Re: [PATCH] x86/entry/64: randomize kernel stack offset upon syscall

From: Andy Lutomirski <luto@kernel.org>
To: Ingo Molnar <mingo@kernel.org>
Cc: "Reshetova, Elena" <elena.reshetova@intel.com>,
	David Laight <David.Laight@aculab.com>,
	Andy Lutomirski <luto@kernel.org>,
	"Theodore Ts'o" <tytso@mit.edu>,
	Eric Biggers <ebiggers3@gmail.com>,
	"ebiggers@google.com" <ebiggers@google.com>,
	"herbert@gondor.apana.org.au" <herbert@gondor.apana.org.au>,
	Peter Zijlstra <peterz@infradead.org>,
	"keescook@chromium.org" <keescook@chromium.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"jpoimboe@redhat.com" <jpoimboe@redhat.com>,
	"jannh@google.com" <jannh@google.com>,
	"Perla, Enrico" <enrico.perla@intel.com>,
	"mingo@redhat.com" <mingo@redhat.com>,
	"bp@alien8.de" <bp@alien8.de>,
	"tglx@linutronix.de" <tglx@linutronix.de>,
	"gregkh@linuxfoundation.org" <gregkh@linuxfoundation.org>,
	"Edgecombe, Rick P" <rick.p.edgecombe@intel.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: Re: [PATCH] x86/entry/64: randomize kernel stack offset upon syscall
Date: Sat, 11 May 2019 15:45:19 -0700	[thread overview]
Message-ID: <CALCETrV1067Es=KEjkz=CtdoT79a2EJg4dJDae6oGDiTaubL1A@mail.gmail.com> (raw)
In-Reply-To: <20190509084352.GA96236@gmail.com>

On Thu, May 9, 2019 at 1:43 AM Ingo Molnar <mingo@kernel.org> wrote:
>
>
> * Reshetova, Elena <elena.reshetova@intel.com> wrote:
>
> > > I find it ridiculous that even with 4K blocked get_random_bytes(),
> > > which gives us 32k bits, which with 5 bits should amortize the RNG
> > > call to something like "once per 6553 calls", we still see 17%
> > > overhead? It's either a measurement artifact, or something doesn't
> > > compute.
> >
> > If you check what happens underneath of get_random_bytes(), there is a
> > fair amount of stuff that is going on, including reseeding CRNG if
> > reseeding interval has passed (see _extract_crng()). It also even
> > attempts to stir in more entropy from rdrand if avalaible:
> >
> > I will look into this whole construction slowly now to investigate. I
> > did't optimize anything yet also (I take 8 bits at the time for
> > offset), but these small optimization won't make performance impact
> > from 17% --> 2%, so pointless for now, need a more radical shift.
>
> So assuming that the 17% overhead primarily comes from get_random_bytes()
> (does it? I don't know), that's incredibly slow for something like the
> system call entry path, even if it's batched.
>

ISTM maybe a better first step would be to make get_random_bytes() be
much faster? :)

--Andy