linux-stable.git/arch/x86/kernel/fpu, branch linux-4.3.y

x86/fpu: Fix 32-bit signal frame handling

2015-12-09T19:34:19+00:00

commit ab6b52947545a5355154f64f449f97af9d05845f upstream.

(This should have gone to LKML originally. Sorry for the extra
 noise, folks on the cc.)

Background:

Signal frames on x86 have two formats:

  1. For 32-bit executables (whether on a real 32-bit kernel or
     under 32-bit emulation on a 64-bit kernel) we have a
    'fpregset_t' that includes the "FSAVE" registers.

  2. For 64-bit executables (on 64-bit kernels obviously), the
     'fpregset_t' is smaller and does not contain the "FSAVE"
     state.

When creating the signal frame, we have to be aware of whether
we are running a 32 or 64-bit executable so we create the
correct format signal frame.

Problem:

save_xstate_epilog() uses 'fx_sw_reserved_ia32' whenever it is
called for a 32-bit executable.  This is for real 32-bit and
ia32 emulation.

But, fpu__init_prepare_fx_sw_frame() only initializes
'fx_sw_reserved_ia32' when emulation is enabled, *NOT* for real
32-bit kernels.

This leads to really wierd situations where 32-bit programs
lose their extended state when returning from a signal handler.
The kernel copies the uninitialized (zero) 'fx_sw_reserved_ia32'
out to userspace in save_xstate_epilog().  But when returning
from the signal, the kernel errors out in check_for_xstate()
when it does not see FP_XSTATE_MAGIC1 present (because it was
zeroed).  This leads to the FPU/XSAVE state being initialized.

For MPX, this leads to the most permissive state and means we
silently lose bounds violations.  I think this would also mean
that we could lose *ANY* FPU/SSE/AVX state.  I'm not sure why
no one has spotted this bug.

I believe this was broken by:

	72a671ced66d ("x86, fpu: Unify signal handling code paths for x86 and x86_64 kernels")

way back in 2012.

Signed-off-by: Dave Hansen 
Cc: Andy Lutomirski 
Cc: Borislav Petkov 
Cc: Brian Gerst 
Cc: Denys Vlasenko 
Cc: H. Peter Anvin 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Cc: dave@sr71.net
Cc: fenghua.yu@intel.com
Cc: yu-cheng.yu@intel.com
Link: http://lkml.kernel.org/r/20151111002354.A0799571@viggo.jf.intel.com
Signed-off-by: Ingo Molnar 
Signed-off-by: Greg Kroah-Hartman

x86/fpu: Fix get_xsave_addr() behavior under virtualization

2015-12-09T19:34:19+00:00

commit a05917b6ba9dc9a95fc42bdcbe3a875e8ad83935 upstream.

KVM uses the get_xsave_addr() function in a different fashion from
the native kernel, in that the 'xsave' parameter belongs to guest vcpu,
not the currently running task.

But 'xsave' is replaced with current task's (host) xsave structure, so
get_xsave_addr() will incorrectly return the bad xsave address to KVM.

Fix it so that the passed in 'xsave' address is used - as intended
originally.

Signed-off-by: Huaitong Han 
Reviewed-by: Dave Hansen 
Cc: Andy Lutomirski 
Cc: Paolo Bonzini 
Cc: Borislav Petkov 
Cc: Fenghua Yu 
Cc: H. Peter Anvin 
Cc: Linus Torvalds 
Cc: Oleg Nesterov 
Cc: Peter Zijlstra 
Cc: Quentin Casasnovas 
Cc: Thomas Gleixner 
Cc: dave.hansen@intel.com
Link: http://lkml.kernel.org/r/1446800423-21622-1-git-send-email-huaitong.han@intel.com
[ Tidied up the changelog. ]
Signed-off-by: Ingo Molnar 
Signed-off-by: Greg Kroah-Hartman

x86/fpu/math-emu: Fix crash in fork()

2015-08-22T08:23:03+00:00

During later stages of math-emu bootup the following crash triggers:

	 math_emulate: 0060:c100d0a8
	 Kernel panic - not syncing: Math emulation needed in kernel
	 CPU: 0 PID: 1511 Comm: login Not tainted 4.2.0-rc7+ #1012
	 [...]
	 Call Trace:
	  [] dump_stack+0x41/0x52
	  [] panic+0x77/0x189
	  [] ? math_error+0x140/0x140
	  [] math_emulate+0xba7/0xbd0
	  [] ? fpu__copy+0x138/0x1c0
	  [] ? __alloc_pages_nodemask+0x12c/0x870
	  [] ? proc_clear_tty+0x40/0x70
	  [] ? session_clear_tty+0x1e/0x30
	  [] ? math_error+0x140/0x140
	  [] do_device_not_available+0x45/0x70
	  [] ? fpu__copy+0x138/0x1c0
	  [] error_code+0x5a/0x60
	  [] ? math_error+0x140/0x140
	  [] ? fpu__copy+0x138/0x1c0
	  [] arch_dup_task_struct+0x25/0x30
	  [] copy_process.part.51+0xea/0x1480
	  [] ? dput+0x175/0x200
	  [] ? no_tty+0x30/0x30
	  [] ? do_vfs_ioctl+0x322/0x540
	  [] _do_fork+0xca/0x340
	  [] ? SyS_rt_sigaction+0x66/0x90
	  [] SyS_clone+0x27/0x30
	  [] sysenter_do_call+0x12/0x12

The reason is the incorrect assumption in fpu_copy(), that FNSAVE
can be executed from math-emu kernels as well.

Don't try to copy the registers, the soft state will be copied
by fork anyway, so the child task inherits the parent task's
soft math state.

With this fix applied math-emu kernels boot up fine on modern
hardware and the 'no387 nofxsr' boot options.

Cc: Andy Lutomirski 
Cc: Bobby Powers 
Cc: Borislav Petkov 
Cc: Brian Gerst 
Cc: Dave Hansen 
Cc: Denys Vlasenko 
Cc: Fenghua Yu 
Cc: H. Peter Anvin 
Cc: Linus Torvalds 
Cc: Oleg Nesterov 
Cc: Peter Zijlstra 
Cc: Quentin Casasnovas 
Cc: Thomas Gleixner 
Signed-off-by: Ingo Molnar

x86/fpu/math-emu: Fix math-emu boot crash

2015-08-22T08:02:04+00:00

On a math-emu bootup the following crash occurs:

	Initializing CPU#0
	------------[ cut here ]------------
	kernel BUG at arch/x86/kernel/traps.c:779!
	invalid opcode: 0000 [#1] SMP
	[...]
	EIP is at do_device_not_available+0xe/0x70
	[...]
	Call Trace:
	 [] error_code+0x5a/0x60
	 [] ? math_error+0x140/0x140
	 [] ? fpu__init_cpu+0x59/0xa0
	 [] cpu_init+0x202/0x330
	 [] ? __native_set_fixmap+0x1f/0x30
	 [] trap_init+0x305/0x346
	 [] start_kernel+0x1a5/0x35d
	 [] i386_start_kernel+0x82/0x86

The reason is that in the following commit:

  b1276c48e91b ("x86/fpu: Initialize fpregs in fpu__init_cpu_generic()")

I failed to consider math-emu's limitation that it cannot execute the
FNINIT instruction in kernel mode.

The long term fix might be to allow math-emu to execute (certain) kernel
mode FPU instructions, but for now apply the safe (albeit somewhat ugly)
fix: initialize the emulation state explicitly without trapping out to
the FPU emulator.

Cc: Andy Lutomirski 
Cc: Borislav Petkov 
Cc: Brian Gerst 
Cc: Dave Hansen 
Cc: Denys Vlasenko 
Cc: Fenghua Yu 
Cc: H. Peter Anvin 
Cc: Linus Torvalds 
Cc: Oleg Nesterov 
Cc: Peter Zijlstra 
Cc: Quentin Casasnovas 
Cc: Thomas Gleixner 
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar

x86/fpu: Disable dependent CPU features on "noxsave"

2015-07-21T06:20:42+00:00

Complete the set of dependent features that need disabling at
once: XSAVEC, AVX-512 and all currently known to the kernel
extensions to it, as well as MPX need to be disabled too.

Signed-off-by: Jan Beulich 
Cc: Dave Hansen 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Link: http://lkml.kernel.org/r/55ACC40D0200007800092E6C@mail.emea.novell.com
Signed-off-by: Ingo Molnar

x86/fpu, sched: Introduce CONFIG_ARCH_WANTS_DYNAMIC_TASK_STRUCT and use it on x86

2015-07-18T01:42:51+00:00

Don't burden architectures without dynamic task_struct sizing
with the overhead of dynamic sizing.

Also optimize the x86 code a bit by caching task_struct_size.

Acked-and-Tested-by: Dave Hansen 
Cc: Andy Lutomirski 
Cc: Borislav Petkov 
Cc: Brian Gerst 
Cc: Dave Hansen 
Cc: Denys Vlasenko 
Cc: Linus Torvalds 
Cc: Oleg Nesterov 
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Link: http://lkml.kernel.org/r/1437128892-9831-3-git-send-email-mingo@kernel.org
Signed-off-by: Ingo Molnar

x86/fpu, sched: Dynamically allocate 'struct fpu'

2015-07-18T01:42:35+00:00

The FPU rewrite removed the dynamic allocations of 'struct fpu'.
But, this potentially wastes massive amounts of memory (2k per
task on systems that do not have AVX-512 for instance).

Instead of having a separate slab, this patch just appends the
space that we need to the 'task_struct' which we dynamically
allocate already.  This saves from doing an extra slab
allocation at fork().

The only real downside here is that we have to stick everything
and the end of the task_struct.  But, I think the
BUILD_BUG_ON()s I stuck in there should keep that from being too
fragile.

Signed-off-by: Dave Hansen 
Cc: Andy Lutomirski 
Cc: Borislav Petkov 
Cc: Brian Gerst 
Cc: Dave Hansen 
Cc: Denys Vlasenko 
Cc: H. Peter Anvin 
Cc: Linus Torvalds 
Cc: Oleg Nesterov 
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Link: http://lkml.kernel.org/r/1437128892-9831-2-git-send-email-mingo@kernel.org
Signed-off-by: Ingo Molnar

x86/fpu: Fix boot crash in the early FPU code

2015-07-04T08:05:56+00:00

Jan Kara and Thomas Gleixner reported boot crashes in the FPU
code:

  general protection fault: 0000 [#1] SMP
  RIP: 0010:[]  [] mxcsr_feature_mask_init+0x1c/0x40

  2b:*  0f ae 85 00 fe ff ff    fxsave -0x200(%rbp)

and bisected it down to the following FPU commit:

   91a8c2a5b43f ("x86/fpu: Clean up and fix MXCSR handling")

The reason is that the on-stack FPU registers state variable,
used by the FXSAVE instruction, did not have the required
minimum alignment of 16 bytes, causing the general protection
fault.

This is most likely a GCC bug in older GCC versions, but the
offending commit also added a bogus extra 32-byte alignment
(which GCC ignored too).

So fix this bug by making the variable static again, but also
mark it __initdata this time, because fpu__init_system_mxcsr()
is now an __init function.

Reported-and-bisected-by: Jan Kara 
Reported-bisected-and-tested-by: Thomas Gleixner 
Cc: Andy Lutomirski 
Cc: Borislav Petkov 
Cc: Brian Gerst 
Cc: Dave Hansen 
Cc: Denys Vlasenko 
Cc: Fenghua Yu 
Cc: H. Peter Anvin 
Cc: Jan Kara 
Cc: Linus Torvalds 
Cc: Oleg Nesterov 
Cc: Peter Zijlstra 
Cc: Quentin Casasnovas 
Cc: Thomas Gleixner 
Link: http://lkml.kernel.org/r/20150704075819.GA9201@gmail.com
Signed-off-by: Ingo Molnar

x86/fpu: Fix double-increment in setup_xstate_features()

2015-06-12T08:48:12+00:00

I noticed that my MPX tracepoints were producing garbage for the
lower and upper bounds:

	mpx_bounds_register_exception: address referenced: 0x00007fffffffccb7 bounds: lower: 0x0 ~upper: 0xffffffffffffffff
	mpx_bounds_register_exception: address referenced: 0x00007fffffffccbf bounds: lower: 0x0 ~upper: 0xffffffffffffffff

This is, of course, bogus because 0x00007fffffffccbf is *within*
the bounds.  I assumed that my instruction decoder was bad and
went looking at it.  But I eventually realized that I was
getting a '0' offset back from xstate_offsets[BNDREGS].

It was being skipped in the initialization, which is obviously
bogus, so remove the extra leaf++.

This also goes an initializes xstate_offsets/sizes[] to -1 so
so that bugs like this will oops instead of silently failing
in interesting ways.

This was introduced by:

	39f1acd ("x86/fpu/xstate: Don't assume the first zero xfeatures zero bit means the end")

Signed-off-by: Dave Hansen 
Cc: Andrew Morton 
Cc: Andy Lutomirski 
Cc: Borislav Petkov 
Cc: Brian Gerst 
Cc: Denys Vlasenko 
Cc: H. Peter Anvin 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Cc: dave@sr71.net
Link: http://lkml.kernel.org/r/20150611193400.2E0B00DB@viggo.jf.intel.com
Signed-off-by: Ingo Molnar

x86/fpu/xstate: Wrap get_xsave_addr() to make it safer

2015-06-09T10:24:29+00:00

The MPX code appears is calling a low-level FPU function
(copy_fpregs_to_fpstate()).  This function is not able to
be called in all contexts, although it is safe to call
directly in some cases.

Although probably correct, the current code is ugly and
potentially error-prone.  So, add a wrapper that calls
the (slightly) higher-level fpu__save() (which is preempt-
safe) and also ensures that we even *have* an FPU context
(in the case that this was called when in lazy FPU mode).

Ingo had this to say about the details about when we need
preemption disabled:

> it's indeed generally unsafe to access/copy FPU registers with preemption enabled,
> for two reasons:
>
>   - on older systems that use FSAVE the instruction destroys FPU register
>     contents, which has to be handled carefully
>
>   - even on newer systems if we copy to FPU registers (which this code doesn't)
>     then we don't want a context switch to occur in the middle of it, because a
>     context switch will write to the fpstate, potentially overwriting our new data
>     with old FPU state.
>
> But it's safe to access FPU registers with preemption enabled in a couple of
> special cases:
>
>   - potentially destructively saving FPU registers: the signal handling code does
>     this in copy_fpstate_to_sigframe(), because it can rely on the signal restore
>     side to restore the original FPU state.
>
>   - reading FPU registers on modern systems: we don't do this anywhere at the
>     moment, mostly to keep symmetry with older systems where FSAVE is
>     destructive.
>
>   - initializing FPU registers on modern systems: fpu__clear() does this. Here
>     it's safe because we don't copy from the fpstate.
>
>   - directly writing FPU registers from user-space memory (!). We do this in
>     fpu__restore_sig(), and it's safe because neither context switches nor
>     irq-handler FPU use can corrupt the source context of the copy (which is
>     user-space memory).
>
> Note that the MPX code's current use of copy_fpregs_to_fpstate() was safe I think,
> because:
>
>  - MPX is predicated on eagerfpu, so the destructive F[N]SAVE instruction won't be
>    used.
>
>  - the code was only reading FPU registers, and was doing it only in places that
>    guaranteed that an FPU state was already active (i.e. didn't do it in
>    kthreads)

Signed-off-by: Dave Hansen 
Reviewed-by: Thomas Gleixner 
Cc: Andrew Morton 
Cc: Andy Lutomirski 
Cc: Dave Hansen 
Cc: Fenghua Yu 
Cc: H. Peter Anvin 
Cc: Linus Torvalds 
Cc: Oleg Nesterov 
Cc: Peter Zijlstra 
Cc: Rik van Riel 
Cc: Suresh Siddha 
Cc: bp@alien8.de
Link: http://lkml.kernel.org/r/20150607183700.AA881696@viggo.jf.intel.com
Signed-off-by: Ingo Molnar