Usage of device-tree for blobs

[Mark Kettenis]

> From: Simon Glass <sjg at chromium.org>
> Date: Thu, 26 Aug 2021 14:00:12 -0600
> 
> Hi Mark,
> 
> On Thu, 26 Aug 2021 at 06:55, Mark Kettenis <mark.kettenis at xs4all.nl> wrote:
> >
> > > From: Simon Glass <sjg at chromium.org>
> > > Date: Wed, 25 Aug 2021 21:15:00 -0600
> > >
> > > Hi Heinrich,
> > >
> > > On Wed, 25 Aug 2021 at 02:05, Heinrich Schuchardt <xypron.glpk at gmx.de> wrote:
> > > >
> > > > Hello Simon,
> > > >
> > > > some boards like qemu-riscv64_defconfig do not use any device-tree at
> > > > build time. A device-tree is only supplied at runtime by the prior boot
> > > > stage (CONFIG_OF_PRIOR_STAGE=y).
> > > >
> > > > In doc/develop/devicetree/intro.rst you suggest to put binary blobs into
> > > > the device-tree.
> > > >
> > > > Could you, please, update the documentation to explain how adding blobs
> > > > to the device-tree works in the different scenarios depending on the
> > > > values of:
> > > >
> > > > CONFIG_OF_EMBED
> > > > CONFIG_OF_SEPARATE
> > > > CONFIG_OF_BOARD
> > > > CONFIG_OF_HOSTFILE
> > > > CONFIG_OF_PRIOR_STAGE
> > > >
> > > > It would be especially important to understand how one can develop a
> > > > board independent feature which works for all of these settings.
> > >
> > > OK I will take a crack at this tomorrow.
> > >
> > > But I think there is a disconnect here, since the only options that
> > > matter within U-Boot are OF_SEPARATE and OF_HOSTFILE, which both use a
> > > u-boot.dtb file. There is nothing tricky here.
> > >
> > > The OF_BOARD business is for when the board does special things.
> > > Presumably signing will do special things too. We cannot really know
> > > what those things are because the board as 'opted out' of the standard
> > > options.
> > >
> > > >
> > > > Please, describe CONFIG_OF_PRIOR_STAGE in
> > > > doc/develop/devicetree/control.rst.
> > >
> > > So I'm not allowed to delete that option? :-) It seems to me to be
> > > extremely sparse on documentation. We need an explanation of what it
> > > means and what effect it has on the build system, U-Boot and some
> > > discussion of how qemu works. It seems to have been added as part of
> > > an unrelated broadcom commit. The tags were incorrect so I doubt
> > > anyone noticed it. Since then it has apparently proved useful
> > > elsewhere, but no one has added more docs.
> > >
> > > So perhaps you can help me with my doc by explaining how
> > > OF_PRIOR_STAGE works in qmeu, why the DT in U-Boot cannot be used, and
> > > which project actually hosts the DT that qemu provides? Armed with
> > > that information, I might be able to make sense of it all.
> >
> > Well, QEMU allows configuration of (emulated/paravirtualised) devices
> > from the command line.  So it is pretty much impossible for U-Boot to
> > provide a DT that matches that configuration without adding a lot of
> > code the dynamically build it from some sort of configuration
> > specification provided by QEMU to U-Boot.  But at that point QEMU
> > might as well provide the DT itself, don't you agree?
> 
> Don't these devices have a DT in U-Boot? Is qemu enabling them (status
> = "okay") or actually inventing them?!

I actually invents them.  See the mail I just sent for a reference to
the code.

> > Anyway, replying to this thread since for the Apple M1 support that
> > I'm working on we're in a somewhat similar situation.  The Asahi Linux
> > project has implemented m1n1 as a bootloader and plans to use U-Boot
> > as a "payload" to implement booting a standard Linux distribution (and
> > other OSes).  In this scenario m1n1 actually provides the DT since it
> > parses Apple's version of the DT (which isn't in the standard DT
> > format) and adds/updates certain properties that depend on the actual
> > hardware it is running on.
> 
> Yes I see. Still I'd like a basic DT in U-Boot, even if just for
> discoverability.

Discoverability of what?

> > For this code I use CONFIG_OF_BOARD and implement
> > board_fdt_blob_setup() to simple return a pointer to the DT that m1n1
> > set up.  But that seems to be exactly what CONFIG_OF_PRIOR_STAGE
> > does...
> 
> Right, and neither is really documented to the level that people can
> understand the purpose. They just come across as hacks to me.

Well, I think U-Boot by its very nature is a giant hack ;).

> I'd like to see an API for passing a info (including DT) to U-Boot. I
> think riscv has it? I suggest we use a register that points to a
> bloblist.

Well, having a generic API will be hard.  It will depend on the
low-level firmware.  On RISC-V the SBI specification defines such an
API.  For the Apple M1 I just use the existing code that makes U-Boot
look like a Linux kernel and use the same API that is used to pass the
DTB to the kernel.  The Raspberry Pi code does something similar.  In
the end it just boils down to passing a pointer to the DT in a
specific register.