[PATCH v5 04/17] arm: mach-k3: Add basic support for J784S4 SoC definition

[Francesco Dolcini]

On Fri, Dec 01, 2023 at 10:37:55AM -0600, Bryan Brattlof wrote:
> On November 30, 2023 thus sayeth Apurva Nandan:
> > Add J784S4 initialization files for initial SPL boot.
> > 
> > Signed-off-by: Hari Nagalla <hnagalla at ti.com>
> > [ add firewall configurations and change the R5 MCU scratchpad ]
> > Signed-off-by: Manorit Chawdhry <m-chawdhry at ti.com>
> > Signed-off-by: Dasnavis Sabiya <sabiya.d at ti.com>
> > Signed-off-by: Apurva Nandan <a-nandan at ti.com>
> > ---
> >  arch/arm/mach-k3/Kconfig                      |  16 +-
> >  arch/arm/mach-k3/Makefile                     |   2 +
> >  arch/arm/mach-k3/arm64-mmu.c                  |  52 +++
> >  arch/arm/mach-k3/include/mach/hardware.h      |   4 +
> >  .../mach-k3/include/mach/j784s4_hardware.h    |  60 ++++
> >  arch/arm/mach-k3/include/mach/j784s4_spl.h    |  47 +++
> >  arch/arm/mach-k3/include/mach/spl.h           |   4 +
> >  arch/arm/mach-k3/j784s4_fdt.c                 |  15 +
> >  arch/arm/mach-k3/j784s4_init.c                | 338 ++++++++++++++++++
> >  arch/arm/mach-k3/r5/Makefile                  |   1 +
> >  10 files changed, 532 insertions(+), 7 deletions(-)
> >  create mode 100644 arch/arm/mach-k3/include/mach/j784s4_hardware.h
> >  create mode 100644 arch/arm/mach-k3/include/mach/j784s4_spl.h
> >  create mode 100644 arch/arm/mach-k3/j784s4_fdt.c
> >  create mode 100644 arch/arm/mach-k3/j784s4_init.c
> > 
> 
> ...
> 
> > +	if (IS_ENABLED(CONFIG_TARGET_J784S4_R5_EVM)) {
> > +		ret = uclass_get_device_by_name(UCLASS_MISC, "msmc", &dev);
> > +		if (ret)
> > +			panic("Probe of msmc failed: %d\n", ret);
> > +
> > +		ret = uclass_get_device(UCLASS_RAM, 0, &dev);
> > +		if (ret)
> > +			panic("DRAM 0 init failed: %d\n", ret);
> > +
> > +		ret = uclass_next_device_err(&dev);
> > +		if (ret)
> > +			panic("DRAM 1 init failed: %d\n", ret);
> > +
> > +		ret = uclass_next_device_err(&dev);
> > +		if (ret)
> > +			panic("DRAM 2 init failed: %d\n", ret);
> > +
> > +		ret = uclass_next_device_err(&dev);
> > +		if (ret)
> > +			panic("DRAM 3 init failed: %d\n", ret);
> > +	}
> 
> I don't know if this is is the right approach. I can see situations 
> where people will not have all these controllers utilized or would want 
> to artificially remove them for some reason. 

Not sure on the exact implication, but yes, you should expect customers
(like myself/my company) integrating the SoC into some board to not
necessarily use all the functionalities, memory controller included.

Francesco