JiaKe

> Thanks!!! Could you provide the Apache Parquet file that has a problem? I'll debug it on my local machine. @kou You can have a try https://github.com/JkSelf/velox/commit/d815f3544a5d677338d2184a8f1de5dcf1d9d83d

@kou Here is the stacktrace: ``` 0# facebook::velox::parquet::PageReader::readPageHeader() in /mnt/DP_disk3/jk/projects/gluten/cpp/build/releases/libvelox.so 1# facebook::velox::parquet::PageReader::seekToPage(long) in /mnt/DP_disk3/jk/projects/gluten/cpp/build/releases/libvelox.so 2# facebook::velox::parquet::PageReader::rowsForPage(facebook::velox::dwio::common::SelectiveColumnReader&, bool, bool, folly::Range&, unsigned long const*&) in /mnt/DP_disk3/jk/projects/gluten/cpp/build/releases/libvelox.so 3# void facebook::velox::parquet::PageReader::readWithVisitor(facebook::velox::dwio::common::ColumnVisitor&) in /mnt/DP_disk3/jk/projects/gluten/cpp/build/releases/libvelox.so 4#...

@pedroerp @xiaoxmeng Can you help to review this PR? Thanks.

@zhouyuan I have fixed the failed unit tests.

> In long term, we need to implement the Spark way. Broadcast hashtable instead of raw table data. @FelixYBW Yes, we will support broadcasting the hash table approach after adding...

> @JkSelf will this solution helpful to the final solution? @FelixYBW Yes, the primary difference between Design 1 and Design 2 is the need for serialization and deserialization of Velox's...

@zhztheplayer Thank you for your review. It appears that the existing tests cover the broadcast changes introduced in this PR. I have added logging to the native broadcast hash table...