benjmarshall / hlsclt Goto Github PK

When configuration options are specified as blank strings "" the program should gracefully handle not merging them into the default confguration structure. This currently throws an error:

NameError: name 'ConfigError' is not define

Currently source and test bench directories must be specified along with individual files. This seems like a duplicate requirement to users.

The tool also forces users into maintaining a separate source directory which they may not want.

This often means that relative paths to include files must be maintained in multiple source files.

A neater solution must exist so this issue is here to discuss the best way forward towards something that makes sense to the various users whilst not making the tool onerously complex.

See #15 for additional context for this issue.

If environment is set in the calling process/cmd the shell=True is needed in build_commands.py

hls_processs = subprocess.run(["vivado_hls", "-f", "run_hls.tcl"],shell=True)

The keep argument currently forces the use of solution1 as the solution folder which is unintended. This should be updated to use the current highest solution number.

Subprocess.run() is now used to call the vivado hls process. This breaks compatibilty with python 2.

Hello,

I'm a happy user of hlsclt. However, I tried to port a project of mine to a bigger part,
part_name = "xczu9eg-ffvc900-2i"
and now the build csim (and every other task) fails with

can't read "lib_specifications": no such variable
    while executing
"source [lindex $::argv 1] "
    ("uplevel" body line 1)
    invoked from within
"uplevel \#0 { source [lindex $::argv 1] } "

I can't make anything of this error. Developing under the GUI works for this part, too (no license issues). Google didn't help. Strangely, a 442-byte binary file named 'c' remains in the work directory after the unsuccessful run.
HLS 2017.4 here.

Best,
Peter

The recent changes trigger a small omission in the build steps. Apparently,
do_end_build_stuff() assumes that the source files are in src/ instead of using the src_dir_name configuration setting.

See this (src_dir_name = "." was specified in the hls_config.py)

Copying the source and config files to solution1
Traceback (most recent call last):
File "/usr/local/bin/hlsclt", line 11, in
sys.exit(cli())
File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 722, in call
return self.main(*args, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 697, in main
rv = self.invoke(ctx)
File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 1066, in invoke
return _process_result(sub_ctx.command.invoke(sub_ctx))
File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 1093, in invoke
return _process_result(rv)
File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 1031, in _process_result
**ctx.params)
File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 535, in invoke
return callback(*args, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/click/decorators.py", line 17, in new_func
return f(get_current_context(), *args, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/hlsclt/build_commands/build_commands.py", line 164, in build_end_callback
do_end_build_stuff(ctx,sub_command_returns,report)
File "/usr/local/lib/python2.7/dist-packages/hlsclt/build_commands/build_commands.py", line 122, in do_end_build_stuff
shutil.copytree("src", destiny_src)
File "/usr/lib/python2.7/shutil.py", line 171, in copytree
names = os.listdir(src)
OSError: [Errno 2] No such file or directory: 'src'

Ubuntu 16.04, HLS 2018.3

With a few recent enhancement suggestions this project will probably grow to a size where some slightly more formal testing will become increasingly useful.

Adding basic tests for key functionality would be a good start. I think initially aiming for a quicker method of testing nothing is broken by 'simple' enhancement PRs.

e.g. --std=c++11

say, I want to add the following command into hls_config.py But, how?
config_dataflow -strict_mode (off | error | warning)

A bug in the function used to determine the solution number to be used for a build. If the - k option is passed the determined current highest existing solution number is incremented by 1. If no solutions exist, ie a fresh project, then passing the - k option creates the first solution as solution 2. This has a knock on effect of incorrectly counting the solution num for the next build call.

how to translate the following into hls_config.py
add_files -tb tb/xxx.c -cflags -lnetpbm

Hi,

would be possible add -cflags to tb files?