Skip to content

R
rocksdb

kvdb / rocksdb
11 个月 前同步成功

通知 0
Star 0
Fork 0
R
rocksdb

体验新版 GitCode，发现更多精彩内容 >>

切换分支/标签
  1. 16 6月, 2018 7 次提交
  3. 14 6月, 2018 4 次提交
    • A
      Check with PosixEnv before opening LOCK file (#3993) · 1f32dc7d
      由 Andrew Kryczka 提交于 
      Summary:
Rebased and resubmitting #1831 on behalf of stevelittle.

The problem is when a single process attempts to open the same DB twice, the second attempt fails due to LOCK file held. If the second attempt had opened the LOCK file, it'll now need to close it, and closing causes the file to be unlocked. Then, any subsequent attempt to open the DB will succeed, which is the wrong behavior.

The solution was to track which files a process has locked in PosixEnv, and check those before opening a LOCK file.

Fixes #1780.
Closes https://github.com/facebook/rocksdb/pull/3993

Differential Revision: D8398984

Pulled By: ajkr

fbshipit-source-id: 2755fe66950a0c9de63075f932f9e15768041918
      1f32dc7d
    • A
      Run manual compaction in stress/crash tests (#3936) · 7497f992
      由 Andrew Kryczka 提交于 
      Summary:
- Add support to `db_stress` for `CompactRange`
- Enable `CompactRange` and `CompactFiles` in crash tests
Closes https://github.com/facebook/rocksdb/pull/3936

Differential Revision: D8230953

Pulled By: ajkr

fbshipit-source-id: 208f9980b5bc8c204b1fa726e83791ad674e21e8
      7497f992
    • A
      Choose unique keys faster in db_stress (#3990) · dd216dd7
      由 Andrew Kryczka 提交于 
      Summary:
db_stress initialization randomly chooses a set of keys to not overwrite. It was doing it separately for each column family. That caused 30+ second initialization times for the non-simple crash tests, which have 10 CFs. This PR:

- reuses the same set of randomly chosen no-overwrite keys across all CFs
- logs a couple more timestamps so we can more easily see initialization time
Closes https://github.com/facebook/rocksdb/pull/3990

Differential Revision: D8393821

Pulled By: ajkr

fbshipit-source-id: d0b263a298df607285ffdd8b0983ff6575cc6c34
      dd216dd7
    • A
      Avoid acquiring SyncPoint mutex when it is disabled (#3991) · a7204018
      由 Andrew Kryczka 提交于 
      Summary:
In `db_stress` profile the vast majority of CPU time is spent acquiring the `SyncPoint` mutex. I mistakenly assumed #3939 had fixed this mutex contention problem by disabling `SyncPoint` processing. But actually the lock was still being acquired just to check whether processing is enabled. We can avoid that overhead by using an atomic to track whether it's enabled.
Closes https://github.com/facebook/rocksdb/pull/3991

Differential Revision: D8393825

Pulled By: ajkr

fbshipit-source-id: 5bc4e3c722ee7304e7a9c2439998c456b05a6897
      a7204018
  5. 13 6月, 2018 4 次提交
    • S
      Fix regression bug of Prev() with upper bound (#3989) · d82f1421
      由 Siying Dong 提交于 
      Summary:
A recent change pushed down the upper bound checking to child iterators. However, this causes the logic of following sequence wrong:
  Seek(key);
  if (!Valid()) SeekToLast();
Because !Valid() may be caused by upper bounds, rather than the end of the iterator. In this case SeekToLast() points to totally wrong places. This can cause wrong results, infinite loops, or segfault in some cases.
This sequence is called when changing direction from forward to backward. And this by itself also implicitly happen during reseeking optimization in Prev().

Fix this bug by using SeekForPrev() rather than this sequuence, as what is already done in prefix extrator case.
Closes https://github.com/facebook/rocksdb/pull/3989

Differential Revision: D8385422

Pulled By: siying

fbshipit-source-id: 429e869990cfd2dc389421e0836fc496bed67bb4
      d82f1421
    • A
      Fix argument mismatch in BlockBasedTableBuilder (#3974) · 9d347332
      由 Andrew Kryczka 提交于 
      Summary:
The sixth argument should be `key_includes_seq` bool, the seventh a `GetContext*`. We were mistakenly passing the `GetContext*` as the sixth argument and relying on the default (nullptr) for the seventh. This would make statistics inaccurate, at least.

Blame: 402b7aa0
Closes https://github.com/facebook/rocksdb/pull/3974

Differential Revision: D8344907

Pulled By: ajkr

fbshipit-source-id: 3ad865a0541d6d30f75dfc726352788118cfe12e
      9d347332
    • S
      Fix a crash in WinEnvIO::GetSectorSize (#3975) · 9c7da963
      由 shpala 提交于 
      Summary:
Fix a crash in `WinEnvIO::GetSectorSize` that happens on old Windows systems (e.g Windows 7).
On old Windows systems that don't support querying StorageAccessAlignmentProperty using IOCTL_STORAGE_QUERY_PROPERTY, the flow calls a different DeviceIoControl with nullptr as lpBytesReturned.
When the code reaches this point, we get an access violation.
Closes https://github.com/facebook/rocksdb/pull/3975

Differential Revision: D8385186

Pulled By: ajkr

fbshipit-source-id: fae4c9b4b0a52c8a10182e1b35bcaa30dc393bbb
      9c7da963
    • F
      Remove restart point from the properties_block (#3970) · 35932753
      由 Fenggang Wu 提交于 
      Summary:
Property block will be read sequentially and cached in a heap located
object, so there's no need for restart points. Thus we set the restart
interval to infinity to save space.
Closes https://github.com/facebook/rocksdb/pull/3970

Differential Revision: D8332586

Pulled By: fgwu

fbshipit-source-id: 899c3267832a81d0f084ec2db6b387332f461134
      35932753
  7. 09 6月, 2018 1 次提交
  9. 08 6月, 2018 3 次提交
    • Y
      Fix build errors. · 3470c758
      由 Yanqin Jin 提交于 
      Summary: Closes https://github.com/facebook/rocksdb/pull/3967

Differential Revision: D8322775

Pulled By: riversand963

fbshipit-source-id: bd73067bd5d3ed4627348f0685bc499359ad6442
      3470c758
    • Z
      Fixed the fprintf of uint64_t by using PRIu64 (#3963) · 23e1d236
      由 Zhichao Cao 提交于 
      Summary:
Fixed the fprintf format of uint64_t by using PRIu64 in file tools/ldb_cmd.cc
Closes https://github.com/facebook/rocksdb/pull/3963

Differential Revision: D8306179

Pulled By: zhichao-cao

fbshipit-source-id: 597dcd55321576801bbf2cf4714736ebc4750a0c
      23e1d236
    • Y
      Refactoring db_stress.cc (#3902) · 0a0860a5
      由 Yanqin Jin 提交于 
      Summary:
We use `db_stress.cc` intensively to test and verify the behavior of RocksDB. Sometimes we need to add new tests for recently added features. Original `StressTest` class provides many general functionality that can be leveraged by other tests. Therefore, in this refactoring PR, I try to identify the general operations as well as operations that future tests most likely want to customize. Future tests can inherit `StressTest` and overriding the virtual functions to test custom logic.
Closes https://github.com/facebook/rocksdb/pull/3902

Differential Revision: D8284607

Pulled By: riversand963

fbshipit-source-id: 019302d04665a2b18334b6d05d04a477168c8ea4
      0a0860a5
  11. 07 6月, 2018 3 次提交
  13. 06 6月, 2018 3 次提交
  15. 05 6月, 2018 4 次提交
    • M
      Extend some tests to format_version=3 (#3942) · d0c38c0c
      由 Maysam Yabandeh 提交于 
      Summary:
format_version=3 changes the format of SST index. This is however not being tested currently since tests only work with the default format_version which is currently 2. The patch extends the most related tests to also test for format_version=3.
Closes https://github.com/facebook/rocksdb/pull/3942

Differential Revision: D8238413

Pulled By: maysamyabandeh

fbshipit-source-id: 915725f55753dd8e9188e802bf471c23645ad035
      d0c38c0c
    • A
      Fix singleton destruction order of PosixEnv and SyncPoint (#3951) · 22101529
      由 Andrew Kryczka 提交于 
      Summary:
Ensure the PosixEnv singleton is destroyed first since its destructor waits for background threads to all complete. This ensures background threads cannot hit sync points after the SyncPoint singleton is destroyed, which was previously possible.
Closes https://github.com/facebook/rocksdb/pull/3951

Differential Revision: D8265295

Pulled By: ajkr

fbshipit-source-id: 7738dd458c5d993a78377dd0420e82badada81ab
      22101529
    • M
      Fix clang analyze · ab2254be
      由 Manuel Ung 提交于 
      Summary:
This fixes the errors as reported here:
https://github.com/facebook/rocksdb/pull/3941#issuecomment-394424043
Closes https://github.com/facebook/rocksdb/pull/3950

Differential Revision: D8263086

Pulled By: lth

fbshipit-source-id: 5e148d489cab2153e5846d16979a0a1f2d677d57
      ab2254be
    • D
      Provide a way to override windows memory allocator with jemalloc for ZSTD · f4b72d70
      由 Dmitri Smirnov 提交于 
      Summary:
Windows does not have LD_PRELOAD mechanism to override all memory allocation functions and ZSTD makes use of C-tuntime calloc. During flushes and compactions default system allocator fragments and the system slows down considerably.

For builds with jemalloc we employ an advanced ZSTD context creation API that re-directs memory allocation to jemalloc. To reduce the cost of context creation on each block we cache ZSTD context within the block based table builder while a new SST file is being built, this will help all platform builds including those w/o jemalloc. This avoids system allocator fragmentation and improves the performance.

The change does not address random reads and currently on Windows reads with ZSTD regress as compared with SNAPPY compression.
Closes https://github.com/facebook/rocksdb/pull/3838

Differential Revision: D8229794

Pulled By: miasantreble

fbshipit-source-id: 719b622ab7bf4109819bc44f45ec66f0dd3ee80d
      f4b72d70
  17. 04 6月, 2018 1 次提交
    • A
      Fix crash test check for direct I/O · 4f297ad0
      由 Andrew Kryczka 提交于 
      Summary:
We need to keep the DB directory around since the direct IO check in "db_crashtest.py" relies on it existing. This PR fixes an issue where it was removed after each stress test run during the second half of whitebox crash testing.
Closes https://github.com/facebook/rocksdb/pull/3946

Differential Revision: D8247998

Pulled By: ajkr

fbshipit-source-id: 4e7cffbdab9b40df125e7842d0d59916e76261d3
      4f297ad0
  19. 02 6月, 2018 5 次提交
    • Z
      Fix test for rocksdb_lite: hide incompatible option kDirectIO · 50d7ac0e
      由 Zhongyi Xie 提交于 
      Summary:
Previous commit https://github.com/facebook/rocksdb/pull/3935 unhide a few test options which includes kDirectIO. However it's not supported by RocksDB lite. Need to hide this option from the lite build.
Closes https://github.com/facebook/rocksdb/pull/3943

Differential Revision: D8242757

Pulled By: miasantreble

fbshipit-source-id: 1edfad3a5d01a46bfb7eedee765981ebe02c500a
      50d7ac0e
    • A
      Copy Get() result when file reads use mmap · fea2b1df
      由 Andrew Kryczka 提交于 
      Summary:
For iterator reads, a `SuperVersion` is pinned to preserve a snapshot of SST files, and `Block`s are pinned to allow `key()` and `value()` to return pointers directly into a RocksDB memory region. This works for both non-mmap reads, where the block owns the memory region, and mmap reads, where the file owns the memory region.

For point reads with `PinnableSlice`, only the `Block` object is pinned. This works for non-mmap reads because the block owns the memory region, so even if the file is deleted after compaction, the memory region survives. However, for mmap reads, file deletion causes the memory region to which the `PinnableSlice` refers to be unmapped.   The result is usually a segfault upon accessing the `PinnableSlice`, although sometimes it returned wrong results (I repro'd this a bunch of times with `db_stress`).

This PR copies the value into the `PinnableSlice` when it comes from mmap'd memory. We can tell whether the `Block` owns its memory using `Block::cachable()`, which is unset when reads do not use the provided buffer as is the case with mmap file reads. When that is false we ensure the result of `Get()` is copied.

This feels like a short-term solution as ideally we'd have the `PinnableSlice` pin the mmap'd memory so we can do zero-copy reads. It seemed hard so I chose this approach to fix correctness in the meantime.
Closes https://github.com/facebook/rocksdb/pull/3881

Differential Revision: D8076288

Pulled By: ajkr

fbshipit-source-id: 31d78ec010198723522323dbc6ea325122a46b08
      fea2b1df
    • A
      Configure direct I/O statically in db_stress · 88c3ee2d
      由 Andrew Kryczka 提交于 
      Summary:
Previously `db_stress` attempted to configure direct I/O dynamically in `SetOptions()` which had multiple problems (ummm must've never been tested):

- It's a DB option so SetDBOptions should've been called instead
- It's not a dynamic option so even SetDBOptions would fail
- It required enabling SyncPoint to mask O_DIRECT since it had no way to detect whether the DB directory was in tmpfs or not. This required locking that consumed ~80% of db_stress CPU.

In this PR I delete the broken dynamic config and instead configure it statically, only enabling it if the DB directory truly supports O_DIRECT.
Closes https://github.com/facebook/rocksdb/pull/3939

Differential Revision: D8238120

Pulled By: ajkr

fbshipit-source-id: 60bb2deebe6c9b54a3f788079261715b4a229279
      88c3ee2d
    • M
      Extend existing unit tests to run with WriteUnprepared as well · 01e3c30d
      由 Manuel Ung 提交于 
      Summary:
As titled.

I have not extended the Compatibility tests because the new WAL markers are still unimplemented.
Closes https://github.com/facebook/rocksdb/pull/3941

Differential Revision: D8238394

Pulled By: lth

fbshipit-source-id: 980e3d44837bbf2cfa64047f9738f559dfac4b1d
      01e3c30d
    • S
      add c api rocksdb_sstfilewriter_file_size · 89b37081
      由 straw 提交于 
      Summary: Closes https://github.com/facebook/rocksdb/pull/3922

Differential Revision: D8208528

Pulled By: ajkr

fbshipit-source-id: d384fe53cf526f2aadc7b79a423ce36dbd3ff224
      89b37081
  21. 01 6月, 2018 5 次提交