Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20200929133814.2834621-1-elver@google.com> <20200929133814.2834621-2-elver@google.com>
 <CAG48ez3+_K6YXoXgKBkB8AMeSQj++Mxi5u2OT--B+mJgE7Cyfg@mail.gmail.com>
 <20201002171959.GA986344@elver.google.com> <CAG48ez0D1+hStZaDOigwbqNqFHJAJtXK+8Nadeuiu1Byv+xp5A@mail.gmail.com>
 <CANpmjNN7s3o4DYbP64iLYo0MeDWciQnKd61njJKLsiZv+ZLQdA@mail.gmail.com>
In-Reply-To: <CANpmjNN7s3o4DYbP64iLYo0MeDWciQnKd61njJKLsiZv+ZLQdA@mail.gmail.com>
From:   Jann Horn <jannh@google.com>
Date:   Sat, 3 Oct 2020 00:27:16 +0200
Message-ID: <CAG48ez0az-Mv1f6EpnQwO6cYQANwx4qCDLa+yda_i15AzciS1Q@mail.gmail.com>
Subject: Re: [PATCH v4 01/11] mm: add Kernel Electric-Fence infrastructure
To:     Marco Elver <elver@google.com>
Cc:     Andrew Morton <akpm@linux-foundation.org>,
        Alexander Potapenko <glider@google.com>,
        "H . Peter Anvin" <hpa@zytor.com>,
        "Paul E . McKenney" <paulmck@kernel.org>,
        Andrey Konovalov <andreyknvl@google.com>,
        Andrey Ryabinin <aryabinin@virtuozzo.com>,
        Andy Lutomirski <luto@kernel.org>,
        Borislav Petkov <bp@alien8.de>,
        Catalin Marinas <catalin.marinas@arm.com>,
        Christoph Lameter <cl@linux.com>,
        Dave Hansen <dave.hansen@linux.intel.com>,
        David Rientjes <rientjes@google.com>,
        Dmitry Vyukov <dvyukov@google.com>,
        Eric Dumazet <edumazet@google.com>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Hillf Danton <hdanton@sina.com>,
        Ingo Molnar <mingo@redhat.com>,
        Jonathan Cameron <Jonathan.Cameron@huawei.com>,
        Jonathan Corbet <corbet@lwn.net>,
        Joonsoo Kim <iamjoonsoo.kim@lge.com>,
        Kees Cook <keescook@chromium.org>,
        Mark Rutland <mark.rutland@arm.com>,
        Pekka Enberg <penberg@kernel.org>,
        Peter Zijlstra <peterz@infradead.org>,
        SeongJae Park <sjpark@amazon.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Vlastimil Babka <vbabka@suse.cz>,
        Will Deacon <will@kernel.org>,
        "the arch/x86 maintainers" <x86@kernel.org>,
        "open list:DOCUMENTATION" <linux-doc@vger.kernel.org>,
        kernel list <linux-kernel@vger.kernel.org>,
        kasan-dev <kasan-dev@googlegroups.com>,
        Linux ARM <linux-arm-kernel@lists.infradead.org>,
        Linux-MM <linux-mm@kvack.org>, SeongJae Park <sjpark@amazon.de>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Fri, Oct 2, 2020 at 11:28 PM Marco Elver <elver@google.com> wrote:
> On Fri, 2 Oct 2020 at 21:32, Jann Horn <jannh@google.com> wrote:
> > > That's another check; we don't want to make this more expensive.
> >
> > Ah, right, I missed that this is the one piece of KFENCE that is
> > actually really hot code until Dmitry pointed that out.
> >
> > But actually, can't you reduce how hot this is for SLUB by moving
> > is_kfence_address() down into the freeing slowpath? At the moment you
> > use it in slab_free_freelist_hook(), which is in the super-hot
> > fastpath, but you should be able to at least move it down into
> > __slab_free()...
> >
> > Actually, you already have hooked into __slab_free(), so can't you
> > just get rid of the check in the slab_free_freelist_hook()?
>
> I missed this bit: the loop that follows wants the free pointer, so I
> currently see how this might work. :-/

reverse call graph:
__slab_free
  do_slab_free
    slab_free
      kmem_cache_free (frees a single non-kmalloc allocation)
      kmem_cache_free_bulk (frees multiple)
      kfree (frees a single kmalloc allocation)
    ___cache_free (frees a single allocation for KASAN)

So the only path for which we can actually loop in __slab_free() is
kmem_cache_free_bulk(); and you've already changed
build_detached_freelist() (which is used by kmem_cache_free_bulk() to
group objects from the same page) to consume KFENCE allocations before
they can ever reach __slab_free(). So we know that if we've reached
__slab_free(), then we are being called with either a single object
(which may be a KFENCE object) or with a list of objects that all
belong to the same page and don't contain any KFENCE allocations.