linux-kernel - Re: [PATCH RFC 01/10] mm: add Kernel Electric-Fence infrastructure

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]

Message-ID: <CANpmjNN7qAtnUmibwGJEnxd+UcjBM1WeocoLeW0SO24NW3SkVA@mail.gmail.com>
Date:   Thu, 10 Sep 2020 17:06:25 +0200
From:   Marco Elver <elver@...gle.com>
To:     Dmitry Vyukov <dvyukov@...gle.com>
Cc:     Alexander Potapenko <glider@...gle.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Catalin Marinas <catalin.marinas@....com>,
        Christoph Lameter <cl@...ux.com>,
        David Rientjes <rientjes@...gle.com>,
        Joonsoo Kim <iamjoonsoo.kim@....com>,
        Mark Rutland <mark.rutland@....com>,
        Pekka Enberg <penberg@...nel.org>,
        "H. Peter Anvin" <hpa@...or.com>,
        "Paul E. McKenney" <paulmck@...nel.org>,
        Andrey Konovalov <andreyknvl@...gle.com>,
        Andrey Ryabinin <aryabinin@...tuozzo.com>,
        Andy Lutomirski <luto@...nel.org>,
        Borislav Petkov <bp@...en8.de>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Eric Dumazet <edumazet@...gle.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Ingo Molnar <mingo@...hat.com>, Jann Horn <jannh@...gle.com>,
        Jonathan Corbet <corbet@....net>,
        Kees Cook <keescook@...omium.org>,
        Peter Zijlstra <peterz@...radead.org>, Qian Cai <cai@....pw>,
        Thomas Gleixner <tglx@...utronix.de>,
        Will Deacon <will@...nel.org>,
        "the arch/x86 maintainers" <x86@...nel.org>,
        "open list:DOCUMENTATION" <linux-doc@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        kasan-dev <kasan-dev@...glegroups.com>,
        Linux ARM <linux-arm-kernel@...ts.infradead.org>,
        Linux-MM <linux-mm@...ck.org>
Subject: Re: [PATCH RFC 01/10] mm: add Kernel Electric-Fence infrastructure

On Thu, 10 Sep 2020 at 16:58, Dmitry Vyukov <dvyukov@...gle.com> wrote:
>
> On Mon, Sep 7, 2020 at 3:41 PM Marco Elver <elver@...gle.com> wrote:
> > +config KFENCE_NUM_OBJECTS
> > +       int "Number of guarded objects available"
> > +       default 255
> > +       range 1 65535
> > +       help
> > +         The number of guarded objects available. For each KFENCE object, 2
> > +         pages are required; with one containing the object and two adjacent
> > +         ones used as guard pages.
>
> Hi Marco,
>
> Wonder if you tested build/boot with KFENCE_NUM_OBJECTS=65535? Can a
> compiler create such a large object?

Indeed, I get a "ld: kernel image bigger than KERNEL_IMAGE_SIZE".
Let's lower it to something more reasonable.

The main reason to have the limit is to constrain random configs and
avoid the inevitable error reports.

> > +config KFENCE_FAULT_INJECTION
> > +       int "Fault injection for stress testing"
> > +       default 0
> > +       depends on EXPERT
> > +       help
> > +         The inverse probability with which to randomly protect KFENCE object
> > +         pages, resulting in spurious use-after-frees. The main purpose of
> > +         this option is to stress-test KFENCE with concurrent error reports
> > +         and allocations/frees. A value of 0 disables fault injection.
>
> I would name this differently. "FAULT_INJECTION" is already taken for
> a different thing, so it's a bit confusing.
> KFENCE_DEBUG_SOMETHING may be a better name.
> It would also be good to make it very clear in the short description
> that this is for testing of KFENCE itself. When I configure syzbot I
> routinely can't figure out if various DEBUG configs detect user
> errors, or enable additional unit tests, or something else.

Makes sense, we'll change the name.

> Maybe it should depend on DEBUG_KERNEL as well?

EXPERT selects DEBUG_KERNEL, so depending on DEBUG_KERNEL doesn't make sense.

> > +/*
> > + * Get the canary byte pattern for @addr. Use a pattern that varies based on the
> > + * lower 3 bits of the address, to detect memory corruptions with higher
> > + * probability, where similar constants are used.
> > + */
> > +#define KFENCE_CANARY_PATTERN(addr) ((u8)0xaa ^ (u8)((unsigned long)addr & 0x7))
>
> (addr) in macro body

Done for v2.

> > +       seq_con_printf(seq,
> > +                      "kfence-#%zd [0x" PTR_FMT "-0x" PTR_FMT
>
> PTR_FMT is only used in this file, should it be declared in report.c?

It's also used by the test.

> Please post example reports somewhere. It's hard to figure out all
> details of the reporting/formatting.

They can be seen in Documentation added later in the series (also
viewable here: https://github.com/google/kasan/blob/kfence/Documentation/dev-tools/kfence.rst)

Thank you!

-- Marco