Summary:
The code is dead in RocksDB as `log::Reader::initial_offset_` is always zero. We should delete it so we don't have to maintain it like in #4359.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4362
Differential Revision: D9817829
Pulled By: ajkr
fbshipit-source-id: 474a2c679e5bd273b40608f3a5332931d9eefe6d
Summary:
Please consider this small PR providing access to the `MemoryUsage::GetApproximateMemoryUsageByType` function in plain C API. Actually I'm working on Go application and now trying to investigate the reasons of high memory consumption (#4313). Go [wrappers](https://github.com/tecbot/gorocksdb) are built on the top of Rocksdb C API. According to the #706, `MemoryUsage::GetApproximateMemoryUsageByType` is considered as the best option to get database internal memory usage stats, but it wasn't supported in C API yet.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4340
Differential Revision: D9655135
Pulled By: ajkr
fbshipit-source-id: a3d2f3f47c143ae75862fbcca2f571ea1b49e14a
Summary:
`RangeDelAggregator::AddTombstones` contained an assertion which stated that, if a range tombstone extended past the largest key in the sstable, then `FileMetaData::largest` must have a sentinel sequence number of `kMaxSequenceNumber`, which implies that the tombstone's end key is safe to truncate. However, `largest` will not be a sentinel key when the next sstable in the level's smallest key is equal to the current sstable's largest key, which caused the assertion to fail.
The assertion must hold for the truncation to be safe, so it has been moved to an additional check on end-key truncation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4356
Differential Revision: D9760891
Pulled By: abhimadan
fbshipit-source-id: 7c20c3885cd919dcd14f291f88fd27aa33defebc
Summary:
TransactionOptions::skip_concurrency_control allows pessimistic transactions to skip the overhead of concurrency control. This could be as an optimization if the application knows that the transaction would not have any conflict with concurrent transactions. It is currently used during recovery assuming (i) application guarantees no conflict between prepared transactions in the WAL (ii) application guarantees that recovered transactions will be rolled back/commit before new transactions start.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4346
Differential Revision: D9759149
Pulled By: maysamyabandeh
fbshipit-source-id: f896e84fa58b0b584be904c7fd3883a41ea3215b
Summary:
In C++ 11, the order of argument and move evaluation in a statement such
as below is unspecified -
foo(a.b).bar(std::move(a))
The compiler is free to evaluate std::move(a) first, and then a.b is unspecified.
In C++ 17, this will be safe if a draft proposal around function
chaining rules is accepted.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4348
Differential Revision: D9688810
Pulled By: anand1976
fbshipit-source-id: e4651d0ca03dcf007e50371a0fc72c0d1e710fb4
Summary:
As you know, almost all compilers support "pragma once" keyword instead of using include guards. To be keep consistency between header files, all header files are edited.
Besides this, try to fix some warnings about loss of data.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4339
Differential Revision: D9654990
Pulled By: ajkr
fbshipit-source-id: c2cf3d2d03a599847684bed81378c401920ca848
Summary:
This is a followup to #4311. Checking `!RangeDelAggregator::IsEmpty()` before opening a dedicated range tombstone SST did not properly prevent empty SSTs from being generated. That's because it relies on `CollapsedRangeDelMap::Size`, which had an underflow bug when the map was empty. This PR fixes that underflow bug.
Also fixed an uninitialized variable in db_stress.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4336
Differential Revision: D9600080
Pulled By: ajkr
fbshipit-source-id: bc6980ca79d2cd01b825ebc9dbccd51c1a70cfc7
Summary:
Basically at the moment it seems it's possible to cause write stall by calling flush (either manually vis DB::Flush(), or from Backup Engine directly calling FlushMemTable() while background flush may be already happening.
One of the ways to fix it is that in DBImpl::CompactRange() we already check for possible stall and delay flush if needed before we actually proceed to call FlushMemTable(). We can simply move this delay logic to separate method and call it from FlushMemTable.
This is draft patch, for first look; need to check tests/update SyncPoints and most certainly would need to add allow_write_stall method to FlushOptions().
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4297
Differential Revision: D9420705
Pulled By: mikhail-antonov
fbshipit-source-id: f81d206b55e1d7b39e4dc64242fdfbceeea03fcc
Summary: For the CURRENT file forged during checkpoint, we were forgetting to `fsync` or `fdatasync` it after its creation. This PR fixes it.
Differential Revision: D9525939
Pulled By: ajkr
fbshipit-source-id: a505483644026ee3f501cfc0dcbe74832165b2e3
Summary:
According to 4848bd0c4e/db/log_reader.cc (L355), the original text is misleading when describing the layout of RecyclableLogHeader.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4315
Differential Revision: D9505284
Pulled By: riversand963
fbshipit-source-id: 79994c37a69e7003f03453e7efc0186feeafa609
Summary:
This PR fixes issue 3842. We drop deletion markers iff
1. We are the bottom most level AND
2. All other occurrences of the key are in the same snapshot range as the delete
I've also enhanced db_stress_test to add an option that does a full compare of the keys. This is done by a single thread (thread # 0). For tests I've run (so far)
make check -j64
db_stress
db_stress --acquire_snapshot_one_in=1000 --ops_per_thread=100000 /* to verify that new code doesnt break existing tests */
./db_stress --compare_full_db_state_snapshot=true --acquire_snapshot_one_in=1000 --ops_per_thread=100000 /* to verify new test code */
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4289
Differential Revision: D9491165
Pulled By: shrikanthshankar
fbshipit-source-id: ce144834f31736c189aaca81bed356ba990331e2
Summary:
RocksDB currently queues individual column family for flushing. This is not sufficient to support the needs of some applications that want to enforce order/dependency between column families, given that multiple foreground and background activities can trigger flushing in RocksDB.
This PR aims to address this limitation. Each flush request is described as a `FlushRequest` that can contain multiple column families. A background flushing thread pops one flush request from the queue at a time and processes it.
This PR does not enable atomic_flush yet, but is a subset of [PR 3752](https://github.com/facebook/rocksdb/pull/3752).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/3952
Differential Revision: D8529933
Pulled By: riversand963
fbshipit-source-id: 78908a21e389a3a3f7de2a79bae0cd13af5f3539
Summary:
I have a PR to start calling `OnTableFileCreated` for empty SSTs: #4307. However, it is a behavior change so should not go into a patch release.
This PR adds back a check to make sure range deletions at least exist before starting file creation. This PR should be safe to backport to earlier versions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4311
Differential Revision: D9493734
Pulled By: ajkr
fbshipit-source-id: f0d43cda4cfd904f133cfe3a6eb622f52a9ccbe8
Summary:
The API comment on `OnTableFileCreationStarted` (b6280d01f9/include/rocksdb/listener.h (L331-L333)) led users to believe a call to `OnTableFileCreationStarted` will always be matched with a call to `OnTableFileCreated`. However, we were skipping the `OnTableFileCreated` call in one case: no error happens but also no file is generated since there's no data.
This PR adds the call to `OnTableFileCreated` for that case. The filename will be "(nil)" and the size will be zero.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4307
Differential Revision: D9485201
Pulled By: ajkr
fbshipit-source-id: 2f077ec7913f128487aae2624c69a50762394df6
Summary:
Memtables are selected for flushing by the flush job. Currently we
have listener which is invoked when memtables for a column family are
flushed. That listener does not indicate which memtable was flushed in
the notification. If clients want to know if particular data in the
memtable was retired, there is no straight forward way to know this.
This method will help users who implement memtablerep factory and extend
interface for memtablerep, to know if the data in the memtable was
retired.
Another option that was tried, was to depend on memtable destructor to
be called after flush to mark that data was persisted. This works all
the time but sometimes there can huge delays between actual flush
happening and memtable getting destroyed. Hence, if anyone who is
waiting for data to persist will have to wait that longer.
It is expected that anyone who is implementing this method to have
return quickly as it blocks RocksDB.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4304
Reviewed By: riversand963
Differential Revision: D9472312
Pulled By: gdrane
fbshipit-source-id: 8e693308dee749586af3a4c5d4fcf1fa5276ea4d
Summary:
We want to sample the file I/O issued by RocksDB and report the function calls. This requires us to include the file paths otherwise it's hard to tell what has been going on.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4039
Differential Revision: D8670178
Pulled By: riversand963
fbshipit-source-id: 97ee806d1c583a2983e28e213ee764dc6ac28f7a
Summary:
User reported (https://github.com/facebook/rocksdb/issues/4168) that when opening RocksDB in read-only mode, some statistics are not correctly reported. After some investigation, we believe the following counters are indeed not reported during Get() call in a read-only DB:
rocksdb.memtable.hit
rocksdb.memtable.miss
rocksdb.number.keys.read
rocksdb.bytes.read
As well as histogram rocksdb.bytes.per.read
and perf context get_read_bytes
This PR will add the necessary counter reporting logic in the Get() call path
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4260
Differential Revision: D9476431
Pulled By: miasantreble
fbshipit-source-id: 7ab409d4e59df05d09ae8b69fe75554e5aa240d6
Summary:
Clang analyze is not happy in two pieces of code, with "Potential memory leak". No idea what the problem but slightly changing the code makes clang happy.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4292
Differential Revision: D9413555
Pulled By: siying
fbshipit-source-id: 9428c9d3664530c72129feefd135ee63d8386137
Summary:
During recovery, RocksDB is able to handle version edits that belong to group commits.
This PR is a subset of [PR 3752](https://github.com/facebook/rocksdb/pull/3752)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/3945
Differential Revision: D8529122
Pulled By: riversand963
fbshipit-source-id: 57cb0f9cc55ecca684a837742d6626dc9c07f37e
Summary:
This PR addresses issue #3865 and implements the following approach to fix it:
- adds `MergeContext::GetOperandsDirectionForward` and `MergeContext::GetOperandsDirectionBackward` to query merge operands in a specific order
- `MergeContext::GetOperands` becomes a shortcut for `MergeContext::GetOperandsDirectionForward`
- pass `MergeContext::GetOperandsDirectionBackward` to `MergeOperator::ShouldMerge` and document the order
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4266
Differential Revision: D9360750
Pulled By: sagar0
fbshipit-source-id: 20cb73ff017760b062ecdcf4382560767086e092
Summary:
Add a unit test to check that iterators release data blocks after it has moved away from it. Verify the same for compaction input iterators.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4170
Differential Revision: D8962513
Pulled By: siying
fbshipit-source-id: 05a5b604d7d29887fb488f2cda7286f554a14407
Summary:
Revert this change. Not generating the OnTableFileCreated() notification for a 0 byte SST on flush breaks the assumption that every OnTableFileCreationStarted() notification is followed by a corresponding OnTableFileCreated().
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4263
Differential Revision: D9285623
Pulled By: anand1976
fbshipit-source-id: 808c3dcd498b4b4f4ed4be947a29a24b2296aa8d
Summary:
In the current trace_and replay, Get an WriteBatch are traced. This pull request track down the Seek() and SeekForPrev() to the trace file. <target_key, timestamp, column_family_id> are write to the file.
Replay of Iterator is not supported in the current implementation.
Tested with trace_analyzer.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4228
Differential Revision: D9201381
Pulled By: zhichao-cao
fbshipit-source-id: 6f9cc9cb3c20260af741bee065ec35c5c96354ab
Summary:
The pair of ROCKSDB_LITE condition inclusion is redundant, it is already inside the #ifndef ROCKSDB_LITE. Remove them to void confusion.
Tested by make asan_check.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4254
Differential Revision: D9281652
Pulled By: zhichao-cao
fbshipit-source-id: 06bf7641ede71391f21f6a3fe37fbd13f0e2a43a
Summary:
Given that index value is a BlockHandle, which is basically an <offset, size> pair we can apply delta encoding on the values. The first value at each index restart interval encoded the full BlockHandle but the rest encode only the size. Refer to IndexBlockIter::DecodeCurrentValue for the detail of the encoding. This reduces the index size which helps using the block cache more efficiently. The feature is enabled with using format_version 4.
The feature comes with a bit of cpu overhead which should be paid back by the higher cache hits due to smaller index block size.
Results with sysbench read-only using 4k blocks and using 16 index restart interval:
Format 2:
19585 rocksdb read-only range=100
Format 3:
19569 rocksdb read-only range=100
Format 4:
19352 rocksdb read-only range=100
Pull Request resolved: https://github.com/facebook/rocksdb/pull/3983
Differential Revision: D8361343
Pulled By: maysamyabandeh
fbshipit-source-id: f882ee082322acac32b0072e2bdbb0b5f854e651
Summary:
We add two subcommands `write_extern_sst` and `ingest_extern_sst` to ldb. This PR avoids changing existing code because we hope to cherry-pick to earlier releases to support compatibility check for external SST file ingestion.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4205
Differential Revision: D9112711
Pulled By: riversand963
fbshipit-source-id: 7cae88380d4de86da8440230e87eca66755648e4
Summary:
In the current code, `error_msg` is pointing to the inner buffer of a temporary std::string object. When `error_msg` is used to construct the error message, that array is already released. This PR will fix this bug by copying the string to a local variable.
Fixes https://github.com/facebook/rocksdb/issues/4239
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4240
Differential Revision: D9204334
Pulled By: miasantreble
fbshipit-source-id: 0ac599e166ae0a4ec413e32d8b8853d7c5fba878
Summary:
The test has become complicated over the years and hard to reason about the corner cases that makes the test flaky. The patch simplifies the test and also fixes some probable synchronization issues.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4235
Differential Revision: D9187995
Pulled By: maysamyabandeh
fbshipit-source-id: 53c7b060f14367e5a9e361014578c26debfe3d27
Summary:
So that we can act accordingly on blob index entries
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4233
Differential Revision: D9190205
Pulled By: yiwu-arbug
fbshipit-source-id: e5b84d5b41e44fa7a76762f1f7b0305369bb3a0c
Summary:
There are two issues with `VisibleToActiveSnapshot`:
1. If there are no snapshots, `oldest_snapshot` will be 0 and `VisibleToActiveSnapshot` will always return true. Since the method is used to decide whether it is safe to delete obsolete files, obsolete file won't be able to delete in this case.
2. The `auto` keyword of `auto snapshots = db_impl_->snapshots()` translate to a copy of `const SnapshotList` instead of a reference. Since copy constructor of `SnapshotList` is not defined, using the copy may yield unexpected result.
Issue 2 actually hide issue 1 from being catch by tests. During test `snapshots.empty()` can return false while it should actually be empty, and `snapshots.oldest()` return an invalid address, making `oldest_snapshot` being some random large number.
The issue was originally reported by BlobDB early adopter at Kuaishou.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4236
Differential Revision: D9188706
Pulled By: yiwu-arbug
fbshipit-source-id: a0f2624b927cf9bf28c1bb534784fee5d106f5ea
Summary:
HashMayMatch is related to AddKey() instead of CreateFilter().
Also applies some minor Fixes#4191#4200#3910
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4202
Differential Revision: D9180945
Pulled By: maysamyabandeh
fbshipit-source-id: 6f07b81c5bb9bda5c0273475b486ba8a030471e6
Summary:
In the past, we assume that a job modifies a single column family. Therefore, a job can create at most one superversion since each superversion corresponds to one column family. This assumption leads to the fact that a `JobContext` has only one member variable called `superversion_context`.
Now we want to support group flush of column families, indicating that each job can create multiple superversions. Therefore, we need to make the following change to accommodate this new feature.
Add a vector of `SuperVersionContext` to `JobContext` to support installing
superversions for multiple column families in one job context.
This PR is a subset of [PR 3752](https://github.com/facebook/rocksdb/pull/3752).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/3949
Differential Revision: D8864895
Pulled By: riversand963
fbshipit-source-id: 5937a48817276370d3c8172db9c8aafc826d97ca
Summary:
The current verification logic does not consider the case in which multiple
threads (foreground and background) may execute `PurgeObsoleteFiles` function
simultaneously. Each invocation will trigger the callback adding elements to
a vector. Then we verify the elements in the vector, which can fail sometimes.
The solution is to give up checking the elements. Instead, we check the number
of OPTIONS file in the database dir.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4218
Differential Revision: D9128727
Pulled By: riversand963
fbshipit-source-id: 2b13b705fb21bc0ddd41940c4ec9b6b0c8d88224
Summary:
`CollapsedRangeDelMap` internally uses seqno zero as a sentinel value to
denote a gap between range tombstones or the end of range tombstones. It
therefore expects to never have consecutive sentinel tombstones.
However, since `DeleteRange` is now supported in `SstFileWriter`, an
ingested file may contain range tombstones, and that ingested file may
be assigned global seqno zero. When such tombstones are added to the
collapsed map, they resemble sentinel tombstones due to having seqno
zero. Then, the invariant mentioned above about never having consecutive
sentinel tombstones can be violated.
The symptom of this violation was dereferencing the `end()` iterator
(#4204). The fix in this PR is to not add range tombstones with seqno
zero to the collapsed map. They're not needed anyways since they can't
possibly cover anything (in case of a key and a range tombstone with the
same seqno, the key is visible).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4216
Differential Revision: D9121716
Pulled By: ajkr
fbshipit-source-id: f5b78a70bea9527354603ea7ac8542a7e2b6a210
Summary:
A framework for tracing and replaying RocksDB operations.
A binary trace file is created by capturing the DB operations, and it can be replayed back at the same rate using db_bench.
- Column-families are supported
- Multi-threaded tracing is supported.
- TraceReader and TraceWriter are exposed to the user, so that tracing to various destinations can be enabled (say, to other messaging/logging services). By default, a FileTraceReader and FileTraceWriter are implemented to capture to a file and replay from it.
- This is not yet ideal to be enabled in production due to large performance overhead, but it can be safely tried out in a shadow setup, say, for analyzing RocksDB operations.
Currently supported DB operations:
- Writes:
-- Put
-- Merge
-- Delete
-- SingleDelete
-- DeleteRange
-- Write
- Reads:
-- Get (point lookups)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/3837
Differential Revision: D7974837
Pulled By: sagar0
fbshipit-source-id: 8ec65aaf336504bc1f6ed0feae67f6ed5ef97a72
Summary:
RocksDB used to store global_seqno in external SST files written by
SstFileWriter. During file ingestion, RocksDB uses `pwrite` to update the
`global_seqno`. Since random write is not supported in some non-POSIX compliant
file systems, external SST file ingestion is not supported on these file
systems. To address this limitation, we no longer update `global_seqno` during
file ingestion. Later RocksDB uses the MANIFEST and other information in table
properties to deduce global seqno for externally-ingested SST files.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4172
Differential Revision: D8961465
Pulled By: riversand963
fbshipit-source-id: 4382ec85270a96be5bc0cf33758ca2b167b05071
Summary:
If crash happen after a hard link established, Recover function may reuse the file number that has already assigned to the internal file, and this will overwrite the external file. To protect the external file, we have to make sure the file number will never being reused.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4099
Differential Revision: D9034092
Pulled By: riversand963
fbshipit-source-id: 3f1a737440b86aa2ef01673e5013aacbb7c33e28
Summary:
995fcf7573 has a bug: ReleaseFileNumberFromPendingOutputs() added is not protected by the DB mutex. Fix it by grabbing the lock for this operation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4189
Differential Revision: D9015447
Pulled By: siying
fbshipit-source-id: b8506e09a96c3f95a6fe32b5ca5fcdb9bee88937
Summary:
92ee3350e0 introduces an out-of-bound check in BlockBasedTableIterator::Valid(). However, this flag is not reset when re-seeking in backward direction. This caused the iterator to be invalide by mistake. Fix it by always resetting the out-of-bound flag in every seek.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4187
Differential Revision: D8996600
Pulled By: siying
fbshipit-source-id: b6235ea614f71381e50e7904c4fb036300604ac1
Summary:
This adds support for writing unprepared batches based on size defined in `TransactionOptions::max_write_batch_size`. This is done by overriding methods that modify data (Put/Delete/SingleDelete/Merge) and checking first if write batch size has exceeded threshold. If so, the write batch is written to DB as an unprepared batch.
Support for Commit/Rollback for unprepared batch is added as well. This has been done by simply extending the WritePrepared Commit/Rollback logic to take care of all unprep_seq numbers either when updating prepare heap, or adding to commit map. For updating the commit map, this logic exists inside `WriteUnpreparedCommitEntryPreReleaseCallback`.
A test change was also made to have transactions unregister themselves when committing without prepare. This is because with write unprepared, there may be unprepared entries (which act similarly to prepared entries) already when a commit is done without prepare.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4104
Differential Revision: D8785717
Pulled By: lth
fbshipit-source-id: c02006e281ec1ce00f628e2a7beec0ee73096a91
Summary:
Currently in `Version::Get` when reporting ticker stats stored in `GetContext`, there is a big for-loop through all `Ticker` which adds unnecessary cost to overall CPU usage. We can optimize by storing only ticker values that are used in `Get()` calls in a new struct `GetContextStats` since only a small fraction of all tickers are used in `Get()` calls. For comparison, with the new approach we only need to visit 17 values while old approach will require visiting 100+ `Ticker`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/3490
Differential Revision: D6969154
Pulled By: miasantreble
fbshipit-source-id: fc27072965a3a94125a3e6883d20dafcf5b84029
Summary:
Lint is not happy with some new code recently committed. Format them.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4161
Differential Revision: D8940582
Pulled By: siying
fbshipit-source-id: c9b43b1ef8c88b5e923911058b44eb77234b36b7
Summary:
Right now we use one hard-coded prefetch size to prefetch data from the tail of the SST files. However, this may introduce a waste for some use cases, while not efficient for others.
Introduce a way to adjust this prefetch size by tracking 32 recent times, and pick a value with which the wasted read is less than 10%
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4156
Differential Revision: D8916847
Pulled By: siying
fbshipit-source-id: 8413f9eb3987e0033ed0bd910f83fc2eeaaf5758
Summary:
PR #3944 introduces group commit of `VersionEdit` in MANIFEST. The
implementation has a bug. When updating the log file number of each column
family, we must consider only `VersionEdit`s that operate on the same column
family. Otherwise, a column family may accidentally set its log file number
higher than actual value, indicating that log files with smaller file number
will be ignored, thus causing some updates to be lost.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4157
Differential Revision: D8916650
Pulled By: riversand963
fbshipit-source-id: 8f456cf688f17bf35ad87b38e30e899aa162f201
Yanqin Jin