This is a note to let you know that I've just added the patch titled
MIPS: Guard against any partial write attempt with PTRACE_SETREGSET
to the 3.18-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
mips-guard-against-any-partial-write-attempt-with-ptrace_setregset.patch
and it can be found in the queue-3.18 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
>From dc24d0edf33c3e15099688b6bbdf7bdc24bf6e91 Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@xxxxxxxx>
Date: Mon, 11 Dec 2017 22:52:15 +0000
Subject: MIPS: Guard against any partial write attempt with PTRACE_SETREGSET
From: Maciej W. Rozycki <macro@xxxxxxxx>
commit dc24d0edf33c3e15099688b6bbdf7bdc24bf6e91 upstream.
Complement commit d614fd58a283 ("mips/ptrace: Preserve previous
registers for short regset write") and ensure that no partial register
write attempt is made with PTRACE_SETREGSET, as we do not preinitialize
any temporaries used to hold incoming register data and consequently
random data could be written.
It is the responsibility of the caller, such as `ptrace_regset', to
arrange for writes to span whole registers only, so here we only assert
that it has indeed happened.
Signed-off-by: Maciej W. Rozycki <macro@xxxxxxxx>
Fixes: 72b22bbad1e7 ("MIPS: Don't assume 64-bit FP registers for FP regset")
Cc: James Hogan <james.hogan@xxxxxxxx>
Cc: Paul Burton <Paul.Burton@xxxxxxxx>
Cc: Alex Smith <alex@xxxxxxxxxxxxxxxx>
Cc: Dave Martin <Dave.Martin@xxxxxxx>
Cc: linux-mips@xxxxxxxxxxxxxx
Cc: linux-kernel@xxxxxxxxxxxxxxx
Patchwork: https://patchwork.linux-mips.org/patch/17926/
Signed-off-by: Ralf Baechle <ralf@xxxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
arch/mips/kernel/ptrace.c | 12 +++++++++++-
1 file changed, 11 insertions(+), 1 deletion(-)
--- a/arch/mips/kernel/ptrace.c
+++ b/arch/mips/kernel/ptrace.c
@@ -497,7 +497,15 @@ static int fpr_set_msa(struct task_struc
return 0;
}
-/* Copy the supplied NT_PRFPREG buffer to the floating-point context. */
+/*
+ * Copy the supplied NT_PRFPREG buffer to the floating-point context.
+ *
+ * We optimize for the case where `count % sizeof(elf_fpreg_t) == 0',
+ * which is supposed to have been guaranteed by the kernel before
+ * calling us, e.g. in `ptrace_regset'. We enforce that requirement,
+ * so that we can safely avoid preinitializing temporaries for
+ * partial register writes.
+ */
static int fpr_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
@@ -505,6 +513,8 @@ static int fpr_set(struct task_struct *t
{
int err;
+ BUG_ON(count % sizeof(elf_fpreg_t));
+
/* XXX fcr31 */
if (sizeof(target->thread.fpu.fpr[0]) == sizeof(elf_fpreg_t))
Patches currently in stable-queue which might be from macro@xxxxxxxx are
queue-3.18/mips-consistently-handle-buffer-counter-with-ptrace_setregset.patch
queue-3.18/mips-disallow-outsized-ptrace_setregset-nt_prfpreg-regset-accesses.patch
queue-3.18/mips-guard-against-any-partial-write-attempt-with-ptrace_setregset.patch
queue-3.18/mips-also-verify-sizeof-elf_fpreg_t-with-ptrace_setregset.patch
queue-3.18/mips-factor-out-nt_prfpreg-regset-access-helpers.patch
queue-3.18/mips-fix-an-fcsr-access-api-regression-with-nt_prfpreg-and-msa.patch