[LAD] Static verification of real-time safety

Mark D. McCurry mark.d.mccurry at gmail.com
Wed Jan 15 06:28:12 UTC 2014

On Tue, Jan 14, 2014 at 08:15:31PM -0500, David Robillard wrote:
> Ah, spfv, right.  That was it.

Yep, the github link is the most up-to-date place for the project, though it
certainly isn't perfect at this point.
For any C project it should behave as you might expect, though there are plenty
of things that confound it with larger C++ projects.
IIRC when it is run on a large portion of the non-* tools, it manages to choke on
some compiler generated code which violates some basic assumptions, like source
code actually existing in a file with some location information (there are some
weird autogenerated methods within the AST as well as many other odd edge cases
of the clang API).

This tool also seems to have a bit of a bad habit of bitrotting, but the clang
developers might have decided to slow down the pace of changing the exposed API
somewhat since I have last checked.
If there are any problems with the current version of the code just open up an
issue and I should be able to at least put out any fires.
Improving what's there is also a possibility, though it depends on how resistant
the clang API is with respect to some issues.
Tracking lambda functions for instance is a bit of a headache as traversing the
AST in search of lambda expressions is a lot tricker than one might expect at
first glance.

Thanks for the continued interest,
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 490 bytes
Desc: not available
URL: <http://lists.linuxaudio.org/pipermail/linux-audio-dev/attachments/20140115/30da3fdb/attachment.pgp>

More information about the Linux-audio-dev mailing list