Summary: Closes https://github.com/facebook/rocksdb/pull/2738

Differential Revision: D5736261

Pulled By: maysamyabandeh

fbshipit-source-id: 49d27e9ccd786c4056a3d586a060fe460ea883ac

Summary:
fixed in #2799
Closes https://github.com/facebook/rocksdb/pull/2805

Differential Revision: D5734324

Pulled By: ajkr

fbshipit-source-id: a285d4e84bf1018dc2257fd6c3e7c075a7243263

Summary:
GCC < 5 + ASAN does not instrument aligned_alloc, which can make ASAN
report false-positive with "free on address which was not malloc" error.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61693

Also suppress leak warning with LRUCache::DisownData().
Closes https://github.com/facebook/rocksdb/pull/2783

Differential Revision: D5696465

Pulled By: yiwu-arbug

fbshipit-source-id: 87c607c002511fa089b18cc35e24909bee0e74b4

Summary:
Since tombstones are not stored in order, we may get a wrong smallest key if we only consider the first added tombstone.
Check https://github.com/facebook/rocksdb/issues/2752 for more details.
Closes https://github.com/facebook/rocksdb/pull/2799

Differential Revision: D5728217

Pulled By: ajkr

fbshipit-source-id: 4a53edb0ca80d2a9fcf10749e52d47d57d6417d3

Summary:
* db/range_del_aggregator.cc (AddTombstone): Avoid a potential
use-after-move bug. The original code would both use and move
`tombstone` in a context where the order of those operations is
not specified.  The fix is to perform the use on a new, preceding
statement.

Author: meyering
Closes https://github.com/facebook/rocksdb/pull/2796

Differential Revision: D5721163

Pulled By: ajkr

fbshipit-source-id: a1d328d6a77a17c6425e8069860a202e615e2f48

Summary:
This if followup of #2195.
Closes https://github.com/facebook/rocksdb/pull/2772

Differential Revision: D5722495

Pulled By: sagar0

fbshipit-source-id: 169d0cef53b03056ea7b9454954a35c707a67d52

Summary:
Adds support for CMake packages: https://cmake.org/cmake/help/v3.9/manual/cmake-packages.7.html#creating-packages.

This allow using RocksDB by other CMake projects this way:

```
cmake_minimum_required(VERSION 3.5)
project(rdbt)

find_package(RocksDB CONFIG)

add_executable(rdbt test.cpp)
target_link_libraries(rdbt PRIVATE RocksDB::rocksdb)
```
Closes https://github.com/facebook/rocksdb/pull/2773

Differential Revision: D5722587

Pulled By: sagar0

fbshipit-source-id: 0d90dc4a77b42a617cdbe1348a370e719c282b87

Summary:
small edit of the language binding file to add the Erlang binding.
Closes https://github.com/facebook/rocksdb/pull/2797

Differential Revision: D5722235

Pulled By: sagar0

fbshipit-source-id: 8ecd74996dad4cac19666783256cfa4d9ce09160

Summary: Closes https://github.com/facebook/rocksdb/pull/2792

Differential Revision: D5715365

Pulled By: maysamyabandeh

fbshipit-source-id: 0837b93ea1d4b0a08dfb3cd0d1feb6e098ef26a4

Summary:
Divide the old snapshots to two lists: a few that fit into a cached array and the rest in a vector, which is expected to be empty in normal cases. The former is to optimize concurrent reads from snapshots without requiring locks. It is done by an array of std::atomic, from which std::memory_order_acquire reads are compiled to simple read instructions in most of the x86_64 architectures.
Closes https://github.com/facebook/rocksdb/pull/2758

Differential Revision: D5660504

Pulled By: maysamyabandeh

fbshipit-source-id: 524fcf9a8e7f90a92324536456912a99aaa6740c

Summary: Closes https://github.com/facebook/rocksdb/pull/2790

Differential Revision: D5711609

Pulled By: maysamyabandeh

fbshipit-source-id: ea103dac013c0a6a031834541ad67e7d95a80fe8

Summary:
Fixing flaky blob_db_test.

To close a blob file, blob db used to add a CloseSeqWrite job to the background thread to close it. Changing file close to be synchronous in order to simplify logic, and fix flaky blob_db_test.
Closes https://github.com/facebook/rocksdb/pull/2787

Differential Revision: D5699387

Pulled By: yiwu-arbug

fbshipit-source-id: dd07a945cd435cd3808fce7ee4ea57817409474a

Summary:
Allow user to reduce number of levels in LSM by issue a full CompactRange() and put the result in a lower level, and then reopen DB with reduced options.num_levels. Previous this will fail on reopen on when recovery replaying the previous MANIFEST and found a historical file was on a higher level than the new options.num_levels. The workaround was after CompactRange(), reopen the DB with old num_levels, which will create a new MANIFEST, and then reopen the DB again with new num_levels.

This patch relax the check of levels during recovery. It allows DB to open if there was a historical file on level > options.num_levels, but was also deleted.
Closes https://github.com/facebook/rocksdb/pull/2740

Differential Revision: D5629354

Pulled By: yiwu-arbug

fbshipit-source-id: 545903f6b36b6083e8cbaf777176aef2f488021d

Summary:
It should hold db mutex while accessing max_total_in_memory_state_.
Closes https://github.com/facebook/rocksdb/pull/2784

Differential Revision: D5696536

Pulled By: yiwu-arbug

fbshipit-source-id: 45430634d7fe11909b38e42e5f169f618681c4ee

Summary: Closes https://github.com/facebook/rocksdb/pull/2788

Differential Revision: D5700189

Pulled By: maysamyabandeh

fbshipit-source-id: 6f043e652093ff904e52f6d35190855781b87673

Summary:
store a zero as the checksum when disabled since it's easier to keep block trailer a fixed length.
Closes https://github.com/facebook/rocksdb/pull/2781

Differential Revision: D5694702

Pulled By: ajkr

fbshipit-source-id: 69cea9da415778ba2b600dfd9d0dfc8cb5188ecd

Summary:
This is the warning that clang considers a bug and has been causing it to fail:

```
table/block_based_table_reader.cc:240:27: warning: Potential leak of memory pointed to by 'block.value'
    for (; biter.Valid(); biter.Next()) {
                          ^~~~~
```

Actually clang just doesn't have enough knowledge to statically determine it's safe. We can teach it using an assert.
Closes https://github.com/facebook/rocksdb/pull/2779

Differential Revision: D5691225

Pulled By: ajkr

fbshipit-source-id: 3f0d545bf44636953b30ee5243c63239e8f16d8e

Summary:
One of the core assumptions of DeleteRange is that files containing portions of the same range tombstone are treated as a single unit from the perspective of compaction picker. Need better tests for this. This PR adds the tests for manual compaction.
Closes https://github.com/facebook/rocksdb/pull/2769

Differential Revision: D5676677

Pulled By: ajkr

fbshipit-source-id: 1b4b3382b300ff7048b872911405fdf900e4fbec

Summary: Closes https://github.com/facebook/rocksdb/pull/2756

Differential Revision: D5660516

Pulled By: maysamyabandeh

fbshipit-source-id: f3f3d3b5f544007a7fbdd78e49e4738b4437c7ee

Summary:
Solves #2632

Added OptimisticTransactionDB to the C API.
Added missing merge operations to Transaction.
Added missing get_for_update operation to transaction

If required I will create tests for this another day.
Closes https://github.com/facebook/rocksdb/pull/2633

Differential Revision: D5600906

Pulled By: yiwu-arbug

fbshipit-source-id: da23e4484433d8f59d471f778ff2ae210e3fe4eb

Summary:
I made another rust binding. 👻

* Use C++ API (instead of C API)
* Try to follow [Rust Guidelines](https://aturon.github.io/README.html)
* Working in progress (the APIs are not stable yet)
Closes https://github.com/facebook/rocksdb/pull/2438

Differential Revision: D5690612

Pulled By: siying

fbshipit-source-id: 11d3956c33b5e5366555afbf3786b782be3046e7

Reviewed By: mzlee

Differential Revision: D5688624

fbshipit-source-id: c726b4e56bd823b994a7b713488fef93c6f796d0

Summary:
Allow `Slice` holding nullptr as a sentinel value but not in comparisons. This new restriction eliminates the need for the manual checks in 39ef9005, while still conforming to glibc's `memcmp` API. Thanks siying for the idea. Users may need to migrate, so mentioned it in HISTORY.md.
Closes https://github.com/facebook/rocksdb/pull/2777

Differential Revision: D5686016

Pulled By: ajkr

fbshipit-source-id: 03a2ca3fd9a0ebade9d0d5686c81d59a9534f563

Summary:
The ::Get from DB is not augmented with an overload method that takes a PinnableSlice instead of a string. Transactions however are not yet upgraded to use the new API. As a result, transaction users such as MyRocks cannot benefit from it. This patch updates the transactional API with a PinnableSlice overload.
Closes https://github.com/facebook/rocksdb/pull/2736

Differential Revision: D5645770

Pulled By: maysamyabandeh

fbshipit-source-id: f6af520df902f842de1bcf99bed3e8dfc43ad96d

Summary:
This is the continuation of https://github.com/facebook/rocksdb/pull/2661 for filter partitions. When pin_l0 is set (along with cache_xxx), then open table open the filter partitions are loaded into the cache and pinned there.
Closes https://github.com/facebook/rocksdb/pull/2766

Differential Revision: D5671098

Pulled By: maysamyabandeh

fbshipit-source-id: 174f24018f1d7f1129621e7380287b65b67d2115

Summary:
it doesn't take nullptr according to its declaration in glibc, and calling it in this way causes our sanitizers (ubsan, clang analyze) to fail.
Closes https://github.com/facebook/rocksdb/pull/2776

Differential Revision: D5683260

Pulled By: ajkr

fbshipit-source-id: 114b137ee188172f96eedc43139255cae7bee80a

Summary:
In Java API, `SstFileWriter.put/merge/delete` takes `Slice` type of key and value, which is a Java wrapper object around C++ Slice object.  The Slice object inherited [ `finalize`](https://github.com/facebook/rocksdb/blob/3c327ac2d0fd50bbd82fe1f1af5de909dad769e6/java/src/main/java/org/rocksdb/AbstractNativeReference.java#L69) method, which [added huge overhead](https://softwareengineering.stackexchange.com/questions/288715/is-overriding-object-finalize-really-bad/288753#288753) to JVM while creating new SstFile.

To address this issue, this PR overload the merge method to take Java byte array instead of the Slice object, and added unit test for it.

We also benchmark these two different merge function, where we could see GC Stall reduced from 50%  to 1%, and the throughput increased from 50MB to 200MB.
Closes https://github.com/facebook/rocksdb/pull/2746

Reviewed By: sagar0

Differential Revision: D5653145

Pulled By: scv119

fbshipit-source-id: b55ea58554b573d0b1c6f6170f8d9223811bc4f5

Summary:
The goal is to reduce the number of histogram buckets, particularly now that we print these histograms for each column family. I chose 1.5 as the factor. We can adjust it later to either make buckets more granular or make fewer buckets.
Closes https://github.com/facebook/rocksdb/pull/2139

Differential Revision: D4872076

Pulled By: ajkr

fbshipit-source-id: 87790d782a605506c3d24190a028cecbd7aa564a

Summary:
- Allow setting custom installation prefix.
- Add option to disable building tests.
Closes https://github.com/facebook/rocksdb/pull/2195

Differential Revision: D5054239

Pulled By: sagar0

fbshipit-source-id: 2de6bef8b7eafed60a830e1796b262f9e6f79da0

Summary:
Changes:
* checks if ASAN mode is on, and uses malloc and free in the constructor and destructor
Closes https://github.com/facebook/rocksdb/pull/2767

Differential Revision: D5671243

Pulled By: armishra

fbshipit-source-id: 8e4ad0f7f163400c4effa8617d3b30134119d802

Summary:
If GC kicks in between

* A Get() reads index entry from base db.
* The Get() read from a blob file

The GC can delete the corresponding blob file, making the key not found. Fortunately we have existing logic to avoid deleting a blob file if it is referenced by a snapshot. So the fix is to explicitly create a snapshot before reading index entry from base db.
Closes https://github.com/facebook/rocksdb/pull/2754

Differential Revision: D5655956

Pulled By: yiwu-arbug

fbshipit-source-id: e4ccbc51331362542e7343175bbcbdea5830f544

Summary:
When out of space, blob db should GC the oldest file. The current implementation GC the newest one instead. Fixing it.
Closes https://github.com/facebook/rocksdb/pull/2757

Differential Revision: D5657611

Pulled By: yiwu-arbug

fbshipit-source-id: 56c30a4c52e6ab04551dda8c5c46006d4070b28d

Summary:
add this counter stat to track usage of deletion-dropping optimization. if usage is low, we can delete it to prevent bugs like #2726.
Closes https://github.com/facebook/rocksdb/pull/2761

Differential Revision: D5665421

Pulled By: ajkr

fbshipit-source-id: 881befa2d199838dac88709e7b376a43d304e3d4

Summary:
https://github.com/facebook/rocksdb/pull/2661 mistakenly updates the version. This patch reverts it.
Closes https://github.com/facebook/rocksdb/pull/2760

Differential Revision: D5662089

Pulled By: maysamyabandeh

fbshipit-source-id: f4735e37921c0ced6081a89080c78ac3728aa8bd

Summary:
it's a new feature that'll be released in 5.8, introduced by PR #2498.
Closes https://github.com/facebook/rocksdb/pull/2759

Differential Revision: D5661923

Pulled By: ajkr

fbshipit-source-id: 9ba9f0d146c453715358ef2dd298aa7765649d7c

Summary:
With this PR, we can measure read-amp for queries where perf_context is enabled as follows:

```
SetPerfLevel(kEnableCount);
Get(1, "foo");
double read_amp = static_cast<double>(get_perf_context()->block_read_byte / get_perf_context()->get_read_bytes);
SetPerfLevel(kDisable);
```

Our internal infra enables perf_context for a sampling of queries. So we'll be able to compute the read-amp for the sample set, which can give us a good estimate of read-amp.
Closes https://github.com/facebook/rocksdb/pull/2749

Differential Revision: D5647240

Pulled By: ajkr

fbshipit-source-id: ad73550b06990cf040cc4528fa885360f308ec12

Summary:
This fixes the existing logic for pinning l0 index partitions. The patch preloads the partitions into block cache and pin them if they belong to level 0 and pin_l0 is set.

The drawback is that it does many small IOs when preloading all the partitions into the cache is direct io is enabled. Working for a solution for that.
Closes https://github.com/facebook/rocksdb/pull/2661

Differential Revision: D5554010

Pulled By: maysamyabandeh

fbshipit-source-id: 1e6f32a3524d71355c77d4138516dcfb601ca7b2

Summary:
Changes:
* extended the wait_txn_map to track additional information
* designed circular buffer to store n latest deadlocks' information
* added test coverage to verify the additional information tracked is accurately stored in the buffer
Closes https://github.com/facebook/rocksdb/pull/2630

Differential Revision: D5478025

Pulled By: armishra

fbshipit-source-id: 2b138de7b5a73f5ca554fc3ff8220a3be49f39e7

Summary:
Clang complain about an cast from uint64_t to int32 in db_stress. Fixing it.
Closes https://github.com/facebook/rocksdb/pull/2755

Differential Revision: D5655947

Pulled By: yiwu-arbug

fbshipit-source-id: cfac10e796e0adfef4727090b50975b0d6e2c9be

Summary:
On initial call to BlobDBImpl::WaStats() `all_periods_write_` would be empty, so it will crash when we call pop_front() at line 1627. Apparently it is mean to pop only when `all_periods_write_.size() > kWriteAmplificationStatsPeriods`.

The whole write amp calculation doesn't seems to be correct and it is not being exposed. Will work on it later.

Test Plan
Change kWriteAmplificationStatsPeriodMillisecs to 1000 (1 second) and run db_bench --use_blob_db for 5 minutes.
Closes https://github.com/facebook/rocksdb/pull/2751

Differential Revision: D5648269

Pulled By: yiwu-arbug

fbshipit-source-id: b843d9a09bb5f9e1b713d101ec7b87e54b5115a4