Gabriel39 commented on code in PR #12214:
URL: https://github.com/apache/doris/pull/12214#discussion_r959194685
##########
be/src/vec/olap/vgeneric_iterators.cpp:
##########
@@ -382,21 +399,36 @@ Status VMergeIterator::next_batch(vectorized::Block*
block) {
_block_row_locations.resize(block_row_max);
}
size_t row_idx = 0;
+ VMergeIteratorContext* pre_ctx = nullptr;
while (block->rows() < block_row_max) {
if (_merge_heap.empty()) break;
auto ctx = _merge_heap.top();
_merge_heap.pop();
if (!ctx->need_skip()) {
Review Comment:
Is it possible to have different `need_skip` in a `VMergeIterator `? If all
of `ctx` in each `VMergeIterator` is `need_skip` or not `need_skip`, maybe we
should move this condition out of this while-loop
##########
be/src/vec/olap/vgeneric_iterators.cpp:
##########
@@ -382,21 +399,36 @@ Status VMergeIterator::next_batch(vectorized::Block*
block) {
_block_row_locations.resize(block_row_max);
}
size_t row_idx = 0;
+ VMergeIteratorContext* pre_ctx = nullptr;
while (block->rows() < block_row_max) {
if (_merge_heap.empty()) break;
auto ctx = _merge_heap.top();
Review Comment:
We should make this comparison inside one block as much as possible instead
of always using a heap to compare rows between different blocks.
--
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: commits-unsubscr...@doris.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@doris.apache.org
For additional commands, e-mail: commits-h...@doris.apache.org
