Vendor numpy median, percentile, and quantile reduction tests#567
Open
brandon-b-miller wants to merge 30 commits intoNVIDIA:mainfrom
Open
Vendor numpy median, percentile, and quantile reduction tests#567brandon-b-miller wants to merge 30 commits intoNVIDIA:mainfrom
median, percentile, and quantile reduction tests#567brandon-b-miller wants to merge 30 commits intoNVIDIA:mainfrom
Conversation
Co-authored-by: Phillip Cloud <417981+cpcloud@users.noreply.github.com>
|
Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
brandon-b-miller
changed the title
median, percentile, and quantile reductionsmedian, percentile, and quantile reduction tests
brandon-b-miller
added
3 - Ready for Review
|
Auto-sync is disabled for ready for review pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
Contributor
Author
|
/ok to test |
Contributor
Greptile SummaryVendors comprehensive tests for Key changes:
Issues found:
Confidence Score: 3/5
Important Files Changed
|
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Additional Comments (1)
-
numba_cuda/numba/cuda/tests/cudapy/test_array_reductions.py, line 25-36 (link)logic: redundant config management -
NRTEnablingCUDATestCasealready managesself.old_nrt_settingandconfig.CUDA_ENABLE_NRTin its ownsetUp/tearDownthis creates duplicate state and could cause issues. remove the NRT-related lines:
7 files reviewed, 2 comments
Comment on lines
5506
to
+5531
| assert rettype.layout in "CF" | ||
| if arytype.layout == rettype.layout: | ||
| # Fast path: memcpy | ||
| cgutils.raw_memcpy( | ||
| builder, dest_data, src_data, ary.nitems, ary.itemsize, align=1 | ||
| ) | ||
|
|
||
| else: | ||
| src_strides = cgutils.unpack_tuple(builder, ary.strides) | ||
| dest_strides = cgutils.unpack_tuple(builder, ret.strides) | ||
| intp_t = context.get_value_type(types.intp) | ||
| src_strides = cgutils.unpack_tuple(builder, ary.strides) | ||
| dest_strides = cgutils.unpack_tuple(builder, ret.strides) | ||
| intp_t = context.get_value_type(types.intp) | ||
|
|
||
| with cgutils.loop_nest(builder, shapes, intp_t) as indices: | ||
| src_ptr = cgutils.get_item_pointer2( | ||
| context, | ||
| builder, | ||
| src_data, | ||
| shapes, | ||
| src_strides, | ||
| arytype.layout, | ||
| indices, | ||
| ) | ||
| dest_ptr = cgutils.get_item_pointer2( | ||
| context, | ||
| builder, | ||
| dest_data, | ||
| shapes, | ||
| dest_strides, | ||
| rettype.layout, | ||
| indices, | ||
| ) | ||
| builder.store(builder.load(src_ptr), dest_ptr) | ||
| with cgutils.loop_nest(builder, shapes, intp_t) as indices: | ||
| src_ptr = cgutils.get_item_pointer2( | ||
| context, | ||
| builder, | ||
| src_data, | ||
| shapes, | ||
| src_strides, | ||
| arytype.layout, | ||
| indices, | ||
| ) | ||
| dest_ptr = cgutils.get_item_pointer2( | ||
| context, | ||
| builder, | ||
| dest_data, | ||
| shapes, | ||
| dest_strides, | ||
| rettype.layout, | ||
| indices, | ||
| ) | ||
| builder.store(builder.load(src_ptr), dest_ptr) |
Contributor
There was a problem hiding this comment.
logic: removed memcpy fast path optimization when source and destination layouts match - this causes performance regression for same-layout copies
the original code had:
if arytype.layout == rettype.layout:
# Fast path: memcpy
cgutils.raw_memcpy(builder, dest_data, src_data, ary.nitems, ary.itemsize, align=1)consider restoring the fast path for performance
brandon-b-miller
added
2 - In Progress
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
Follow up #523