David--Cléris Timothée

**Describe the motivation for the feature request** Hello, as in the MacOS CI, the include path to openMP must be passed down as a CXX flags when compiling with acpp....

Add source location info for accessor related warnings

10

**Describe the motivation for the feature request** Currently those two warnings ``` [AdaptiveCpp Warning] kernel_operation: Could not find embedded pointer in kernel blob for this requirement; do you have unnecessary...

[CUDA] std::bad_alloc on system with multiple GPUs

3

**Bug summary** On a DGX system with 4 A100 : Launching a basic SYCL kernel on the second GPU (order given by `.get_devices();`) create a crash with ``` terminate called...

[Cuda][Rocm?] memory leak in Ipc cache handling

6

## Background information Using openmpi direct GPU communication with CUDA, a memory growth is observed throughout the duration of the run. After tracing the issue using the ompi logs `--mca...

Crash when calling too many MPI_Probe ?

4

Hi, Yesterday I encountered issue with MPI_Probe(s) when used on a large scale (on Aurora with ). To clarify the context the error was reproduced with 1536 process, each sending/receiving...

[Algorithms] Data race in In-Place Exclusive Scan Algorithm

3

**Bug summary** When using `adaptivecpp::algorithms::exclusive_scan` i get on all backends either a segfault, either a corrupted output. Here is an example (index, acpp output, expected one) ``` Debug: 16374 81309...

Controlling the output width and height of matplotlib animations as mp4

5

Hi, First of all, thank you for this wonderful project — it has been extremely helpful for building our example documentation. We develop a hydrodynamics simulation code and include many...