Re: [PATCH 1/2] remoteproc: Add userspace char device driver

From: rishabhb@codeaurora.org
To: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Bjorn Andersson <bjorn.andersson@linaro.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	linux-remoteproc <linux-remoteproc@vger.kernel.org>,
	Ohad Ben-Cohen <ohad@wizery.com>,
	psodagud@codeaurora.org, tsoni@codeaurora.org,
	Siddharth Gupta <sidgup@codeaurora.org>,
	linux-remoteproc-owner@vger.kernel.org
Subject: Re: [PATCH 1/2] remoteproc: Add userspace char device driver
Date: Tue, 31 Mar 2020 10:38:24 -0700	[thread overview]
Message-ID: <30f8b41df8831d19ce6efd0a28862708@codeaurora.org> (raw)
In-Reply-To: <CANLsYkxEA66kGZh1rToSn79fpgPHqEjMZsSw74Rx3OLwG2k35w@mail.gmail.com>

On 2020-03-31 09:47, Mathieu Poirier wrote:
> On Mon, 30 Mar 2020 at 16:45, Bjorn Andersson
> <bjorn.andersson@linaro.org> wrote:
>> 
>> On Mon 30 Mar 15:12 PDT 2020, Mathieu Poirier wrote:
>> [..]
>> > > +   struct rproc *rproc;
>> > > +
>> > > +   rproc = container_of(inode->i_cdev, struct rproc, char_dev);
>> > > +   if (!rproc)
>> > > +           return -EINVAL;
>> > > +
>> > > +   rproc_shutdown(rproc);
>> >
>> > The scenario I see here is that a userspace program will call
>> > open(/dev/rproc_xyz, SOME_OPTION) when it is launched.  The file stays open
>> > until either the application shuts down, in which case it calls close() or it
>> > crashes.  In that case the system will automatically close all file descriptors
>> > that were open by the application, which will also call rproc_shutdown().
>> >
>> > To me the same functionality can be achieved with the current functionality
>> > provided by sysfs.
>> >
>> > When the application starts it needs to read
>> > "/sys/class/remoteproc/remoteprocX/state".  If the state is "offline" then
>> > "start" should be written to "/sys/.../state".  If the state is "running" the
>> > application just crashed and got restarted.  In which case it needs to stop the
>> > remote processor and start it again.
>> >
>> 
>> A case when this would be useful is the Qualcomm modem, which relies 
>> on
>> disk access through a "remote file system service" [1].
>> 
>> Today we register the service (a few layers ontop of rpmsg/GLINK) then
>> find the modem remoteproc and write "start" into the state sysfs file.
>> 
>> When we get a signal for termination we write "stop" into state to 
>> stop
>> the remoteproc before exiting.
>> 
>> There is however no way for us to indicate to the modem that rmtfs 
>> just
>> died, e.g. a kill -9 on the process will result in the modem continue
>> and the next IO request will fail which in most cases will be fatal.
I have this scenario written down in the cover letter for this patchset
  "[PATCH 0/2] Add character device interface to remoteproc"
I'll add it to the commit text as well.
> 
> The modem will crash when attempting an IO while rmtfs is down?
> 
>> 
>> So instead having rmtfs holding /dev/rproc_foo open would upon its
>> termination cause the modem to be stopped automatically, and as the
>> system respawns rmtfs the modem would be started anew and the two 
>> sides
>> would be synced up again.
> 
> I have a better idea of what is going on now - thanks for writing this 
> up.
> 
> I would make this feature a kernel configurable option as some people
> may not want it.  I also think having "/dev/remoteprocX" is fine, so
> no need to change anything currently visible in sysfs.
Ok. Makes sense.
> 
> Mathieu
> 
>> 
>> [1] https://github.com/andersson/rmtfs
>> 
>> Regards,
>> Bjorn