Thread (2 messages) flat view 2 messages, 2 authors, 2020-04-20

Re: [RFC PATCH v1 40/50] arch/*/include/asm/stackprotector.h: Use get_random_canary() consistently

From: Will Deacon <will@kernel.org>
Date: 2020-04-20 20:42:47
Also in: linux-arm-kernel, linux-mips, linux-sh, lkml

On Tue, Dec 10, 2019 at 12:35:14AM -0500, George Spelvin wrote:
... in boot_init_stack_canary().

This is the archetypical example of where the extra security of
get_random_bytes() is wasted.  The canary is only important as
long as it's stored in __stack_chk_guard.

It's also a great example of code that has been copied around
a lot and not updated.

Remove the XOR with LINUX_VERSION_CODE as it's pointless; the inclusion
of utsname() in init_std_data in the random seeding obviates it.

The XOR with the TSC on x86 and mtfb() on powerPC were left in,
as I haven't proved them redundant yet.  For those, we call
get_random_long(), xor, and mask manually.

FUNCTIONAL CHANGE: mips and xtensa were changed from 64-bit
get_random_long() to 56-bit get_random_canary() to match the
others, in accordance with the logic in CANARY_MASK.

(We could do 1 bit better and zero *one* of the two high bytes.)

Signed-off-by: George Spelvin <redacted>
Cc: Russell King <linux@armlinux.org.uk>
Cc: linux-arm-kernel@lists.infradead.org
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Will Deacon <will@kernel.org>
Cc: Ralf Baechle <redacted>
Cc: Paul Burton <paulburton@kernel.org>
Cc: James Hogan <jhogan@kernel.org>
Cc: linux-mips@vger.kernel.org
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <redacted>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: linuxppc-dev@lists.ozlabs.org
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Rich Felker <dalias@libc.org>
Cc: linux-sh@vger.kernel.org
Cc: Thomas Gleixner <redacted>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc:  "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Cc: Chris Zankel <chris@zankel.net>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: linux-xtensa@linux-xtensa.org
---
 arch/arm/include/asm/stackprotector.h     | 9 +++------
 arch/arm64/include/asm/stackprotector.h   | 8 ++------
 arch/mips/include/asm/stackprotector.h    | 7 ++-----
 arch/powerpc/include/asm/stackprotector.h | 6 ++----
 arch/sh/include/asm/stackprotector.h      | 8 ++------
 arch/x86/include/asm/stackprotector.h     | 4 ++--
 arch/xtensa/include/asm/stackprotector.h  | 7 ++-----
 7 files changed, 15 insertions(+), 34 deletions(-)
Just found this kicking around in the depths of my inbox. Is the series
dead?

Will
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help