Seth R. Johnson
Seth R. Johnson
@elliottbiondo Could you try running the `models/analysis/orange-complexity.ipynb` notebook from the `benchmarks` repo? Hopefully it's clear enough how to run, and under `Load results` you can see the relevant failures with:...
Weird, so maybe it's some sort of compiler optimization? 😕
Here's another false positive in #2048: https://app.codecov.io/gh/celeritas-project/celeritas/pull/2048 I *know* every loop here and conditional is being tested in PolySolid.test.cc. And what's up with the "missing" line of coverage? @pcanal Is...
We definitely do not have the time and overhead to wait for code coverage to give us back a bunch of false positive and then add ugly markup to the...
Aha, very sneaky! Good diagnosis.
Are there flags that can be added to silence *all* exception code at once, e.g., https://app.codecov.io/gh/celeritas-project/celeritas/blob/develop/src%2Fcorecel%2Fdata%2Fdetail%2FPinnedAllocatorImpl.cc ; or a way to have "don't test this line" flags propagate through macros...
@pcanal Here's what I really want to avoid: This PR (#2051) reports itself as having *failed*. Even though codecov is not one of the "required" checks, this failure obscures the...
Tagging #2093
@osanstrong I know you're allergic to LLM bots but the kind of refactoring to replace `0.25` -> `constexpr real_type quarter{0.25};` and `q2 * q2 -> ipow(q2)` is an ideal use...
@osanstrong @elliottbiondo When an ARM test fails when the x86_64 build succeeds, it's almost always in some complex numerical function (as is the case here) and indicates a numerical instability...