Skip to main content

Package to automate the synthesis of vivado HLS components

Project description

vhlsrs

It means Vivado HLS run synthesis.

This package contains a script that runs vivado synthesis in a (somewhat) reproducible way.

It also allow to search for the minimum pipeline depth design for a given clock period constrain.

Usage

Experience description

The experience to run should be described in a .ini.

A complete experiment description looks like :

[EXP_NAME]
comp_path=path/to/component/file.cpp
top_level_comp= Name of top level function
period= clock period in ns (float)
part= part code name (default is xc7k160tfbg484)
standard= c++ standard to use. (default is c++11)
includes=comma separated list of directories containing includes to use during the synthesis. optionnal.
defines=Key=comma separated list of key=value preprocessor macro that will be defined during synthesis
keep_env=True/False Should the synthesis environment be kept after the synthesis is performed. Default is False.
ip_lib= Name of the ip library to use when exporting the ip (optionnal)
ip_version= Version to use when exporting the ip (optionnal)
ip_descr= description that will be exported along the vivado ip (optionnal)
ip_vendor= vendor name for the exported vivado ip (optionnal)
latency= ltency constraint for the initial design (optionnal)
hdl= verilog|VHDL
no_pipeline=True|False if true the design will not be pipelined (default false)
no_inline=True|False if true the function call will not be inlined (default false)
ctrl_hs=True|False Block protocol will be ap_ctrl_hs if true, ap_ctrl_none else. Default false.
exp_type=hls|syn|impl until which phase is the experiment to be performed

Running a one shot synthesis

python -m vhlsrs impl [expdescr.ini]

will implement the experiments described in expdescr.ini either with a latency constraint if given or with the minimal latency estimated by the tool.

Minimizing the latency

As vivado HLS is sometimes not very good at timing estimation, it can lead to overpessimistic latency.

the python -m vhlsrs minimize [expdescr.ini]

will attempt to find the lowest pipeline depth that still respect the clock constraints by dichotomy. If a latency constraint is provided, it will start the exploration with a designed with this constraint. Otherwise, it will start the exploration with the latency estimated by the tool.

Project details


Release history Release notifications | RSS feed

Download files

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

Source Distribution

vhlsrs-0.0.95.tar.gz (10.7 kB view details)

Uploaded Source

Built Distribution

vhlsrs-0.0.95-py3-none-any.whl (11.3 kB view details)

Uploaded Python 3

File details

Details for the file vhlsrs-0.0.95.tar.gz.

File metadata

  • Download URL: vhlsrs-0.0.95.tar.gz
  • Upload date:
  • Size: 10.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.6.1 requests/2.25.0 setuptools/49.2.1 requests-toolbelt/0.9.1 tqdm/4.54.1 CPython/3.9.1

File hashes

Hashes for vhlsrs-0.0.95.tar.gz
Algorithm Hash digest
SHA256 cf0d14e7af50454cc106538b48a3e809c3bed6922963a2d7d614e5c386705a4c
MD5 afb57932516b986169099f48be4aef0c
BLAKE2b-256 7bc63598fd718da593557659448efeecf1552d22b23ea4cd8ba45f205991443f

See more details on using hashes here.

File details

Details for the file vhlsrs-0.0.95-py3-none-any.whl.

File metadata

  • Download URL: vhlsrs-0.0.95-py3-none-any.whl
  • Upload date:
  • Size: 11.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.6.1 requests/2.25.0 setuptools/49.2.1 requests-toolbelt/0.9.1 tqdm/4.54.1 CPython/3.9.1

File hashes

Hashes for vhlsrs-0.0.95-py3-none-any.whl
Algorithm Hash digest
SHA256 41d9a620447a63c14280841fed725c97f0b1d95aef090bf982d25652db0da472
MD5 e70ffc9933bf5621c3dafbd354b41c45
BLAKE2b-256 be45c4d8a86439af205e5d0d590d80f28facf5bac34dbce4050eacbb178cc46a

See more details on using hashes here.

Supported by

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