Repository navigation
perf: compute try-mode integer +, -, and * in one pass - #7698
Draft
mbutrovich wants to merge 2 commits into
Draft
mbutrovich wants to merge 2 commits into
mbutrovich wants to merge 2 commits into
Conversation
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Which issue does this PR close?
Closes #7695.
Rationale for this change
See #7695.
What changes are included in this PR?
+,-, and*onInt8throughInt64use a new kernel. One pass computes the wrapping values and whether any row overflowed. A second pass, which builds the validity of the rows that didn't overflow, runs only when one did.Columnar.rst).benches/checked_arithmetic.rsgets try-mode cases forInt32andInt64addandmul. The inputs have no overflow, one overflow per batch, random overflows in about 1% of rows, and overflows only under nulls. There are also a scalar case and a chainedtry_add(try_add(a, b), c). The ANSI cases are the control, since this PR doesn't touch them.How are these changes tested?
The tests are in
native/spark-expr/src/math_funcs/checked_arithmetic.rs.test_try_integer_arithmetic_matches_i128compares every row with exacti128arithmetic for each type and op. It sweeps pairs of edge values in the array, scalar, and null-scalar layouts. It also moves one overflowing row through sliced batches of 300 rows, with the row valid and with it null, and with overflowing values under the other nulls. It also passes againstmain, so it checks that the new kernel changes no result.Benchmarks
TODO
AI usage disclosure
I wrote this PR with help from Claude but understand and support the core approach, implementation, and test coverage.