Skip to content

Conversation

@aelovikov-intel
Copy link
Contributor

No description provided.

@aelovikov-intel aelovikov-intel requested a review from a team as a code owner June 16, 2022 22:42
Apparently, different versions do it differently...
size_t OneElemSize =
IsTreeReduction ? sizeof(typename Reduction::result_type) : 0;
uint32_t NumConcurrentWorkGroups =
#ifdef __SYCL_REDUCTION_NUM_CONCURRENT_WORKGROUPS
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why did you remove this macro read here?
Moving it to reduction.cpp is not the same as having it here.
The idea of this macro was to let user (and SYCL developers) experiment with it during compilation of the test/bench.
clang++ -fsycl my_bench.cpp -D__SYCL_REDUCTION_NUM_CONCURRENT_WORKGROUPS=16

Moving it to reduction.cpp means that variable would be ignored during my_bench.cpp compilation. Changing that var would require rebuild of libsycl.so

@aelovikov-intel
Copy link
Contributor Author

@intel/llvm-gatekeepers , PR is ready for the merge.

@pvchupin pvchupin merged commit 2475844 into intel:sycl Jun 21, 2022
@aelovikov-intel aelovikov-intel deleted the reduction branch June 21, 2022 22:44
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants