[meta-xilinx] [meta-xilinx-bsp][PATCH 1/2] zcu106-zynqmp.conf: Add support for ZCU106 Evaluation Kit

[Manjukumar Harthikote Matha]

Hi Luca,

> -----Original Message-----
> From: Luca Ceresoli [mailto:luca at lucaceresoli.net]
> Sent: Monday, May 28, 2018 10:38 AM
> To: Manjukumar Harthikote Matha <MANJUKUM at xilinx.com>; meta-
> xilinx at yoctoproject.org
> Cc: Devarsh Thakkar <DEVARSHT at xilinx.com>; Devarsh Thakkar
> <DEVARSHT at xilinx.com>
> Subject: Re: [meta-xilinx] [meta-xilinx-bsp][PATCH 1/2] zcu106-zynqmp.conf: Add
> support for ZCU106 Evaluation Kit
> 
> Hi Manjukumar, Devarsh,
> 
> I'm glad to see zcu106 support coming! However I have a few questions about this
> patch, see below.
> 
> On 28/05/2018 09:40, Manjukumar Matha wrote:
> > From: Devarsh Thakkar <devarsh.thakkar at xilinx.com>
> >
> > The ZCU106 Evaluation Kit enables designers to jumpstart designs for
> > video conferencing, surveillance, Advanced Driver Assisted Systems
> > (ADAS) and streaming and encoding applications. This kit features a
> > Zynq® UltraScale+™ MPSoC EV device and supports all major peripherals
> > and interfaces, enabling development for a wide range of applications.
> > The included ZU7EV device is equipped with a quad-core ARM®
> > Cortex™-A53 applications processor, dual-core Cortex-R5 real-time
> > processor,
> > Mali™-400 MP2 graphics processing unit, 4KP60 capable H.264/H.265
> > video codec, and 16nm FinFET+ programmable logic.
> 
> I find this marketing-style paragraph rather useless in this context.

It's better to have a description of board capabilities.

> Why not replacing it with a short, technical description of how the board boots
> (U-Boot SPL, ATF etc), which device drivers are enabled etc?
> 

I can edit some info to commit message

> > This patch adds machine configuration file for ZCU106 Evaluation Kit
> > with required setting of board specific yocto variables needed for
> > compilation of bootloader, kernel and device-tree
> >
> > Signed-off-by: Devarsh Thakkar <devarsht at xilinx.com>
> > Tested-by: Maulik Desai <maulik.desai at xilinx.com>
> > Signed-off-by: Manjukumar Matha
> > <manjukumar.harthikote-matha at xilinx.com>
> > ---
> >  meta-xilinx-bsp/conf/machine/zcu106-zynqmp.conf | 33
> > +++++++++++++++++++++++++
> >  1 file changed, 33 insertions(+)
> >  create mode 100644 meta-xilinx-bsp/conf/machine/zcu106-zynqmp.conf
> >
> > diff --git a/meta-xilinx-bsp/conf/machine/zcu106-zynqmp.conf
> > b/meta-xilinx-bsp/conf/machine/zcu106-zynqmp.conf
> > new file mode 100644
> > index 0000000..42ac479
> > --- /dev/null
> > +++ b/meta-xilinx-bsp/conf/machine/zcu106-zynqmp.conf
> > @@ -0,0 +1,33 @@
> > +#@TYPE: Machine
> > +#@NAME: zcu106-zynqmp
> > +#@DESCRIPTION: Machine support for ZCU106 Evaluation Board.
> > +
> > +require conf/machine/include/tune-zynqmp.inc
> > +require conf/machine/include/machine-xilinx-default.inc
> > +require conf/machine/include/machine-xilinx-board.inc
> > +include conf/machine/include/zynqmp-pmu-config.inc
> > +
> > +MACHINE_FEATURES = "rtc ext2 ext3 vfat usbhost"
> > +
> > +UBOOT_MACHINE = "xilinx_zynqmp_zcu106_revA_defconfig"
> > +SPL_BINARY = "spl/boot.bin"
> > +
> > +SERIAL_CONSOLE = "115200 ttyPS0"
> > +SERIAL_CONSOLES_CHECK = "${SERIAL_CONSOLES}"
> > +
> > +KERNEL_DEVICETREE = "xilinx/zynqmp-zcu106-revA.dtb"
> > +
> > +PREFERRED_PROVIDER_virtual/kernel ?= "linux-xlnx"
> > +PREFERRED_PROVIDER_virtual/bootloader ?= "u-boot-xlnx"
> > +PREFERRED_PROVIDER_virtual/pmu-firmware ?= "zynqmp-pmu-pmu-firmware"
> > +
> > +EXTRA_IMAGEDEPENDS += " \
> > +		u-boot-zynq-uenv \
> > +		arm-trusted-firmware \
> > +		virtual/pmu-firmware \
> > +		virtual/boot-bin \
> > +		"
> > +
> > +IMAGE_BOOT_FILES += "uEnv.txt atf-uboot.ub ${KERNEL_IMAGETYPE}-zynqmp-
> zcu106-revA.dtb"
> > +
> > +MACHINE_HWCODECS = "libomxil-xlnx"
> 
> Is this machine configuration supposed to boot with this patch?
> How is the configuration object passed to the PMUFW?
> 

SPL flow still have a gap on PMUFW config object load AFAIK. This would be similar to existing zcu102. The only workaround I am aware is of Mike's patch in meta-topic

> If this patch is not yet enough to boot, I think it's worth mentioning it in the
> commit message.
> 

SPL gap for SDboot for all MPSoC devices is a known fact. 

> Also, this patch won't build without patch 2 of this series. The two patches should
> be squashed together if we want a bisectable commit. To me, the two patches
> together implement a single change: add zcu106 support.
>

I thought about it, wasn’t sure if I should squash both are not. Will squash them and send a V2

Thanks,
Manju