On Sat, Dec 24, 2022 at 12:03:52PM +0000, Paul Barker wrote: > On 20/12/2022 15:55, Rob Herring wrote: > > On Wed, Nov 23, 2022 at 05:50:06PM +0000, Paul Barker wrote: > >> Add properties to the Authenta SPI flash device node to enable access by > >> a UEFI application using a fixed GUID. > >> > >> Signed-off-by: Paul Barker <paul.bar...@sancloud.com> > >> --- > >> arch/arm/dts/am335x-sancloud-bbe-lite-u-boot.dtsi | 13 ++++++++++--- > >> arch/arm/dts/am335x-sancloud-bbe-lite.dts | 2 +- > >> 2 files changed, 11 insertions(+), 4 deletions(-) > >> > >> diff --git a/arch/arm/dts/am335x-sancloud-bbe-lite-u-boot.dtsi > >> b/arch/arm/dts/am335x-sancloud-bbe-lite-u-boot.dtsi > >> index 01c105ebb383..6c4ff67f9a4b 100644 > >> --- a/arch/arm/dts/am335x-sancloud-bbe-lite-u-boot.dtsi > >> +++ b/arch/arm/dts/am335x-sancloud-bbe-lite-u-boot.dtsi > >> @@ -38,7 +38,14 @@ > >> > >> &spi0 { > >> u-boot,dm-pre-reloc; > >> - channel@0 { > >> - u-boot,dm-pre-reloc; > >> - }; > >> +}; > >> + > >> +&authenta_flash { > >> + u-boot,dm-pre-reloc; > >> + > >> + u-boot,uefi-spi-vendor = "micron"; > >> + u-boot,uefi-spi-part-number = "mt25ql128abb"; > > > > Looks like a compatible string. Yet, the flash node compatible string, > > micron,spi-authenta, is not documented (though in use for spidev). So > > use a compatible string for the flash that is specific to the flash > > model. I assume there is some reason the specific model is needed? > > For context, the UEFI Platform Initialization (PI) spec defines > EFI_SPI_PART, EFI_SPI_PERIPHERAL and EFI_SPI_IO_PROTOCOL structures. > I'm referencing v1.7 Errata A. See https://uefi.org/specifications for > downloads. > > The EFI_SPI_PART structure has "Vendor" and "PartNumber" fields. We need > something to put in those fields and the device tree is the best place > to store the data. These properties are in the `-u-boot.dtsi` file so > they won't be submitted to the Linux kernel.
Well, IMHO, this doesn't belong in U-Boot only for forever. Just like other bindings/properties that we're working on getting merged upstream and so that there is really Just One Device Tree, this should go upstream I believe. This might be another case of starts in -u-boot.dtsi while things get sorted out. -- Tom
signature.asc
Description: PGP signature