-
This ticket will track potential things to cleanup in our CI.
A lot of this can be based on what is present in https://github.com/G-Research/fasttrackml/tree/main/.github/workflows
## New Struct…
-
### Question
Hi m4rs-mt and team, kudos for this library!
I'm facing an issue trying to migrate a resampling computation from CPU to GPU.
I simplified the code below as much as I could to repro…
-
First, I appreciate the developpers for this amazing work.
I am currently using GPU to perform some image processing. I want to calculate the true image coordinates by iterating the value in order …
-
I expected the new `Int128` data type to just work, even if not necessarily performant. However, I have found an unexpected issue.
The following kernel does not work on Cuda, and generates the wron…
MoFtZ updated
7 months ago
-
Whenever I run a kernel with ILGPU, these exceptions are thrown:
> System.BadImageFormatException
HResult=0x8007000B
Message=Metadata header too small.
Source=System.Reflection.Metadata
…
-
**Version Used**:
.NET 8.0 (SDK 8.0.100, Windows x64 Installer)
**Steps to Reproduce**:
1. Create a `net8.0` project.
2. Create a C# file (`Context.cs`) containing a class having a Property…
MoFtZ updated
11 months ago
-
Not sure if it is a bug or a missing feature, but it appears to be impossible for multiple GPUs to use the same pinned memory on host.
Code:
```
var startTime = DateTime.Now;
…
-
In ILGPU 1.5.1 on .NET 7, I get a `NullReferenceException` when I try to pass an empty `ArrayView` to an OpenCL kernel. Below is a minimal reproducer:
```cs
using ILGPU;
using ILGPU.Runtime;
v…
-
It would be nice if the generic kernel load methods could work with type inference, so that you don't have to spell out the kernel argument types twice. Unfortunately, it seems method groups aren't i…
-
Hi, So according to Intel, in devices where the the cpu and integrated gpu use the same physical memory, one can simply set a buffer as "cpu" or as "gpu" without needing to copy to another buffer for …