Engin Kayraklioglu
Engin Kayraklioglu
```chpl var Arr = [1,2,3]; on Locales[1] { writeln(Arr.localAccess[0]); } ``` does not produce any error message with checks enabled, but it should. Whereas ```chpl use BlockDist; var Arr =...
Resolves https://github.com/chapel-lang/chapel/issues/25727 The following code would fail because of an ALA bug: ```chpl use BlockDist; var A: [blockDist.createDomain({1..100})] int = 5; var db = {1..50}; var B: [blockDist.createDomain(db)] int =...
This came about when we were discussing some performance results @jeremiah-corrado is observing in his Navier-Stokes experiments. It looks like boundary condition updates can really use AVE support for Stencil-distributed...
While working on a GPU/non-GPU portability branch, I realized that `assertOnGpu` causes segfaults during codegen. This is because `chpl_assert_on_gpu` runtime function is in `runtime/include/gpu/chpl-gen-gpu-common.h`. That file isn't included unless you...
GPUs have several characteristics that can be queried by the application. But [there a lot of them](https://docs.nvidia.com/cuda/cuda-runtime-api/structcudaDeviceProp.html#structcudaDeviceProp), and most of them should be irrelevant to the user. Which ones are...
We have some logic in the compiler where GPU transforms could fail much later due to various reasons. An example is unsupported reductions. Here's an example with `&&` reduction: ```chpl...
Reported in https://chapel.discourse.group/t/complex-sine-broken-with-gpu-enabled-chapel/35562 GPU kernels don't support complex today, but the issue has nothing to do with GPU kernels, where: ```chpl use Math only sin; writeln(sin(0.0+1.0i)); ``` fails compilation with...
I've stumbled into this several times in the past. It could be really frustrating and confusing, but at the same time it could also be related to my setup as...
The runtime doesn't seem to report the correct number of devices in this config. ```chpl for loc in Locales do on loc { writeln(here, " ", here.gpus.size); } ``` reports...
This was asked in Gitter before, but I can't find the context. Today, I wanted to confirm `gpu.itersPerThread` behavior in the following code: ```chpl on here.gpus[0] { foreach i in...