Skip to content

support loading device from json#480

Merged
yrabbit merged 3 commits intoYosysHQ:masterfrom
mia8973:patch-1000
Apr 6, 2026
Merged

support loading device from json#480
yrabbit merged 3 commits intoYosysHQ:masterfrom
mia8973:patch-1000

Conversation

@mia8973
Copy link
Copy Markdown
Contributor

@mia8973 mia8973 commented Mar 5, 2026

Co-Authored-By: Claude Opus 4.6 noreply@anthropic.com

@mia8973 mia8973 force-pushed the patch-1000 branch 4 times, most recently from 7d49274 to 54ff532 Compare March 5, 2026 19:25
@mia8973 mia8973 marked this pull request as draft March 5, 2026 19:25
@mia8973 mia8973 marked this pull request as ready for review March 5, 2026 19:30
@mia8973 mia8973 force-pushed the patch-1000 branch 2 times, most recently from 50d6315 to 193e1a4 Compare March 5, 2026 19:31
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
@mia8973 mia8973 had a problem deploying to Hardware in the loop March 5, 2026 20:16 — with GitHub Actions Failure
@mia8973 mia8973 had a problem deploying to Hardware in the loop March 5, 2026 20:16 — with GitHub Actions Failure
@yrabbit
Copy link
Copy Markdown
Collaborator

yrabbit commented Mar 5, 2026

Hmm... pnr[‘modules’][‘top’][‘settings’][‘packer.partno’] was supposed to be used to select a device, but then it turned out that Gowin's most detailed PARTNO is not a unique identifier: GW1N-9C and GW1N-9 are completely different chips, but you can't tell that from the PARTNO.

The same story with GW2A-18 and GW2A-18C.

You can easily see this if you try to compile blinky-clkdiv-tangnano9k.fs in the examples directory without specifying a device for gowin_pack.

@mia8973
Copy link
Copy Markdown
Contributor Author

mia8973 commented Mar 6, 2026

it looks like it was meant to be supported by the parsing of the part number out of the device isn't complete enough i think that you dcan just use the partdb field insteaed i tried it and it workd

@yrabbit
Copy link
Copy Markdown
Collaborator

yrabbit commented Mar 6, 2026

packer.chipdb

This will naturally work because it is effectively the name of the chip base 😄

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
setup.py Outdated
'msgspec',
'numpy',
'fastcrc',
'pillow',
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Usually, the set of Python libraries used remains unchanged. This is because Apicula is often included as part of a large package, and the feasibility of including/replacing third-party packages must be proven with blood, sweat, and tears.

You can use anything locally (after all, it's just a Python file), but you need to have very good reasons for changing the distribution set.

@yrabbit yrabbit temporarily deployed to Hardware in the loop April 6, 2026 10:51 — with GitHub Actions Inactive
@yrabbit yrabbit had a problem deploying to Hardware in the loop April 6, 2026 10:51 — with GitHub Actions Failure
@yrabbit yrabbit merged commit fb3e138 into YosysHQ:master Apr 6, 2026
18 of 28 checks passed
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.

2 participants