Allow full merge when root of history for a key is reached (#4909)

Summary:
Previously compaction was not collapsing operands for a first
key on a layer, even in cases when it was its root of history. Some
tests (CompactionJobTest.NonAssocMerge) was actually accounting
for that bug,
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4909

Differential Revision: D13781169

Pulled By: finik

fbshipit-source-id: d2de353ecf05bec39b942cd8d5b97a8dc445f336

db/compaction_job_test.cc

@@ -456,8 +456,7 @@ TEST_F(CompactionJobTest, NonAssocMerge) {
 
   auto expected_results =
       mock::MakeMockFile({{KeyStr("a", 0U, kTypeValue), "3,4,5"},
-                          {KeyStr("b", 2U, kTypeMerge), "2"},
-                          {KeyStr("b", 1U, kTypeMerge), "1"}});
+                          {KeyStr("b", 2U, kTypeValue), "1,2"}});
 
   SetLastSequence(5U);
   auto files = cfd_->current()->storage_info()->LevelFiles(0);
@@ -484,7 +483,7 @@ TEST_F(CompactionJobTest, MergeOperandFilter) {
 
   auto expected_results =
       mock::MakeMockFile({{KeyStr("a", 0U, kTypeValue), test::EncodeInt(8U)},
-                          {KeyStr("b", 2U, kTypeMerge), test::EncodeInt(2U)}});
+                          {KeyStr("b", 2U, kTypeValue), test::EncodeInt(2U)}});
 
   SetLastSequence(5U);
   auto files = cfd_->current()->storage_info()->LevelFiles(0);

db/db_range_del_test.cc

@@ -1138,6 +1138,14 @@ TEST_F(DBRangeDelTest, KeyAtOverlappingEndpointReappears) {
   options.target_file_size_base = kFileBytes;
   Reopen(options);
 
+  // Push dummy data to L3 so that our actual test files on L0-L2
+  // will not be considered "bottommost" level, otherwise compaction
+  // may prevent us from creating overlapping user keys
+  // as on the bottommost layer MergeHelper
+  ASSERT_OK(db_->Merge(WriteOptions(), "key", "dummy"));
+  ASSERT_OK(db_->Flush(FlushOptions()));
+  MoveFilesToLevel(3);
+
   Random rnd(301);
   const Snapshot* snapshot = nullptr;
   for (int i = 0; i < kNumFiles; ++i) {
@@ -1159,8 +1167,9 @@ TEST_F(DBRangeDelTest, KeyAtOverlappingEndpointReappears) {
   std::string value;
   ASSERT_TRUE(db_->Get(ReadOptions(), "key", &value).IsNotFound());
 
-  ASSERT_OK(db_->CompactRange(CompactRangeOptions(), nullptr /* begin_key */,
-                              nullptr /* end_key */));
+  dbfull()->TEST_CompactRange(0 /* level */, nullptr /* begin */,
+                              nullptr /* end */, nullptr /* column_family */,
+                              true /* disallow_trivial_move */);
   ASSERT_EQ(0, NumTableFilesAtLevel(0));
   // Now we have multiple files at L1 all containing a single user key, thus
   // guaranteeing overlap in the file endpoints.

db/merge_helper.cc

@@ -282,22 +282,24 @@ Status MergeHelper::MergeUntil(InternalIterator* iter,
     return Status::OK();
   }
 
-  // We are sure we have seen this key's entire history if we are at the
-  // last level and exhausted all internal keys of this user key.
-  // NOTE: !iter->Valid() does not necessarily mean we hit the
-  // beginning of a user key, as versions of a user key might be
-  // split into multiple files (even files on the same level)
-  // and some files might not be included in the compaction/merge.
+  // We are sure we have seen this key's entire history if:
+  // at_bottom == true (this does not necessarily mean it is the bottommost
+  // layer, but rather that we are confident the key does not appear on any of
+  // the lower layers, at_bottom == false doesn't mean it does appear, just
+  // that we can't be sure, see Compaction::IsBottommostLevel for details)
+  // AND
+  // we have either encountered another key or end of key history on this
+  // layer.
   //
-  // There are also cases where we have seen the root of history of this
-  // key without being sure of it. Then, we simply miss the opportunity
+  // When these conditions are true we are able to merge all the keys
+  // using full merge.
+  //
+  // For these cases we are not sure about, we simply miss the opportunity
   // to combine the keys. Since VersionSet::SetupOtherInputs() always makes
   // sure that all merge-operands on the same level get compacted together,
   // this will simply lead to these merge operands moving to the next level.
-  //
-  // So, we only perform the following logic (to merge all operands together
-  // without a Put/Delete) if we are certain that we have seen the end of key.
-  bool surely_seen_the_beginning = hit_the_next_user_key && at_bottom;
+  bool surely_seen_the_beginning = (hit_the_next_user_key || !iter->Valid())
+                                    && at_bottom;
   if (surely_seen_the_beginning) {
     // do a final merge with nullptr as the existing value and say
     // bye to the merge type (it's now converted to a Put)

db/merge_helper_test.cc

@@ -130,7 +130,7 @@ TEST_F(MergeHelperTest, SingleOperand) {
 
   AddKeyVal("a", 50, kTypeMerge, test::EncodeInt(1U));
 
-  ASSERT_TRUE(Run(31, true).IsMergeInProgress());
+  ASSERT_TRUE(Run(31, false).IsMergeInProgress());
   ASSERT_FALSE(iter_->Valid());
   ASSERT_EQ(test::KeyStr("a", 50, kTypeMerge), merge_helper_->keys()[0]);
   ASSERT_EQ(test::EncodeInt(1U), merge_helper_->values()[0]);

include/rocksdb/merge_operator.h

@@ -108,6 +108,23 @@ class MergeOperator {
     Slice& existing_operand;
   };
 
+  // This function applies a stack of merge operands in chrionological order
+  // on top of an existing value. There are two ways in which this method is
+  // being used:
+  // a) During Get() operation, it used to calculate the final value of a key
+  // b) During compaction, in order to collapse some operands with the based
+  //    value.
+  //
+  // Note: The name of the method is somewhat misleading, as both in the cases
+  // of Get() or compaction it may be called on a subset of operands:
+  // K:    0    +1    +2    +7    +4     +5      2     +1     +2
+  //                              ^
+  //                              |
+  //                          snapshot
+  // In the example above, Get(K) operation will call FullMerge with a base
+  // value of 2 and operands [+1, +2]. Compaction process might decide to
+  // collapse the beginning of the history up to the snapshot by performing
+  // full Merge with base value of 0 and operands [+1, +2, +7, +3].
   virtual bool FullMergeV2(const MergeOperationInput& merge_in,
                            MergeOperationOutput* merge_out) const;
 
@@ -182,11 +199,11 @@ class MergeOperator {
   //       consistent MergeOperator between DB opens.
   virtual const char* Name() const = 0;
 
-  // Determines whether the MergeOperator can be called with just a single
+  // Determines whether the PartialMerge can be called with just a single
   // merge operand.
-  // Override and return true for allowing a single operand. Both FullMergeV2
-  // and PartialMerge/PartialMergeMulti should be overridden and implemented
-  // correctly to handle a single operand.
+  // Override and return true for allowing a single operand. PartialMerge
+  // and PartialMergeMulti should be overridden and implemented
+  // correctly to properly handle a single operand.
   virtual bool AllowSingleOperand() const { return false; }
 
   // Allows to control when to invoke a full merge during Get.