feat: Update merge_hll to use HSA with Memory Pool #15642
+223
−135
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
✅ Deploy Preview for meta-velox canceled.
|
meta-cla
bot
added
the
CLA Signed
|
@natashasehgal has exported this pull request. If you are a Meta employee, you can view the originating Diff in D87932002. |
…ator#15641) Summary: Add MemoryPool support to SimpleFunction `initialize()` method to enable memory tracking for scalar functions (e.g merge_hll) that use HashStringAllocator. Some types that require this include Hyperloglog, KHyperloglog, SetDigest, SfmSketch etc. **Main Change** - Modified SimpleFunctionAdapter to pass MemoryPool from ExprCompiler to UDF initialize() - Two signatures in UDFHolder::initialize() for backward compatibility - - New signature: initialize(vector<TypePtr>, QueryConfig, MemoryPool*, constantArgs...) - Old signature: initialize(vector<TypePtr>, QueryConfig, constantArgs...) **Issue:** The initial straightforward implementation failed because some UDFs (e.g., Spark's https://www.internalfb.com/code/fbsource/fbcode/velox/functions/sparksql/Rand.h?lines=30 function) use template initialize() methods, that was matching fourth MemoryPool argument: ```cpp template <typename TInput> void initialize(vector<TypePtr>&, QueryConfig&, const TInput* param); V2: Contains Template Detection via SFINAE Compile-time template detection that probes initialize() with a DummyProbeType. If the method accepts this unrelated type, it's a template and is excluded from receiving MemoryPool. Differential Revision: D87932003
…#15642) Summary: Use SimpleFunction instead of VectorFunction in `merge_hll` utilizing the new MemoryPool infrastructure in Simple Function The HashStringAllocator is created from MemoryPool, to get performance of HSA with serialial processing, and use query's memory from MemoryPool for proper memory tracking Simple function is cleaner and more maintainable code with auto null handling Differential Revision: D87932002
natashasehgal
force-pushed
the
export-D87932002
branch
from
9606dad to
4298ce7
Compare
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.
Summary:
Use SimpleFunction instead of VectorFunction in
merge_hllutilizing the new MemoryPool infrastructure in Simple FunctionThe HashStringAllocator is created from MemoryPool, to get performance of HSA with serialial processing, and use query's memory from MemoryPool for proper memory tracking
Simple function is cleaner and more maintainable code with auto null handling
Differential Revision: D87932002