Skip to content

Avoid using sycl/ext/intel/math.hpp on non-Intel devices #2439

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 5 commits into from
May 8, 2025
Merged

Avoid using sycl/ext/intel/math.hpp on non-Intel devices #2439

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
e429dfa
Avoid using sycl::ext::intel::math::cyl_bessel_i0 on CUDA
vlad-perevezentsev May 8, 2025
c34a498
Update CHANGELOG.md
vlad-perevezentsev May 8, 2025
a9032cc
Use sycl::ext::intel::math only for Intel devices
vlad-perevezentsev May 8, 2025
94fdd5b
Update CHANGELOG.md
vlad-perevezentsev May 8, 2025
a4d5409
Address remark
vlad-perevezentsev May 8, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,6 +28,7 @@ This release achieves 100% compliance with Python Array API specification (revis
* Updated Python Array API specification version supported to `2024.12` [#2416](https://github.com/IntelPython/dpnp/pull/2416)
* Removed `einsum_call` keyword from `dpnp.einsum_path` signature [#2421](https://github.com/IntelPython/dpnp/pull/2421)
* Changed `"max dimensions"` to `None` in array API capabilities [#2432](https://github.com/IntelPython/dpnp/pull/2432)
* Added CUDA compatibility for `dpnp.i0` and `dpnp.kaiser` functions [#2439](https://github.com/IntelPython/dpnp/pull/2439)

### Fixed

Expand Down
9 changes: 7 additions & 2 deletions dpnp/backend/extensions/window/kaiser.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,11 @@
#define __SYCL_COMPILER_BESSEL_I0_SUPPORT 20241208L
#endif

#if __SYCL_COMPILER_VERSION >= __SYCL_COMPILER_BESSEL_I0_SUPPORT
// The <sycl/ext/intel/math.hpp> header and related types like
// _iml_half_internal are not compatible with NVIDIA GPUs and cause compilation
// errors when building with -fsycl-targets=nvptx64-nvidia-cuda.
#if !defined(__NVPTX__) && \
(__SYCL_COMPILER_VERSION >= __SYCL_COMPILER_BESSEL_I0_SUPPORT)
#include <sycl/ext/intel/math.hpp>
#endif

Expand Down Expand Up @@ -74,7 +78,8 @@ class KaiserFunctor

void operator()(sycl::id<1> id) const
{
#if __SYCL_COMPILER_VERSION >= __SYCL_COMPILER_BESSEL_I0_SUPPORT
#if !defined(__NVPTX__) && \
(__SYCL_COMPILER_VERSION >= __SYCL_COMPILER_BESSEL_I0_SUPPORT)
using sycl::ext::intel::math::cyl_bessel_i0;
#else
using dpnp::kernels::i0::impl::cyl_bessel_i0;
Expand Down
9 changes: 7 additions & 2 deletions dpnp/backend/kernels/elementwise_functions/i0.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,11 @@
#define __SYCL_COMPILER_BESSEL_I0_SUPPORT 20241208L
#endif

#if __SYCL_COMPILER_VERSION >= __SYCL_COMPILER_BESSEL_I0_SUPPORT
// The <sycl/ext/intel/math.hpp> header and related types like
// _iml_half_internal are not compatible with NVIDIA GPUs and cause compilation
// errors when building with -fsycl-targets=nvptx64-nvidia-cuda.
#if !defined(__NVPTX__) && \
(__SYCL_COMPILER_VERSION >= __SYCL_COMPILER_BESSEL_I0_SUPPORT)
#include <sycl/ext/intel/math.hpp>
#endif

Expand Down Expand Up @@ -253,7 +257,8 @@ struct I0Functor

resT operator()(const argT &x) const
{
#if __SYCL_COMPILER_VERSION >= __SYCL_COMPILER_BESSEL_I0_SUPPORT
#if !defined(__NVPTX__) && \
(__SYCL_COMPILER_VERSION >= __SYCL_COMPILER_BESSEL_I0_SUPPORT)
using sycl::ext::intel::math::cyl_bessel_i0;
#else
using impl::cyl_bessel_i0;
Expand Down
Loading