Re: [PATCH v3 14/21] fpga: dfl: add fpga manager platform driver for FME

From: Moritz Fischer
Date: Tue Feb 06 2018 - 01:44:36 EST


On Mon, Feb 05, 2018 at 10:25:27PM -0600, Alan Tull wrote:
> On Mon, Feb 5, 2018 at 8:17 PM, Wu Hao <hao.wu@xxxxxxxxx> wrote:
> > On Mon, Feb 05, 2018 at 11:21:52AM -0600, Alan Tull wrote:
> >> On Sun, Feb 4, 2018 at 4:05 AM, Wu Hao <hao.wu@xxxxxxxxx> wrote:
> >> > On Sat, Feb 03, 2018 at 11:41:24AM +0100, Moritz Fischer wrote:
> >> >> Hi Hao,
> >> >>
> >> >> On Fri, Feb 02, 2018 at 04:26:26PM -0800, Luebbers, Enno wrote:
> >> >> > Hi Hao, Alan,
> >> >> >
> >> >> > On Fri, Feb 02, 2018 at 05:42:13PM +0800, Wu Hao wrote:
> >> >> > > On Thu, Feb 01, 2018 at 04:00:36PM -0600, Alan Tull wrote:
> >> >> > > > On Mon, Nov 27, 2017 at 12:42 AM, Wu Hao <hao.wu@xxxxxxxxx> wrote:
> >> >> > > >
> >> >> > > > Hi Hao,
> >> >> > > >
> >> >> > > > A few comments below. Besides that, looks good.
> >> >> > > >
> >> >> > > > > This patch adds fpga manager driver for FPGA Management Engine (FME). It
> >> >> > > > > implements fpga_manager_ops for FPGA Partial Reconfiguration function.
> >> >> > > > >
> >> >> > > > > Signed-off-by: Tim Whisonant <tim.whisonant@xxxxxxxxx>
> >> >> > > > > Signed-off-by: Enno Luebbers <enno.luebbers@xxxxxxxxx>
> >> >> > > > > Signed-off-by: Shiva Rao <shiva.rao@xxxxxxxxx>
> >> >> > > > > Signed-off-by: Christopher Rauer <christopher.rauer@xxxxxxxxx>
> >> >> > > > > Signed-off-by: Kang Luwei <luwei.kang@xxxxxxxxx>
> >> >> > > > > Signed-off-by: Xiao Guangrong <guangrong.xiao@xxxxxxxxxxxxxxx>
> >> >> > > > > Signed-off-by: Wu Hao <hao.wu@xxxxxxxxx>
> >> >> > > > > ----
> >> >> > > > > v3: rename driver to dfl-fpga-fme-mgr
> >> >> > > > > implemented status callback for fpga manager
> >> >> > > > > rebased due to fpga api changes
> >> >> > > > > ---
> >> >> > > > > .../ABI/testing/sysfs-platform-fpga-dfl-fme-mgr | 8 +
> >> >> > > > > drivers/fpga/Kconfig | 6 +
> >> >> > > > > drivers/fpga/Makefile | 1 +
> >> >> > > > > drivers/fpga/fpga-dfl-fme-mgr.c | 318 +++++++++++++++++++++
> >> >> > > > > drivers/fpga/fpga-dfl.h | 39 ++-
> >> >> > > > > 5 files changed, 371 insertions(+), 1 deletion(-)
> >> >> > > > > create mode 100644 Documentation/ABI/testing/sysfs-platform-fpga-dfl-fme-mgr
> >> >> > > > > create mode 100644 drivers/fpga/fpga-dfl-fme-mgr.c
> >> >> > > > >
> >> >> > > > > diff --git a/Documentation/ABI/testing/sysfs-platform-fpga-dfl-fme-mgr b/Documentation/ABI/testing/sysfs-platform-fpga-dfl-fme-mgr
> >> >> > > > > new file mode 100644
> >> >> > > > > index 0000000..2d4f917
> >> >> > > > > --- /dev/null
> >> >> > > > > +++ b/Documentation/ABI/testing/sysfs-platform-fpga-dfl-fme-mgr
> >> >> > > > > @@ -0,0 +1,8 @@
> >> >> > > > > +What: /sys/bus/platform/devices/fpga-dfl-fme-mgr.0/interface_id
> >> >> > > > > +Date: November 2017
> >> >> > > > > +KernelVersion: 4.15
> >> >> > > > > +Contact: Wu Hao <hao.wu@xxxxxxxxx>
> >> >> > > > > +Description: Read-only. It returns interface id of partial reconfiguration
> >> >> > > > > + hardware. Userspace could use this information to check if
> >> >> > > > > + current hardware is compatible with given image before FPGA
> >> >> > > > > + programming.
> >> >> > > >
> >> >> > > > I'm a little confused by this. I can understand that the PR bitstream
> >> >> > > > has a dependency on the FPGA's static image, but I don't understand
> >> >> > > > the dependency of the bistream on the hardware that is used to program
> >> >> > > > the bitstream to the FPGA.
> >> >> > >
> >> >> > > Sorry for the confusion, the interface_id is used to indicate the version of
> >> >> > > the hardware for partial reconfiguration (it's part of the static image of
> >> >> > > the FPGA device). Will improve the description on this.
> >> >>
> >> >> I'm not sure userland should be making the call on whether what you're
> >> >> trying to load is compatible or not.

I guess I was worried that you might take down a PR region that you
could've avoided. On second thought unless the kernel could extract that
info somehow from the bitstream before programming this is probably
anyway not possible, since the only one that knows the id -> bitstream
mapping is userland.

Hao's comments further down outline well enough that the hardware can
deal with incompatible bitstreams. So ignore my comments ;-)

> >>
> >> Could you explain more about what your concern was about this (unless
> >> Hao has covered it below)?
> >>
> >> It makes sense to me in this use case at least since userspace has a
> >> pile of images and is choosing which one to load.
> >>
> >> >> Isn't there a way to check this in
> >> >> your PR reconfiguration handler in-kernel?
> >> >
> >> > Hi Moritz
> >> >
> >> > Actually with current driver interface, doing a partial reconfiguration with an
> >> > incompatible image, then driver will report PR failure with error code
> >> > FPGA_MGR_STATUS_INCOMPATIBLE_IMAGE_ERR as hardware checks it, but anyway user
> >> > needs to know hardware interface_id information to find or re-generated correct
> >> > images. I think it's more flexible to leave it to userspace on using this
> >> > information exposed by driver. : )
> >> >
> >> > Thanks
> >> > Hao
> >> >
> >> >>
> >> >> > >
> >> >> >
> >> >> > The interface_id expresses the compatibility of the static region with PR
> >> >> > bitstreams generated for it. It changes every time a new static region is
> >> >> > generated.
> >>
> >> In the near future the DFL framework will be used with SoC's that have
> >> a hard FPGA PR manager (that's not part of the static region). The
> >> hard FPGA manager driver won't know anything about the static region.
> >>
> >> >> >
> >> >> > Would it make more sense to have the interface_id exposed as part of the FME
> >> >> > device (which represents the static region)? I'm not sure - it kind of also
> >> >> > makes sense here, where you would have all the information in one place (if the
> >> >> > interface_id matches, I can use this component to program a bitstream).
> >>
> >> According to the intel-fpga.txt document, the identifier for the
> >> static image is at
> >>
> >> /sys/class/fpga_region/regionX/fpga-dfl-fme.n/bitstream_id
> >
> > Hi Alan
>
> Hi Hao,
>
> >
> > This bitstream_id refects the full static region version. As you know, PR is
> > only a sub feature of the FME functional unit, it's possible that we have
> > different static region (different bitstream_id) but it has the exact same
> > PR sub feature under the FME, only changes on other sub features or function
> > units.
>
> OK, thanks for that explanation. That makes sense but could have
> easily been different. Please document this somewhere.
>
> Alan

Moritz