FLiT
FLiT copied to clipboard
PRUNERS
A project to quickly detect discrepancies in floating point computation across hardware, compilers, libraries and software.
Many projects have documented what others can do to contribute to a project. FLiT does not yet have this. - As of the addressing of issue #182, we will have...
This is related to issue #95. In this project, we use terms like "ground-truth", "good", "bad", "trouble", etc. These adjectives imply something about the correctness of the result. However, FLiT...
We would like to use pylint to evaluate how our code is doing stylistically. If there are rules we do not want to require, then we should create a `pylint.rc`...
The bisect biggest feature (with the `--biggest` flag) is implemented from issue #172 (pull reqest #197). But we are on a tight schedule for the next paper project. We want...
The `flit_bisect.py` file is starting to get unwieldy because it is getting too large at approximately 1800 lines. Much of that is doctest strings, but still. So one could either...
The values in the `DistributivityOfMultiplication` test within the litmus tests do not demonstrate variability. Since `inputGen` is not working, it is not easy to find variability inputs for this test...
Maybe not all printing should go through the logging module, but we have this situation that every message we print to the console, we are doing either the exact same...
Right now in the `Makefile` generated by `flit`, during the linking phase, the compiler flags are used. The question is do we want or need the compiler flags there at...
Broken off from issue #91, the bisect functionality is missing unit testing or verification testing.
Even after making the timing be more of a first-class citizen, it is clear that there is overhead unaccounted in the timing. For a completely empty tests, the timing overhead...