-
**Describe the bug**
I wrote some application, comparing SYCL and OpenMP offloading
performance using the oneAPI products on devcloud where everything
works happily with OpenCL. I am failing to
r…
-
Hi everyone,
we are currently investigating the use of AdaptiveCPP for our use case of implementing a highly parallelizable algorithm. So far, we are truly impressed by this project, great work and b…
-
The [C++ draft for cmath](https://eel.is/c++draft/c.math#cmath.syn-3) states that:
> For each function with at least one parameter of type floating-point-type other than abs, the implementation als…
-
The SYCL specification relies on the `std::size_t` and `std::ptrdiff_t` type aliases, but does not currently impose requirements that they be defined on the host and device to either the same type, or…
-
Some comments, questions and feedback about the SYCL USM proposal, very interesting work!
### 1. Use context instead of device?
All malloc functions target a specific device in the system. Howe…
-
Imagine a device-side function with the following signature:
```c++
void foo(int* vec);
```
I don't know if `vec` comes from global, local, constant or private memory. However, inside `foo` I'd …
-
The Rodinia benchmarks have been ported to SYCL and are available publicly at: https://github.com/zjin-lcf/Rodinia_SYCL.
I have forked the repository and modified the Makefiles to be able to compile …
-
I'm currently trying to assert the status of expectation we can have on this setup for DeepSpeech, relying on TensorFlow with ComputeCpp. I have been able to cross-build the driver, and most of the `T…
-
Some math manglings from xocc's SPIR libraries appear to be incorrect or missing. In the sense that math functions correctly translated to their SPIR mangled names (found in: https://github.com/Khrono…
-
## Description
Replace all HIP calls with `hipper` calls. Continues #427.
## Motivation and context
HIP is complex, out of our control, gets in the way, and often breaks things. There is …