Re: [PATCH 1/2] Documentation: riscv: Add early boot document

[Alexandre Ghiti]

On Tue, Jun 20, 2023 at 9:18 AM Sunil V L <sunilvl@xxxxxxxxxxxxxxxx> wrote:
>
> Hey Alex,
>
> Many thanks for writing this up!
>
> On Mon, Jun 19, 2023 at 11:49:13AM +0200, Alexandre Ghiti wrote:
> > @Sunil V L Something about ACPI is more than welcome :)
> >
> > And thanks to @Björn Töpel and @Atish Kumar Patra for helping in
> > writing this document!
> >
> > On Mon, Jun 19, 2023 at 11:47 AM Alexandre Ghiti <alexghiti@xxxxxxxxxxxx> wrote:
> > >
> > > This document describes the constraints and requirements of the early
> > > boot process in a RISC-V kernel.
> > >
> > > Szigned-off-by: Alexandre Ghiti <alexghiti@xxxxxxxxxxxx>
> > > ---
> > >  Documentation/riscv/boot-image-header.rst |   3 -
> > >  Documentation/riscv/boot.rst              | 181 ++++++++++++++++++++++
> > >  Documentation/riscv/index.rst             |   1 +
> > >  3 files changed, 182 insertions(+), 3 deletions(-)
> > >  create mode 100644 Documentation/riscv/boot.rst
> > >
> > > diff --git a/Documentation/riscv/boot-image-header.rst b/Documentation/riscv/boot-image-header.rst
> > > index d7752533865f..a4a45310c4c4 100644
> > > --- a/Documentation/riscv/boot-image-header.rst
> > > +++ b/Documentation/riscv/boot-image-header.rst
> > > @@ -7,9 +7,6 @@ Boot image header in RISC-V Linux
> > >
> > >  This document only describes the boot image header details for RISC-V Linux.
> > >
> > > -TODO:
> > > -  Write a complete booting guide.
> > > -
> > >  The following 64-byte header is present in decompressed Linux kernel image::
> > >
> > >         u32 code0;                /* Executable code */
> > > diff --git a/Documentation/riscv/boot.rst b/Documentation/riscv/boot.rst
> > > new file mode 100644
> > > index 000000000000..b02230818b79
> > > --- /dev/null
> > > +++ b/Documentation/riscv/boot.rst
> > > @@ -0,0 +1,181 @@
> > > +.. SPDX-License-Identifier: GPL-2.0
> > > +
> > > +=============================================
> > > +Early boot requirements/constraints on RISC-V
> > > +=============================================
> > > +
> > > +:Author: Alexandre Ghiti <alexghiti@xxxxxxxxxxxx>
> > > +:Date: 23 May 2023
> > > +
> > > +This document describes what the RISC-V kernel expects from the previous stages
> > > +and the firmware, but also the constraints that any developer must have in mind
> > > +when touching the early boot process, e.g. before the final virtual mapping is
> > > +setup.
> > > +
> > > +Pre-kernel boot (Expectations from firmware)
> > > +============================================
> > > +
> > > +Registers state
> > > +---------------
> > > +
> > > +The RISC-V kernel expects:
> > > +
> > > +  * `$a0` to contain the hartid of the current core.
> > > +  * `$a1` to contain the address of the device tree in memory.
> > > +
> > > +CSR state
> > > +---------
> > > +
> > > +The RISC-V kernel expects:
> > > +
> > > +  * `$satp = 0`: the MMU must be disabled.
> > > +
> > > +Reserved memory for resident firmware
> > > +-------------------------------------
> > > +
> > > +The RISC-V kernel expects the firmware to mark any resident memory with the
> > > +`no-map` flag, thus the kernel won't map those regions in the direct mapping
> > > +(avoiding issues with hibernation, speculative accesses and probably other
> > > +subsystems).
> > > +
> > > +Kernel location
> > > +---------------
> > > +
> > > +The RISC-V kernel expects to be placed at a PMD boundary (2MB for rv64 and 4MB
> > > +for rv32). Note though that the EFI stub will physically relocate the kernel if
> > > +that's not the case.
> > > +
> > > +Device-tree
> > > +-----------
> > > +
> > > +The RISC-V kernel always expects a device tree, it is:
> > > +
> In general, the firmware can pass either DT or ACPI (not both at the
> same time) to the OS. So, I think the statement that kernel always
> expects DT probably causes confusion. Can we mention some thing like
> below?
>
> The firmware can pass either DeviceTree or ACPI tables to the RISC-V
> kernel.
>
> The DeviceTree is either passed directly to the kernel from the
> previous stage using the `$a1` register, or when booting with UEFI, it
> can be passed using the EFI configuration table.
>
> The ACPI tables are passed to the kernel using the EFI configuration
> table. In this case, a tiny DeviceTree is still created by the EFI
> stub. Please refer to "EFI stub and device" tree section below for
> details of the tiny DeviceTree.
>

Great, this is way better so I'll go with that:

Hardware description
--------------------

The firmware can pass either a devicetree or ACPI tables to the RISC-V kernel.

The devicetree is either passed directly to the kernel from the previous stage
using the `$a1` register, or when booting with UEFI, it can be passed
using the
EFI configuration table.

The ACPI tables are passed to the kernel using the EFI configuration table. In
this case, a tiny devicetree is still created by the EFI stub. Please refer to
"EFI stub and devicetree" tree section below for details about this devicetree.

Thanks!

> Thanks,
> Sunil