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: <YRo58c+JGOvec7tc@elver.google.com> <20210816145945.GB121345@rowland.harvard.edu>
 <YRqfJz/lpUaZpxq7@elver.google.com> <20210816192109.GC121345@rowland.harvard.edu>
 <20210816205057.GN4126399@paulmck-ThinkPad-P17-Gen-1> <20210817122816.GA12746@willie-the-truck>
In-Reply-To: <20210817122816.GA12746@willie-the-truck>
From:   Marco Elver <elver@google.com>
Date:   Tue, 17 Aug 2021 15:27:23 +0200
Message-ID: <CANpmjNMZxiyFbua2ck=0u7cJnHhtG4RY=Db=ry3COEED_sm7Xw@mail.gmail.com>
Subject: Re: LKMM: Read dependencies of writes ordered by dma_wmb()?
To:     Will Deacon <will@kernel.org>
Cc:     "Paul E. McKenney" <paulmck@kernel.org>,
        Alan Stern <stern@rowland.harvard.edu>,
        Boqun Feng <boqun.feng@gmail.com>,
        Andrea Parri <parri.andrea@gmail.com>,
        Mark Rutland <mark.rutland@arm.com>,
        Dmitry Vyukov <dvyukov@google.com>, kasan-dev@googlegroups.com,
        linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, 17 Aug 2021 at 14:28, Will Deacon <will@kernel.org> wrote:
> Just on this bit...
>
> On Mon, Aug 16, 2021 at 01:50:57PM -0700, Paul E. McKenney wrote:
> > 5.    The dma_mb(), dma_rmb(), and dma_wmb() appear to be specific
> >       to ARMv8.
>
> These are useful on other architectures too! IIRC, they were added by x86 in
> the first place. They're designed to be used with dma_alloc_coherent()
> allocations where you're sharing something like a ring buffer with a device
> and they guarantee accesses won't be reordered before they become visible
> to the device. They _also_ provide the same ordering to other CPUs.

Ah, good you pointed it out again. Re-reading memory-barriers.txt and
it does also say these provide order for other CPUs...

> I gave a talk at LPC about some of this, which might help (or might make
> things worse...):
>
> https://www.youtube.com/watch?v=i6DayghhA8Q

Nice, thank you!

> Ignore the bits about mmiowb() as we got rid of that.
>
> Will