Add Launch Bounds #144
Open
Add Launch Bounds #144
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
aacostadiaz
reviewed
Oct 21, 2024
aacostadiaz
reviewed
Oct 21, 2024
| @@ -53,6 +53,7 @@ if(NOT "${DPCPP_SYCL_ARCH}" STREQUAL "") | |||
| if("${DPCPP_SYCL_TARGET}" STREQUAL "nvptx64-nvidia-cuda") | |||
| list(APPEND DPCPP_FLAGS "-Xsycl-target-backend") | |||
| list(APPEND DPCPP_FLAGS "--cuda-gpu-arch=${DPCPP_SYCL_ARCH}") | |||
| list(APPEND DPCPP_FLAGS "-fgpu-inline-threshold=1000000;") | |||
There was a problem hiding this comment.
If this still needed if we use __attribute__((always_inline)) inline?
There was a problem hiding this comment.
I remember that back in portFFT, we had to increase the inline threshold, in combination with always_inline,
hence I went ahead with this, and have not tested it individually.
I can try without it, and will remove it if not required
There was a problem hiding this comment.
Ran without the inline-threshold...
it apparently is required
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Small PR to add Launch bounds to the cuda launch.
Also force inlines the
device_kernelwhich leads to nocallinstructions.This increases the performance in the benchmarks, making the SYCL path, roughly equal or better than the CUDA one