Track the memtype for RAM pages in page struct instead of using the
memtype list. This avoids the explosion in the number of entries in
memtype list (of the order of 20,000 with AGP) and makes the PAT
tracking simpler.

We are using PG_arch_1 bit in page->flags.

We still use the memtype list for non RAM pages.
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Interrupt context no longer splits large page in cpa(). So we can do away
with cpa memory pool code.
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: arjan@linux.intel.com
Cc: venkatesh.pallipadi@intel.com
Cc: jeremy@goop.org
Signed-off-by: NIngo Molnar <mingo@elte.hu>

remove USER from the PTE/PDE attributes for the very early identity
mapping. We overwrite these mappings with KERNEL attribute later
in the boot. Just being paranoid here as there is no need for USER bit
to be set.

If this breaks something(don't know the history), then we can simply drop
this change.
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: arjan@linux.intel.com
Cc: venkatesh.pallipadi@intel.com
Cc: jeremy@goop.org
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: arjan@linux.intel.com
Cc: venkatesh.pallipadi@intel.com
Cc: jeremy@goop.org
Signed-off-by: NIngo Molnar <mingo@elte.hu>

include/linux/iommu-helper.h has no header guards, which breaks
sparc64 build. Add them.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Cc: Joerg Roedel <joerg.roedel@amd.com>
Cc: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

This one fixes http://bugzilla.kernel.org/show_bug.cgi?id=11602.

A more generic fix for drives which cannot autoclose tray will follow.
Signed-off-by: NBorislav Petkov <petkovbb@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
[bart: add an extra parentheses for consistency with the rest of kernel code]
Signed-off-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

include/linux/stacktrace.h:13: warning:
 'struct task_struct' declared inside parameter list

(This might be a hard error on sparc64, which uses this header and has
-Werror)
Reported-by: N"Randy.Dunlap" <rdunlap@xenotime.net>
Acked-by: NIngo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Rework of SMTC support to make it work with the new clock event system,
allowing "tickless" operation, and to make it compatible with the use of
the "wait_irqoff" idle loop.  The new clocking scheme means that the
previously optional IPI instant replay mechanism is now required, and has
been made more robust.
Signed-off-by: NKevin D. Kissell <kevink@paralogos.com>
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Signed-off-by: NKevin D. Kissell <kevink@paralogos.com>
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Though from a hardware perspective it would be sensible to use only a
32-bit unsigned int type Linux defines interrupt flags to be stored in
an unsigned long and nothing else.
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

The previous patch db203d53 ("mm:
tiny-shmem fix lock ordering: mmap_sem vs i_mutex") to fix the lock
ordering in tiny-shmem breaks shared anonymous and IPC memory on NOMMU
architectures because it was using the expanding truncate to signal ramfs
to allocate a physically contiguous RAM backing the inode (otherwise it is
unusable for "memory mapping" it to userspace).

However do_truncate is what caused the lock ordering error, due to it
taking i_mutex.  In this case, we can actually just call ramfs directly to
allocate memory for the mapping, rather than go via truncate.
Acked-by: NDavid Howells <dhowells@redhat.com>
Acked-by: NHugh Dickins <hugh@veritas.com>
Signed-off-by: NNick Piggin <npiggin@suse.de>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix inotify lock order reversal with mmap_sem due to holding locks over
copy_to_user.
Signed-off-by: NNick Piggin <npiggin@suse.de>
Reported-by: N"Daniel J Blueman" <daniel.blueman@gmail.com>
Tested-by: N"Daniel J Blueman" <daniel.blueman@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NRobert Richter <robert.richter@amd.com>
Cc: oprofile-list <oprofile-list@lists.sourceforge.net>
Cc: Barry Kasindorf <barry.kasindorf@amd.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Since call to function sctp_sf_abort_violation() need paramter 'arg' with
'struct sctp_chunk' type, it will read the chunk type and chunk length from
the chunk_hdr member of chunk. But call to sctp_sf_violation_paramlen()
always with 'struct sctp_paramhdr' type's parameter, it will be passed to
sctp_sf_abort_violation(). This may cause kernel panic.

   sctp_sf_violation_paramlen()
     |-- sctp_sf_abort_violation()
        |-- sctp_make_abort_violation()

This patch fixed this problem. This patch also fix two place which called
sctp_sf_violation_paramlen() with wrong paramter type.
Signed-off-by: NWei Yongjun <yjwei@cn.fujitsu.com>
Signed-off-by: NVlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The number of BIOSes that have an option to enable the IOMMU, or fix
anything about its configuration, is vanishingly small.  There's no good
reason to punish quiet boot for this.
Signed-off-by: NAdam Jackson <ajax@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Impact: per CPU hrtimers can be migrated from a dead CPU

The hrtimer code has no knowledge about per CPU timers, but we need to
prevent the migration of such timers and warn when such a timer is
active at migration time.

Explicitely mark the timers as per CPU and use a more understandable
mode descriptor for the interrupts safe unlocked callback mode, which
is used by hrtimer_sleeper and the scheduler code.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

Impact: during migration active hrtimers can be seen as inactive

The migration code removes the hrtimers from the queues of the dead
CPU and sets the state temporary to INACTIVE. The enqueue code sets it
to ACTIVE/PENDING again.

Prevent that the wrong state can be seen by using a separate migration
state bit.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

reduce size by 8 bytes from 1160 to 1152 allowing it to fit in 1 fewer
cachelines.
Signed-off-by: NRichard Kennedy <richard@rsk.demon.co.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

On user request (through sysfs), the IDLE IMMEDIATE command with UNLOAD
FEATURE as specified in ATA-7 is issued to the device and processing of
the request queue is stopped thereafter until the specified timeout
expires or user space asks to resume normal operation. This is supposed
to prevent the heads of a hard drive from accidentally crashing onto the
platter when a heavy shock is anticipated (like a falling laptop
expected to hit the floor). In fact, the whole port stops processing
commands until the timeout has expired in order to avoid any resets due
to failed commands on another device.
Signed-off-by: NElias Oltmanns <eo@nebensachen.de>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Add a function to check an ATA device's id for head unload support as
specified in ATA-7.
Signed-off-by: NElias Oltmanns <eo@nebensachen.de>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Explanation taken from the comment of ata_slave_link_init().

 In libata, a port contains links and a link contains devices.  There
 is single host link but if a PMP is attached to it, there can be
 multiple fan-out links.  On SATA, there's usually a single device
 connected to a link but PATA and SATA controllers emulating TF based
 interface can have two - master and slave.

 However, there are a few controllers which don't fit into this
 abstraction too well - SATA controllers which emulate TF interface
 with both master and slave devices but also have separate SCR
 register sets for each device.  These controllers need separate links
 for physical link handling (e.g. onlineness, link speed) but should
 be treated like a traditional M/S controller for everything else
 (e.g. command issue, softreset).

 slave_link is libata's way of handling this class of controllers
 without impacting core layer too much.  For anything other than
 physical link handling, the default host link is used for both master
 and slave.  For physical link handling, separate @ap->slave_link is
 used.  All dirty details are implemented inside libata core layer.
 From LLD's POV, the only difference is that prereset, hardreset and
 postreset are called once more for the slave link, so the reset
 sequence looks like the following.

 prereset(M) -> prereset(S) -> hardreset(M) -> hardreset(S) ->
 softreset(M) -> postreset(M) -> postreset(S)

 Note that softreset is called only for the master.  Softreset resets
 both M/S by definition, so SRST on master should handle both (the
 standard method will work just fine).

As slave_link excludes PMP support and only code paths which deal with
the attributes of physical link are affected, all the changes are
localized to libata.h, libata-core.c and libata-eh.c.

 * ata_is_host_link() updated so that slave_link is considered as host
   link too.

 * iterator extended to iterate over the slave_link when using the
   underbarred version.

 * force param handling updated such that devno 16 is mapped to the
   slave link/device.

 * ata_link_on/offline() updated to return the combined result from
   master and slave link.  ata_phys_link_on/offline() are the direct
   versions.

 * EH autopsy and report are performed separately for master slave
   links.  Reset is udpated to implement the above described reset
   sequence.

Except for reset update, most changes are minor, many of them just
modifying dev->link to ata_dev_phys_link(dev) or using phys online
test instead.

After this update, LLDs can take full advantage of per-dev SCR
registers by simply turning on slave link.
Signed-off-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

* Add ATA_EH_ALL_ACTIONS.

* Make sata_link_{on|off}_line() return bool instead of int.
Signed-off-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Implement __ata_port_next_link() and reimplement
__ata_port_for_each_link() and ata_port_for_each_link() using it.
This removes relatively large inlined code and makes iteration easier
to extend.
Signed-off-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Logically, SCR access ops should take @link; however, there was no
compelling reason to convert all SCR access ops when adding @link
abstraction as there's one-to-one mapping between a port and a non-PMP
link.  However, that assumption won't hold anymore with the scheduled
addition of slave link.

Make SCR access ops per-link.
Signed-off-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJeff Garzik <jgarzik@redhat.com>

Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

On x86_64 the gdb serial register structure defines the PS (also known
as eflags), CS and SS registers as 4 bytes entities.

This patch splits the x86_64 regnames enum into a 32 and 64 version to
account for the 32 bit entities in the gdb serial packets.

Also the program counter is properly filled in for the sleeping
threads.
Signed-off-by: NJason Wessel <jason.wessel@windriver.com>

The BX and DX registers in the gdb serial register packet need to be
flipped for gdb to receive the correct data.
Signed-off-by: NJason Wessel <jason.wessel@windriver.com>

This patch exports nommu_alloc_coherent (renamed
dma_generic_alloc_coherent). GART needs this function.
Signed-off-by: NFUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Move asm-arm/cnt32_to_63.h to include/linux/ so that MN10300 can make
use of it too.
Signed-off-by: NDavid Howells <dhowells@redhat.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

9p trans modules aren't refcounted nor were they unregistered
properly.  Fix it.

* Add 9p_trans_module->owner and reference the module on each trans
  instance creation and put it on destruction.

* Protect v9fs_trans_list with a spinlock.  This isn't strictly
  necessary as the list is manipulated only during module loading /
  unloading but it's a good idea to make the API safe.

* Unregister trans modules when the corresponding module is being
  unloaded.

* While at it, kill unnecessary EXPORT_SYMBOL on p9_trans_fd_init().
Signed-off-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NEric Van Hensbergen <ericvh@gmail.com>

When we use > 4KB's page size the original definition is not consistent
with PGDIR_SIZE. For exeample, if we use 16KB page size the PGDIR_SHIFT is
(14-2) + 14 = 26, PGDIR_SIZE is 2^26，so the PTRS_PER_PGD should be:

	2^32/2^26 = 2^6

but the original definition of PTRS_PER_PGD is 4096 (PGDIR_ORDER = 0).

So, this definition needs to be consistent with the PGDIR_SIZE.

And the new definition is consistent with the PGD init in pagetable_init().
Signed-off-by: NDajie Tan <jiankemeng@gmail.com>
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

linux/time.h conflicts with time.h from glibc

It breaks building smbmount from samba.  It's regression introduced by
commit 76308da1 (" smb.h: uses struct timespec but didn't include
linux/time.h").
Signed-off-by: NKirill A. Shutemov <kirill@shutemov.name>
Cc: <stable@kernel.org>             [2.6.26.x]
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Impact: System hang when AMD C1E machines switch into C2/C3

AMD C1E enabled systems do not work with normal ACPI C-states 
even if the BIOS is advertising them. Limit the C-states to 
C1 for the ACPI processor idle code.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

Impact: hang which happens across CPU offline/online on AMD C1E systems.

When a CPU goes offline then the corresponding bit in the broadcast
mask is cleared. For AMD C1E enabled CPUs we do not reenable the
broadcast when the CPU comes online again as we do not clear the
corresponding bit in the c1e_mask, which keeps track which CPUs
have been switched to broadcast already. So on those !$@#& machines
we never switch back to broadcasting after a CPU offline/online cycle.

Clear the bit when the CPU plays dead.
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>

This patch against tip/x86/iommu virtually reverts
2842e5bf. But just reverting the
commit breaks AMD IOMMU so this patch also includes some fixes.

The above commit adds new two options to x86 IOMMU generic kernel boot
options, fullflush and nofullflush. But such change that affects all
the IOMMUs needs more discussion (all IOMMU parties need the chance to
discuss it):

http://lkml.org/lkml/2008/9/19/106Signed-off-by: NFUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

"export iommu_area_reserve helper funciton" patch converted all the
users of set_bit_string, GART, Calgary and AMD IOMMU drivers, to use
iommu_area_reserve helper function. Now we can remove unused
set_bit_string function.
Signed-off-by: NFUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

There's a small window when NMI watchdog is being set up that if any NMIs
are triggered, the NMI code will make make use of not initalized wd_ops
elements:
	void setup_apic_nmi_watchdog(void *unused)
	{
		if (__get_cpu_var(wd_enabled))
			return;

		/* cheap hack to support suspend/resume */
		/* if cpu0 is not active neither should the other cpus */
		if (smp_processor_id() != 0 && atomic_read(&nmi_active) <= 0)
			return;

		switch (nmi_watchdog) {
		case NMI_LOCAL_APIC:
			/* enable it before to avoid race with handler */
-->			__get_cpu_var(wd_enabled) = 1;
-->			if (lapic_watchdog_init(nmi_hz) < 0) {
(...)
	asmlinkage notrace __kprobes void default_do_nmi(struct pt_regs *regs)
	{
	(...)
			if (nmi_watchdog_tick(regs, reason))
				return;
(...)
	notrace __kprobes int
	nmi_watchdog_tick(struct pt_regs *regs, unsigned reason)
	{
	(...)
		if (!__get_cpu_var(wd_enabled))
			return rc;
		switch (nmi_watchdog) {
		case NMI_LOCAL_APIC:
			rc |= lapic_wd_event(nmi_hz);
(...)
int lapic_wd_event(unsigned nmi_hz)
{
	struct nmi_watchdog_ctlblk *wd = &__get_cpu_var(nmi_watchdog_ctlblk);
	u64 ctr;

-->	rdmsrl(wd->perfctr_msr, ctr);

and wd->*_msr will be initialized on each processor type specific setup, after
enabling NMIs for PMIs. Since the counter was just set, the chances of an
performance counter generated NMI is minimal, but any other unknown NMI would
trigger the problem. This patch fixes the problem by setting everything up
before enabling performance counter generated NMIs and will set wd_enabled
using a callback function.
Signed-off-by: NAristeu Rozanski <aris@redhat.com>
Acked-by: NDon Zickus <dzickus@redhat.com>
Acked-by: NPrarit Bhargava <prarit@redhat.com>
Acked-by: NVivek Goyal <vgoyal@redhat.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

x86 has set_bit_string() that does the exact same thing that
set_bit_area() in lib/iommu-helper.c does.

This patch exports set_bit_area() in lib/iommu-helper.c as
iommu_area_reserve(), converts GART, Calgary, and AMD IOMMU to use it.
Signed-off-by: NFUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Remove some magic numbers and split the pte_root using standard
functions.
Signed-off-by: NJoerg Roedel <joerg.roedel@amd.com>
Signed-off-by: NIngo Molnar <mingo@elte.hu>