Remove the cpuset hooks that defined sched domains depending on the setting
of the 'cpu_exclusive' flag.

The cpu_exclusive flag can only be set on a child if it is set on the
parent.

This made that flag painfully unsuitable for use as a flag defining a
partitioning of a system.

It was entirely unobvious to a cpuset user what partitioning of sched
domains they would be causing when they set that one cpu_exclusive bit on
one cpuset, because it depended on what CPUs were in the remainder of that
cpusets siblings and child cpusets, after subtracting out other
cpu_exclusive cpusets.

Furthermore, there was no way on production systems to query the
result.

Using the cpu_exclusive flag for this was simply wrong from the get go.

Fortunately, it was sufficiently borked that so far as I know, almost no
successful use has been made of this.  One real time group did use it to
affectively isolate CPUs from any load balancing efforts.  They are willing
to adapt to alternative mechanisms for this, such as someway to manipulate
the list of isolated CPUs on a running system.  They can do without this
present cpu_exclusive based mechanism while we develop an alternative.

There is a real risk, to the best of my understanding, of users
accidentally setting up a partitioned scheduler domains, inhibiting desired
load balancing across all their CPUs, due to the nonobvious (from the
cpuset perspective) side affects of the cpu_exclusive flag.

Furthermore, since there was no way on a running system to see what one was
doing with sched domains, this change will be invisible to any using code.
Unless they have real insight to the scheduler load balancing choices, they
will be unable to detect that this change has been made in the kernel's
behaviour.

Initial discussion on lkml of this patch has generated much comment.  My
(probably controversial) take on that discussion is that it has reached a
rough concensus that the current cpuset cpu_exclusive mechanism for
defining sched domains is borked.  There is no concensus on the
replacement.  But since we can remove this mechanism, and since its
continued presence risks causing unwanted partitioning of the schedulers
load balancing, we should remove it while we can, as we proceed to work the
replacement scheduler domain mechanisms.
Signed-off-by: NPaul Jackson <pj@sgi.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Christoph Lameter <clameter@engr.sgi.com>
Cc: Dinakar Guniguntala <dino@in.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add code to connect to the DCA driver and provide cpu tags for use by
drivers that would like to use Direct Cache Access hints.

    [Adrian Bunk]                Several Kconfig cleanup items
    [Andrew Morten, Chris Leech] Fix for using cpu_physical_id() even when
			         built for uni-processor
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Direct Cache Access (DCA) is a method for warming the CPU cache before data
is used, with the intent of lessening the impact of cache misses.  This
patch adds a manager and interface for matching up client requests for DCA
services with devices that offer DCA services.

In order to use DCA, a module must do bus writes with the appropriate tag
bits set to trigger a cache read for a specific CPU.  However, different
CPUs and chipsets can require different sets of tag bits, and the methods
for determining the correct bits may be simple hardcoding or may be a
hardware specific magic incantation.  This interface is a way for DCA
clients to find the correct tag bits for the targeted CPU without needing
to know the specifics.

    [Dave Miller] use DEFINE_SPINLOCK()
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add support for MSI and MSI-X interrupt handling, including the ability
to choose the desired interrupt method.
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
[bunk@kernel.org: drivers/dma/ioat_dma.c: make 3 functions static]
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Split the general PCI startup from the DMA handling code in order to
prepare for adding support for DCA services and future versions of the
ioatdma device.

    [Rusty Russell] Removal of __unsafe() usage.
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Take care of a bunch of little code nits in ioatdma files
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Rename the ioatdma.c file in preparation for splitting into multiple files,
which will allow for easier adding new functionality.
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add device ids for new revs of the Intel I/OAT DMA engine
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
Cc: Miles Bader <miles@gnu.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In a stock 2.6.22.6 kernel, poweroff a user mode linux guest (2.6.22.6 running
in skas0 mode) will halt the host linux.  I think the reason is the kernel
thread abort because of a bug.  Then the sys_reboot in process of user mode
linux guest is not trapped by the user mode linux kernel and is executed by
host.  I think it is better to make sure all of our children process to quit
when user mode linux kernel abort.

[ jdike - the kernel process needs to ignore SIGTERM, plus the waitpid/kill
loop is needed to make sure that all of our children are dead before the
kernel exits ]
Signed-off-by: NLepton Wu <ytht.net@gmail.com>
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Style fixes for the rest of the drivers.  arch/um/drivers should be pretty
CodingStyle-compliant now.

Except for the ubd driver, which will have to be treated separately.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The calculation of CONFIG_STUB_CODE and CONFIG_STUB_DATA didn't take into
account anything but 3G/1G and 2G/2G, leaving the other vmsplits out in the
cold.

I'd rather not duplicate the four known host vmsplit cases for each of these
symbols.  I'd also like to calculate them based on the highest userspace
address.

The Kconfig language seems not to allow calculation of hex constants, so I
moved this to as-layout.h.  CONFIG_STUB_CODE, CONFIG_STUB_DATA, and
CONFIG_STUB_START are now gone.  In their place are STUB_CODE, STUB_DATA, and
STUB_START in as-layout.h.

i386 and x86_64 seem to differ as to whether an unadorned constant is an int
or a long, so I cast them to unsigned long so they can be printed
consistently.  However, they are also used in stub.S, where C types don't work
so well.  So, there are ASM_ versions of these constants for use in stub.S.  I
also ifdef-ed the non-asm-friendly portion of as-layout.h.

With this in place, most of the rest of this patch is changing CONFIG_STUB_*
to STUB_*, except in stub.S, where they are changed to ASM_STUB_*.

defconfig has the old symbols deleted.

I also print these addresses out in case there is any problem mapping them on
the host.

The two stub.S files had some trailing whitespace, so that is cleaned up here.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Handle memory allocation failures when reading packets.

We have to read something from the host, even if we can't allocate any
memory.  If we don't, the host side of the device may fill up and stop
delivering interrupts because no new packets can be queued.

A single sk_buff is allocated whenever an MTU is seen which is larger
than any seen earlier.  This is used to read packets if there is a
memory allocation failure.

The large MTU check is done from eth_configure, which is called when a
interface is added to the system.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

A bunch of MTU-related cleanups in the network code.

First, there is the addition of the notion of a maximally-sized packet, which
is the MTU plus headers.  This is used to size the skb that will receive a
packet.  This allows ether_adjust_skb to go away, as it was used to resize the
skb after it was allocated.

Since the skb passed into the low-level read routine is no longer resized, and
possibly reallocated, there, they (and the write routines) don't need to get
an sk_buff **.  They just need the sk_buff * now.  The callers of
ether_adjust_skb still need to do the skb_put, so that's now inlined.

The MAX_PACKET definitions in most of the drivers are gone.

The set_mtu methods were all the same and did nothing, so they can be
removed.

The ethertap driver had a typo which doubled the size of the packet rather
than adding two bytes to it.  It also wasn't defining its setup_size, causing
a zero-byte kmalloc and crash when the invalid pointer returned from kmalloc
was dereferenced.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Style and other non-functional changes in the UML networking code, including
	include tidying
	style violations
	copyright updates
	printks getting severities
	userspace code calling libc directly rather than using the os_*
wrappers

There's also a exit path cleanup in the pcap driver.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

There are various uses of powers of 1000, plus the odd BILLION constant in the
time code.  However, there are perfectly good definitions of *SEC_PER_*SEC in
linux/time.h which can be used instaed.

These are replaced directly in kernel code.  Userspace code imports those
constants as UM_*SEC_PER_*SEC and uses these.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Now that ITIMER_REAL is no longer used, there is no need for any use of
SIGALRM whatsoever.  This patch removes all mention of it.

In addition, real_alarm_handler took a signal argument which is now always
SIGVTALRM.  So, that is gone.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Put back an implementation of timeval_to_ns in arch/um/os-Linux/time.c.
tglx pointed out in his review of tickless support that there was a
perfectly good implementation of it in linux/time.h.  The problem is that
this is userspace code which can't pull in kernel headers and there doesn't
seem to be a libc version.

So, I'm copying the version from linux/time.h rather than resurrecting my
version.  This causes some declaration changes as it now returns a signed
value rather than an unsigned value.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Now, the idle loop now longer needs SIGALRM firing - it can just sleep for the
requisite amount of time and fake a timer interrupt when it finishes.

Any use of ITIMER_REAL now goes away.  disable_timer only turns off
ITIMER_VIRTUAL.  switch_timers is no longer needed, so it, and all calls, goes
away.

disable_timer now returns the amount of time remaining on the timer.
default_idle uses this to tell idle_sleep how long to sleep.  idle_sleep will
call alarm_handler if nanosleep returns 0, which is the case if it didn't
return early due to an interrupt.  Otherwise, it just returns.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Enable tickless support.

CONFIG_TICK_ONESHOT and CONFIG_NO_HZ are enabled.

itimer_clockevent gets CLOCK_EVT_FEAT_ONESHOT and an implementation of
.set_next_event.

CONFIG_UML_REAL_TIME_CLOCK goes away because it only makes sense when there is
a clock ticking away all the time.  timer_handler now just calls do_IRQ once
without trying to figure out how many ticks to emulate.

The idle loop now needs to turn ticking on and off.

Userspace ticks keep happening as usual.  However, the userspace loop keep
track of when the next wakeup should happen and suppresses process ticks until
that happens.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add clocksource support.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Enable CONFIG_GENERIC_CLOCKEVENTS.

timer_irq gets its name changed to timer_handler, and becomes the recipient of
timer signals.

The clock_event_device is set up to imitate the current ticking clock, i.e.
CLOCK_EVT_FEAT_ONESHOT is not enabled yet.

disable_timer now doesn't ignore SIGALRM and SIGVTALRM because that breaks
delay calibration.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Enable CONFIG_GENERIC_TIME.

As a side-effect of this, the UML implementations of do_gettimeofday and
do_settimeofday go away, as these are provided by generic code.  set_time also
goes away since it was only used by do_settimeofday.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Move timer signal initialization from init_irq_signals to a new function,
timer_init.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

set_interval took a timer type as an argument, but it always specified a
virtual timer.  So, it is not needed, and it is gone, and set_interval is
simplified appropriately.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix up the switching between virtual and real timers.  The idle loop sleeps,
so the timer at that point must be real time.  At all other times, the timer
must be virtual.  Even when userspace is running, and the kernel is asleep,
the virtual timer is correct because the process timer will be running and the
process timer will be firing.

The timer switch used to be in the context switch and timer handler code.
This is moved to the idle loop and the signal handler, making it much more
clear why it is happening.

switch_timers now returns the old timer type so that it may be restored.  The
signal handler uses this in order to restore the previous timer type when it
returns.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Eliminate hz() since its only purpose was to provide a kernel-space constant
to userspace code.  This can be done instead by providing the constant
directly through kernel_constants.h.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Make mconsole parameter parsing slightly more idiomatic.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

arch/um/os-Linux/tt.c is no longer used.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

vmalloc() returns a void pointer, so casting to (void *) is pretty pointless.
Signed-off-by: NJesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The floating point fields in the pt_regs register file aren't used, so they
are deleted.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Tidy the code affected by the floating point fixes.

A bunch of unused stuff is gone, including two sigcontext.c files,
which turned out to be entirely unneeded.

There are the usual fixes -
	whitespace and style cleanups
	copyright updates
	emacs formatting comments gone
	include cleanups
	adding severities to printks
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix the passing of printk output back to the mconsole client.  The existing
code was somewhat confused, accumulating output in a buffer, but writing it
out entirely whenever a new chunk was added.  This is fixed.

The earlier include cleanups caused linux/sysrq.h to not be included - this is
fixed by adding the include back, under CONFIG_MAGIC_SYSRQ.

CONFIG_MAGIC_SYSRQ is also defaulted to on in defconfig.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix core dumping of floating point state.  ELF_CORE_COPY_FPREGS gets a
definitions, and as a result, dump_fpu no longer needs to exist.  Also,
elf_fpregset_t needed a real definition.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Handle floating point state better in ptrace.  The code now correctly
distinguishes between PTRACE_[GS]ETFPREGS and PTRACE_[GS]ETFPXREGS.  The FPX
requests get handed off to arch-specific code because that's not generic.

get_fpregs, set_fpregs, set_fpregs, and set_fpxregs needed real
implementations.

Something here exposed a missing include in asm/page.h, which needed
linux/types.h in order to get gfp_t, so that's fixed here.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Handle floating point state in across signals correctly.  UML/i386 needs to
know whether the host does PTRACE_[GS]ETFPXREGS, so an arch_init_registers
hook is added, which on x86_64 does nothing.

UML doesn't save and restore floating point registers on kernel entry and
exit, so they need to be copied between the host process and the sigcontext.
save_fpx_registers and restore_fpx_registers are added for this purpose.
save_fp_registers and restore_fp_registers already exist.

There was a bunch of floating point state conversion code in
arch/um/sys-i386/ptrace.c which isn't needed there, but is needed in signal.c,
so it is moved over.

The i386 code now distinguishes between fp and fpx state and handles them
correctly.  The x86_64 code just needs to copy state as-is between the host
process and the stack.  There are also some fixes there to pass the correct
address of the floating point state around.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Stop including asm/user.h from libc - it seems to be disappearing from
distros.  It's replaced with sys/user.h which defines user_fpregs_struct and
user_fpxregs_struct instead of user_i387_struct and struct user_fxsr_struct on
i386.

As a bonus, on x86_64, I get to dump some stupid typedefs which were needed in
order to get asm/user.h to compile.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Style fixes in hostfs.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Get rid of an empty if statement which might look like a bug to a
casual reader.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Tidy the tlb flushing code.

With tt mode gone, there is no reason to have the capability to have
called directly from do_mmap, do_mprotect, and do_munmap, rather than
calling a function pointer that it is given.

There was a large amount of data that was passed from function to
function, being used at the lowest level, without being changed.  This
stuff is now encapsulated in a structure which is initialized at the
top layer and passed down.  This simplifies the code, reduces the
amount of code needed to pass the parameters around, and saves on
stack space.

A somewhat more subtle change is the meaning of the current operation
index.  It used to start at -1, being pre-incremented when adding an
operation.  It now starts at 0, being post-incremented, with
associated adjustments of +/- 1 on comparisons.

In addition, tlb.h contained a couple of declarations which had no
users outside of tlb.c, so they could be moved or deleted.
Signed-off-by: NJeff Dike <jdike@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>