[casper] Vivado 2019.1.1 Single Port Memory and dual port memory Issues

[Adam Isaacson]

Dear Casperites,

We have recently ported the toolflow to work with Matlab R2018a and Vivado
2019.1.1. We noticed a bug with our F-engine that has finally been tracked
down and SARAO does not wish for you to fall into the same black hole.

There are actually two bugs - one minor and one major:

1) *Minor*: If you try and simulate asymmetric dual port memories in
Simulink then you will not get sensible data. It does work on the chip
though. This is only a simulation bug.

2) *Major*: if you use a single port memory with the enable line then this
will apply to you. Please note the enable line does not work as it should,
especially if you are asserting/deasserting asynchronously. If you choose a
single port memory with a latency of 3 then the output registers that are
added internally to give you the desired latency do not have enable lines
when they should have enable lines. The PFB that we are using in our
F-engine happens to use two single port memories used as delay lines with
enable lines.

There is a workaround for point 2), which we are using. Set the latency of
the single port RAM to 1 and add external registers with enable lines. This
will do the trick- see git commit below. I have been working very closely
with Xilinx and they will be releasing a fix for the above points in Vivado
2020.1.

If you wish to learn more about our bug then please checkout
https://github.com/ska-sa/mlib_devel (devel branch) githash: 793d05b.

Here is hoping that you avoid the trap!

Kind regards,

Adam Isaacson
South African Radio Astronomy Observatory (SARAO)
Hardware Manager
Cell: (+27) 825639602
Tel:  (+27) 215067300
email: aisaac...@ska.ac.za

-- 
You received this message because you are subscribed to the Google Groups 
"casper@lists.berkeley.edu" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to casper+unsubscr...@lists.berkeley.edu.
To view this discussion on the web visit 
https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CADTJ%3DnEC%2B5r6V7DysMrap0OOzFPEEaYgjtwH42qZuUPjuKeb0w%40mail.gmail.com.

Re: [casper] SKARAB Microblaze Source Code

[Adam Isaacson]

Hi Lewis,

My apologies, the repo was private. It is now public, so you should be able
to access it now.

https://github.com/ska-sa/skarab_microblaze_software

Kind regards,

Adam Isaacson
South African Radio Astronomy Observatory (SARAO)
Hardware Manager
Cell: (+27) 825639602
Tel:  (+27) 215067300
email: aisaac...@ska.ac.za



On Wed, Dec 18, 2019 at 11:53 PM 'Lewis McLaughlin' via
casper@lists.berkeley.edu  wrote:

> Hello,
>
> I am looking to inspect the source code for the microblaze on skarab.
> However the following link results in a 404 - not found:
>
> https://github.com/ska-sa/skarab_microblaze_software
>
> Is this repo private or does it no longer exist?
>
> Many thanks,
> Lewis.
>
> --
> You received this message because you are subscribed to the Google Groups "
> casper@lists.berkeley.edu" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to casper+unsubscr...@lists.berkeley.edu.
> To view this discussion on the web visit
> https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/7da891ae-ff15-4e62-9228-9cb20398f998%40lists.berkeley.edu
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"casper@lists.berkeley.edu" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to casper+unsubscr...@lists.berkeley.edu.
To view this discussion on the web visit 
https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CADTJ%3DnEwXj0H8BsPKOys6-5wFYfpQPyQsyfX_744ovOF8Jq1eg%40mail.gmail.com.