Wolfgang Walter reported this oops on his via C3 using padlock for
AES-encryption:
##################################################################
BUG: unable to handle kernel NULL pointer dereference at 000001f0
IP: [<c01028c5>] __switch_to+0x30/0x117
*pde = 00000000
Oops: 0002 [#1] PREEMPT
Modules linked in:
Pid: 2071, comm: sleep Not tainted (2.6.26 #11)
EIP: 0060:[<c01028c5>] EFLAGS: 00010002 CPU: 0
EIP is at __switch_to+0x30/0x117
EAX: 00000000 EBX: c0493300 ECX: dc48dd00 EDX: c0493300
ESI: dc48dd00 EDI: c0493530 EBP: c04cff8c ESP: c04cff7c
DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 0068
Process sleep (pid: 2071, ti=c04ce000 task=dc48dd00 task.ti=d2fe6000)
Stack: dc48df30 c0493300 00000000 00000000 d2fe7f44 c03b5b43 c04cffc8 00000046
c0131856 0000005a dc472d3c c0493300 c0493470 d983ae00 00002696 00000000
c0239f54 00000000 c04c4000 c04cffd8 c01025fe c04f3740 00049800 c04cffe0
Call Trace:
[<c03b5b43>] ? schedule+0x285/0x2ff
[<c0131856>] ? pm_qos_requirement+0x3c/0x53
[<c0239f54>] ? acpi_processor_idle+0x0/0x434
[<c01025fe>] ? cpu_idle+0x73/0x7f
[<c03a4dcd>] ? rest_init+0x61/0x63
=======================
Wolfgang also found out that adding kernel_fpu_begin() and kernel_fpu_end()
around the padlock instructions fix the oops.
Suresh wrote:
These padlock instructions though don't use/touch SSE registers, but it behaves
similar to other SSE instructions. For example, it might cause DNA faults
when cr0.ts is set. While this is a spurious DNA trap, it might cause
oops with the recent fpu code changes.
This is the code sequence that is probably causing this problem:
a) new app is getting exec'd and it is somewhere in between
start_thread() and flush_old_exec() in the load_xyz_binary()
b) At pont "a", task's fpu state (like TS_USEDFPU, used_math() etc) is
cleared.
c) Now we get an interrupt/softirq which starts using these encrypt/decrypt
routines in the network stack. This generates a math fault (as
cr0.ts is '1') which sets TS_USEDFPU and restores the math that is
in the task's xstate.
d) Return to exec code path, which does start_thread() which does
free_thread_xstate() and sets xstate pointer to NULL while
the TS_USEDFPU is still set.
e) At the next context switch from the new exec'd task to another task,
we have a scenarios where TS_USEDFPU is set but xstate pointer is null.
This can cause an oops during unlazy_fpu() in __switch_to()
Now:
1) This should happen with or with out pre-emption. Viro also encountered
similar problem with out CONFIG_PREEMPT.
2) kernel_fpu_begin() and kernel_fpu_end() will fix this problem, because
kernel_fpu_begin() will manually do a clts() and won't run in to the
situation of setting TS_USEDFPU in step "c" above.
3) This was working before the fpu changes, because its a spurious
math fault which doesn't corrupt any fpu/sse registers and the task's
math state was always in an allocated state.
With out the recent lazy fpu allocation changes, while we don't see oops,
there is a possible race still present in older kernels(for example,
while kernel is using kernel_fpu_begin() in some optimized clear/copy
page and an interrupt/softirq happens which uses these padlock
instructions generating DNA fault).
This is the failing scenario that existed even before the lazy fpu allocation
changes:
0. CPU's TS flag is set
1. kernel using FPU in some optimized copy routine and while doing
kernel_fpu_begin() takes an interrupt just before doing clts()
2. Takes an interrupt and ipsec uses padlock instruction. And we
take a DNA fault as TS flag is still set.
3. We handle the DNA fault and set TS_USEDFPU and clear cr0.ts
4. We complete the padlock routine
5. Go back to step-1, which resumes clts() in kernel_fpu_begin(), finishes
the optimized copy routine and does kernel_fpu_end(). At this point,
we have cr0.ts again set to '1' but the task's TS_USEFPU is stilll
set and not cleared.
6. Now kernel resumes its user operation. And at the next context
switch, kernel sees it has do a FP save as TS_USEDFPU is still set
and then will do a unlazy_fpu() in __switch_to(). unlazy_fpu()
will take a DNA fault, as cr0.ts is '1' and now, because we are
in __switch_to(), math_state_restore() will get confused and will
restore the next task's FP state and will save it in prev tasks's FP state.
Remember, in __switch_to() we are already on the stack of the next task
but take a DNA fault for the prev task.
This causes the fpu leakage.
Fix the padlock instruction usage by calling them inside the
context of new routines irq_ts_save/restore(), which clear/restore cr0.ts
manually in the interrupt context. This will not generate spurious DNA
in the context of the interrupt which will fix the oops encountered and
the possible FPU leakage issue.
Reported-and-bisected-by: Wolfgang Walter <wolfgang.walter@stwm.de>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
On Thu, Jul 24, 2008 at 03:43:44PM -0700, Linus Torvalds wrote:
> So how about this patch as a starting point? This is the RightThing(tm) to
> do regardless, and if it then makes it easier to do some other cleanups,
> we should do it first. What do you think?
restore_fpu_checking() calls init_fpu() in error conditions.
While this is wrong(as our main intention is to clear the fpu state of
the thread), this was benign before commit 92d140e21f ("x86: fix taking
DNA during 64bit sigreturn").
Post commit 92d140e21f, live FPU registers may not belong to this
process at this error scenario.
In the error condition for restore_fpu_checking() (especially during the
64bit signal return), we are doing init_fpu(), which saves the live FPU
register state (possibly belonging to some other process context) into
the thread struct (through unlazy_fpu() in init_fpu()). This is wrong
and can leak the FPU data.
For the signal handler restore error condition in restore_i387(), clear
the fpu state present in the thread struct(before ultimately sending a
SIGSEGV for badframe).
For the paranoid error condition check in math_state_restore(), send a
SIGSEGV, if we fail to restore the state.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: <stable@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Suresh Siddha wants to fix a possible FPU leakage in error conditions,
but the fact that save/restore_i387() are inlines in a header file makes
that harder to do than necessary. So start off with an obvious cleanup.
This just moves the x86-64 version of save/restore_i387() out of the
header file, and moves it to the only file that it is actually used in:
arch/x86/kernel/signal_64.c. So exposing it in a header file was wrong
to begin with.
[ Side note: I'd like to fix up some of the games we play with the
32-bit version of these functions too, but that's a separate
matter. The 32-bit versions are shared - under different names
at that! - by both the native x86-32 code and the x86-64 32-bit
compatibility code ]
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix the math emulation that got broken with the recent lazy allocation of FPU
area. init_fpu() need to be added for the math-emulation path aswell
for the FPU area allocation.
math emulation enabled kernel booted fine with this, in the presence
of "no387 nofxsr" boot param.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: hpa@zytor.com
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
If the task never used fpu, initialize the fpu before restoring the FP
state from the signal handler context. This will allocate the fpu
state, if the task never needed it before.
Reported-and-bisected-by: Eric Sesterhenn <snakebyte@gmx.de>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Tested-by: Eric Sesterhenn <snakebyte@gmx.de>
Cc: Frederik Deweerdt <deweerdt@free.fr>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Only allocate the FPU area when the application actually uses FPU, i.e., in the
first lazy FPU trap. This could save memory for non-fpu using apps.
for example: on my system after boot, there are around 300 processes, with
only 17 using FPU.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Split the FPU save area from the task struct. This allows easy migration
of FPU context, and it's generally cleaner. It also allows the following
two optimizations:
1) only allocate when the application actually uses FPU, so in the first
lazy FPU trap. This could save memory for non-fpu using apps. Next patch
does this lazy allocation.
2) allocate the right size for the actual cpu rather than 512 bytes always.
Patches enabling xsave/xrstor support (coming shortly) will take advantage
of this.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Use the _ASM_EXTABLE macro from <asm/asm.h>, instead of open-coding
__ex_table entires in include/asm-x86/i387.h.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This removes a bunch of dead code that is no longer needed now
that the user_regset interfaces are being used for all these jobs.
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This removes all the old code that is no longer used after
the i387 unification and cleanup. The i387_64.h is renamed
to i387.h with no changes, but since it replaces the nonempty
one-line stub i387.h it looks like a big diff and not a rename.
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This revamps the i387 code to be shared across 32-bit, 64-bit,
and 32-on-64. It does so by consolidating the code in one place
based on the user_regset accessor interfaces. This switches
32-bit to using the i387_64.h header and 64-bit to using the
i387.c that was previously i387_32.c, but that's what took the
least cleanup in each file. Here i387.h is stubbed to always
include i387_64.h rather than renaming the file, to keep this
diff smaller and easier to read.
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Move the headers to include/asm-x86 and fixup the
header install make rules
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Suresh Siddha