We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

Showing results for 
Search instead for 
Did you mean: 
Observer seb.mitchison
Registered: ‎03-15-2018

Multiple Implementation Runs Targetting Multiple Devices


I am using Trenz boards to test a development design. Due to limited board availability I have got Trenz TE0820 boards with Zynq Ultrascale+ 2cg, 2eg, 3cg and 3eg devices on them distributed throughout the department. I am trying to use a single project with 4 synthesis runs and 4 implementation runs to target the 4 different devices. The packaging and pinout on the device is the same so the only difference is the part number. I have created four synthesis and implementation runs which target the correct Trenz TE0820 board definition (provided by Trenz) for the 2cg, 2eg, 3cg or 3eg device.

My top level project is configured to a Trenz TE0820 3cg board definition (provided by Trenz) as this is the first board my project worked with. When I check the warnings on the synth_3cg and impl_3cg runs they are as they were before when I was synthesizing and implementing just a single run type. When I check the warnings on the synth_2cg/impl_2cg, synth_2eg/impl_2eg and synth_3eg/impl_3eg runs I get the following warning.

[Project 1-153] The current project device 'xczu2cg-sfvc784-1-e' does not match with the device on the 'TRENZ.BIZ:TE0820_3CG_1E:PART0:1.0' board part. A device change to match the device on 'TRENZ.BIZ:TE0820_3CG_1E:PART0:1.0' board part is being done. Please upgrade the IP in the project via the upgrade_ip command or by selecting Reports => Reports IP Status.

This sounds like it is reverting my 2cg, 2eg and 3eg runs to be the default 3cg setup which would make the different runs pointless. I also can't seem to update the IP from the 3CG type as report_ip_status always states it is up to date which doesn't seem correct for non 3cg runs. However, if I open the implemented designs I can see the percentage of resources utilized differ between 2 and 3 devices (as expected) as if it has used the specified devices. Is there something I am missing here?

I have been using (https://www.xilinx.com/support/documentation/sw_manuals/xilinx2018_2/ug892-vivado-design-flows-overview.pdf) as a reference for this work but it doesn't mention much detail on using multiple devices. I really don't want to manage 4 separate projects for this but am not sure it is really using my specifications.




0 Kudos
1 Reply
Registered: ‎07-18-2018

Re: Multiple Implementation Runs Targetting Multiple Devices

I might try two things. I would try making the IP not be built OOC, but part of the synth run, it will make runs take longer, but I think it should prevent the part mismatch between the IP and the current run.

Another option is to do a non project flow. This is a setup where that might make more sense. You can with the same sources run each step for each device, save the checkpoint, and then complete the next step and repeat.

You wouldn't get to manage them all in a project, but you'd have extra control over the builds and should be able to avoid the mismatching of parts.



0 Kudos