From e03d0b7d216f02e489236978a9641c43b9d6d00c Mon Sep 17 00:00:00 2001
From: Ke Jia <ke.a.jia@intel.com>
Date: Mon, 22 Sep 2025 17:13:11 +0800
Subject: [PATCH 1/2] Call HdfsFileSystem#close() method to close the hdfs
 connection prior to the jvm termination

---
 cpp/velox/compute/VeloxBackend.cc | 24 ++++++++++++++++++++++++
 cpp/velox/compute/VeloxBackend.h  | 20 +-------------------
 2 files changed, 25 insertions(+), 19 deletions(-)

diff --git a/cpp/velox/compute/VeloxBackend.cc b/cpp/velox/compute/VeloxBackend.cc
index 04faab8f3ac..b20be533381 100644
--- a/cpp/velox/compute/VeloxBackend.cc
+++ b/cpp/velox/compute/VeloxBackend.cc
@@ -331,4 +331,28 @@ VeloxBackend* VeloxBackend::get() {
   }
   return instance_.get();
 }
+
+void VeloxBackend::tearDown() {
+  for (const auto& [_, filesystem] : facebook::velox::filesystems::registeredFilesystems) {
+    filesystem->close();
+  }
+  // Destruct IOThreadPoolExecutor will join all threads.
+  // On threads exit, thread local variables can be constructed with referencing global variables.
+  // So, we need to destruct IOThreadPoolExecutor and stop the threads before global variables get destructed.
+  ioExecutor_.reset();
+  globalMemoryManager_.reset();
+
+  // dump cache stats on exit if enabled
+  if (dynamic_cast<facebook::velox::cache::AsyncDataCache*>(asyncDataCache_.get())) {
+    LOG(INFO) << asyncDataCache_->toString();
+    for (const auto& entry : std::filesystem::directory_iterator(cachePathPrefix_)) {
+      if (entry.path().filename().string().find(cacheFilePrefix_) != std::string::npos) {
+        LOG(INFO) << "Removing cache file " << entry.path().filename().string();
+        std::filesystem::remove(cachePathPrefix_ + "/" + entry.path().filename().string());
+      }
+    }
+    asyncDataCache_->shutdown();
+  }
+}
+
 } // namespace gluten
diff --git a/cpp/velox/compute/VeloxBackend.h b/cpp/velox/compute/VeloxBackend.h
index be923c22d16..94e7ec93fba 100644
--- a/cpp/velox/compute/VeloxBackend.h
+++ b/cpp/velox/compute/VeloxBackend.h
@@ -56,25 +56,7 @@ class VeloxBackend {
     return globalMemoryManager_.get();
   }
 
-  void tearDown() {
-    // Destruct IOThreadPoolExecutor will join all threads.
-    // On threads exit, thread local variables can be constructed with referencing global variables.
-    // So, we need to destruct IOThreadPoolExecutor and stop the threads before global variables get destructed.
-    ioExecutor_.reset();
-    globalMemoryManager_.reset();
-
-    // dump cache stats on exit if enabled
-    if (dynamic_cast<facebook::velox::cache::AsyncDataCache*>(asyncDataCache_.get())) {
-      LOG(INFO) << asyncDataCache_->toString();
-      for (const auto& entry : std::filesystem::directory_iterator(cachePathPrefix_)) {
-        if (entry.path().filename().string().find(cacheFilePrefix_) != std::string::npos) {
-          LOG(INFO) << "Removing cache file " << entry.path().filename().string();
-          std::filesystem::remove(cachePathPrefix_ + "/" + entry.path().filename().string());
-        }
-      }
-      asyncDataCache_->shutdown();
-    }
-  }
+  void tearDown();
 
  private:
   explicit VeloxBackend(

From 907fdd5bfd417c1a93e069d1e7ba3eab66c5e70b Mon Sep 17 00:00:00 2001
From: Ke Jia <ke.a.jia@intel.com>
Date: Mon, 22 Sep 2025 19:37:17 +0800
Subject: [PATCH 2/2] Fix the compile issue

---
 cpp/velox/compute/VeloxBackend.cc                   | 4 ++++
 cpp/velox/operators/writer/VeloxParquetDataSource.h | 1 -
 2 files changed, 4 insertions(+), 1 deletion(-)

diff --git a/cpp/velox/compute/VeloxBackend.cc b/cpp/velox/compute/VeloxBackend.cc
index b20be533381..e26cd852b61 100644
--- a/cpp/velox/compute/VeloxBackend.cc
+++ b/cpp/velox/compute/VeloxBackend.cc
@@ -44,6 +44,7 @@
 #include "velox/connectors/hive/HiveDataSource.h"
 #include "velox/connectors/hive/storage_adapters/abfs/RegisterAbfsFileSystem.h" // @manual
 #include "velox/connectors/hive/storage_adapters/gcs/RegisterGcsFileSystem.h" // @manual
+#include "velox/connectors/hive/storage_adapters/hdfs/HdfsFileSystem.h"
 #include "velox/connectors/hive/storage_adapters/hdfs/RegisterHdfsFileSystem.h" // @manual
 #include "velox/connectors/hive/storage_adapters/s3fs/RegisterS3FileSystem.h" // @manual
 #include "velox/dwio/orc/reader/OrcReader.h"
@@ -333,9 +334,12 @@ VeloxBackend* VeloxBackend::get() {
 }
 
 void VeloxBackend::tearDown() {
+#ifdef ENABLE_HDFS
   for (const auto& [_, filesystem] : facebook::velox::filesystems::registeredFilesystems) {
     filesystem->close();
   }
+#endif
+
   // Destruct IOThreadPoolExecutor will join all threads.
   // On threads exit, thread local variables can be constructed with referencing global variables.
   // So, we need to destruct IOThreadPoolExecutor and stop the threads before global variables get destructed.
diff --git a/cpp/velox/operators/writer/VeloxParquetDataSource.h b/cpp/velox/operators/writer/VeloxParquetDataSource.h
index 8bbbd5ef061..4dde5c69aa8 100644
--- a/cpp/velox/operators/writer/VeloxParquetDataSource.h
+++ b/cpp/velox/operators/writer/VeloxParquetDataSource.h
@@ -39,7 +39,6 @@
 #include "velox/connectors/hive/storage_adapters/gcs/GcsFileSystem.h"
 #endif
 #ifdef ENABLE_HDFS
-#include "velox/connectors/hive/storage_adapters/hdfs/HdfsFileSystem.h"
 #include "velox/connectors/hive/storage_adapters/hdfs/HdfsUtil.h"
 #endif
 #ifdef ENABLE_ABFS