pitrou commented on code in PR #45202:
URL: https://github.com/apache/arrow/pull/45202#discussion_r1908280590
##########
cpp/src/parquet/column_writer.cc:
##########
@@ -1468,42 +1468,43 @@ class TypedColumnWriterImpl : public ColumnWriterImpl,
public TypedColumnWriter<
// which case we call back to the dense write path)
std::shared_ptr<::arrow::Array> preserved_dictionary_;
- int64_t WriteLevels(int64_t num_values, const int16_t* def_levels,
+ int64_t WriteLevels(int64_t num_levels, const int16_t* def_levels,
const int16_t* rep_levels) {
+ // Update histograms now, to maximize cache efficiency.
Review Comment:
I'm not sure and I don't see any difference on our mini-benchmark. But
`WriteDefinitionLevels` also writes to a destination buffer, so perhaps some of
the levels will be evicted from the CPU cache.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
