM pages in ioremap()
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00
Alexander van Heukelum
320a6b2efc x86: reserve end-of-conventional-memory to 1MB, 64-bit
This patch is an add-on to the 64-bit ebda patch. It makes
the functions reserve_ebda_region (renamed from reserve_ebda)
and copy_e820_map equal to the 32-bit versions of the previous
patch.

Changes:

Use u64 and u32 for local variables in copy_e820_map.

The amount of conventional memory and the start of the EBDA are
detected by reading the BIOS data area directly. Paravirtual
environments do not provide this area, so we bail out early
in that case. They will just have to set up a correct memory
map to start with.

Add a safety net for zeroed out BIOS data area.

Signed-off-by: Alexander van Heukelum <heukelum@fastmail.fm>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-04-17 17:40:52 +02:00