Commit Graph

5985 Commits

Author SHA1 Message Date
Andrew Kryczka
243975d5da More accurate error status for BackupEngine::Open
Summary:
Some users are assuming NotFound means the backup does not
exist at the provided path, which is a reasonable assumption. We need to
stop returning NotFound for system errors.

Depends on #1644
Closes https://github.com/facebook/rocksdb/pull/1645

Differential Revision: D4312233

Pulled By: ajkr

fbshipit-source-id: 5343c10
2016-12-12 13:24:21 -08:00
Andrew Kryczka
f0c509e2c8 Return finer-granularity status from Env::GetChildren*
Summary:
It'd be nice to use the error status type to distinguish
between user error and system error. For example, GetChildren can fail
listing a backup directory's contents either because a bad path was provided
(user error) or because an operation failed, e.g., a remote storage service
call failed (system error). In the former case, we want to continue and treat
the backup directory as empty; in the latter case, we want to immediately
propagate the error to the caller.

This diff uses NotFound to indicate user error and IOError to indicate
system error. Previously IOError indicated both.
Closes https://github.com/facebook/rocksdb/pull/1644

Differential Revision: D4312157

Pulled By: ajkr

fbshipit-source-id: 51b4f24
2016-12-12 12:54:13 -08:00
Andrew Kryczka
dc64f46b1c Add db_bench option for stderr logging
Summary:
The info log file ("LOG") is stored in the db directory by default. When the db is on a distributed env, this is unnecessarily slow. So, I added an option to db_bench to just print the info log messages to stderr.
Closes https://github.com/facebook/rocksdb/pull/1641

Differential Revision: D4309348

Pulled By: ajkr

fbshipit-source-id: 1e6f851
2016-12-09 16:54:14 -08:00
Jonathan Lee
2cabdb8f44 Increase buffer size
Summary:
When compiling with GCC>=7.0.0, "db/internal_stats.cc" fails to compile as the data being written to the buffer potentially exceeds its size.

This fix simply doubles the size of the buffer, thus accommodating the max possible data size.
Closes https://github.com/facebook/rocksdb/pull/1635

Differential Revision: D4302162

Pulled By: yiwu-arbug

fbshipit-source-id: c76ad59
2016-12-09 11:54:22 -08:00
Jonathan Lee
4a17b47bb5 Remove unnecessary header include
Summary:
Remove "util/testharness.h" from list of includes for "db/db_filesnapshot.cc", as it wasn't being used and thus caused an extraneous dependency on gtest.
Closes https://github.com/facebook/rocksdb/pull/1634

Differential Revision: D4302146

Pulled By: yiwu-arbug

fbshipit-source-id: e900c0b
2016-12-09 11:54:21 -08:00
Mike Kolupaev
8c2b921fdf Fixed a crash in debug build in flush_job.cc
Summary:
It was doing `&range_del_iters[0]` on an empty vector. Even though the resulting pointer is never dereferenced, it's still bad for two reasons:
* the practical reason: it crashes with `std::out_of_range` exception in our debug build,
* the "C++ standard lawyer" reason: it's undefined behavior because, in `std::vector` implementation, it probably "dereferences" a null pointer, which is invalid even though it doesn't actually read the pointed memory, just converts a pointer into a reference (and then flush_job.cc converts it back to pointer); nullptr references are undefined behavior.
Closes https://github.com/facebook/rocksdb/pull/1612

Differential Revision: D4265625

Pulled By: al13n321

fbshipit-source-id: db26fb9
2016-12-09 10:39:12 -08:00
Islam AbdelRahman
20ce081fae Fix issue where IngestExternalFile insert blocks in block cache with g_seqno=0
Summary:
When we Ingest an external file we open it to read some metadata and first/last key
during doing that we insert blocks into the block cache with global_seqno = 0

If we move the file (did not copy it) into the DB, we will use these blocks with the wrong seqno in the read path
Closes https://github.com/facebook/rocksdb/pull/1627

Differential Revision: D4293332

Pulled By: yiwu-arbug

fbshipit-source-id: 3ce5523
2016-12-08 13:39:18 -08:00
Andrew Kryczka
5241e0dbfc fix db_bench argument type
Summary: Closes https://github.com/facebook/rocksdb/pull/1633

Differential Revision: D4298161

Pulled By: yiwu-arbug

fbshipit-source-id: 2c7af35
2016-12-08 11:09:14 -08:00
Jonathan Lee
c04f6a0b4c Specify shell in makefile
Summary:
The second variable "SHELL" simply tells make explicitly which shell to use, instead of allowing it to default to "/bin/sh", which may or may not be Bash.

However, simply defining the second variable by itself causes make to throw an error concerning a circular definition, as it would be attempting to use the "shell" command while simultaneously trying to set which shell to use. Thus, the first variable "BASH_EXISTS" is defined such that make already knows about "/path/to/bash" before trying to use it to set "SHELL".

A more technically correct solution would be to edit the makefile itself to make it compatible with non-bash shells (see the original Issue discussion for details). However, as it seems very few of the people working on this project were building with non-bash shells, I figured this solution would be good enough.
Closes https://github.com/facebook/rocksdb/pull/1631

Differential Revision: D4295689

Pulled By: yiwu-arbug

fbshipit-source-id: e4f9532
2016-12-07 22:24:15 -08:00
zhangjinpeng1987
45c7ce1377 CompactRangeOptions C API
Summary:
Add C API for CompactRangeOptions.
Closes https://github.com/facebook/rocksdb/pull/1596

Differential Revision: D4252339

Pulled By: yiwu-arbug

fbshipit-source-id: f768f93
2016-12-07 17:54:14 -08:00
Andrew Kryczka
2c2ba68247 db_stress support for range deletions
Summary:
made db_stress capable of adding range deletions to its db and verifying their correctness. i'll make db_crashtest.py use this option later once the collapsing optimization (https://github.com/facebook/rocksdb/pull/1614) is committed because currently it slows down the test too much.
Closes https://github.com/facebook/rocksdb/pull/1625

Differential Revision: D4293939

Pulled By: ajkr

fbshipit-source-id: d3beb3a
2016-12-07 13:09:24 -08:00
Andrew Kryczka
b821984d31 DeleteRange read path end-to-end tests
Summary: Closes https://github.com/facebook/rocksdb/pull/1592

Differential Revision: D4246260

Pulled By: ajkr

fbshipit-source-id: ce03fa2
2016-12-07 12:54:17 -08:00
Artemiy Kolesnikov
2f4fc539c6 Compaction::IsTrivialMove relaxing
Summary:
IsTrivialMove returns true if no input file overlaps with output_level+1 with more than max_compaction_bytes_ bytes.
Closes https://github.com/facebook/rocksdb/pull/1619

Differential Revision: D4278338

Pulled By: yiwu-arbug

fbshipit-source-id: 994c001
2016-12-07 11:54:11 -08:00
dhruba borthakur
1dce75b2e2 Update USERS.md
Summary: Closes https://github.com/facebook/rocksdb/pull/1630

Differential Revision: D4293200

Pulled By: IslamAbdelRahman

fbshipit-source-id: 9c81f85
2016-12-07 11:24:14 -08:00
dhruba borthakur
304b3c706b Update USERS.md
Summary: Closes https://github.com/facebook/rocksdb/pull/1629

Differential Revision: D4293183

Pulled By: IslamAbdelRahman

fbshipit-source-id: 759ea92
2016-12-07 11:24:14 -08:00
Andrew Kryczka
fa50fffaf5 Option to expand range tombstones in db_bench
Summary:
When enabled, this option replaces range tombstones with a sequence of
point tombstones covering the same range. This can be used to A/B test perf of
range tombstones vs sequential point tombstones, and help us find the cross-over
point, i.e., the size of the range above which range tombstones outperform point
tombstones.
Closes https://github.com/facebook/rocksdb/pull/1594

Differential Revision: D4246312

Pulled By: ajkr

fbshipit-source-id: 3b00b23
2016-12-06 18:09:14 -08:00
Yi Wu
c26a4d8e8a Fix compile error in trasaction_lock_mgr.cc
Summary:
Fix error on mac/windows build since they don't recognize `uint`.
Closes https://github.com/facebook/rocksdb/pull/1624

Differential Revision: D4287139

Pulled By: yiwu-arbug

fbshipit-source-id: b7cc88f
2016-12-06 14:39:16 -08:00
Islam AbdelRahman
ed8fbdb560 Add EventListener::OnExternalFileIngested() event
Summary:
Add EventListener::OnExternalFileIngested() to allow user to subscribe to external file ingestion events
Closes https://github.com/facebook/rocksdb/pull/1623

Differential Revision: D4285844

Pulled By: IslamAbdelRahman

fbshipit-source-id: 0b95a88
2016-12-06 14:09:17 -08:00
Manuel Ung
2005c88a75 Implement non-exclusive locks
Summary:
This is an implementation of non-exclusive locks for pessimistic transactions. It is relatively simple and does not prevent starvation (ie. it's possible that request for exclusive access will never be granted if there are always threads holding shared access). It is done by changing `KeyLockInfo` to hold an set a transaction ids, instead of just one, and adding a flag specifying whether this lock is currently held with exclusive access or not.

Some implementation notes:
- Some lock diagnostic functions had to be updated to return a set of transaction ids for a given lock, eg. `GetWaitingTxn` and `GetLockStatusData`.
- Deadlock detection is a bit more complicated since a transaction can now wait on multiple other transactions. A BFS is done in this case, and deadlock detection depth is now just a limit on the number of transactions we visit.
- Expirable transactions do not work efficiently with shared locks at the moment, but that's okay for now.
Closes https://github.com/facebook/rocksdb/pull/1573

Differential Revision: D4239097

Pulled By: lth

fbshipit-source-id: da7c074
2016-12-05 17:39:17 -08:00
Yi Wu
0b0f235724 Mention IngestExternalFile changes in HISTORY.md
Summary:
I hit the land button too fast and didn't include the line.
Closes https://github.com/facebook/rocksdb/pull/1622

Differential Revision: D4281316

Pulled By: yiwu-arbug

fbshipit-source-id: c7b38e0
2016-12-05 16:09:11 -08:00
Yi Wu
23db48e8d8 Update HISTORY.md for 5.0 branch
Summary:
These changes are included in the new branch-cut.
Closes https://github.com/facebook/rocksdb/pull/1621

Differential Revision: D4281015

Pulled By: yiwu-arbug

fbshipit-source-id: d88858b
2016-12-05 15:39:11 -08:00
Mike Kolupaev
beb36d9c1e Fixed CompactionFilter::Decision::kRemoveAndSkipUntil
Summary:
Embarassingly enough, the first time I tried to use my new feature in logdevice it crashed with this assertion failure:

  db/pinned_iterators_manager.h:30: void rocksdb::PinnedIteratorsManager::StartPinning(): Assertion `pinning_enabled == false' failed

The issue was that `pinned_iters_mgr_.StartPinning()` was called but `pinned_iters_mgr_.ReleasePinnedData()` wasn't.
Closes https://github.com/facebook/rocksdb/pull/1611

Differential Revision: D4265622

Pulled By: al13n321

fbshipit-source-id: 747b10f
2016-12-05 15:24:11 -08:00
Islam AbdelRahman
67f37cf198 Allow user to specify a CF for SST files generated by SstFileWriter
Summary:
Allow user to explicitly specify that the generated file by SstFileWriter will be ingested in a specific CF.
This allow us to persist the CF id in the generated file
Closes https://github.com/facebook/rocksdb/pull/1615

Differential Revision: D4270422

Pulled By: IslamAbdelRahman

fbshipit-source-id: 7fb954e
2016-12-05 14:24:16 -08:00
Anton Safonov
9053fe2a5c Made delete_obsolete_files_period_micros option dynamic
Summary:
Made delete_obsolete_files_period_micros option dynamic. It can be updating using DB::SetDBOptions().
Closes https://github.com/facebook/rocksdb/pull/1595

Differential Revision: D4246569

Pulled By: tonek

fbshipit-source-id: d23f560
2016-12-05 14:24:16 -08:00
Islam AbdelRahman
edde954e7b fix clang build
Summary:
override is missing for FilterV2
Closes https://github.com/facebook/rocksdb/pull/1606

Differential Revision: D4263832

Pulled By: IslamAbdelRahman

fbshipit-source-id: d8b337a
2016-12-01 18:39:10 -08:00
Yi Wu
56281f3a97 Add memtable_insert_with_hint_prefix_size option to db_bench
Summary:
Add memtable_insert_with_hint_prefix_size option to db_bench
Closes https://github.com/facebook/rocksdb/pull/1604

Differential Revision: D4260549

Pulled By: yiwu-arbug

fbshipit-source-id: cee5ef7
2016-12-01 16:54:16 -08:00
Islam AbdelRahman
4a21b1402c Cache heap::downheap() root comparison (optimize heap cmp call)
Summary:
Reduce number of comparisons in heap by caching which child node in the first level is smallest (left_child or right_child)
So next time we can compare directly against the smallest child

I see that the total number of calls to comparator drops significantly when using this optimization

Before caching (~2mil key comparison for iterating the DB)
```
$ DEBUG_LEVEL=0 make db_bench -j64 && ./db_bench --benchmarks="readseq" --db="/dev/shm/heap_opt" --use_existing_db --disable_auto_compactions --cache_size=1000000000  --perf_level=2
readseq      :       0.338 micros/op 2959201 ops/sec;  327.4 MB/s user_key_comparison_count = 2000008
```
After caching (~1mil key comparison for iterating the DB)
```
$ DEBUG_LEVEL=0 make db_bench -j64 && ./db_bench --benchmarks="readseq" --db="/dev/shm/heap_opt" --use_existing_db --disable_auto_compactions --cache_size=1000000000 --perf_level=2
readseq      :       0.309 micros/op 3236801 ops/sec;  358.1 MB/s user_key_comparison_count = 1000011
```

It also improves
Closes https://github.com/facebook/rocksdb/pull/1600

Differential Revision: D4256027

Pulled By: IslamAbdelRahman

fbshipit-source-id: 76fcc66
2016-12-01 13:39:14 -08:00
Islam AbdelRahman
e39d080871 Fix travis (compile for clang < 3.9)
Summary:
Travis fail because it uses clang 3.6 which don't recognize
`__attribute__((__no_sanitize__("undefined")))`
Closes https://github.com/facebook/rocksdb/pull/1601

Differential Revision: D4257175

Pulled By: IslamAbdelRahman

fbshipit-source-id: fb4d1ab
2016-12-01 10:09:22 -08:00
Igor Canadi
3f407b065c Kill flashcache code in RocksDB
Summary:
Now that we have userspace persisted cache, we don't need flashcache anymore.
Closes https://github.com/facebook/rocksdb/pull/1588

Differential Revision: D4245114

Pulled By: igorcanadi

fbshipit-source-id: e2c1c72
2016-12-01 10:09:22 -08:00
fangchenliaohui
b77007df8b Bug: paralle_group status updated in WriteThread::CompleteParallelWorker
Summary:
Multi-write thread may update the status of the parallel_group in
WriteThread::CompleteParallelWorker if the status of Writer is not ok!
When copy write status to the paralle_group, the write thread just hold the
mutex of the the writer processed by itself. it is useless. The thread
should held the the leader of the parallel_group instead.
Closes https://github.com/facebook/rocksdb/pull/1598

Differential Revision: D4252335

Pulled By: siying

fbshipit-source-id: 3864cf7
2016-12-01 09:54:11 -08:00
Mike Kolupaev
247d0979aa Support for range skips in compaction filter
Summary:
This adds the ability for compaction filter to say "drop this key-value, and also drop everything up to key x". This will cause the compaction to seek input iterator to x, without reading the data. This can make compaction much faster when large consecutive chunks of data are filtered out. See the changes in include/rocksdb/compaction_filter.h for the new API.

Along the way this diff also adds ability for compaction filter changing merge operands, similar to how it can change values; we're not going to use this feature, it just seemed easier and cleaner to implement it than to document that it's not implemented :)

The diff is not as big as it may seem, about half of the lines are a test.
Closes https://github.com/facebook/rocksdb/pull/1599

Differential Revision: D4252092

Pulled By: al13n321

fbshipit-source-id: 41e1e48
2016-12-01 07:09:15 -08:00
Panagiotis Ktistakis
96fcefbf1d c api: expose option for dynamic level size target
Summary: Closes https://github.com/facebook/rocksdb/pull/1587

Differential Revision: D4245923

Pulled By: yiwu-arbug

fbshipit-source-id: 6ee7291
2016-11-30 11:24:14 -08:00
zhangjinpeng1987
00197cff39 Add C API to set base_backgroud_compactions
Summary:
Add C API to set base_backgroud_compactions
Closes https://github.com/facebook/rocksdb/pull/1571

Differential Revision: D4245709

Pulled By: yiwu-arbug

fbshipit-source-id: 792c6b8
2016-11-30 11:09:13 -08:00
Andrew Kryczka
5b219eccb5 deleterange end-to-end test improvements for lite/robustness
Summary: Closes https://github.com/facebook/rocksdb/pull/1591

Differential Revision: D4246019

Pulled By: ajkr

fbshipit-source-id: 0c4aa37
2016-11-29 12:24:13 -08:00
Anirban Rahut
aad1191765 pass rocksdb oncall to mysql_mtr_filter otherwise tasks get created w…
Summary:
…rong owner

 mysql_mtr_filter script needs proper oncall
Closes https://github.com/facebook/rocksdb/pull/1586

Differential Revision: D4245150

Pulled By: anirbanr-fb

fbshipit-source-id: fd8577c
2016-11-29 12:09:12 -08:00
Andrew Kryczka
e333528991 DeleteRange write path end-to-end tests
Summary: Closes https://github.com/facebook/rocksdb/pull/1578

Differential Revision: D4241171

Pulled By: ajkr

fbshipit-source-id: ce5fd83
2016-11-29 11:09:22 -08:00
Siying Dong
7784980fcd Fix mis-reporting of compaction read bytes to the base level
Summary:
In dynamic leveled compaction, when calculating read bytes, output level bytes may be wronglyl calculated as input level inputs. Fix it.
Closes https://github.com/facebook/rocksdb/pull/1475

Differential Revision: D4148412

Pulled By: siying

fbshipit-source-id: f2f475a
2016-11-29 11:09:22 -08:00
Islam AbdelRahman
3c6b49ed66 Fix implicit conversion between int64_t to int
Summary:
Make conversion explicit, implicit conversion breaks the build
Closes https://github.com/facebook/rocksdb/pull/1589

Differential Revision: D4245158

Pulled By: IslamAbdelRahman

fbshipit-source-id: aaec00d
2016-11-29 10:54:15 -08:00
Siying Dong
b3b875657f Remove unused assignment in db/db_iter.cc
Summary:
"make analyze" complains the assignment is not useful. Remove it.
Closes https://github.com/facebook/rocksdb/pull/1581

Differential Revision: D4241697

Pulled By: siying

fbshipit-source-id: 178f67a
2016-11-29 09:09:14 -08:00
Andrew Kryczka
4f6e89b1d0 Fix range deletion covering key in same SST file
Summary:
AddTombstones() needs to be before t->Get(), oops :'(
Closes https://github.com/facebook/rocksdb/pull/1576

Differential Revision: D4241041

Pulled By: ajkr

fbshipit-source-id: 781ceea
2016-11-28 22:54:13 -08:00
Islam AbdelRahman
a2bf265a39 Avoid intentional overflow in GetL0ThresholdSpeedupCompaction
Summary:
99c052a34f fixes integer overflow in GetL0ThresholdSpeedupCompaction() by checking if int become -ve.
UBSAN will complain about that since this is still an overflow, we can fix the issue by simply using int64_t
Closes https://github.com/facebook/rocksdb/pull/1582

Differential Revision: D4241525

Pulled By: IslamAbdelRahman

fbshipit-source-id: b3ae21f
2016-11-28 18:39:13 -08:00
Islam AbdelRahman
52fd1ff2c2 disable UBSAN for functions with intentional -ve shift / overflow
Summary:
disable UBSAN for functions with intentional left shift on -ve number / overflow

These functions are
rocksdb:: Hash
FixedLengthColBufEncoder::Append
FaultInjectionTest:: Key
Closes https://github.com/facebook/rocksdb/pull/1577

Differential Revision: D4240801

Pulled By: IslamAbdelRahman

fbshipit-source-id: 3e1caf6
2016-11-28 17:54:12 -08:00
Islam AbdelRahman
1886c435b9 Fix CompactionJob::Install division by zero
Summary:
Fix CompactionJob::Install division by zero
Closes https://github.com/facebook/rocksdb/pull/1580

Differential Revision: D4240794

Pulled By: IslamAbdelRahman

fbshipit-source-id: 7286721
2016-11-28 16:54:16 -08:00
Islam AbdelRahman
63c30de80d fix options_test ubsan
Summary:
Having -ve value for max_write_buffer_number does not make sense and cause us to do a left shift on a -ve value number
Closes https://github.com/facebook/rocksdb/pull/1579

Differential Revision: D4240798

Pulled By: IslamAbdelRahman

fbshipit-source-id: bd6267e
2016-11-28 16:39:14 -08:00
Islam AbdelRahman
13e66a8f51 Fix compaction_job.cc division by zero
Summary:
Fix division by zero in compaction_job.cc
Closes https://github.com/facebook/rocksdb/pull/1575

Differential Revision: D4240818

Pulled By: IslamAbdelRahman

fbshipit-source-id: a8bc757
2016-11-28 16:39:13 -08:00
Andrew Kryczka
01eabf7375 Fix double-counted deletion stat
Summary:
Both the single deletion and the value are included in compaction outputs, so no need to update the stat for the value's deletion yet, otherwise it'd be double-counted.
Closes https://github.com/facebook/rocksdb/pull/1574

Differential Revision: D4241181

Pulled By: ajkr

fbshipit-source-id: c9aaa15
2016-11-28 15:54:12 -08:00
Andrew Kryczka
7ffb10fc1a DeleteRange compaction statistics
Summary:
- "rocksdb.compaction.key.drop.range_del" - number of keys dropped during compaction due to a range tombstone covering them
- "rocksdb.compaction.range_del.drop.obsolete" - number of range tombstones dropped due to compaction to bottom level and no snapshot saving them
- s/CompactionIteratorStats/CompactionIterationStats/g since this class is no longer specific to CompactionIterator -- it's also updated for range tombstone iteration during compaction
- Move the above class into a separate .h file to avoid circular dependency.
Closes https://github.com/facebook/rocksdb/pull/1520

Differential Revision: D4187179

Pulled By: ajkr

fbshipit-source-id: 10c2103
2016-11-28 11:54:12 -08:00
Mike Kolupaev
236d4c67e9 Less linear search in DBIter::Seek() when keys are overwritten a lot
Summary:
In one deployment we saw high latencies (presumably from slow iterator operations) and a lot of CPU time reported by perf with this stack:

```
  rocksdb::MergingIterator::Next
  rocksdb::DBIter::FindNextUserEntryInternal
  rocksdb::DBIter::Seek
```

I think what's happening is:
1. we create a snapshot iterator,
2. we do lots of Put()s for the same key x; this creates lots of entries in memtable,
3. we seek the iterator to a key slightly smaller than x,
4. the seek walks over lots of entries in memtable for key x, skipping them because of high sequence numbers.

CC IslamAbdelRahman
Closes https://github.com/facebook/rocksdb/pull/1413

Differential Revision: D4083879

Pulled By: IslamAbdelRahman

fbshipit-source-id: a83ddae
2016-11-28 10:24:11 -08:00
Siying Dong
cd7c4143d7 Improve Write Stalling System
Summary:
Current write stalling system has the problem of lacking of positive feedback if the restricted rate is already too low. Users sometimes stack in very low slowdown value. With the diff, we add a positive feedback (increasing the slowdown value) if we recover from slowdown state back to normal. To avoid the positive feedback to keep the slowdown value to be to high, we add issue a negative feedback every time we are close to the stop condition. Experiments show it is easier to reach a relative balance than before.

Also increase level0_stop_writes_trigger default from 24 to 32. Since level0_slowdown_writes_trigger default is 20, stop trigger 24 only gives four files as the buffer time to slowdown writes. In order to avoid stop in four files while 20 files have been accumulated, the slowdown value must be very low, which is amost the same as stop. It also doesn't give enough time for the slowdown value to converge. Increase it to 32 will smooth out the system.
Closes https://github.com/facebook/rocksdb/pull/1562

Differential Revision: D4218519

Pulled By: siying

fbshipit-source-id: 95e4088
2016-11-23 09:24:15 -08:00
Yi Wu
dfb6fe6755 Unified InlineSkipList::Insert algorithm with hinting
Summary:
This PR is based on nbronson's diff with small
modifications to wire it up with existing interface. Comparing to
previous version, this approach works better for inserting keys in
decreasing order or updating the same key, and impose less restriction
to the prefix extractor.

---- Summary from original diff ----

This diff introduces a single InlineSkipList::Insert that unifies
the existing sequential insert optimization (prev_), concurrent insertion,
and insertion using externally-managed insertion point hints.

There's a deep symmetry between insertion hints (cursors) and the
concurrent algorithm.  In both cases we have partial information from
the recent past that is likely but not certain to be accurate.  This diff
introduces the struct InlineSkipList::Splice, which encodes predecessor
and successor information in the same form that was previously only used
within a single call to InsertConcurrently.  Splice holds information
about an insertion point that can be used to levera
Closes https://github.com/facebook/rocksdb/pull/1561

Differential Revision: D4217283

Pulled By: yiwu-arbug

fbshipit-source-id: 33ee437
2016-11-22 14:09:13 -08:00