04-08-2019 01:50 AM
I am trying to adjust the ZCU102 TRD for a custom board. I have built the block design in Vivado and exported the .hdf file with the DPU IP v1.3.0 in it. Now I have built a Petalinux project with the template flow and found a cached version of a DPU integration tutorial from Xilinx, where I'm stuck on the following points:
1) add a recipe for OpenCV v3.1 as this is the version needed by the DPU
-> I have modified the .bbappend in $PETALINUX/components/yocto/source/aarch64/layers/meta-openembedded/meta-oe/recipes-support/opencv and changed the preferred version to install from 3.3 to 3.1. Is this how you're supposed to do it?
2) Add a bbappend for the protobuf package to change the branch that its source is pulled from. This is needed due to the OpenCV v3.1 change.
-> don't really know how to do that as I don't know which version of protobuf is needed for opencv3.1; the default bbappend for protobuf is version 3.4.1
3) Add a bbappend to modify the LINUX_VERSION_EXTENSION of the kernel. This is needed to make the pre-built dpu kernel module (dpu.ko) "version magic" match the kernel that we build.
4) As far as I understand I also need to add dpu.ko, libdputils.so, libn2cube.so to the root file system. However, I could only find those files specifically for the supported reference boards. The documentation of the DPU IP mentions that it should be available for all Zynq 7000 and Zynq MPSoC devices.
5) I also need add the dnndk to the recipes.
-> I added IMAGE_INSTALL_append = " dnndk" to my petalinux-image.bbappend. However, I don't know how to add the source for this package to Petalinux to find this package.
So any help with any of these things would be very much appreciated. Or maybe an update can be given as to when the updated DPU Integration tutorial will go online again. I have zero experience with Petalinux, so coming up with a solution to any of these is quite challenging and google isn't helping that much.
Thank you very much!
04-22-2019 02:53 PM
04-08-2019 09:21 AM
The udpated tutorial and all of the files needed to complete the flow should be avaiable in the next week. I would recommend waiting until it is avaiable so you can see the implementation details of all of the steps mentioned below.
Once it's available, I'll post an update on the forum.
04-16-2019 12:36 AM
I have the same problem.
The errro is " nothing provides libopencv_core.so.3.1()(64bit) needed by dnndk-1.0-r0.ultra96_zynqmp".
I wonder if you could help me check it out.
04-22-2019 02:53 PM
04-23-2019 12:08 AM - edited 04-23-2019 04:01 AM
Thanks @terryo for your announcement
he tutorial seems to be available for zedboard or pynq if dnndk-lib packages is used at https://github.com/hirayaku/DAC2018-TGIIF/tree/master/prerequisites/dnndk-lib/pkgs
The pacakges seem to be too old to compatible with dpu ip
04-25-2019 02:36 AM
04-25-2019 01:40 PM
Well, it looks like a section of the tutorial went missing when it was published. There is an entire sectino on building the applications in Xilinx SDK. I have someone looking into it, hope to have it fixed today.