Now that the allocation users have been split off into a separate
function, init_fpu() has become local to i387.c: make it static.
Reviewed-by: NBorislav Petkov <bp@alien8.de>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NIngo Molnar <mingo@kernel.org>

Most init_fpu() users don't want the register-saving aspect of the
function, they are calling it for 'current' and when FPU registers
are not allocated and initialized yet.

Split out a simplified API that does just that (and add debug-checks
for these conditions): fpstate_alloc_init().

Use it where appropriate.
Reviewed-by: NBorislav Petkov <bp@alien8.de>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NIngo Molnar <mingo@kernel.org>

This function is a misnomer on two levels:

1) it doesn't really manipulate TS on modern CPUs anymore, its
   primary purpose is to save FPU state, used:

      - when executing fork()/clone(): to copy current FPU state
        to the child's FPU state.

      - when handling math exceptions: to generate the math error
        si_code in the signal frame.

2) even on legacy CPUs it doesn't actually 'unlazy', if then
   it lazies the FPU state: as a side effect of the old FNSAVE
   instruction which clears (destroys) FPU state it's necessary
   to set CR0::TS.

So rename it to fpu__save() to better reflect its purpose.
Reviewed-by: NBorislav Petkov <bp@alien8.de>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: NIngo Molnar <mingo@kernel.org>

math_state_restore() can race with kernel_fpu_begin() if irq comes
right after __thread_fpu_begin(), __save_init_fpu() will overwrite
fpu->state we are going to restore.

Add 2 simple helpers, kernel_fpu_disable() and kernel_fpu_enable()
which simply set/clear in_kernel_fpu, and change math_state_restore()
to exclude kernel_fpu_begin() in between.

Alternatively we could use local_irq_save/restore, but probably these
new helpers can have more users.

Perhaps they should disable/enable preemption themselves, in this case
we can remove preempt_disable() in __restore_xstate_sig().
Signed-off-by: NOleg Nesterov <oleg@redhat.com>
Reviewed-by: NRik van Riel <riel@redhat.com>
Cc: matt.fleming@intel.com
Cc: bp@suse.de
Cc: pbonzini@redhat.com
Cc: luto@amacapital.net
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Suresh Siddha <sbsiddha@gmail.com>
Link: http://lkml.kernel.org/r/20150115192028.GD27332@redhat.comSigned-off-by: NThomas Gleixner <tglx@linutronix.de>

interrupted_kernel_fpu_idle() tries to detect if kernel_fpu_begin()
is safe or not. In particular it should obviously deny the nested
kernel_fpu_begin() and this logic looks very confusing.

If use_eager_fpu() == T we rely on a) __thread_has_fpu() check in
interrupted_kernel_fpu_idle(), and b) on the fact that _begin() does
__thread_clear_has_fpu().

Otherwise we demand that the interrupted task has no FPU if it is in
kernel mode, this works because __kernel_fpu_begin() does clts() and
interrupted_kernel_fpu_idle() checks X86_CR0_TS.

Add the per-cpu "bool in_kernel_fpu" variable, and change this code
to check/set/clear it. This allows to do more cleanups and fixes, see
the next changes.

The patch also moves WARN_ON_ONCE() under preempt_disable() just to
make this_cpu_read() look better, this is not really needed. And in
fact I think we should move it into __kernel_fpu_begin().
Signed-off-by: NOleg Nesterov <oleg@redhat.com>
Reviewed-by: NRik van Riel <riel@redhat.com>
Cc: matt.fleming@intel.com
Cc: bp@suse.de
Cc: pbonzini@redhat.com
Cc: luto@amacapital.net
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Suresh Siddha <sbsiddha@gmail.com>
Link: http://lkml.kernel.org/r/20150115191943.GB27332@redhat.comSigned-off-by: NThomas Gleixner <tglx@linutronix.de>

Preemption is disabled between kernel_fpu_begin/end() and as such
it is not a good idea to use these routines in kvm_load/put_guest_fpu()
which can be very far apart.

kvm_load/put_guest_fpu() routines are already called with
preemption disabled and KVM already uses the preempt notifier to save
the guest fpu state using kvm_put_guest_fpu().

So introduce __kernel_fpu_begin/end() routines which don't touch
preemption and use them instead of kernel_fpu_begin/end()
for KVM's use model of saving/restoring guest FPU state.

Also with this change (and with eagerFPU model), fix the host cr0.TS vm-exit
state in the case of VMX. For eagerFPU case, host cr0.TS is always clear.
So no need to worry about it. For the traditional lazyFPU restore case,
change the cr0.TS bit for the host state during vm-exit to be always clear
and cr0.TS bit is set in the __vmx_load_host_state() when the FPU
(guest FPU or the host task's FPU) state is not active. This ensures
that the host/guest FPU state is properly saved, restored
during context-switch and with interrupts (using irq_fpu_usable()) not
stomping on the active FPU state.
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Link: http://lkml.kernel.org/r/1348164109.26695.338.camel@sbsiddha-desk.sc.intel.com
Cc: Avi Kivity <avi@redhat.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Fundamental model of the current Linux kernel is to lazily init and
restore FPU instead of restoring the task state during context switch.
This changes that fundamental lazy model to the non-lazy model for
the processors supporting xsave feature.

Reasons driving this model change are:

i. Newer processors support optimized state save/restore using xsaveopt and
xrstor by tracking the INIT state and MODIFIED state during context-switch.
This is faster than modifying the cr0.TS bit which has serializing semantics.

ii. Newer glibc versions use SSE for some of the optimized copy/clear routines.
With certain workloads (like boot, kernel-compilation etc), application
completes its work with in the first 5 task switches, thus taking upto 5 #DNA
traps with the kernel not getting a chance to apply the above mentioned
pre-load heuristic.

iii. Some xstate features (like AMD's LWP feature) don't honor the cr0.TS bit
and thus will not work correctly in the presence of lazy restore. Non-lazy
state restore is needed for enabling such features.

Some data on a two socket SNB system:
 * Saved 20K DNA exceptions during boot on a two socket SNB system.
 * Saved 50K DNA exceptions during kernel-compilation workload.
 * Improved throughput of the AVX based checksumming function inside the
   kernel by ~15% as xsave/xrstor is faster than the serializing clts/stts
   pair.

Also now kernel_fpu_begin/end() relies on the patched
alternative instructions. So move check_fpu() which uses the
kernel_fpu_begin/end() after alternative_instructions().
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Link: http://lkml.kernel.org/r/1345842782-24175-7-git-send-email-suresh.b.siddha@intel.com
Merge 32-bit boot fix from,
Link: http://lkml.kernel.org/r/1347300665-6209-4-git-send-email-suresh.b.siddha@intel.com
Cc: Jim Kukunas <james.t.kukunas@linux.intel.com>
Cc: NeilBrown <neilb@suse.de>
Cc: Avi Kivity <avi@redhat.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Disintegrate asm/system.h for X86.
Signed-off-by: NDavid Howells <dhowells@redhat.com>
Acked-by: NH. Peter Anvin <hpa@zytor.com>
cc: x86@kernel.org

While various modules include <asm/i387.h> to get access to things we
actually *intend* for them to use, most of that header file was really
pretty low-level internal stuff that we really don't want to expose to
others.

So split the header file into two: the small exported interfaces remain
in <asm/i387.h>, while the internal definitions that are only used by
core architecture code are now in <asm/fpu-internal.h>.

The guiding principle for this was to expose functions that we export to
modules, and leave them in <asm/i387.h>, while stuff that is used by
task switching or was marked GPL-only is in <asm/fpu-internal.h>.

The fpu-internal.h file could be further split up too, especially since
arch/x86/kvm/ uses some of the remaining stuff for its module.  But that
kvm usage should probably be abstracted out a bit, and at least now the
internal FPU accessor functions are much more contained.  Even if it
isn't perhaps as contained as it _could_ be.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1202211340330.5354@i5.linux-foundation.orgSigned-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Instead of exporting the very low-level internals of the FPU state
save/restore code (ie things like 'fpu_owner_task'), we should export
the higher-level interfaces.

Inlining these things is pointless anyway: sure, sometimes the end
result is small, but while 'stts()' can result in just three x86
instructions, those are not cheap instructions (writing %cr0 is a
serializing instruction and a very slow one at that).

So the overhead of a function call is not noticeable, and we really
don't want random modules mucking about with our internal state save
logic anyway.

So this unexports 'fpu_owner_task', and instead uninlines and exports
the actual functions that modules can use: fpu_kernel_begin/end() and
unlazy_fpu().
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
Link: http://lkml.kernel.org/r/alpine.LFD.2.02.1202211339590.5354@i5.linux-foundation.orgSigned-off-by: NH. Peter Anvin <hpa@linux.intel.com>

This makes us recognize when we try to restore FPU state that matches
what we already have in the FPU on this CPU, and avoids the restore
entirely if so.

To do this, we add two new data fields:

 - a percpu 'fpu_owner_task' variable that gets written any time we
   update the "has_fpu" field, and thus acts as a kind of back-pointer
   to the task that owns the CPU.  The exception is when we save the FPU
   state as part of a context switch - if the save can keep the FPU
   state around, we leave the 'fpu_owner_task' variable pointing at the
   task whose FP state still remains on the CPU.

 - a per-thread 'last_cpu' field, that indicates which CPU that thread
   used its FPU on last.  We update this on every context switch
   (writing an invalid CPU number if the last context switch didn't
   leave the FPU in a lazily usable state), so we know that *that*
   thread has done nothing else with the FPU since.

These two fields together can be used when next switching back to the
task to see if the CPU still matches: if 'fpu_owner_task' matches the
task we are switching to, we know that no other task (or kernel FPU
usage) touched the FPU on this CPU in the meantime, and if the current
CPU number matches the 'last_cpu' field, we know that this thread did no
other FP work on any other CPU, so the FPU state on the CPU must match
what was saved on last context switch.

In that case, we can avoid the 'f[x]rstor' entirely, and just clear the
CR0.TS bit.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This inlines what is usually just a couple of instructions, but more
importantly it also fixes the theoretical error case (can that FPU
restore really ever fail? Maybe we should remove the checking).

We can't start sending signals from within the scheduler, we're much too
deep in the kernel and are holding the runqueue lock etc.  So don't
bother even trying.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This makes sure we clear the FPU usage counter for newly created tasks,
just so that we start off in a known state (for example, don't try to
preload the FPU state on the first task switch etc).

It also fixes a thinko in when we increment the fpu_counter at task
switch time, introduced by commit 34ddc81a ("i387: re-introduce FPU
state preloading at context switch time").  We should increment the
*new* task fpu_counter, not the old task, and only if we decide to use
that state (whether lazily or preloaded).
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

After all the FPU state cleanups and finally finding the problem that
caused all our FPU save/restore problems, this re-introduces the
preloading of FPU state that was removed in commit b3b0870e ("i387:
do not preload FPU state at task switch time").

However, instead of simply reverting the removal, this reimplements
preloading with several fixes, most notably

 - properly abstracted as a true FPU state switch, rather than as
   open-coded save and restore with various hacks.

   In particular, implementing it as a proper FPU state switch allows us
   to optimize the CR0.TS flag accesses: there is no reason to set the
   TS bit only to then almost immediately clear it again.  CR0 accesses
   are quite slow and expensive, don't flip the bit back and forth for
   no good reason.

 - Make sure that the same model works for both x86-32 and x86-64, so
   that there are no gratuitous differences between the two due to the
   way they save and restore segment state differently due to
   architectural differences that really don't matter to the FPU state.

 - Avoid exposing the "preload" state to the context switch routines,
   and in particular allow the concept of lazy state restore: if nothing
   else has used the FPU in the meantime, and the process is still on
   the same CPU, we can avoid restoring state from memory entirely, just
   re-expose the state that is still in the FPU unit.

   That optimized lazy restore isn't actually implemented here, but the
   infrastructure is set up for it.  Of course, older CPU's that use
   'fnsave' to save the state cannot take advantage of this, since the
   state saving also trashes the state.

In other words, there is now an actual _design_ to the FPU state saving,
rather than just random historical baggage.  Hopefully it's easier to
follow as a result.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This moves the bit that indicates whether a thread has ownership of the
FPU from the TS_USEDFPU bit in thread_info->status to a word of its own
(called 'has_fpu') in task_struct->thread.has_fpu.

This fixes two independent bugs at the same time:

 - changing 'thread_info->status' from the scheduler causes nasty
   problems for the other users of that variable, since it is defined to
   be thread-synchronous (that's what the "TS_" part of the naming was
   supposed to indicate).

   So perfectly valid code could (and did) do

	ti->status |= TS_RESTORE_SIGMASK;

   and the compiler was free to do that as separate load, or and store
   instructions.  Which can cause problems with preemption, since a task
   switch could happen in between, and change the TS_USEDFPU bit. The
   change to TS_USEDFPU would be overwritten by the final store.

   In practice, this seldom happened, though, because the 'status' field
   was seldom used more than once, so gcc would generally tend to
   generate code that used a read-modify-write instruction and thus
   happened to avoid this problem - RMW instructions are naturally low
   fat and preemption-safe.

 - On x86-32, the current_thread_info() pointer would, during interrupts
   and softirqs, point to a *copy* of the real thread_info, because
   x86-32 uses %esp to calculate the thread_info address, and thus the
   separate irq (and softirq) stacks would cause these kinds of odd
   thread_info copy aliases.

   This is normally not a problem, since interrupts aren't supposed to
   look at thread information anyway (what thread is running at
   interrupt time really isn't very well-defined), but it confused the
   heck out of irq_fpu_usable() and the code that tried to squirrel
   away the FPU state.

   (It also caused untold confusion for us poor kernel developers).

It also turns out that using 'task_struct' is actually much more natural
for most of the call sites that care about the FPU state, since they
tend to work with the task struct for other reasons anyway (ie
scheduling).  And the FPU data that we are going to save/restore is
found there too.

Thanks to Arjan Van De Ven <arjan@linux.intel.com> for pointing us to
the %esp issue.

Cc: Arjan van de Ven <arjan@linux.intel.com>
Reported-and-tested-by: NRaphael Prevost <raphael@buro.asia>
Acked-and-tested-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Tested-by: NPeter Anvin <hpa@zytor.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception is
pending.  In order to not leak FIP state from one process to another, we
need to do a floating point load after the fxsave of the old process,
and before the fxrstor of the new FPU state.  That resets the state to
the (uninteresting) kernel load, rather than some potentially sensitive
user information.

We used to do this directly after the FPU state save, but that is
actually very inconvenient, since it

 (a) corrupts what is potentially perfectly good FPU state that we might
     want to lazy avoid restoring later and

 (b) on x86-64 it resulted in a very annoying ordering constraint, where
     "__unlazy_fpu()" in the task switch needs to be delayed until after
     the DS segment has been reloaded just to get the new DS value.

Coupling it to the fxrstor instead of the fxsave automatically avoids
both of these issues, and also ensures that we only do it when actually
necessary (the FP state after a save may never actually get used).  It's
simply a much more natural place for the leaked state cleanup.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Yes, taking the trap to re-load the FPU/MMX state is expensive, but so
is spending several days looking for a bug in the state save/restore
code.  And the preload code has some rather subtle interactions with
both paravirtualization support and segment state restore, so it's not
nearly as simple as it should be.

Also, now that we no longer necessarily depend on a single bit (ie
TS_USEDFPU) for keeping track of the state of the FPU, we migth be able
to do better.  If we are really switching between two processes that
keep touching the FP state, save/restore is inevitable, but in the case
of having one process that does most of the FPU usage, we may actually
be able to do much better than the preloading.

In particular, we may be able to keep track of which CPU the process ran
on last, and also per CPU keep track of which process' FP state that CPU
has.  For modern CPU's that don't destroy the FPU contents on save time,
that would allow us to do a lazy restore by just re-enabling the
existing FPU state - with no restore cost at all!
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This creates three helper functions that do the TS_USEDFPU accesses, and
makes everybody that used to do it by hand use those helpers instead.

In addition, there's a couple of helper functions for the "change both
CR0.TS and TS_USEDFPU at the same time" case, and the places that do
that together have been changed to use those.  That means that we have
fewer random places that open-code this situation.

The intent is partly to clarify the code without actually changing any
semantics yet (since we clearly still have some hard to reproduce bug in
this area), but also to make it much easier to use another approach
entirely to caching the CR0.TS bit for software accesses.

Right now we use a bit in the thread-info 'status' variable (this patch
does not change that), but we might want to make it a full field of its
own or even make it a per-cpu variable.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Touching TS_USEDFPU without touching CR0.TS is confusing, so don't do
it.  By moving it into the callers, we always do the TS_USEDFPU next to
the CR0.TS accesses in the source code, and it's much easier to see how
the two go hand in hand.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Commit 5b1cbac3 ("i387: make irq_fpu_usable() tests more robust")
added a sanity check to the #NM handler to verify that we never cause
the "Device Not Available" exception in kernel mode.

However, that check actually pinpointed a (fundamental) race where we do
cause that exception as part of the signal stack FPU state save/restore
code.

Because we use the floating point instructions themselves to save and
restore state directly from user mode, we cannot do that atomically with
testing the TS_USEDFPU bit: the user mode access itself may cause a page
fault, which causes a task switch, which saves and restores the FP/MMX
state from the kernel buffers.

This kind of "recursive" FP state save is fine per se, but it means that
when the signal stack save/restore gets restarted, it will now take the
'#NM' exception we originally tried to avoid.  With preemption this can
happen even without the page fault - but because of the user access, we
cannot just disable preemption around the save/restore instruction.

There are various ways to solve this, including using the
"enable/disable_page_fault()" helpers to not allow page faults at all
during the sequence, and fall back to copying things by hand without the
use of the native FP state save/restore instructions.

However, the simplest thing to do is to just allow the #NM from kernel
space, but fix the race in setting and clearing CR0.TS that this all
exposed: the TS bit changes and the TS_USEDFPU bit absolutely have to be
atomic wrt scheduling, so while the actual state save/restore can be
interrupted and restarted, the act of actually clearing/setting CR0.TS
and the TS_USEDFPU bit together must not.

Instead of just adding random "preempt_disable/enable()" calls to what
is already excessively ugly code, this introduces some helper functions
that mostly mirror the "kernel_fpu_begin/end()" functionality, just for
the user state instead.

Those helper functions should probably eventually replace the other
ad-hoc CR0.TS and TS_USEDFPU tests too, but I'll need to think about it
some more: the task switching functionality in particular needs to
expose the difference between the 'prev' and 'next' threads, while the
new helper functions intentionally were written to only work with
'current'.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The check for save_init_fpu() (introduced in commit 5b1cbac3: "i387:
make irq_fpu_usable() tests more robust") was the wrong way around, but
I hadn't noticed, because my "tests" were bogus: the FPU exceptions are
disabled by default, so even doing a divide by zero never actually
triggers this code at all unless you do extra work to enable them.

So if anybody did enable them, they'd get one spurious warning.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Some code - especially the crypto layer - wants to use the x86
FP/MMX/AVX register set in what may be interrupt (typically softirq)
context.

That *can* be ok, but the tests for when it was ok were somewhat
suspect.  We cannot touch the thread-specific status bits either, so
we'd better check that we're not going to try to save FP state or
anything like that.

Now, it may be that the TS bit is always cleared *before* we set the
USEDFPU bit (and only set when we had already cleared the USEDFP
before), so the TS bit test may actually have been sufficient, but it
certainly was not obviously so.

So this explicitly verifies that we will not touch the TS_USEDFPU bit,
and adds a few related sanity-checks.  Because it seems that somehow
AES-NI is corrupting user FP state.  The cause is not clear, and this
patch doesn't fix it, but while debugging it I really wanted the code to
be more obviously correct and robust.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

It was marked asmlinkage for some really old and stale legacy reasons.
Fix that and the equally stale comment.

Noticed when debugging the irq_fpu_usable() bugs.
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This patch changes fields in cpustat from a structure, to an
u64 array. Math gets easier, and the code is more flexible.
Signed-off-by: NGlauber Costa <glommer@parallels.com>
Reviewed-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Paul Tuner <pjt@google.com>
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1322498719-2255-2-git-send-email-glommer@parallels.comSigned-off-by: NIngo Molnar <mingo@elte.hu>

On 32bit systems without SSE (that is, they use FSAVE/FRSTOR for FPU
context switches), FPU exceptions in user mode cause Oopses, BUGs,
recursive faults and other nasty things:

fpu exception: 0000 [#1]
last sysfs file: /sys/power/state
Modules linked in: psmouse evdev pcspkr serio_raw [last unloaded: scsi_wait_scan]

Pid: 1638, comm: fxsave-32-excep Not tainted 2.6.35-07798-g58a992b9-dirty #633 VP3-596B-DD/VT82C597
EIP: 0060:[<c1003527>] EFLAGS: 00010202 CPU: 0
EIP is at math_error+0x1b4/0x1c8
EAX: 00000003 EBX: cf9be7e0 ECX: 00000000 EDX: cf9c5c00
ESI: cf9d9fb4 EDI: c1372db3 EBP: 00000010 ESP: cf9d9f1c
DS: 007b ES: 007b FS: 0000 GS: 00e0 SS: 0068
Process fxsave-32-excep (pid: 1638, ti=cf9d8000 task=cf9be7e0 task.ti=cf9d8000)
Stack:
00000000 00000301 00000004 00000000 00000000 cf9d3000 cf9da8f0 00000001
<0> 00000004 cf9b6b60 c1019a6b c1019a79 00000020 00000242 000001b6 cf9c5380
<0> cf806b40 cf791880 00000000 00000282 00000282 c108a213 00000020 cf9c5380
Call Trace:
[<c1019a6b>] ? need_resched+0x11/0x1a
[<c1019a79>] ? should_resched+0x5/0x1f
[<c108a213>] ? do_sys_open+0xbd/0xc7
[<c108a213>] ? do_sys_open+0xbd/0xc7
[<c100353b>] ? do_coprocessor_error+0x0/0x11
[<c12d5965>] ? error_code+0x65/0x70
Code: a8 20 74 30 c7 44 24 0c 06 00 03 00 8d 54 24 04 89 d9 b8 08 00 00 00 e8 9b 6d 02 00 eb 16 8b 93 5c 02 00 00 eb 05 e9 04 ff ff ff <9b> dd 32 9b e9 16 ff ff ff 81 c4 84 00 00 00 5b 5e 5f 5d c3 c6
EIP: [<c1003527>] math_error+0x1b4/0x1c8 SS:ESP 0068:cf9d9f1c

This usually continues in slight variations until the system is reset.

This bug was introduced by commit 58a992b9:
	x86-32, fpu: Rewrite fpu_save_init()
Signed-off-by: NHans Rosenfeld <hans.rosenfeld@amd.com>
Link: http://lkml.kernel.org/r/1302106003-366952-1-git-send-email-hans.rosenfeld@amd.comSigned-off-by: NH. Peter Anvin <hpa@zytor.com>

Checkin d7acb92f made use of fxsaveq
in fpu_fxsave() if the assembler supports it; this adds
fxsaveq/fxrstorq to fxrstor_checking() and fxsave_user() as well.
Reported-by: NLinus Torvalds <torvalds@linux-foundation.org>
LKML-Reference: <AANLkTi=RKyHLNTq6iomZOXkc6Zw1j9iAgsq8388XmzwN@mail.gmail.com>
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>

Kbuild allows for us to probe for the existence of specific constructs
in the assembler, use them to find out if we can use fxsave64 and
permit the compiler to generate better code.
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Make 64-bit use the 32-bit version of fpu_save_init().  Remove
unused clear_fpu_state().
Signed-off-by: NBrian Gerst <brgerst@gmail.com>
Acked-by: NPekka Enberg <penberg@kernel.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1283563039-3466-13-git-send-email-brgerst@gmail.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Rewrite fpu_save_init() to prepare for merging with 64-bit.
Signed-off-by: NBrian Gerst <brgerst@gmail.com>
Acked-by: NPekka Enberg <penberg@kernel.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1283563039-3466-12-git-send-email-brgerst@gmail.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

The PSHUFB_XMM5_* macros are no longer used.
Signed-off-by: NBrian Gerst <brgerst@gmail.com>
Acked-by: NPekka Enberg <penberg@kernel.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1283563039-3466-11-git-send-email-brgerst@gmail.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Remove ifdefs for code that the compiler can optimize away on 64-bit.
Signed-off-by: NBrian Gerst <brgerst@gmail.com>
Acked-by: NPekka Enberg <penberg@kernel.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1283563039-3466-10-git-send-email-brgerst@gmail.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Use the "R" constraint (legacy register) instead of listing all the
possible registers.  Clean up the comments as well.
Signed-off-by: NBrian Gerst <brgerst@gmail.com>
Acked-by: NPekka Enberg <penberg@kernel.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1283563039-3466-8-git-send-email-brgerst@gmail.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Consolidates code and fixes the below race for 64-bit.

commit 9fa2f37bfeb798728241cc4a19578ce6e4258f25
Author: torvalds <torvalds>
Date:   Tue Sep 2 07:37:25 2003 +0000

    Be a lot more careful about TS_USEDFPU and preemption

    We had some races where we testecd (or set) TS_USEDFPU together
    with sequences that depended on the setting (like clearing or
    setting the TS flag in %cr0) and we could be preempted in between,
    which screws up the FPU state, since preemption will itself change
    USEDFPU and the TS flag.

    This makes it a lot more explicit: the "internal" low-level FPU
    functions ("__xxxx_fpu()") all require preemption to be disabled,
    and the exported "real" functions will make sure that is the case.

    One case - in __switch_to() - was switched to the non-preempt-safe
    internal version, since the scheduler itself has already disabled
    preemption.

    BKrev: 3f5448b5WRiQuyzAlbajs3qoQjSobw
Signed-off-by: NBrian Gerst <brgerst@gmail.com>
Acked-by: NPekka Enberg <penberg@kernel.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1283563039-3466-6-git-send-email-brgerst@gmail.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

__save_init_fpu() is identical for 32-bit and 64-bit.
Signed-off-by: NBrian Gerst <brgerst@gmail.com>
Acked-by: NPekka Enberg <penberg@kernel.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1283563039-3466-5-git-send-email-brgerst@gmail.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Commit e2e75c91 merged the math exception handler, allowing both 32-bit
and 64-bit to handle math exceptions from kernel mode.  Switch to using
the 64-bit version of tolerant_fwait() without fnclex, which simply
ignores the exception if one is still pending from userspace.
Signed-off-by: NBrian Gerst <brgerst@gmail.com>
Acked-by: NPekka Enberg <penberg@kernel.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1283563039-3466-4-git-send-email-brgerst@gmail.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

Also add some constants.
Signed-off-by: NSheng Yang <sheng@linux.intel.com>
Signed-off-by: NAvi Kivity <avi@redhat.com>

As xsave also supports other than fpu features, it should be
initialized independently of the fpu. This patch moves this out of fpu
initialization.

There is also a lot of cross referencing between fpu and xsave
code. This patch reduces this by making xsave_cntxt_init() and
init_thread_xstate() static functions.

The patch moves the cpu_has_xsave check at the beginning of
xsave_init(). All other checks may removed then.
Signed-off-by: NRobert Richter <robert.richter@amd.com>
LKML-Reference: <1279731838-1522-2-git-send-email-robert.richter@amd.com>
Acked-by: NSuresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

xsaveopt is a more optimized form of xsave specifically designed
for the context switch usage. xsaveopt doesn't save the state that's not
modified from the prior xrstor. And if a specific feature state gets
modified to the init state, then xsaveopt just updates the header bit
in the xsave memory layout without updating the corresponding memory
layout.
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20100719230205.604014179@sbs-t61.sc.intel.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

With xsaveopt, if a processor implementation discern that a processor state
component is in its initialized state it may modify the corresponding bit in
the xsave_hdr.xstate_bv as '0', with out modifying the corresponding memory
layout. Hence wHile presenting the xstate information to the user, we always
ensure that the memory layout of a feature will be in the init state if the
corresponding header bit is zero. This ensures the consistency and avoids the
condition of the user seeing some some stale state in the memory layout during
signal handling, debugging etc.
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <20100719230205.351459480@sbs-t61.sc.intel.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>

fxsave/xsave doesn't touch all the bytes in the memory layout used by
these instructions. Specifically SW reserved (bytes 464..511) fields
in the fxsave frame and the reserved fields in the xsave header.

To present a clean context for the signal handling, just clear these fields
instead of clearing the complete fxsave/xsave memory layout, when we dump these
registers directly to the user signal frame.

Also avoid the call to second xrstor (which inits the state not passed
in the signal frame) in restore_user_xstate() if all the state has already
been restored by the first xrstor.

These changes improve the performance of signal handling(by ~3-5% as measured
by the lat_sig).
Signed-off-by: NSuresh Siddha <suresh.b.siddha@intel.com>
LKML-Reference: <1277249017.2847.85.camel@sbs-t61.sc.intel.com>
Signed-off-by: NH. Peter Anvin <hpa@linux.intel.com>