[meta-xilinx] failure : SD to u-boot SPL to u-boot on zcu102-zynqmp

[Jean-Francois Dagenais]

> On Mar 17, 2017, at 03:16, Nathan Rossi <nathan at nathanrossi.com> wrote:
> 
> On 17 March 2017 at 05:27, Jean-Francois Dagenais
> <jeff.dagenais at gmail.com> wrote:
>> 
>> 
>> Is it ok for me not to care about the PMU FW or ATF at this point of our
>> development?
> 
> For PMU Firmware, sure you can probably ignore it and use the
> xilinx-v2016.4 kernel and u-boot. ATF is needed though since a psci
> implementation is needed that can handle cpu bringup.

Ok, help me make sure I understand from having read your whole answer. I only
need to provide an alternate (TBD) PMU FW because I am using u-boot-xlnx/master
and something was added since u-boot xilinx-v2016.4 that requires it?

And did you just misprint "xilinx-v2016.4 kernel" instead of "u-boot"?

Did you mean I should use u-boot (upstream) instead of u-boot-xlnx?

You can tell I am confused! It'll get better within a few weeks! ;) Until
then, please put more info that's not too much trouble. :)

> It is ok for the U-Boot build to succeed without a psu_init_gpl, since
> it is common to use FSBL as a loader. Which is normally just loading
> the full U-Boot so SPL is not needed in that case. But the meta-xilinx
> layer does have a hard fail (for zynq at least, but will be for zynqmp
> too) if you try to build/deploy SPL (SPL_BINARY = "spl/boot.bin" is
> set) and nothing is providing the ps*_init_gpl files.

I'm no expert on u-boot (yet ;) but I think this smells trouble. Maybe not for
meta-xilinx supported builds, but for integrators such as myself and all the
other OEMs which will use meta-xilinx as a base.

I understand about an SPL-less build. Perhaps the Makefile could inspect
CONFIG_SPL_BUILD and fail if the psu_init_gpl files aren't found. You don't get
very far with a "psu_init"-less SPL, but much better if failure occurs at build
time. I can can attempt a patch in board/xilinx/zynqmp/Makefile unless you think
its a bad idea.

> 
> On a side note, you should be able to just copy the psu_init_gpl files
> from master u-boot-xlnx and use them in the xilinx-v2016.4 version
> (which doesn't have the pmufw requirements).

My first tries were with u-boot-xlnx (v2016.4) and the SPL almost didn't start
at all. It may be related to 7d355473f34a (mmc: sdhci: zynqmp: Add support of
SD3.0) not being there yet. I did not try exactly your idea though. I will get
to it soon if nothing else works.

Can I not change something in the defconfig to remove the extra PMUFW dependency?

> 
>> 
>> At this point of our early development, all I want is a command line in linux...
>> ASAP in order to unlock my hardware department.
>> 
>> I will deal with optimizing the boot and improving the power management, secure
>> boot and all these low-level subjects later.
>> 
>> Like even the PMU FW, I have the Vivado suite running, I did follow blindly some
>> instructions to generate the default PMU FW a few weeks ago. If I could hack
>> this as a binary blob into my bitbake at this point I would be happy.
>> 
>> What do you recommend?
> 
> If your just trying to bring up a system for development/testing, do
> what ever you need to get it working. Since its likely this stuff will
> change and or be more polished by the time you need to set it up
> properly.

Thing is I have no idea what to do next! SD/eMMC support in meta-xilinx seems so
rough on zynqmp...

I guess I need to fall back to the documentation in order to better understand the
whole ATF/PMUFW/Low-level boot process.

Thanks for all the help!