chapel
chapel copied to clipboard
a Productive Parallel Programming Language
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...
[This is a longstanding issue that I'm capturing on GitHub since I couldn't find it here] When iterating over multiple multidimensional arrays using zippered iteration, performance currently falls off a...
### Summary of Feature **Description:** Currently, there are many choices to be made when writing `init=` procs. These decisions are related to code cleanliness, semantics, and performance. I think it...
### Summary of Feature **Description:** Being able to have chapel/python file reading and writing via Python's pickle system. **Is this issue currently blocking your progress?** Yes
### Summary of Problem **Description:** I'd like to `min reduce` arrays of integers (so that the result is an array of minimums in each position). I am led to think...
### Summary of Feature **Description:** Right now, there is no way to open/save png or jpeg images. The image module only supports bmp images.
### Summary of Feature **Description:** It would be really useful if chapel had an IO utility for reading and writing from zip and tar.gz files.
### Summary of Problem **Description:** I've run into an internal compiler error when calling walkDirs. It turns out that I wanted to call findFiles rather than walkDirs anyway. But I...
PR #25636 adds a `proc sort` that accepts a `region` argument to indicate that only a portion of the array should be sorted. This is intended as a performance improvement...
Today, we support a `CHPL_TARGET_PLATFORM` variable that sometimes tells us a lot about the target platform if it's something specific like an HPE Cray EX or Cray XC system, but...