Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[device] CW340 support #195

Draft
wants to merge 1 commit into
base: master
Choose a base branch
from
Draft

[device] CW340 support #195

wants to merge 1 commit into from

Conversation

nasahlpa
Copy link
Member

This PR adds experimental CW340 support with a minimal running capture script.

This PR adds experimental CW340 support with a minimal running
capture script.

Signed-off-by: Pascal Nasahl <[email protected]>
@Blebowski
Copy link

Hi,
may I mask what is the status and plans here ? Is CW340 supposed to replace the CW310 for the purposes
of OT SCA ?

I see that the main board being used is CW310 currently. However, this-one features 7-series Kintex,
an older device that will eventually go out of support by Xilinx. The CW340 has Ultrascale which is newer
FPGA family with better clocking architecture (AFAICT better suited for ASIC emulation purposes).

Further, the NewAE web-page claims that CW340:

The CW340 is designed specifically for working with the OpenTitan design, either in FPGA emulation or with a production silicon device.

@nasahlpa
Copy link
Member Author

nasahlpa commented Feb 3, 2025

Hi @Blebowski,

Thanks for your interest in ot-sca!

Currently, we are not planning to support CW340 in ot-sca. The reason is that capturing power traces on CW340 is much more challenging (i.e. more noise) due to the larger FPGA and the larger design. Hence, when capturing traces, we use the smaller OpenTitan design for CW310 and even remove some IPs that we are not measuring to minimize the noise.

However, adding support for CW340 by yourself should be quite straightforward - this draft PR should give you an idea how this could work.

@Blebowski
Copy link

Thanks @nasahlpa.

My thinking was precisely opposite. Since ultra-scale has better clocking, it would allow to emulate
e.g. per-IP clock gate enable as on the ASIC, and therefore not to cause noise from other IPs.

But I guess that Open Titan does not have per-peripheral manual gating(that needs to be functionally enabled
for peripheral to work) and relies only on inserted clock gates ?

@Blebowski
Copy link

So it is good to have the information instead of going completely wrong direction :)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants