[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAPcyv4iYVcCZxdd5wHDf2av8F=E1F7UGTX9uVzX68KkrJ1ZuQQ@mail.gmail.com>
Date: Wed, 17 Jan 2018 11:54:12 -0800
From: Dan Williams <dan.j.williams@...el.com>
To: Al Viro <viro@...iv.linux.org.uk>
Cc: Alan Cox <alan@...ux.intel.com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
linux-arch@...r.kernel.org, Andi Kleen <ak@...ux.intel.com>,
Kees Cook <keescook@...omium.org>,
kernel-hardening@...ts.openwall.com,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
"the arch/x86 maintainers" <x86@...nel.org>,
Ingo Molnar <mingo@...hat.com>,
"H. Peter Anvin" <hpa@...or.com>,
Thomas Gleixner <tglx@...utronix.de>,
Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH v3 8/9] x86: use __uaccess_begin_nospec and ASM_IFENCE in
get_user paths
On Wed, Jan 17, 2018 at 10:52 AM, Al Viro <viro@...iv.linux.org.uk> wrote:
> On Wed, Jan 17, 2018 at 02:17:26PM +0000, Alan Cox wrote:
[..]
> Incidentally, what about copy_to_iter() and friends? They
> check iov_iter flavour and go either into the "copy to kernel buffer"
> or "copy to userland" paths. Do we need to deal with mispredictions
> there? We are calling a bunch of those on read()...
>
Those should be protected by the conversion of __uaccess_begin to
__uaccess_begin_nospec that includes the lfence.
Powered by blists - more mailing lists