Commit graph

600 commits

Author SHA1 Message Date
Jeremy Fitzhardinge
5ba7c91341 x86: unify pud_present
Impact: cleanup

Unify and demacro pud_present.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
2009-02-06 12:31:07 -08:00
Jeremy Fitzhardinge
7c683851d9 x86: unify pte_present
Impact: cleanup

Unify and demacro pte_present.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
2009-02-06 12:28:37 -08:00
Jeremy Fitzhardinge
8de01da35e x86: unify pte_same
Impact: cleanup

Unify and demacro pte_same.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
2009-02-06 12:26:08 -08:00
Jeremy Fitzhardinge
a034a010f4 x86: unify pte_none
Impact: cleanup

Unify and demacro pte_none.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
2009-02-06 12:26:08 -08:00
Ingo Molnar
4f179d1218 x86, numaq: cleanups
Also move xquad_portio over to where it's allocated.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-05 22:30:14 +01:00
Ingo Molnar
9d45cf9e36 Merge branch 'x86/urgent' into x86/apic
Conflicts:
	arch/x86/mach-default/setup.c

Semantic merge:
	arch/x86/kernel/irqinit_32.c

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-05 22:30:01 +01:00
Yinghai Lu
c5e9548203 x86: move default_ipi_xx back to ipi.c
Impact: cleanup

only leave _default_ipi_xx etc in .h

Beyond the cleanup factor, this saves a bit of code size as well:

    text	   data	    bss	    dec	            hex	filename
 7281931	1630144	1463304	10375379	 9e50d3	vmlinux.before
 7281753	1630144	1463304	10375201	 9e5021	vmlinux.after

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-05 22:27:56 +01:00
Ingo Molnar
65a4e574d2 smp, generic: introduce arch_disable_smp_support() instead of disable_ioapic_setup()
Impact: cleanup

disable_ioapic_setup() in init/main.c is ugly as the function is
x86-specific. The #ifdef inline prototype there is ugly too.

Replace it with a generic arch_disable_smp_support() function - which
has a weak alias for non-x86 architectures and for non-ioapic x86 builds.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-05 22:27:54 +01:00
Jeremy Fitzhardinge
b534816b55 x86: don't apply __supported_pte_mask to non-present ptes
On an x86 system which doesn't support global mappings,
__supported_pte_mask has _PAGE_GLOBAL clear, to make sure it never
appears in the PTE.  pfn_pte() and so on will enforce it with:

static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
{
	return __pte((((phys_addr_t)page_nr << PAGE_SHIFT) |
		      pgprot_val(pgprot)) & __supported_pte_mask);
}

However, we overload _PAGE_GLOBAL with _PAGE_PROTNONE on non-present
ptes to distinguish them from swap entries.  However, applying
__supported_pte_mask indiscriminately will clear the bit and corrupt the
pte.

I guess the best fix is to only apply __supported_pte_mask to present
ptes.  This seems like the right solution to me, as it means we can
completely ignore the issue of overlaps between the present pte bits and
the non-present pte-as-swap entry use of the bits.

__supported_pte_mask contains the set of flags we support on the
current hardware.  We also use bits in the pte for things like
logically present ptes with no permissions, and swap entries for
swapped out pages.  We should only apply __supported_pte_mask to
present ptes, because otherwise we may destroy other information being
stored in the ptes.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-02-04 21:33:09 -08:00
Hiroshi Shimamoto
18114f6135 x86: uaccess: use errret as error value in __put_user_size()
Impact: cleanup

In __put_user_size() macro errret is used for error value.
But if size is 8, errret isn't passed to__put_user_asm_u64().
This behavior is inconsistent.

Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2009-02-04 17:28:21 -08:00
H. Peter Anvin
327641da8e Merge branch 'core/percpu' into x86/paravirt 2009-02-04 16:58:26 -08:00
Huang Ying
f5deb79679 x86: kexec: Use one page table in x86_64 machine_kexec
Impact: reduce kernel BSS size by 7 pages, improve code readability

Two page tables are used in current x86_64 kexec implementation. One
is used to jump from kernel virtual address to identity map address,
the other is used to map all physical memory. In fact, on x86_64,
there is no conflict between kernel virtual address space and physical
memory space, so just one page table is sufficient. The page table
pages used to map control page are dynamically allocated to save
memory if kexec image is not loaded. ASM code used to map control page
is replaced by C code too.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-02-03 18:29:18 -08:00
Jeremy Fitzhardinge
0eb592dbba x86/paravirt: return full 64-bit result
Impact: Bug fix

A hunk went missing in the original patch, and callee-save callsites were
not marked as returning the upper 32-bit of result, causing Badness.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-02-03 16:05:00 -08:00
Ingo Molnar
063f8913af x86: document 64-bit and 32-bit function call convention ABI
- also clean up the calling.h file a tiny bit

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-02-03 19:48:18 +01:00
Jaswinder Singh Rajput
15c554439f headers_check fix: x86, setup.h
fix the following 'make headers_check' warning:

  usr/include/asm/setup.h:16: extern's make no sense in userspace
  usr/include/asm/setup.h:17: extern's make no sense in userspace
  usr/include/asm/setup.h:23: extern's make no sense in userspace
  usr/include/asm/setup.h:24: extern's make no sense in userspace
  usr/include/asm/setup.h:51: extern's make no sense in userspace
  usr/include/asm/setup.h:52: extern's make no sense in userspace

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
2009-02-02 23:27:48 +05:30
Jaswinder Singh Rajput
8d4b498119 headers_check fix: x86, prctl.h
fix the following 'make headers_check' warning:

  usr/include/asm/prctl.h:10: extern's make no sense in userspace

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
2009-02-02 23:27:09 +05:30
Jeremy Fitzhardinge
e584f559c7 x86/paravirt: don't restore second return reg
Impact: bugfix

In the 32-bit calling convention, %eax:%edx is used to return 64-bit
values.  Don't save and restore %edx around wrapped functions, or they
can't return a full 64-bit result.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-02-02 08:06:30 -08:00
Jeremy Fitzhardinge
11e3a840cd x86: split loading percpu segments from loading gdt
Impact: split out a function, no functional change

Xen needs to be able to access percpu data from very early on.  For
various reasons, it cannot also load the gdt at that time.   It does,
however, have a pefectly functional gdt at that point, so there's no
pressing need to reload the gdt.

Split the function to load the segment registers off, so Xen can call
it directly.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-31 14:28:54 +09:00
Brian Gerst
552be871e6 x86: pass in cpu number to switch_to_new_gdt()
Impact: cleanup, prepare for xen boot fix.

Xen needs to call this function very early to setup the GDT and
per-cpu segments.  Remove the call to smp_processor_id() and just
pass in the cpu number.

Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-31 14:28:50 +09:00
Ingo Molnar
d8106d2e24 x86, vm86: clean up invalid_vm86_irq()
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-31 04:21:26 +01:00
Ingo Molnar
009eb3fe14 x86, irq: describe NR_IRQ sizing details, clean up
Impact: cleanup

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-31 04:21:25 +01:00
Ingo Molnar
c379698fda x86, irq_vectors.h: remove needless includes
Reduce include file dependencies a bit - remove the two headers
that are included in irq_vectors.h.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-31 04:21:24 +01:00
Ingo Molnar
9fc2e79d4f x86, irq: add IRQ layout comments
Describe the layout of x86 trap/exception/IRQ vectors and clean
up indentation and other small details.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-31 04:21:23 +01:00
Ingo Molnar
3e92ab3d7e x86, irqs, voyager: remove Voyager quirk
Remove a Voyager complication from the generic irq_vectors.h header.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-31 04:21:22 +01:00
Ingo Molnar
ed74ca6d5a x86, voyager: move Voyager-specific defines to voyager.h
They dont belong into the generic headers.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-31 04:21:21 +01:00
Ingo Molnar
647ad94fc0 x86, apic: clean up spurious vector sanity check
Move the spurious vector sanity check to the place where it's
defined - out of a .c file.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-31 04:21:20 +01:00
Ingo Molnar
5da690d29f x86, apic: unify the APIC vector enumeration
Most of the vector layout on 32-bit and 64-bit is identical now,
so eliminate the duplicated enumeration of the vectors.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-31 04:21:19 +01:00
Ingo Molnar
193c81b979 x86, irq: add LOCAL_PERF_VECTOR
Add a slot for the performance monitoring interrupt. Not yet used
by any subsystem - but the hardware has it. (This eases integration
with performance monitoring code.)

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-31 02:23:27 +01:00
Jeremy Fitzhardinge
4767afbf1f x86/paravirt: fix missing callee-save call on pud_val
Impact: Fix build when CONFIG_PARAVIRT_DEBUG is enabled

Fix missed convertion to using callee-saved calls for pud_val, which
causes a compile error when CONFIG_PARAVIRT_DEBUG is enabled.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2009-01-30 14:52:39 -08:00
Jeremy Fitzhardinge
da5de7c22e x86/paravirt: use callee-saved convention for pte_val/make_pte/etc
Impact: Optimization

In the native case, pte_val, make_pte, etc are all just identity
functions, so there's no need to clobber a lot of registers over them.

(This changes the 32-bit callee-save calling convention to return both
EAX and EDX so functions can return 64-bit values.)

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-01-30 14:51:45 -08:00
Jeremy Fitzhardinge
791bad9d28 x86/paravirt: implement PVOP_CALL macros for callee-save functions
Impact: Optimization

Functions with the callee save calling convention clobber many fewer
registers than the normal C calling convention.  Implement variants of
PVOP_V?CALL* accordingly.  This only bothers with functions up to 3
args, since functions with more args may as well use the normal
calling convention.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-01-30 14:51:45 -08:00
Jeremy Fitzhardinge
ecb93d1ccd x86/paravirt: add register-saving thunks to reduce caller register pressure
Impact: Optimization

One of the problems with inserting a pile of C calls where previously
there were none is that the register pressure is greatly increased.
The C calling convention says that the caller must expect a certain
set of registers may be trashed by the callee, and that the callee can
use those registers without restriction.  This includes the function
argument registers, and several others.

This patch seeks to alleviate this pressure by introducing wrapper
thunks that will do the register saving/restoring, so that the
callsite doesn't need to worry about it, but the callee function can
be conventional compiler-generated code.  In many cases (particularly
performance-sensitive cases) the callee will be in assembler anyway,
and need not use the compiler's calling convention.

Standard calling convention is:
	 arguments	    return	scratch
x86-32	 eax edx ecx	    eax		?
x86-64	 rdi rsi rdx rcx    rax		r8 r9 r10 r11

The thunk preserves all argument and scratch registers.  The return
register is not preserved, and is available as a scratch register for
unwrapped callee code (and of course the return value).

Wrapped function pointers are themselves wrapped in a struct
paravirt_callee_save structure, in order to get some warning from the
compiler when functions with mismatched calling conventions are used.

The most common paravirt ops, both statically and dynamically, are
interrupt enable/disable/save/restore, so handle them first.  This is
particularly easy since their calls are handled specially anyway.

XXX Deal with VMI.  What's their calling convention?

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-01-30 14:51:45 -08:00
Jeremy Fitzhardinge
9104a18dcd x86/paravirt: selectively save/restore regs around pvops calls
Impact: Optimization

Each asm paravirt-ops call says what registers are available for
clobbering.  This patch makes use of this to selectively save/restore
registers around each pvops call.  In many cases this significantly
shrinks code size.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-01-30 14:51:44 -08:00
Jeremy Fitzhardinge
41edafdb78 x86/pvops: add a paravirt_ident functions to allow special patching
Impact: Optimization

Several paravirt ops implementations simply return their arguments,
the most obvious being the make_pte/pte_val class of operations on
native.

On 32-bit, the identity function is literally a no-op, as the calling
convention uses the same registers for the first argument and return.
On 64-bit, it can be implemented with a single "mov".

This patch adds special identity functions for 32 and 64 bit argument,
and machinery to recognize them and replace them with either nops or a
mov as appropriate.

At the moment, the only users for the identity functions are the
pagetable entry conversion functions.

The result is a measureable improvement on pagetable-heavy benchmarks
(2-3%, reducing the pvops overhead from 5 to 2%).

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-01-30 14:51:44 -08:00
H. Peter Anvin
9b7ed8faa0 Merge branch 'core/percpu' into x86/paravirt 2009-01-30 14:50:57 -08:00
Jaswinder Singh Rajput
7cff3608d2 headers_check fix: x86, swab.h
fix the following 'make headers_check' warnings:

  usr/include/asm/swab.h:4: include of <linux/types.h> is preferred over <asm/types.h>
  usr/include/asm/swab.h:7: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
2009-01-31 00:19:32 +05:30
Jaswinder Singh Rajput
2de548faa7 headers_check fix: x86, sigcontext32.h
fix the following 'make headers_check' warning:

  usr/include/asm/sigcontext32.h:20: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
2009-01-31 00:18:58 +05:30
Jaswinder Singh Rajput
e59afe6a21 headers_check fix: x86, sigcontext.h
fix the following 'make headers_check' warnings:

  usr/include/asm/sigcontext.h:5: include of <linux/types.h> is preferred over <asm/types.h>
  usr/include/asm/sigcontext.h:24: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
2009-01-31 00:18:30 +05:30
Jaswinder Singh Rajput
420ab35eef headers_check fix: x86, ptrace-abi.h
fix the following 'make headers_check' warnings:

  usr/include/asm/ptrace-abi.h:86: include of <linux/types.h> is preferred over <asm/types.h>
  usr/include/asm/ptrace-abi.h:93: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
2009-01-31 00:18:03 +05:30
Jaswinder Singh Rajput
d122072cc0 headers_check fix: x86, mtrr.h
fix the following 'make headers_check' warning:

  usr/include/asm/mtrr.h:61: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
2009-01-31 00:17:39 +05:30
Jaswinder Singh Rajput
999b697b9d headers_check fix: x86, mce.h
fix the following 'make headers_check' warnings:

  usr/include/asm/mce.h:7: include of <linux/types.h> is preferred over <asm/types.h>
  usr/include/asm/mce.h:29: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
2009-01-31 00:17:13 +05:30
Jaswinder Singh Rajput
cef3767852 headers_check fix: x86, kvm.h
fix the following 'make headers_check' warnings:

  usr/include/asm/kvm.h:9: include of <linux/types.h> is preferred over <asm/types.h>
  usr/include/asm/kvm.h:16: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
2009-01-31 00:16:44 +05:30
Jaswinder Singh Rajput
999721ca6d headers_check fix: x86, e820.h
fix the following 'make headers_check' warning:

  usr/include/asm/e820.h:44: found __[us]{8,16,32,64} type without #include <linux/types.h>

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
2009-01-31 00:16:22 +05:30
Ingo Molnar
c43e0e46ad Merge branch 'linus' into core/percpu
Conflicts:
	kernel/irq/handle.c
2009-01-30 18:23:30 +01:00
Yinghai Lu
43f39890db x86: seperate default_send_IPI_mask_sequence/allbutself from logical
Impact: 32-bit should use logical version

there are two version: for default_send_IPI_mask_sequence/allbutself
one in ipi.h and one in ipi.c for 32bit

it seems .h version overwrote ipi.c for a while.

restore it so 32 bit could use its old logical version.
also remove dupicated functions in .c

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-30 15:21:24 +01:00
Yinghai Lu
36ef4944ee x86, apic unification: remove left over files
Impact: cleanup

remove unused files

Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-30 15:21:23 +01:00
Randy Dunlap
5872fb94f8 Documentation: move DMA-mapping.txt to Doc/PCI/
Move DMA-mapping.txt to Documentation/PCI/.

DMA-mapping.txt was supposed to be moved from Documentation/ to
Documentation/PCI/.  The 00-INDEX files in those two directories
were updated, along with a few other text files, but the file
itself somehow escaped being moved, so move it and update more
text files and source files with its new location.

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
cc:	Jesse Barnes <jbarnes@virtuousgeek.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-01-29 18:19:29 -08:00
Hiroshi Shimamoto
019a136966 x86: uaccess: fix compilation error on CONFIG_M386
In case of !CONFIG_X86_WP_WORKS_OK, __put_user_size_ex() is not defined.
Add macros for !CONFIG_X86_WP_WORKS_OK case.

Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-29 21:29:02 +01:00
Ingo Molnar
550fe4f198 x86/Voyager: remove X86_FIND_SMP_CONFIG Kconfig quirk
x86/Voyager had this Kconfig quirk:

 config X86_FIND_SMP_CONFIG
	def_bool y
	depends on X86_MPPARSE || X86_VOYAGER

Which splits off the find_smp_config() callback into a build-time quirk.

Voyager should use the existing x86_quirks.mach_find_smp_config() callback
to introduce SMP-config quirks. NUMAQ-32 and VISWS already use this.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-29 14:17:04 +01:00
Ingo Molnar
f095df0a0c x86/Voyager: remove X86_BIOS_REBOOT Kconfig quirk
Voyager has this Kconfig quirk:

config X86_BIOS_REBOOT
	bool
	depends on !X86_VOYAGER
	default y

Voyager should use the existing machine_ops.emergency_restart reboot
quirk mechanism instead of a build-time quirk.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-29 14:17:03 +01:00
Ingo Molnar
c0b5842a45 x86: generalize boot_cpu_id
x86/Voyager can boot on non-zero processors. While that can probably
be fixed by properly remapping the physical CPU IDs, keep boot_cpu_id
for now for easier transition - and expand it to all of x86.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-29 14:17:01 +01:00
Ingo Molnar
3e5095d152 x86: replace CONFIG_X86_SMP with CONFIG_SMP
The x86/Voyager subarch used to have this distinction between
 'x86 SMP support' and 'Voyager SMP support':

 config X86_SMP
	bool
	depends on SMP && ((X86_32 && !X86_VOYAGER) || X86_64)

This is a pointless distinction - Voyager can (and already does) use
smp_ops to implement various SMP quirks it has - and it can be extended
more to cover all the specialities of Voyager.

So remove this complication in the Kconfig space.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-29 14:17:00 +01:00
Ingo Molnar
6bda2c8b32 x86: remove subarchitecture support
Remove the 32-bit subarchitecture support code.

All subarchitectures but Voyager have been converted. Voyager will be
done later or will be removed.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-29 14:16:52 +01:00
Ingo Molnar
1164dd0099 x86: move mach-default/*.h files to asm/
We are getting rid of subarchitecture support - move the hook files
to asm/. (These are now stale and should be replaced with more explicit
runtime mechanisms - but the transition is simpler this way.)

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-29 14:16:51 +01:00
Ingo Molnar
d53e2f2855 x86, smp: remove mach_ipi.h
Move mach_ipi.h definitions into genapic.h.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-29 14:16:49 +01:00
Suresh Siddha
83d7aeabe4 x86: remove mach_apic.h, fix
Use apic_read() instead of open-coded mmio.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-29 14:16:43 +01:00
Ingo Molnar
1dcdd3d15e x86: remove mach_apic.h
Spread mach_apic.h definitions into genapic.h. (with some knock-on effects
on smp.h and apic.h.)

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-29 14:16:42 +01:00
Jeremy Fitzhardinge
a448720ca3 x86: unify asm/io.h: IO_SPACE_LIMIT
Impact: Cleanup (trivial unification)

Move common define IO_SPACE_LIMIT from <asm/io_*.h> to <asm/io.h>.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
2009-01-28 22:34:17 -08:00
Ingo Molnar
b11b867f78 x86, summit: consolidate code
Consolidate all the Summit code into a single file:
arch/x86/kernel/summit_32.c.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:38 +01:00
Ingo Molnar
5a44632f77 x86, numaq: consolidate code
Move all the NUMAQ subarch definitions into numaq.c. With this it
ceases to depend on build-time subarch features.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:37 +01:00
Ingo Molnar
328386d7ab x86, smp: refactor ->wake_cpu
- remove macro wrappers

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:37 +01:00
Ingo Molnar
1f75ed0c13 x86: remove mach_apicdef.h
Move its definitions into apic.h.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:36 +01:00
Ingo Molnar
b2af018ff2 x86: remove mach_mpspec.h
Move its definitions into mpspec.h.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:36 +01:00
Ingo Molnar
fb5b33c9f6 x86: eliminate asm/mach-*/mach_mpparse.h
Move the definition to mpparse.h.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:35 +01:00
Ingo Molnar
0939e4fd35 x86, smp: eliminate asm/mach-default/mach_wakecpu.h
Spread mach_wakecpu.h's definitions into apic.h and genapic.h
and remove mach_wakecpu.h.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:35 +01:00
Ingo Molnar
018e047f3a x86, ES7000: consolidate the APIC code
Consolidate all the ES7000 APIC code into arch/x86/mach-generic/es7000.c.

With this ES7000 ceases to rely on any subarchitecture include files.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:34 +01:00
Ingo Molnar
25dc004903 x86, smp: refactor ->inquire_remote_apic() methods
Nothing exciting - a few subarches dont want APIC remote reads to
be performed - the others are content with the default method.

 - extend the generic code to handle NULL methods

 - clear out dummy methods and replace them with NULL

 - clean up: remove wrapper macros, etc.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:34 +01:00
Ingo Molnar
3d5f597e93 x86, smp: remove ->restore_NMI_vector()
Nothing actually restores the NMI vector - so remove this
logic altogether.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:34 +01:00
Ingo Molnar
7bd06ec63a x86, smp: refactor ->store/restore_NMI_vector() methods
Only NUMAQ does something substantial here, because it initializes
via NMIs (not via INIT as standard SMP startup) - so it needs to
store and restore the NMI vector.

 - extend the generic code to handle NULL methods

 - clear out dummy methods and replace them with NULL

 - clean up: remove wrapper macros, etc.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:33 +01:00
Ingo Molnar
333344d943 x86, smp: refactor ->smp_callin_clear_local_apic() methods
Only NUMAQ does something substantial here, because it initializes
via NMIs (not via INIT as standard SMP startup) - so it needs to
reset the APIC.

 - extend the generic code to handle NULL methods

 - clear out dummy methods and replace them with NULL

 - clean up: remove wrapper macros, etc.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:33 +01:00
Ingo Molnar
a965936643 x86, smp: refactor ->wait_for_init_deassert()
- spread out the namespace on a per APIC driver basis

 - handle a NULL ->wait_for_init_deassert() as a 'dont wait' default method

 - remove NUMAQ and Summit handlers

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:32 +01:00
Ingo Molnar
abfa584c8d x86: set ->trampoline_phys_low/high on 64-bit too
64-bit x86 has zero for ->trampoline_phys_low/high, but the smpboot
code can use these values - so it's better to set them up to their
correct values.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:32 +01:00
Ingo Molnar
6f177c01db x86, smp: clean up ->trampoline_phys_low/high handling
- spread out the namespace on a per apic driver basis

 - remove wrapper macros

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:31 +01:00
Ingo Molnar
dac5f4121d x86, apic: untangle the send_IPI_*() jungle
Our send_IPI_*() methods and definitions are a twisted mess: the same
symbol is defined to different things depending on .config details,
in a non-transparent way.

 - spread out the quirks into separately named per apic driver methods

 - prefix the standard PC methods with default_

 - get rid of wrapper macro obfuscation

 - clean up various details

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:31 +01:00
Ingo Molnar
debccb3e77 x86, apic: refactor ->cpu_mask_to_apicid*()
- spread out the namespace on a per driver basis

 - clean up the functions

 - get rid of macros

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:30 +01:00
Ingo Molnar
94af187552 x86, apic: get rid of *_APIC_ID_MASK definitions
Impact: cleanup

Remove the *_APIC_ID_MASK subarch definitions and move them straight
to the genapic driver initialization code.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:30 +01:00
Ingo Molnar
5b8127277b x86, apic: refactor ->apic_id_mask & APIC_ID_MASK
- spread out the namespace on a per driver basis

 - get rid of wrapper macros

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:29 +01:00
Ingo Molnar
ca6c8ed464 x86, apic: refactor ->get_apic_id() & GET_APIC_ID()
- spread out the namespace on a per driver basis

 - get rid of macro wrappers

 - small cleanups

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:29 +01:00
Ingo Molnar
9c7642470e x86: consolidate the ->mps_oem_check() code
- spread out the mps_oem_check() namespace on a per APIC driver basis

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:28 +01:00
Ingo Molnar
5f836405ef x86, smp: clean up mps_oem_check()
Impact: cleanup

 - allow NULL ->mps_oem_check() entries

 - clean up the code flow

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:27 +01:00
Ingo Molnar
cb8cc442dc x86, apic: refactor ->phys_pkg_id()
Refactor the ->phys_pkg_id() methods:

 - namespace separation

 - macro wrapper removal

 - open-coded calls to the methods in the generic code

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:27 +01:00
Ingo Molnar
d4c9a9f3d4 x86, apic: unify phys_pkg_id()
- unify the call signature of 64-bit to that of 32-bit

 - clean up the types all around

 - clean up namespace contamination

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:26 +01:00
Ingo Molnar
b0b20e5a3a x86, es7000: clean up es7000_enable_apic_mode()
- eliminate the needless es7000_enable_apic_mode() complication which
  was not apparent prior the namespace cleanups

- clean up the control flow in es7000_enable_apic_mode()

- other cleanups

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:26 +01:00
Ingo Molnar
4904033302 x86: refactor ->enable_apic_mode() subarch methods
Only ES7000 has a real ->enable_apic_mode() method, the other
subarchitectures define it but keep it empty.

So mark the vector as NULL, extend the generic code to handle
NULL -setup_portio_remap() entries and remove all the empty
handlers.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:26 +01:00
Ingo Molnar
a27a621001 x86: refactor ->check_phys_apicid_present() subarch methods
- spread out the namespace to per driver methods

 - extend it to 64-bit as well so that we can use
   apic->check_phys_apicid_present() unconditionally

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:25 +01:00
Ingo Molnar
d83093b504 x86: refactor ->setup_portio_remap() subarch methods
Only NUMAQ has a real ->setup_portio_remap() method, the other
subarchitectures define it but keep it empty.

So mark the vector as NULL, extend the generic code to handle
NULL -setup_portio_remap() entries and remove all the empty
handlers.

Also move the NUMAQ method from the header file into the
 apic driver .c file.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:25 +01:00
Ingo Molnar
8058714a41 x86, apic: clean up ->apicid_to_cpu_present()
- separate the namespace

 - remove macros

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:24 +01:00
Ingo Molnar
a21769a446 x86, apic: clean up ->cpu_present_to_apicid()
- separate the namespace

 - remove macros

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:24 +01:00
Ingo Molnar
5257c5111c x86, apic: clean up ->cpu_to_logical_apicid()
- separate the namespace

 - remove macros

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:23 +01:00
Ingo Molnar
3f57a318c3 x86, apic: clean up ->apicid_to_node()
- separate the namespace

 - remove macros

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:23 +01:00
Ingo Molnar
33a201fac6 x86, apic: streamline the ->multi_timer_check() quirk
only NUMAQ uses this quirk: to prevent the timer IRQ from being added
on secondary nodes.

All other genapic templates can have a NULL ->multi_timer_check()
callback.

Also, extend the generic code to treat a NULL pointer accordingly.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:22 +01:00
Ingo Molnar
72ce016583 x86, apic: clean up ->setup_apic_routing()
- separate the namespace

 - remove macros

 - remove namespace clash on 64-bit

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:22 +01:00
Ingo Molnar
d190cb87c4 x86, apic: clean up ->ioapic_phys_id_map()
- separate the namespace

 - remove macros

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:21 +01:00
Ingo Molnar
a5c4329622 x86, apic: clean up ->init_apic_ldr()
- separate the namespace

 - remove macros

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:21 +01:00
Ingo Molnar
e2d40b1878 x86, apic: clean up ->vector_allocation_domain()
- separate the namespace

 - remove macros

 - move the default vector-allocation-domain to mach-generic

 - fix whitespace damage

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:20 +01:00
Ingo Molnar
2e867b17cc x86, apic: remove no_balance_irq and no_ioapic_check flags
These flags are completely unused. (the in-kernel IRQ balancer has
been removed from the upstream kernel.)

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:20 +01:00
Ingo Molnar
d1d7cae8fd x86, apic: clean up check_apicid*() callbacks
Clean up these methods - to make it clearer which function is
used in which case.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:19 +01:00
Ingo Molnar
bdb1a9b62f x86, apic: rename genapic::apic_destination_logical to genapic::dest_logical
This field name was unreasonably long - shorten it.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:19 +01:00
Ingo Molnar
0b06e734bf x86: clean up the APIC_DEST_LOGICAL logic
Impact: cleanup

The bigsmp and es7000 subarchitectures un-defined APIC_DEST_LOGICAL in
a rather nasty way by re-defining it to zero. That is infinitely
fragile and makes it very hard to see what to code really does in
a given context. The very same constant has different meanings and
values - depending on which subarch is enabled.

Untangle this mess by never undefining the constant, but instead
propagating the right values into the genapic driver templates.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:18 +01:00
Ingo Molnar
08125d3eda x86: rename ->ESR_DISABLE to ->disable_esr
the ->ESR_DISABLE shouting variant was used to enable the esr_disable
macro wrappers. Those ugly macros are removed now so we can rename
->ESR_DISABLE to ->disable_esr

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:18 +01:00
Ingo Molnar
f6f52baf26 x86: clean up esr_disable() methods
Impact: cleanup

Most subarchitectures want to disable the APIC ESR (Error Status Register),
because they generally have hardware hacks that wrap standard CPUs into
a bigger system and hence the APIC bus is quite non-standard and weirdnesses
(lockups) have been seen with ESR reporting.

Remove the esr_disable macros and put the desired flag into each
subarchitecture's genapic template directly.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:17 +01:00
Ingo Molnar
fe402e1f2b x86, apic: clean up / remove TARGET_CPUS
Impact: cleanup

use apic->target_cpus() directly instead of the TARGET_CPUS wrapper.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:17 +01:00
Ingo Molnar
0a9cc20b9c x86, apic: clean up target_cpus methods
Impact: cleanup

Clean up all the target_cpus() namespace overlap that exists
between bigsmp, es7000, mach-default, numaq and summit - by
separating the different functions into different names.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:17 +01:00
Ingo Molnar
7fe732862d x86, apic: remove IRQ_DEST_MODE
Remove leftover definition.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:16 +01:00
Ingo Molnar
38bd77a6c3 x86, apic: remove ES7000_IRQ_DELIVERY_MODE and ES7000_IRQ_DEST_MODE
Impact: cleanup

They were only used in a single place and obscured the apic_es7000
driver template.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:16 +01:00
Ingo Molnar
d8a3539e64 x86, apic: remove BIGSMP_IRQ_DELIVERY_MODE and BIGSMP_IRQ_DEST_MODE
Impact: cleanup

They were only used in a single place and obscured the apic_bigsmp
driver template.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:15 +01:00
Ingo Molnar
1b1bcb3ff4 x86, apic: remove NUMAQ_IRQ_DELIVERY_MODE and NUMAQ_IRQ_DEST_MODE
Impact: cleanup

They were only used in a single place and obscured the apic_numaq template.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:15 +01:00
Ingo Molnar
82daea6b08 x86, apic: remove SUMMIT_IRQ_DELIVERY_MODE and SUMMIT_IRQ_DEST_MODE
Impact: cleanup

They were only used in a single place and obscured the apic_summit template.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:14 +01:00
Ingo Molnar
dcafa4a8c9 x86, apic: remove DEFAULT_IRQ_DELIVERY_MODE and DEFAULT_IRQ_DEST_MODE
Impact: cleanup

They were only used in a single place and obscured the apic_default template.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:14 +01:00
Ingo Molnar
9b5bc8dc12 x86, apic: remove IRQ_DEST_MODE / IRQ_DELIVERY_MODE
Remove the wrapper macros IRQ_DEST_MODE and IRQ_DELIVERY_MODE.

The typical 32-bit and the 64-bit build all dereference via the genapic,
so it's pointless to hide that indirection via these ugly macros.

Furthermore, it also obscures subarchitecture details.

So replace it with apic->irq_dest_mode / etc. accesses.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:13 +01:00
Ingo Molnar
f8987a1093 x86, genapic: rename int_delivery_mode, et. al.
int_delivery_mode is supposed to mean 'interrupt delivery mode', but
it's quite a misnomer as 'int' we usually think of as an integer type ...

The standard naming for such attributes is 'irq' - so rename the following
fields and macros:

 int_delivery_mode => irq_delivery_mode
 INT_DELIVERY_MODE => IRQ_DELIVERY_MODE
 int_dest_mode     => irq_dest_mode
 INT_DEST_MODE     => IRQ_DEST_MODE

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:13 +01:00
Ingo Molnar
7ed248daa5 x86: clean up apic->apic_id_registered() methods
Impact: cleanup

x86 subarchitectures each defined a "apic_id_registered()" method,
which could be an inline function depending on which subarch we build
for, and which was also the name of a genapic field.

Untangle this namespace spaghetti by giving each of the instances
a separate name.

Also remove wrapper macro obfuscation.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:12 +01:00
Ingo Molnar
306db03b0d x86: clean up apic->acpi_madt_oem_check methods
Impact: refactor code

x86 subarchitectures each defined a "acpi_madt_oem_check()" method,
which could be an inline function, or an extern, or a static function,
and which was also the name of a genapic field.

Untangle this namespace spaghetti by setting ->acpi_madt_oem_check()
to NULL on those subarchitectures that have no detection quirks,
and rename the other ones (summit, es7000) that do.

Also change default_acpi_madt_oem_check() to handle NULL entries,
and clean its control flow up as well.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:12 +01:00
Ingo Molnar
9a6801da55 x86: remove APIC_INIT / APICFUNC / IPIFUNC
The APIC_INIT() / APICFUNC / IPIFUNC macros were ugly and obfuscated
the true identity of various APIC driver methods.

Now that they are not used anymore, remove them.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:11 +01:00
Ingo Molnar
c8d46cf06d x86: rename 'genapic' to 'apic'
Rename genapic-> to apic-> references because in a future chagne we'll
open-code all the indirect calls (instead of obscuring them via macros),
so we want this reference to be as short as possible.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:06 +01:00
Ingo Molnar
6781d948cc x86, genapic: provide IPI callbacks unconditionally
64-bit x86 uses the IPI callbacks even on UP - so provide them
generally.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:06 +01:00
Ingo Molnar
505deeb1a2 x86, genapic: cleanups
Unify genapic.h some more.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:05 +01:00
Ingo Molnar
ced733ec0b x86, genapic: finish unification
Unify remaining bits of genapic_32/64.h.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:05 +01:00
Ingo Molnar
ef7471b13f x86, genapic: unify struct genapic
Move over the (now identical) struct genapic definitions from
genapic_32/64.h to genapic.h.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:04 +01:00
Ingo Molnar
943d0f74d4 x86, genapic: refactor genapic_32.h
Impact: pre unification cleanup

Make genapic_32.h similar to genapic_64.h: reorder fields, unify types
and bring in new entries.

No existing functionality is affected.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:04 +01:00
Ingo Molnar
6e7a59944a x86, genapic: refactor genapic_64.h
Impact: pre unification cleanup

Make genapic_64.h similar to genapic_32.h: reorder fields, unify types
and bring in new entries.

No existing functionality is affected.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-28 23:20:03 +01:00
Ingo Molnar
74b6eb6b93 Merge branches 'x86/asm', 'x86/cleanups', 'x86/cpudetect', 'x86/debug', 'x86/doc', 'x86/header-fixes', 'x86/mm', 'x86/paravirt', 'x86/pat', 'x86/setup-v2', 'x86/subarch', 'x86/uaccess' and 'x86/urgent' into x86/core 2009-01-28 23:13:53 +01:00
Ingo Molnar
4369f1fb7c Merge branch 'tj-percpu' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/misc into core/percpu
Conflicts:
	arch/x86/kernel/setup_percpu.c

Semantic conflict:

	arch/x86/kernel/cpu/common.c

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-27 12:03:24 +01:00
Ingo Molnar
3ddeb51d9c Merge branch 'linus' into core/percpu
Conflicts:
	arch/x86/kernel/setup_percpu.c
2009-01-27 12:01:51 +01:00
Brian Gerst
2697fbd5fa x86: load new GDT after setting up boot cpu per-cpu area
Impact: sync 32 and 64-bit code

Merge load_gs_base() into switch_to_new_gdt().  Load the GDT and
per-cpu state for the boot cpu when its new area is set up.

Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-27 12:56:48 +09:00
Brian Gerst
1825b8edc2 x86: remove extra barriers from load_gs_base()
Impact: optimization

mb() generates an mfence instruction, which is not needed here.  Only
a compiler barrier is needed, and that is handled by the memory clobber
in the wrmsrl function.

Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-27 12:56:48 +09:00
Brian Gerst
b2d2f4312b x86: initialize per-cpu GDT segment in per-cpu setup
Impact: cleanup

Rename init_gdt() to setup_percpu_segment(), and move it to
setup_percpu.c.

Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-27 12:56:48 +09:00
Brian Gerst
2f2f52bad7 x86: move setup_cpu_local_masks()
Impact: Code movement, no functional change.

Move setup_cpu_local_masks() to kernel/cpu/common.c, where the
masks are defined.

Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-27 12:56:47 +09:00
Brian Gerst
6470aff619 x86: move 64-bit NUMA code
Impact: Code movement, no functional change.

Move the 64-bit NUMA code from setup_percpu.c to numa_64.c

Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-27 12:56:47 +09:00
Linus Torvalds
810ee58de2 Merge branch 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip: (29 commits)
  xen: unitialised return value in xenbus_write_transaction
  x86: fix section mismatch warning
  x86: unmask CPUID levels on Intel CPUs, fix
  x86: work around PAGE_KERNEL_WC not getting WC in iomap_atomic_prot_pfn.
  x86: use standard PIT frequency
  xen: handle highmem pages correctly when shrinking a domain
  x86, mm: fix pte_free()
  xen: actually release memory when shrinking domain
  x86: unmask CPUID levels on Intel CPUs
  x86: add MSR_IA32_MISC_ENABLE bits to <asm/msr-index.h>
  x86: fix PTE corruption issue while mapping RAM using /dev/mem
  x86: mtrr fix debug boot parameter
  x86: fix page attribute corruption with cpa()
  Revert "x86: signal: change type of paramter for sys_rt_sigreturn()"
  x86: use early clobbers in usercopy*.c
  x86: remove kernel_physical_mapping_init() from init section
  fix: crash: IP: __bitmap_intersects+0x48/0x73
  cpufreq: use work_on_cpu in acpi-cpufreq.c for drv_read and drv_write
  work_on_cpu: Use our own workqueue.
  work_on_cpu: don't try to get_online_cpus() in work_on_cpu.
  ...
2009-01-26 09:47:28 -08:00
Frederic Weisbecker
2d4d57db69 x86: micro-optimize __raw_read_trylock()
The current version of __raw_read_trylock starts with decrementing the lock
and read its new value as a separate operation after that.

That makes 3 dereferences (read, write (after sub), read) whereas
a single atomic_dec_return does only two pointers dereferences (read, write).

Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-26 14:06:36 +01:00
Ingo Molnar
e1b4d11436 x86: use standard PIT frequency
the RDC and ELAN platforms use slighly different PIT clocks, resulting in
a timex.h hack that changes PIT_TICK_RATE during build time. But if a
tester enables any of these platform support .config options, the PIT
will be miscalibrated on standard PC platforms.

So use one frequency - in a subsequent patch we'll add a quirk to allow
x86 platforms to define different PIT frequencies.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-25 16:57:47 +01:00
H. Peter Anvin
75a048119e x86: handle PAT more like other CPU features
Impact: Cleanup

When PAT was originally introduced, it was handled specially for a few
reasons:

- PAT bugs are hard to track down, so we wanted to maintain a
  whitelist of CPUs.
- The i386 and x86-64 CPUID code was not yet unified.

Both of these are now obsolete, so handle PAT like any other features,
including ordinary feature blacklisting due to known bugs.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2009-01-23 18:07:45 -08:00
H. Peter Anvin
b1882e68d1 x86: clean up stray space in <asm/processor.h>
Impact: Whitespace cleanup only

Clean up a stray space character in arch/x86/include/asm/processor.h.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2009-01-23 17:54:17 -08:00
Hiroshi Shimamoto
fe40c0af3c x86: uaccess: introduce try and catch framework
Impact: introduce new uaccess exception handling framework

Introduce {get|put}_user_try and {get|put}_user_catch as new uaccess exception
handling framework.
{get|put}_user_try begins exception block and {get|put}_user_catch(err) ends
the block and gets err if an exception occured in {get|put}_user_ex() in the
block. The exception is stored thread_info->uaccess_err.

The example usage of this framework is below;
int func()
{
	int err = 0;

	get_user_try {
		get_user_ex(...);
		get_user_ex(...);
		:
	} get_user_catch(err);

	return err;
}

Note: get_user_ex() is not clear the value when an exception occurs, it's
different from the behavior of __get_user(), but I think it doesn't matter.

Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2009-01-23 17:17:36 -08:00
Peter Zijlstra
42ef73fe13 x86, mm: fix pte_free()
On -rt we were seeing spurious bad page states like:

Bad page state in process 'firefox'
page:c1bc2380 flags:0x40000000 mapping:c1bc2390 mapcount:0 count:0
Trying to fix it up, but a reboot is needed
Backtrace:
Pid: 503, comm: firefox Not tainted 2.6.26.8-rt13 #3
[<c043d0f3>] ? printk+0x14/0x19
[<c0272d4e>] bad_page+0x4e/0x79
[<c0273831>] free_hot_cold_page+0x5b/0x1d3
[<c02739f6>] free_hot_page+0xf/0x11
[<c0273a18>] __free_pages+0x20/0x2b
[<c027d170>] __pte_alloc+0x87/0x91
[<c027d25e>] handle_mm_fault+0xe4/0x733
[<c043f680>] ? rt_mutex_down_read_trylock+0x57/0x63
[<c043f680>] ? rt_mutex_down_read_trylock+0x57/0x63
[<c0218875>] do_page_fault+0x36f/0x88a

This is the case where a concurrent fault already installed the PTE and
we get to free the newly allocated one.

This is due to pgtable_page_ctor() doing the spin_lock_init(&page->ptl)
which is overlaid with the {private, mapping} struct.

union {
    struct {
        unsigned long private;
        struct address_space *mapping;
    };
    spinlock_t ptl;
    struct kmem_cache *slab;
    struct page *first_page;
};

Normally the spinlock is small enough to not stomp on page->mapping, but
PREEMPT_RT=y has huge 'spin'locks.

But lockdep kernels should also be able to trigger this splat, as the
lock tracking code grows the spinlock to cover page->mapping.

The obvious fix is calling pgtable_page_dtor() like the regular pte free
path __pte_free_tlb() does.

It seems all architectures except x86 and nm10300 already do this, and
nm10300 doesn't seem to use pgtable_page_ctor(), which suggests it
doesn't do SMP or simply doesnt do MMU at all or something.

Signed-off-by: Peter Zijlstra <a.p.zijlsta@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: <stable@kernel.org>
2009-01-23 18:42:06 +01:00
Brian Gerst
2de3a5f795 x86: make irq_cpustat_t fields conditional
Impact: shrink size of irq_cpustat_t when possible

Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-23 11:03:32 +09:00
Brian Gerst
22da7b3df3 x86: merge hardirq_{32,64}.h into hardirq.h
Impact: cleanup

Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-23 11:03:31 +09:00
Brian Gerst
658a9a2c34 x86: sync hardirq_{32,64}.h
Impact: better code generation and removal of unused field for 32bit

In general, use the 64-bit version.

Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-23 11:03:31 +09:00
Brian Gerst
3819cd489e x86: remove include of apic.h from hardirq_64.h
Impact: cleanup

APIC definitions aren't needed here.  Remove the include and fix
up the fallout.

tj: added include to mce_intel_64.c.

Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-23 11:03:29 +09:00
Brian Gerst
03d2989df9 x86: remove idle_timestamp from 32bit irq_cpustat_t
Impact: bogus irq_cpustat field removed

idle_timestamp is left over from the removed irqbalance code.

Signed-off-by: Brian Gerst <brgerst@gmail.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2009-01-23 11:03:28 +09:00
Jeremy Fitzhardinge
6522869c34 x86: add pte_set_flags/clear_flags for pte flag manipulation
It's not necessary to deconstruct and reconstruct a pte every time its
flags are being updated.  Introduce pte_set_flags and pte_clear_flags
to set and clear flags in a pte.  This allows the flag manipulation
code to be inlined, and avoids calls via paravirt-ops.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-22 23:35:21 +01:00
Jeremy Fitzhardinge
ab897d2013 x86/pvops: remove pte_flags pvop
pte_flags() was introduced as a new pvop in order to extract just the
flags portion of a pte, which is a potentially cheaper operation than
extracting the page number as well.  It turns out this operation is
not needed, because simply using a mask to extract the flags from a
pte is sufficient for all current users.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-22 23:35:20 +01:00
venkatesh.pallipadi@intel.com
d639bab8da x86 PAT: ioremap_wc should take resource_size_t parameter
Impact: fix/extend ioremap_wc() beyond 4GB aperture on 32-bit

ioremap_wc() was taking in unsigned long parameter, where as it should take
64-bit resource_size_t parameter like other ioremap variants.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-22 11:53:42 +01:00
H. Peter Anvin
bdf21a49ba x86: add MSR_IA32_MISC_ENABLE bits to <asm/msr-index.h>
Impact: None (new bit definitions currently unused)

Add bit definitions for the MSR_IA32_MISC_ENABLE MSRs to
<asm/msr-index.h>.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2009-01-21 15:13:53 -08:00
Nick Piggin
03b486322e x86: make UV support configurable
Make X86 SGI Ultraviolet support configurable. Saves about 13K of text size
on my modest config.

   text    data     bss     dec     hex filename
6770537 1158680  694356 8623573  8395d5 vmlinux
6757492 1157664  694228 8609384  835e68 vmlinux.nouv

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-21 13:00:42 +01:00
Hiroshi Shimamoto
cc86c9e0dc x86: uaccess: rename __put_user_u64() to __put_user_asm_u64()
Impact: cleanup

rename __put_user_u64() to __put_user_asm_u64() like __get_user_asm_u64().

Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-21 12:34:44 +01:00
Hiroshi Shimamoto
4d5d783896 x86: uaccess: fix style problems
Impact: cleanup

Fix coding style problems in arch/x86/include/asm/uaccess.h.

Signed-off-by: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-21 12:34:44 +01:00
Ingo Molnar
198030782c Merge branch 'x86/mm' into core/percpu
Conflicts:
	arch/x86/mm/fault.c
2009-01-21 10:39:51 +01:00
Ingo Molnar
3eb3963fd1 Merge branch 'cpus4096' into core/percpu
Conflicts:
	arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c
	arch/x86/kernel/tlb_32.c

Merge it here because both the cpumask changes and the ongoing percpu
work is touching the TLB code. The percpu changes take precedence, as
they eliminate tlb_32.c altogether.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
2009-01-21 10:14:17 +01:00