pymtl-users

Hi, Is it possible to make pymtl3 to unroll range loop and emit the Verilog instead of outputting a

Glad to hear it is working! -c On Aug 23, 2023, at 6:43 AM, Jihyun Kim <kimjh...@gmail.com>

Hi David, There is no support for this out of the box ... you could either write your own PyMTL3 pass

Hmmm … not sure if we have ever tried that. Sounds like you figured out a way to make it work though?

Hi, Chris/Peitian, Yes, this is exactly the error and fix. Great catch. I know python list

Hi Damian, But you don't have any "classic" documentation eg with explanations what

We will probably post on this list but you can of course still use PyMTL with the older version of

You could check out this paper: https://www.csl.cornell.edu/~cbatten/pdfs/jiang-mamba-dac2018.pdf If

Hi David, PyMTL3 is definitely still active although most of the users are all at Cornell currently.

I think you need to write your own tracing functions then? The line tracing is setup to output

Hi Alex, Hmmm ... the code you wrote does not model real hardware. The equivalent in Verilog would be

I downloaded Python 3.7 and now it works fine . Thank you so much. Best regards, Marcel Le mardi 25

Hi Victor, Thanks for reaching out to us! We have identified this as an uncaught bug introduced in an

Thanks for the report. I finally got a chance to fix it on master. On Fri, Feb 19, 2021 at 10:28 PM

Hi, we have slightly revamped the way to do these pass configurations. You can take a look here https

Hi Enze, Right ... we usually use the "instantiate components to create test benches"

Hi Enze, Thanks for reaching out! Unfortunately, I don't think there is an easy path to

Hi! Yes, PyMTL3 is perfect for developing accelerator generators. Checkout the OpenCGRA project from

Please note that PyMTL3 currently does not officially support Python > 3.7. Since this assertion

Right... it is possible to check for comments first before searching for the regular expressions. On

Thank-you Peitian I think it worked! Best, Sebastian On Thursday, November 19, 2020 at 2:34:06 PM UTC

I have seen this before on an alpha version of PyMTL3 -- we saw really weird shared library issues

It should work fine even without textwave=True though? That is just there to give you the cute little

Hi Victor, It's because currently after applying the default pass group to enable simulation of

Hi Shunning, You were right, it was the problem caused by Python 3.8 Victor On Sunday, September 13,

sure, thanks for your reply. On Thursday, August 6, 2020 at 9:03:24 AM UTC-4 silicon....@gmail.com

Hi Peitian, Thanks for your reply. After removing all the size casting, my code works! On Wednesday,

Thank you very much! That workds, the correct code is: s.need_to_push_zero.out ==Bits1(0) On Saturday

See the previous email. Also as a shorthand you can use b32 instead of Bits32. Now you can also

Hi Chris, Thanks for your answer. It was really helpful. And yes, I also felt manipulating CL models