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: <000001d5824d$c8b2a060$5a17e120$@codeaurora.org>
 <CACT4Y+aAicvQ1FYyOVbhJy62F4U6R_PXr+myNghFh8PZixfYLQ@mail.gmail.com>
 <CANpmjNOx7fuLLBasdEgnOCJepeufY4zo_FijsoSg0hfVgN7Ong@mail.gmail.com>
 <002801d58271$f5d01db0$e1705910$@codeaurora.org> <CANpmjNPVK00wsrpcVPFjudpqE-4-AVnZY0Pk-WMXTtqZTMXoOw@mail.gmail.com>
In-Reply-To: <CANpmjNPVK00wsrpcVPFjudpqE-4-AVnZY0Pk-WMXTtqZTMXoOw@mail.gmail.com>
From:   Marco Elver <elver@google.com>
Date:   Mon, 15 Jun 2020 14:30:20 +0200
Message-ID: <CANpmjNM9RhZ_V7vPBLp146m_JRqajeHgRT3h3gSBz3OH4Ya_Yg@mail.gmail.com>
Subject: Re: KCSAN Support on ARM64 Kernel
To:     sgrover@codeaurora.org
Cc:     Dmitry Vyukov <dvyukov@google.com>,
        kasan-dev <kasan-dev@googlegroups.com>,
        LKML <linux-kernel@vger.kernel.org>,
        "Paul E. McKenney" <paulmck@linux.ibm.com>,
        Andrea Parri <parri.andrea@gmail.com>,
        Alan Stern <stern@rowland.harvard.edu>,
        Mark Rutland <mark.rutland@arm.com>,
        Will Deacon <will@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, 14 Oct 2019 at 11:31, Marco Elver <elver@google.com> wrote:
> My plan was to send patches upstream within the month.
[...]
> On Mon, 14 Oct 2019 at 11:30, <sgrover@codeaurora.org> wrote:
[...]
> > When can we expect upstream of KCSAN on kernel mainline. Any timeline?
[...]
> > > > Can you please tell me if KCSAN is supported on ARM64 now? Can I ju=
st rebase the KCSAN branch on top of our let=E2=80=99s say android mainline=
 kernel, enable the config and run syzkaller on that for finding race condi=
tions?
[...]
> > KCSAN does not yet have ARM64 support. Once it's upstream, I would expe=
ct that Mark's patches (from repo linked in LKML thread) will just cleanly =
apply to enable ARM64 support.

Just FYI, KCSAN is in mainline now. I believe porting it to other
architectures has also become much simpler due to its reworked
ONCE/atomic support relying on proper compiler instrumentation instead
of other tricks.

Thanks,
-- Marco