Re: [PATCH 4/4] configs: am64x_evm_a53_defconfig: Enable NAND

[Tom Rini <trini@konsulko.com>]

[-- Attachment #1: Type: text/plain, Size: 2264 bytes --]

On Wed, Jan 10, 2024 at 06:06:45PM +0200, Roger Quadros wrote:
> +Lukasz & Mattijs
> 
> On 10/01/2024 11:34, Roger Quadros wrote:
> > 
> > 
> > On 09/01/2024 22:00, Francesco Dolcini wrote:
> >> On Tue, Jan 09, 2024 at 02:54:00PM -0500, Tom Rini wrote:
> >>> On Tue, Jan 09, 2024 at 01:18:59PM -0600, Nishanth Menon wrote:
> >>>> On 14:26-20240109, Roger Quadros wrote:
> >>>>>  CONFIG_CMD_PMIC=y
> >>>>>  CONFIG_CMD_REGULATOR=y
> >>>>> +CONFIG_CMD_MTDPARTS=y
> >>>>> +CONFIG_MTDIDS_DEFAULT="nand0=omap2-nand.0"
> >>>>> +CONFIG_MTDPARTS_DEFAULT="mtdparts=omap2-nand.0:2m(NAND.tiboot3),2m(NAND.tispl),2m(NAND.tiboot3.backup),4m(NAND.u-boot),256k(NAND.u-boot-env),256k(NAND.u-boot-env.backup),-(NAND.file-system)"
> >>>>
> >>>> Why not handle this as device tree partitions?
> >>>
> >>> I honestly forget what the preferred way of defining and passing NAND
> >>> partition information is these days. It might even be the funny case
> >>> that passing as cmdline args is "best" rather than fixed-partitions
> >>> binding?
> >>
> >> According to past discussions [1] doing the fixup in U-Boot is not advised.
> >>
> >> Using the command line or having the partition fixed in the DT are both
> >> valid options.
> >>
> >> [1] https://lore.kernel.org/all/20230206224838.75963-1-francesco@dolcini.it/
> >>
> >> Francesco
> >>
> > 
> > This was not about passing mtdparts to kernel but about getting 'mtdparts' command
> > to work at u-boot. I need to figure out why OF partition parser didn't work here.
> > 
> > For a start I didn't have CONFIG_MTD_PARTITIONS set. Maybe I'm missing something more.
> > 
> 
> The issue was the NAND driver was not setting chip->dev and chip->ofnode correctly.
> Now 'mtd list' shows partitions from the DT.
> But, 'dfu 0 nand list' still fails like so.
> 	"mtdparts variable not set, see 'help mtdparts'"
> 
> Looks like dfu_nand driver doesn't fall back to OF partitions if mtdparts environment
> is not defined.
> 
> Should we add OF partitions support to dfu_nand driver or is it deprecated
> in favor of dfu_mtd driver?

I would rather not have to populate the env variable as that will lead
to confusion later on I fear (make a change and it's not populated
onward).

-- 
Tom

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 659 bytes --]