]> git.ipfire.org Git - thirdparty/kernel/stable.git/commit
powerpc/kprobe: Fix oops when kprobed on 'stdu' instruction
authorRavi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Tue, 11 Apr 2017 05:08:13 +0000 (10:38 +0530)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Thu, 27 Apr 2017 07:12:59 +0000 (09:12 +0200)
commit1136723a6cf0f5e0ca21c89b8df4394c7759a160
treeaa0aaf85fe00ec8562a6790f4043a91f2eed7a3a
parenta6db433483db4013aead6e3bbdf5b5ebe99be72d
powerpc/kprobe: Fix oops when kprobed on 'stdu' instruction

commit 9e1ba4f27f018742a1aa95d11e35106feba08ec1 upstream.

If we set a kprobe on a 'stdu' instruction on powerpc64, we see a kernel
OOPS:

  Bad kernel stack pointer cd93c840 at c000000000009868
  Oops: Bad kernel stack pointer, sig: 6 [#1]
  ...
  GPR00: c000001fcd93cb30 00000000cd93c840 c0000000015c5e00 00000000cd93c840
  ...
  NIP [c000000000009868] resume_kernel+0x2c/0x58
  LR [c000000000006208] program_check_common+0x108/0x180

On a 64-bit system when the user probes on a 'stdu' instruction, the kernel does
not emulate actual store in emulate_step() because it may corrupt the exception
frame. So the kernel does the actual store operation in exception return code
i.e. resume_kernel().

resume_kernel() loads the saved stack pointer from memory using lwz, which only
loads the low 32-bits of the address, causing the kernel crash.

Fix this by loading the 64-bit value instead.

Fixes: be96f63375a1 ("powerpc: Split out instruction analysis part of emulate_step()")
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Reviewed-by: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Reviewed-by: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com>
[mpe: Change log massage, add stable tag]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
arch/powerpc/kernel/entry_64.S