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: <159408711335.2385045.2567600405906448375.stgit@dwillia2-desk3.amr.corp.intel.com>
 <159408717289.2385045.14094866475168644020.stgit@dwillia2-desk3.amr.corp.intel.com>
 <20200709150051.GA17342@infradead.org> <20200709153854.GY23821@mellanox.com>
In-Reply-To: <20200709153854.GY23821@mellanox.com>
From:   "Rafael J. Wysocki" <rafael@kernel.org>
Date:   Thu, 9 Jul 2020 17:43:52 +0200
Message-ID: <CAJZ5v0hW79nEe1P5YqNx0o0jHAMuPw-i02GY3TCvG_uFu48ETg@mail.gmail.com>
Subject: Re: [PATCH v2 11/12] PM, libnvdimm: Add 'mem-quiet' state and
 callback for firmware activation
To:     Jason Gunthorpe <jgg@mellanox.com>
Cc:     Christoph Hellwig <hch@infradead.org>,
        Dan Williams <dan.j.williams@intel.com>,
        "linux-nvdimm@lists.01.org" <linux-nvdimm@lists.01.org>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        "Rafael J. Wysocki" <rafael@kernel.org>,
        Doug Ledford <dledford@redhat.com>,
        Pavel Machek <pavel@ucw.cz>, Len Brown <len.brown@intel.com>,
        ACPI Devel Maling List <linux-acpi@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Jul 9, 2020 at 5:39 PM Jason Gunthorpe <jgg@mellanox.com> wrote:
>
> On Thu, Jul 09, 2020 at 04:00:51PM +0100, Christoph Hellwig wrote:
> > On Mon, Jul 06, 2020 at 06:59:32PM -0700, Dan Williams wrote:
> > > The runtime firmware activation capability of Intel NVDIMM devices
> > > requires memory transactions to be disabled for 100s of microseconds.
> > > This timeout is large enough to cause in-flight DMA to fail and other
> > > application detectable timeouts. Arrange for firmware activation to be
> > > executed while the system is "quiesced", all processes and device-DMA
> > > frozen.
> > >
> > > It is already required that invoking device ->freeze() callbacks is
> > > sufficient to cease DMA. A device that continues memory writes outside
> > > of user-direction violates expectations of the PM core to be to
> > > establish a coherent hibernation image.
> > >
> > > That said, RDMA devices are an example of a device that access memory
> > > outside of user process direction.
>
> Are you saying freeze doesn't work for some RDMA drivers? That would
> be a driver bug, I think.
>
> The consequences of doing freeze are pretty serious, but it should
> still stop DMA.

Yes, it should.  The "freeze" callbacks are expected to prevent any
DMA transfers from being carried out after they have been executed.

Thanks!