linux-kernel - RE: [PATCH] m68k: Implement "current_stack

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 24 Feb 2022 09:12:51 +0000
From:   David Laight <David.Laight@...LAB.COM>
To:     'Geert Uytterhoeven' <geert@...ux-m68k.org>,
        Kees Cook <keescook@...omium.org>
CC:     linux-m68k <linux-m68k@...ts.linux-m68k.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        "linux-hardening@...r.kernel.org" <linux-hardening@...r.kernel.org>
Subject: RE: [PATCH] m68k: Implement "current_stack_pointer"

From: Geert Uytterhoeven
> Sent: 24 February 2022 08:59
...
> > +register unsigned long current_stack_pointer __asm__("sp");
> 
> I don't know what HARDENED_USERCOPY does, so I don't know if you need
> "usp" (see rdusp()) or "sp"?

HARDENED_USERCOPY significantly slows down some systems calls
(especially things like sendmsg()) by trying to run-time verify
that the kernel buffer doesn't overrun a stack frame or kmalloc()ed
buffer.

I've got measurable improvements by either using __copy_to/from_user()
(which skips the tests) or user_access_begin() and __get_user().

At the moment the code for reading a compat iovec[] is actually
faster than that for a native one!

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)