This repository has been archived by the owner on Mar 28, 2023. It is now read-only.
[SYCL] Add tests for span reductions #1009
Merged
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
Tests the following: - Only spans with static extents are available - Kernels with a single reduction span work - Kernels with a reduction pack containing a span work Signed-off-by: John Pennycook <john.pennycook@intel.com>
v-klochkov
reviewed
May 3, 2022
| Q.parallel_for(range<1>{N}, [=](id<1> I) { Output[I] = Identity; }).wait(); | ||
|
|
||
| // Perform generalized "histogram" with N bins | ||
| // TODO: Test Q.parallel_for when code_location is fixed |
There was a problem hiding this comment.
The q.parallel_for issue is already resolved: intel/llvm#4352
There was a problem hiding this comment.
Ah, thanks. I've updated the test to run with both handler and queue in 1c63fab.
v-klochkov
reviewed
May 3, 2022
SYCL/Reduction/reduction_span.cpp
Outdated
| // `Group algorithms are not supported on host device.` on Nvidia. | ||
| // XFAIL: hip_nvidia | ||
|
|
||
| // TODO: test disabled due to sporadic fails in level_zero:gpu RT. |
There was a problem hiding this comment.
Maybe try turning it on initially? I see many fixes and attempts to fix/address those sporadic fails are done, but I did not see tests to be enabled after that.
We can disable the test when/if it starts failing.
There was a problem hiding this comment.
Ok. I've reenabled these tests for now; see a7eb1e1.
|
The tests are good. |
Signed-off-by: John Pennycook <john.pennycook@intel.com>
v-klochkov
approved these changes
May 3, 2022
Comment on lines
+204
to
+223
| test<16, int, std::plus<int>, sycl::nd_range<1>, submission_mode::handler>( | ||
| Q, {24, 8}, 0, 1); | ||
| test<16, float, std::plus<float>, sycl::nd_range<1>, | ||
| submission_mode::handler>(Q, {24, 8}, 0, 1); | ||
| test<16, int, std::multiplies<int>, sycl::nd_range<1>, | ||
| submission_mode::handler>(Q, {24, 8}, 1, 2); | ||
| test<16, int, std::bit_or<int>, sycl::nd_range<1>, submission_mode::handler>( | ||
| Q, {24, 8}, 0, 1); | ||
| test<16, CustomType, CustomBinaryOperation, sycl::nd_range<1>, | ||
| submission_mode::handler>(Q, {24, 8}, CustomType{0}, CustomType{1}); | ||
| test<16, int, std::plus<int>, sycl::nd_range<1>, submission_mode::queue>( | ||
| Q, {24, 8}, 0, 1); | ||
| test<16, float, std::plus<float>, sycl::nd_range<1>, submission_mode::queue>( | ||
| Q, {24, 8}, 0, 1); | ||
| test<16, int, std::multiplies<int>, sycl::nd_range<1>, | ||
| submission_mode::queue>(Q, {24, 8}, 1, 2); | ||
| test<16, int, std::bit_or<int>, sycl::nd_range<1>, submission_mode::queue>( | ||
| Q, {24, 8}, 0, 1); | ||
| test<16, CustomType, CustomBinaryOperation, sycl::nd_range<1>, | ||
| submission_mode::queue>(Q, {24, 8}, CustomType{0}, CustomType{1}); |
There was a problem hiding this comment.
Minor comment:
It seems like the lines 204-213 and 214-223 have only 1 difference - the submission_mode.
If so, it could be easier/shorter to add 2 lines near the line 182, i.e.
template <size_t N, typename T, typename BinaryOperation, typename Range>
void test(queue Q, Range Rng, T Identity, T Value) {
test1<N, T, BinaryOperation, Range, submission_mode::handler>(Q, Rng, Identity, Value);
test2<N, T, BinaryOperation, Range, , submission_mode::handler>(Q, Rng, Identity, Value);
test1<N, T, BinaryOperation, Range, , submission_mode::queue>(Q, Rng, Identity, Value);
test2<N, T, BinaryOperation, Range, submission_mode::queue>(Q, Rng, Identity, Value);
}
myler
pushed a commit
to myler/llvm-test-suite
that referenced
this pull request
Jun 17, 2022
* [SYCL] Add tests for span reductions Tests the following: - Only spans with static extents are available - Kernels with a single reduction span work - Kernels with a reduction pack containing a span work * [SYCL] Reenable sporadically failing tests
aelovikov-intel
pushed a commit
to aelovikov-intel/llvm
that referenced
this pull request
Mar 27, 2023
* [SYCL] Add tests for span reductions Tests the following: - Only spans with static extents are available - Kernels with a single reduction span work - Kernels with a reduction pack containing a span work * [SYCL] Reenable sporadically failing tests
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
Tests the following:
Tests intel/llvm#6019
Signed-off-by: John Pennycook john.pennycook@intel.com