Skip to main content

No project description provided

Project description

Design Philosophy

This is an much more efficient implementation of canal. It is rewritten from scratch to utilize kratos’ efficient hardware generator framework.

Original canal description copied below:

In canal, the hardware is represented as a directed graph (DiGraph). In principle each node in the graph will be turn into a hardware component based on the property of the node. The edge will be transformed into a wire.

One thing to notice is that staged generation happens on the graph level. That is, passes have to be operated on the DiGraph, rather than the hardware generator itself. This allows better analysis and reduce the computation overhead since magma is slow when constructing large hardware. In canal, different bit width routing network will have different DiGraph by design. This allows us to share the common passes on different DiGraph while constructing on the same interconnect

How to install

pip install kcanal

Relationship with CGRA_PNR (Thunder/Cyclone)

The internal DiGraph representation is identical to Cyclone. Due to the requirement of pure-Python implementation, canal re-implemented the graph representation with more utility functions. The PnR information is interchangeable with Thunder and Cyclone and thus can be used in place route directly.

There is a python utility package called archipelago, that can take Interconnect object in canal and perform place and route directly. To install archipelago, you can use

pip install archipelago

It will use pre-built Python native wheels to speed up place and route.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

kcanal-0.0.1-py3-none-any.whl (31.0 kB view details)

Uploaded Python 3

File details

Details for the file kcanal-0.0.1-py3-none-any.whl.

File metadata

  • Download URL: kcanal-0.0.1-py3-none-any.whl
  • Upload date:
  • Size: 31.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.9.13

File hashes

Hashes for kcanal-0.0.1-py3-none-any.whl
Algorithm Hash digest
SHA256 aed27858544770d5829d15ae8b1a93f70f02846be39a9c8a81eef6cc2ae58c65
MD5 fdb7c1d403eb8c4280bda2b8f81ea0e4
BLAKE2b-256 b667e614cebc38c49c2496dd40b4fa22920f37416196e299b9a58e34c7dc4f1b

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page