[GLUTEN-12569][CORE] Add enhanced DSv2 filter test suites for Spark 4.2 - #13023
Draft
manoj-ragupathy wants to merge 23 commits into
Draft
manoj-ragupathy wants to merge 23 commits into
manoj-ragupathy wants to merge 23 commits into
Conversation
Spark 4.2 added parameters to three case classes, which silently breaks positional extractor patterns: - CharType gained a collation parameter. - AppendDataExec / OverwriteByExpressionExec gained tableName and transaction. Match on type (and on the named field where a value is needed) instead, which is stable across all supported Spark versions. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: a4b4451c-d099-4965-871d-12781176d082 (cherry picked from commit b2a23c0)
Spark 4.2 hoisted checkAnswer, checkDataset, assertCached and friends out of QueryTest into a new QueryTestBase trait, which SharedSparkSession now mixes in. Because GlutenQueryTest declared its own copies of those members, every suite mixing both inherited conflicting definitions. Extending QueryTest (an abstract class up to 4.1, a trait in 4.2) makes Gluten's versions genuine overrides on every supported Spark version, and removes the duplicated-member conflict. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: a4b4451c-d099-4965-871d-12781176d082 (cherry picked from commit a0c9f51)
- Spark 4.2 bumps Netty to 4.2.13, which removed PlatformDependent.allocateDirectNoCleaner. Use ByteBuffer.allocateDirect, which works on every supported version. - Spark 4.2's spark-catalyst ships a patched copy of datasketches ResourceImpl, tripping the ban-duplicate-classes enforcer. Both artifacts are 'provided', so nothing extra is packaged; ignore it. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: a4b4451c-d099-4965-871d-12781176d082 (cherry picked from commit 5d4853d)
…rk 4.2
Spark 4.2 adds `getBinaryView` to `SpecializedGetters`, which makes
`PlaceholderRow` (and any other `InternalRowSparkCompatible` subclass) fail to
compile as non-abstract:
BatchCarrierRow.scala:110: error: class PlaceholderRow needs to be abstract.
Missing implementation for member of trait SpecializedGetters:
def getBinaryView(x$1: Int): org.apache.spark.unsafe.types.BinaryView
Handled with the existing cross-version seam: `SpecializedGettersSparkCompatible`
gains a `getBinaryView` stub and `InternalRowSparkCompatible` overrides it. The
`Nothing` return type conforms to whichever concrete type the running Spark
version declares, so one definition covers the whole 3.x/4.x matrix -- exactly
how getVariant/getGeography/getGeometry are already handled.
Verified with clean builds (stale target/ classes previously masked this):
spark-3.4, spark-3.5, spark-4.0, spark-4.1 and spark-4.2 all BUILD SUCCESS.
(cherry picked from commit 7210d79)
Spark 4.2 moved postDriverMetrics to SupportsCustomDriverMetrics and made the reported task metrics an explicit argument. Move doPostDriverMetrics out of the version-agnostic BatchScanExecTransformer into each BatchScanExecShim so the call site can differ per version. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: 8ae7d6bd-f561-417a-a635-40fe24dc67a5
Two Spark 4.2 renames that leak into version-agnostic code: - SampleExec.seed became Option[Long], resolved lazily via resolvedSeed. - KeyGroupedPartitioning was renamed to KeyedPartitioning. Both are now hidden behind SparkShims (getSampleSeed, isKeyGroupedPartitioning) so gluten-substrait and backends-velox stay version-agnostic. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: 8ae7d6bd-f561-417a-a635-40fe24dc67a5
Adds the `spark-4.2` Maven profile and a `shims/spark42` module cloned from `shims/spark41` with identifiers renamed. No behavioural change to existing Spark versions: `spark-4.2` is opt-in and no other profile is touched. - root pom.xml: new `spark-4.2` profile (spark.version 4.2.0, arrow 19.0.0 to track Spark 4.2's own Arrow, JDK 17+ / Scala 2.13 enforcers); `spark-4.2` added to the spark-version requireActiveProfile list. - shims/pom.xml: `spark-4.2` profile activates module `spark42`. - shims/spark42: 37 files, renamed spark41 -> spark42 / Spark41 -> Spark42. Shim selection needs no hardcoded version: SparkShimDescriptor.DESCRIPTOR is derived from the SPARK_COMPILE_VERSION build property. This commit does not compile on its own; the Spark 4.2 source-incompatibilities are fixed in the following commits so that each can be reverted independently. (cherry picked from commit c4e06e1)
Spark 4.2 reworked `MemoryStream` factory overloads: the `apply[A](numPartitions: Int)(encoder, sparkSession)` form no longer exists. The remaining explicit-session overload is `apply[A: Encoder](sparkSession: SparkSession, numPartitions: Int)`. Switch the Spark 4.2 shim to that overload; the encoder is supplied by the existing `A: Encoder` context bound. (cherry picked from commit 252505b)
Spark 4.2 dropped the trailing `profiler: Option[String]` parameter from both `PythonUDFRunner.writeUDFs` overloads. Drop the `None` argument in the Spark 4.2 shim. `ArgumentMetadata` itself is unchanged between 4.1 and 4.2. (cherry picked from commit 5b2558a)
Spark 4.2 reworked the columnar vector API: - `WritableColumnVector` gains three abstract members — `putBytes(int,int,ByteBuffer,int)`, `putShortsFromIntsLittleEndian(...)` and `getBytesAsBinaryView(...)`. Added stub overrides to `WritableColumnVectorShim` following the file's existing style. - `GeographyVal` and `GeometryVal` were removed in 4.2 and `ColumnVector` instead exposes `getBinaryView` (`ColumnVector.java:297`). `ArrowColumnarArray` drops the two geo overrides and delegates `getBinaryView` to `data`. Verified against the 4.1.1 and 4.2.0 source trees: GeographyVal/GeometryVal exist only in 4.1.1, BinaryView only in 4.2.0. These are independent of the SPJ port and were masked until Scala compiled, since scala-compile-first runs before javac. (cherry picked from commit 4d14a83)
Spark 4.2 replaced the storage-partitioned-join (SPJ) plumbing on `BatchScanExec`: - `StoragePartitionJoinParams` and the `KeyGroupedPartitionedScan` trait were removed; the constructor now takes `keyGroupedPartitioning: Option[Seq[Expression]]`. - `KeyGroupedPartitioning` was replaced by `KeyedPartitioning`, which exposes `partitionKeys: Seq[InternalRowComparableWrapper]`, `keyOrdering`, `keyRowOrdering` and `toGrouped` instead of `partitionValues`/`expressions`. - `filteredPartitions` became `Seq[Option[InputPartition]]`, runtime-filter pushdown moved to `PushDownUtils.pushRuntimeFilters`, and `postDriverMetrics` now requires the task-metrics argument. Changes are confined to `shims/spark42`; `gluten-core`, `gluten-substrait` and the backends are untouched. `Spark42Shims` and `BatchScanExecShim` keep the exact signatures of their spark41 counterparts, so the `keyGroupedPartitioning` propagation added in apache#12567 (ScanTransformerFactory) needs no change. KNOWN LIMITATION (Spark 4.2 only, marked DEGRADED in the code): `getCommonPartitionValues` returns `None`, and `orderPartitions` no longer applies `joinKeyPositions`, `reducers` or partially-clustered replication. Spark 4.2 moved these refinements off the scan node into `EnsureRequirements`/`GroupPartitionsExec`, so there is no scan-level equivalent to reproduce. The base fully-clustered SPJ path is unaffected. This is currently unreachable: the only callers of `getCommonPartitionValues` are `IcebergScanTransformer` and `PaimonScanTransformer`, and neither Iceberg nor Paimon publishes Spark 4.2 artifacts yet (both 404 on Maven Central, vs 200 for their 4.1 builds); both profiles are also activeByDefault=false. Gluten never populates `joinKeyPositions`/`reducers` anywhere. To be revisited when Iceberg/Paimon ship Spark 4.2 support. (cherry picked from commit 687d477)
Prepares CI for Spark 4.2 without yet adding the UT jobs (those follow with gluten-ut/spark42): - install-spark-resources.sh: new `4.2)` case installing Spark 4.2.0. As with 4.0/4.1, Spark does not publish a `-scala2.13` binary, so the 2.12 tarball is installed and `assembly/target/scala-2.12` is renamed to `scala-2.13`. Paths are spark42-specific (cf. apache#11973, where the 4.1 case wrongly referenced spark40). - Dockerfile.centos{8,9}-dynamic-build: bake Spark 4.2 resources into the CI image, which is where /opt/shims/sparkNN/spark_home comes from. - velox_backend_x86.yml: add a `shims42` change-detection flag, wired in all four required places (outputs block, both scheduled/dispatch flag loops, and the path-match line). Missing any one of these makes Spark 4.2 jobs silently skip while the PR still reports green. Verified: spark-4.2.0-bin-hadoop3.tgz and spark-4.2.0.tgz both return HTTP 200 from the mirror the installer actually uses. (cherry picked from commit 68e19d4)
Mirrors the existing spark-4.1 profile so the micro-benchmark tool can be built against Spark 4.2.0 / Scala 2.13. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: a4b4451c-d099-4965-871d-12781176d082 (cherry picked from commit e2540dd)
Implements the Spark 4.2 side of the shim contracts introduced earlier: doPostDriverMetrics on BatchScanExecShim, plus getSampleSeed and isKeyGroupedPartitioning on Spark42Shims. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: 8ae7d6bd-f561-417a-a635-40fe24dc67a5
…ark42 The CI license-header check inspects newly added files. Replace the short Apache form carried over from shims/spark41 with the canonical ASF header. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: 8ae7d6bd-f561-417a-a635-40fe24dc67a5
Pure rename so that git preserves file history for the Spark 4.2 test suites. gluten-ut/spark41 is restored unchanged in the next commit; this commit and the next are a history-preserving pair. NOTE FOR MERGE: this PR must be rebase-merged, not squash-merged, or the rename/restore pair collapses and `git log --follow` no longer works for the gluten-ut/spark42 files. (cherry picked from commit 1f06cb8)
Restores gluten-ut/spark41 unchanged after the previous rename commit. The rename/restore pair makes git attribute the gluten-ut/spark42 tree to the spark41 history, so `git log --follow` works on the new Spark 4.2 suites while Spark 4.1 support is left completely untouched. Must be rebase-merged, not squash-merged. (cherry picked from commit 52b3021)
Adapts the moved test module to Spark 4.2 and activates it: - gluten-ut/pom.xml: `spark-4.2` profile activates module `spark42`. - gluten-ut/spark42/pom.xml: artifactId `gluten-ut-spark42`, name updated. - GlutenPlanStabilitySuite: golden-file resource dir and the documented regeneration command now point at spark42 / `-Pspark-4.2` / /opt/shims/spark42/spark_home; the "replace with the previous version" hint now refers back to Spark 4.1. Only 2 of the 1853 files in the tree referenced the Spark version at all, so the suites themselves are carried over unchanged; Spark 4.2-specific test adjustments follow separately. (cherry picked from commit 65926da)
- KeyGroupedPartitioning -> KeyedPartitioning (the suite class itself keeps its upstream name); collectShuffles/collectAllShuffles are now protected upstream, so widen the overrides. - BroadcastHashJoinExec gained isSkewJoin and DataSourceV2ScanRelation gained pushedFilters; stop matching them positionally. - HashedRelationSuite and StreamingJoinSuite gained abstract members and were split upstream into on-heap/off-heap and VCF/non-VCF variants. Mirror that split and register the new suites in VeloxTestSettings. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: a4b4451c-d099-4965-871d-12781176d082 (cherry picked from commit 014892a)
Adds spark-test-spark42 and spark-test-spark42-slow, cloned from the spark41 jobs. The -Pdelta lane is omitted because delta-spark does not yet publish a Spark 4.2 build. These jobs only pass once the CI container image has been rebuilt with install-spark-resources.sh 4.2 baked in (dev/docker/Dockerfile.centos*). Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: a4b4451c-d099-4965-871d-12781176d082 (cherry picked from commit aa236e5)
…t/spark42 The CI license-header check inspects newly added files. Move the ASF header directly after the XML prologue, where the checker expects it. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: 8ae7d6bd-f561-417a-a635-40fe24dc67a5
…Spark 4.2 Wrap Spark 4.2's DataSourceV2EnhancedRuntimePartitionFilterSuite with the Gluten test trait and enable it for Velox. This adds 15 upstream tests for the iterative runtime partition-filter path used by BatchScanExec. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: 8ae7d6bd-f561-417a-a635-40fe24dc67a5 (cherry picked from commit ed12775)
…or Spark 4.2 Wrap Spark 4.2's DataSourceV2EnhancedPartitionFilterSuite and DataSourceV2EnhancedDeleteFilterSuite with the Gluten test trait and enable them for Velox, completing coverage of the three new enhanced DSv2 filter suites. This adds 37 upstream tests covering first/second-pass partition filter pushdown, untranslatable and nested partition predicates, post-scan filter retention, and delete filter handling. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com> Copilot-Session: 8ae7d6bd-f561-417a-a635-40fe24dc67a5 (cherry picked from commit 6a84498)
|
Run Gluten Clickhouse CI on x86 |
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
1 participant
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.
What changes are proposed in this pull request?
Final step of Spark 4.2.x support (#12569): widen coverage of the test module added in #13022.
Spark 4.2.0 adds a set of "Enhanced" DSv2 filter suites that exercise the redesigned SPJ / runtime-filtering paths. Since those paths are exactly what the 4.2 shim had to rewrite (see the
BatchScanExecShimrework and thegetCommonPartitionValuescaveat in #13021), they are the highest-value suites to run under Gluten rather than the ones most likely to already pass.GlutenDataSourceV2EnhancedPartitionFilterSuiteGlutenDataSourceV2EnhancedRuntimePartitionFilterSuiteGlutenDataSourceV2EnhancedDeleteFilterSuiteEach is the standard 7-line wrapper — extends the upstream suite, mixes in
GlutenSQLTestsTrait— and all three are registered inVeloxTestSettings. That registration is the part worth reviewing: a suite absent from it never runs at all.Part of #12569.
How was this patch tested?
BUILD SUCCESS. I confirmed by reflection over the compiled classes that the three suites actually inherit 28 / 15 / 9 test cases, so they are genuinely running upstream's tests and not silently empty.Same caveat as #13022: these are compile-verified only — no
libgluten.solocally — and they cannot run in CI until the image is rebuilt with the Spark 4.2 resources. I fully expect some of the 52 to needexcludeGlutenTest/ignoreentries once CI gives a real signal, and will follow up with those once the lanes are live.Was this patch authored or co-authored using generative AI tooling?
Generated-by: GitHub Copilot CLI (Claude Opus 5)