github-actions[bot] commented on code in PR #33397:
URL: https://github.com/apache/doris/pull/33397#discussion_r1556335915


##########
be/src/vec/exec/skewed_partition_rebalancer.cpp:
##########
@@ -45,13 +45,40 @@
           _partition_data_size_since_last_rebalance_per_task(partition_count, 
0),
           _estimated_task_bucket_data_size_since_last_rebalance(task_count * 
task_bucket_count, 0),
           _partition_assignments(partition_count) {
+    if (task_addresses != nullptr) {
+        CHECK(task_addresses->size() == task_count);
+        _task_addresses = *task_addresses;
+        for (int i = 0; i < _task_addresses.size(); ++i) {
+            auto it = 
_assigned_address_to_task_buckets_num.find(_task_addresses[i]);
+            if (it == _assigned_address_to_task_buckets_num.end()) {
+                
_assigned_address_to_task_buckets_num.insert({_task_addresses[i], 0});
+            }
+        }
+    } else {
+        
_assigned_address_to_task_buckets_num.insert({TASK_BUCKET_ADDRESS_NOT_SET, 0});
+    }
+
     std::vector<int> task_bucket_ids(task_count, 0);
 
     for (int partition = 0; partition < partition_count; partition++) {
         int task_id = partition % task_count;
         int bucket_id = task_bucket_ids[task_id]++ % task_bucket_count;
-        TaskBucket task_bucket(task_id, bucket_id, task_bucket_count);
+        TaskBucket task_bucket(
+                task_id, bucket_id, task_bucket_count,
+                (_task_addresses.empty()) ? TASK_BUCKET_ADDRESS_NOT_SET : 
_task_addresses[task_id]);
         _partition_assignments[partition].emplace_back(std::move(task_bucket));
+
+        for (int i = 0; i < _partition_assignments[partition].size(); ++i) {
+            auto it = _assigned_address_to_task_buckets_num.find(
+                    _partition_assignments[partition][i].task_address);
+            if (it != _assigned_address_to_task_buckets_num.end()) {
+                
_assigned_address_to_task_buckets_num[_partition_assignments[partition][i]
+                                                              .task_address]++;
+            } else {
+                LOG(FATAL) << "__builtin_unreachable";

Review Comment:
   warning: use of undeclared identifier 'FATAL' [clang-diagnostic-error]
   ```cpp
                   LOG(FATAL) << "__builtin_unreachable";
                       ^
   ```
   



##########
be/src/vec/exec/skewed_partition_rebalancer.cpp:
##########
@@ -45,13 +45,40 @@ SkewedPartitionRebalancer::SkewedPartitionRebalancer(
           _partition_data_size_since_last_rebalance_per_task(partition_count, 
0),
           _estimated_task_bucket_data_size_since_last_rebalance(task_count * 
task_bucket_count, 0),
           _partition_assignments(partition_count) {
+    if (task_addresses != nullptr) {
+        CHECK(task_addresses->size() == task_count);

Review Comment:
   warning: use of undeclared identifier 'CHECK' [clang-diagnostic-error]
   ```cpp
           CHECK(task_addresses->size() == task_count);
           ^
   ```
   



##########
be/src/vec/exec/skewed_partition_rebalancer.cpp:
##########
@@ -45,13 +45,40 @@
           _partition_data_size_since_last_rebalance_per_task(partition_count, 
0),
           _estimated_task_bucket_data_size_since_last_rebalance(task_count * 
task_bucket_count, 0),
           _partition_assignments(partition_count) {
+    if (task_addresses != nullptr) {
+        CHECK(task_addresses->size() == task_count);
+        _task_addresses = *task_addresses;
+        for (int i = 0; i < _task_addresses.size(); ++i) {
+            auto it = 
_assigned_address_to_task_buckets_num.find(_task_addresses[i]);
+            if (it == _assigned_address_to_task_buckets_num.end()) {
+                
_assigned_address_to_task_buckets_num.insert({_task_addresses[i], 0});
+            }
+        }
+    } else {
+        
_assigned_address_to_task_buckets_num.insert({TASK_BUCKET_ADDRESS_NOT_SET, 0});
+    }
+
     std::vector<int> task_bucket_ids(task_count, 0);
 
     for (int partition = 0; partition < partition_count; partition++) {
         int task_id = partition % task_count;
         int bucket_id = task_bucket_ids[task_id]++ % task_bucket_count;
-        TaskBucket task_bucket(task_id, bucket_id, task_bucket_count);
+        TaskBucket task_bucket(
+                task_id, bucket_id, task_bucket_count,
+                (_task_addresses.empty()) ? TASK_BUCKET_ADDRESS_NOT_SET : 
_task_addresses[task_id]);
         _partition_assignments[partition].emplace_back(std::move(task_bucket));
+
+        for (int i = 0; i < _partition_assignments[partition].size(); ++i) {

Review Comment:
   warning: use range-based for loop instead [modernize-loop-convert]
   
   ```suggestion
           for (auto & i : _partition_assignments[partition]) {
   ```
   
   be/src/vec/exec/skewed_partition_rebalancer.cpp:72:
   ```diff
   -                     _partition_assignments[partition][i].task_address);
   +                     i.task_address);
   ```
   
   be/src/vec/exec/skewed_partition_rebalancer.cpp:74:
   ```diff
   -                 
_assigned_address_to_task_buckets_num[_partition_assignments[partition][i]
   +                 _assigned_address_to_task_buckets_num[i
   ```
   



##########
be/src/vec/exec/skewed_partition_rebalancer.cpp:
##########
@@ -45,13 +45,40 @@
           _partition_data_size_since_last_rebalance_per_task(partition_count, 
0),
           _estimated_task_bucket_data_size_since_last_rebalance(task_count * 
task_bucket_count, 0),
           _partition_assignments(partition_count) {
+    if (task_addresses != nullptr) {
+        CHECK(task_addresses->size() == task_count);
+        _task_addresses = *task_addresses;
+        for (int i = 0; i < _task_addresses.size(); ++i) {
+            auto it = 
_assigned_address_to_task_buckets_num.find(_task_addresses[i]);

Review Comment:
   warning: use range-based for loop instead [modernize-loop-convert]
   
   ```suggestion
           for (auto & _task_addresse : _task_addresses) {
               auto it = 
_assigned_address_to_task_buckets_num.find(_task_addresse);
   ```
   
   be/src/vec/exec/skewed_partition_rebalancer.cpp:53:
   ```diff
   -                 
_assigned_address_to_task_buckets_num.insert({_task_addresses[i], 0});
   +                 
_assigned_address_to_task_buckets_num.insert({_task_addresse, 0});
   ```
   



##########
be/src/vec/exec/skewed_partition_rebalancer.h:
##########
@@ -60,9 +60,12 @@ class SkewedPartitionRebalancer {
     struct TaskBucket {
         int task_id;
         int id;
+        std::string task_address;
 
-        TaskBucket(int task_id_, int bucket_id_, int task_bucket_count_)
-                : task_id(task_id_), id(task_id_ * task_bucket_count_ + 
bucket_id_) {}
+        TaskBucket(int task_id_, int bucket_id_, int task_bucket_count_, 
std::string task_address_)

Review Comment:
   warning: pass by value and use std::move [modernize-pass-by-value]
   
   be/src/vec/exec/skewed_partition_rebalancer.h:51:
   ```diff
   - #include <vector>
   + #include <utility>
   + #include <vector>
   ```
   
   be/src/vec/exec/skewed_partition_rebalancer.h:67:
   ```diff
   -                   task_address(task_address_) {}
   +                   task_address(std::move(task_address_)) {}
   ```
   



-- 
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]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to