-
# Describe the bug
`std::atomic::fetch_add` is supposed to always work, without undefined behavior. If you have floating-point exceptions unmasked, and `fetch_add` causes a floating-point exception,…
-
fpp-check currently accepts this code:
```
constant a = 1e10000
```
It should emit an error, because 1e10000 is not a finite F64 value.
-
The team has done some comparisons to show that floating-point processing is faster than doubles (on all platforms and compilers, ranging from 25% to roughly 50% faster).
There are still some prom…
kcgen updated
2 months ago
-
### Bug Description
When executing the `fsqrt.d` instruction on a double-precision floating-point value representing positive infinity (0x7ff0000000000000), the Overflow (OF) flag in the fflags reg…
-
### Feature
Generated functions can (presently) only accept and return floating-point values for scalar or matrix arguments. There are use cases for passing integer and boolean values.
For exam…
-
[This](https://github.com/KhronosGroup/OpenGL-Registry/pull/534#issuecomment-1297111898) comment has a file that mentions all the cases I found but with links to related spec/extension.
---
> …
-
I tried this code:
```rust
#[kani::proof]
fn test_rem() {
let dividend: f32 = kani::any::().into();
let divisor: f32 = kani::any::().into();
kani::assume(divisor != 0.0);
let …
-
### 🐛 Describe the bug
A simple slice assignment called inside a module causes a floating point exception when the module is exported into the ONNX via `torch.onnx.export`.
* The full error messag…
-
To replicate: Go to star system referenced in the attached file. Target the first planet orbiting the black hole (BH1b1), hit autopilot, and "enter low orbit around BH1b1".
I re-implemented the std…
-
## Issue Details
I tried to tinker around with the examples of cgal and when I modified statements like
## Source Code
```c++
typedef CGAL::Box_intersection_d::Box_d Box;
```
to use `long in…