Compare commits

...

298 Commits

Author SHA1 Message Date
sdong 4d6ba85469 Disable error as warning 2019-11-05 11:03:59 -08:00
sdong c1ba8bee23 Add one more #include<functional> 2019-11-05 11:03:58 -08:00
sdong f6acd9661d Add some include<functional> 2019-10-31 14:29:25 -07:00
sdong 06b24a3eb3 [FB Internal] Point to the latest tool chain. 2019-10-31 14:29:24 -07:00
sdong 1a79de2ae0 [fb only] revert unintended change of USE_SSE
The previuos change that use gcc-5 set USE_SSE to wrong flag by mistake. Fix it.
2017-07-17 22:22:06 -07:00
sdong f1b6077755 [FB Only] use gcc-5 2017-07-17 21:51:02 -07:00
Islam AbdelRahman 8f52326bd5 backport fbcode gcc path fix 2016-06-08 13:07:36 -07:00
Islam AbdelRahman 70ebf2275b Fix BackupableDBTest
Summary: Fix BackupableDBTest.NoDoubleCopy and BackupableDBTest.DifferentEnvs by mocking the db files in db_env instead of backup_env_

Test Plan: make check -j64

Reviewers: sdong, andrewkr

Reviewed By: andrewkr

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D57273
2016-05-18 16:02:56 -07:00
Islam AbdelRahman 8fffeb27d5 bump version to 4.7.1 2016-05-18 15:56:48 -07:00
Andrew Kryczka 8df2bc008e Retrieve file size from proper Env
Summary:
When db_env_ != backup_env_, InsertPathnameToSizeBytes() would
use the wrong Env during backup creation. This happened because this function
used backup_env_ instead of db_env_ to get WAL/data file sizes.

This diff adds an argument to InsertPathnameToSizeBytes() indicating which Env
to use.

Test Plan: ran @anirbanb's BackupTestTool

Reviewers: sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D57159
2016-05-18 15:55:32 -07:00
sdong ec03b510a6 Rename options.compaction_measure_io_stats to options.report_bg_io_stats and include flush too.
Summary: It is useful to print out IO stats in flush jobs too. Extend options.compaction_measure_io_stats to flush jobs and raname it.

Test Plan: Try db_bench and see the stats are printed out.

Reviewers: yhchiang

Reviewed By: yhchiang

Subscribers: kradhakrishnan, yiwu, IslamAbdelRahman, leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56769
2016-04-15 11:53:01 -07:00
sdong e5d21bfbf1 Fix option settable tests
Summary: In option settable tests, bytes for pointers are not all skipped, so that they may be the same as the special character and cause false positive.

Test Plan: Run the test. Manually verify the issue is not there any more.

Reviewers: IslamAbdelRahman, andrewkr

Reviewed By: IslamAbdelRahman

Subscribers: kradhakrishnan, yiwu, yhchiang, leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56553
2016-04-15 11:52:48 -07:00
sdong e189801f76 Don't run DBOptionsAllFieldsSettable under valgrind
Summary: Test DBOptionsAllFieldsSettable sometimes fails under valgrind. Move option settable tests to a separate test file and disable it in valgrind..

Test Plan: Run valgrind test and make sure the test doesn't run.

Reviewers: andrewkr, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: kradhakrishnan, yiwu, yhchiang, leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56529
2016-04-15 11:52:39 -07:00
sdong fd728c781c BlockBasedTable::PrefixMayMatch() to skip index checking if we can't find a filter block.
Summary:
In the case where we can't find a filter block, there is not much benefit of doing the binary search and see whether the index key has the prefix. With the change, we blindly return true if we can't get the filter.
It also fixes missing row cases for reverse comparator with full bloom.

Test Plan: Add a test case that used to fail.

Reviewers: yhchiang, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: kradhakrishnan, yiwu, hermanlee4, yoshinorim, leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56697
2016-04-14 09:59:05 -07:00
sdong 7761a5faa2 BlockBasedTable::PrefixMayMatch: no need to find data block after full bloom checking
Summary:
Full block checking should be a good enough indication of prefix existance. No need to further check data block.
This also fixes wrong results when using prefix bloom and reverse bitwise comparator.

Test Plan: Will add a unit test.

Reviewers: yhchiang, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: hermanlee4, yoshinorim, yiwu, kradhakrishnan, leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56625
2016-04-14 09:58:51 -07:00
Islam AbdelRahman 027f9f8cf6 Release RocksDB 4.7
Summary: Bump the version and update HISTORY.md

Test Plan: none

Reviewers: yhchiang, sdong, andrewkr

Reviewed By: andrewkr

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56469
2016-04-13 15:10:59 -07:00
Andrew Kryczka 114a1b8792 Fix build errors for windows
Summary:
- Need to use unsigned long long for 64-bit literals on windows
- Need size_t for backup meta-file length since clang doesn't let us assign size_t to int

Test Plan: backupable_db_test and options_test

Reviewers: IslamAbdelRahman, yhchiang, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D56391
2016-04-08 13:09:19 -07:00
Islam AbdelRahman 0522990358 Improve sst_dump help message
Summary:
Current Message

```
sst_dump [--command=check|scan|none|raw] [--verify_checksum] --file=data_dir_OR_sst_file [--output_hex] [--input_key_hex] [--from=<user_key>] [--to=<user_key>] [--read_num=NUM] [--show_properties] [--show_compression_sizes] [--show_compression_sizes [--set_block_size=<block_size>]]
```
New message

```
sst_dump --file=<data_dir_OR_sst_file> [--command=check|scan|raw]
    --file=<data_dir_OR_sst_file>
      Path to SST file or directory containing SST files

    --command=check|scan|raw
        check: Iterate over entries in files but dont print anything except if an error is encounterd (default command)
        scan: Iterate over entries in files and print them to screen
        raw: Dump all the table contents to <file_name>_dump.txt

    --output_hex
      Can be combined with scan command to print the keys and values in Hex

    --from=<user_key>
      Key to start reading from when executing check|scan

    --to=<user_key>
      Key to stop reading at when executing check|scan

    --read_num=<num>
      Maximum number of entries to read when executing check|scan

    --verify_checksum
      Verify file checksum when executing check|scan

    --input_key_hex
      Can be combined with --from and --to to indicate that these values are encoded in Hex

    --show_properties
      Print table properties after iterating over the file

    --show_compression_sizes
      Independent command that will recreate the SST file using 16K block size with different
      compressions and report the size of the file using such compression

    --set_block_size=<block_size>
      Can be combined with --show_compression_sizes to set the block size that will be used
      when trying different compression algorithms
```

Test Plan: none

Reviewers: yhchiang, andrewkr, kradhakrishnan, yiwu, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56325
2016-04-08 12:05:02 -07:00
sdong 0930e5e99f Update comments on include/rocksdb/perf_context.h
Summary: Some grammer mistakes in code comments in include/rocksdb/perf_context.h. Also polish it a liitlebit.

Test Plan: Not needed

Reviewers: IslamAbdelRahman, yhchiang, yiwu, andrewkr

Reviewed By: andrewkr

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56307
2016-04-08 11:27:08 -07:00
Shu Zhang 3b977bcdc8 instructing people to use java7 for hdfs (#1063) 2016-04-07 14:34:28 -07:00
sdong 1518b733eb Change default number of cache shard bit to be 6 and max_file_opening_threads to be 16.
Summary: Cache shard bit 4 is sometimes too small and 6 is a more common value picked by users. Make that default. It shouldn't hurt much to change options.max_file_opening_threads default to be 16, which will reduce the worst case DB open time.

Test Plan: Run all existing tests.

Reviewers: IslamAbdelRahman, yhchiang, kradhakrishnan, igor, andrewkr

Reviewed By: andrewkr

Subscribers: andrewkr, MarkCallaghan, leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D55047
2016-04-07 13:55:10 -07:00
flabby ada88b63f5 fix wrong assignment of level0_stop_writes_trigger in spatialdb (#1061) 2016-04-07 09:02:28 -07:00
Andrew Kryczka 2391ef7214 Embed column family name in SST file
Summary:
Added the column family name to the properties block. This property
is omitted only if the property is unavailable, such as when RepairDB()
writes SST files.

In a next diff, I will change RepairDB to use this new property for
deciding to which column family an existing SST file belongs. If this
property is missing, it will add it to the "unknown" column family (same
as its existing behavior).

Test Plan:
New unit test:

  $ ./db_table_properties_test --gtest_filter=DBTablePropertiesTest.GetColumnFamilyNameProperty

Reviewers: IslamAbdelRahman, yhchiang, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D55605
2016-04-06 23:10:32 -07:00
Igor Canadi ab4c62332e Don't use version in the error message
Summary: We use object `v` in the error message, which is not initialized if the edit is column family manipulation. This doesn't provide much useful info, so this diff is removing it. Instead, it dumps actual VersionEdit contents.

Test Plan: compiles. would be great to get tests in version_set_test.cc that cover cases where a file write fails

Reviewers: sdong, yhchiang, andrewkr

Reviewed By: andrewkr

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D56349
2016-04-06 15:00:15 -07:00
Islam AbdelRahman d02eb8d00b Fix unused variable warning
Summary:
My last diff introduced a warning when compiling under release mode
https://reviews.facebook.net/D55539

fix the warning

Test Plan:
DEBUG_LEVEL=0 make db_bench
make check

Reviewers: sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56295
2016-04-04 11:10:46 -07:00
Igor Canadi 9278097f86 Merge pull request #1056 from facebook/igorcanadi-patch-1
Create LANGUAGE-BINDINGS.md
2016-04-04 08:08:52 -07:00
Aaron Gao cc87075d63 No need to limit to 20 files in UpdateAccumulatedStats() if options.max_open_files=-1
Summary:
There is a hardcoded constraint in our statistics collection that prevents reading properties from more than 20 SST files. This means our statistics will be very inaccurate for databases with > 20 files since additional files are just ignored. The purpose of constraining the number of files used is to bound the I/O performed during statistics collection, since these statistics need to be recomputed every time the database reopened.

However, this constraint doesn't take into account the case where option "max_open_files" is -1. In that case, all the file metadata has already been read, so MaybeInitializeFileMetaData() won't incur any I/O cost. so this diff gets rid of the 20-file constraint in case max_open_files == -1.

Test Plan:
write into unit test db/db_properties_test.cc - "ValidateSampleNumber".
We generate 20 files with 2 rows and 10 files with 1 row.
If max_open_files !=-1, the `rocksdb.estimate-num-keys` should be (10*1 + 10*2)/20 * 30 = 45. Otherwise, it should be the ground truth, 50.
{F1089153}

Reviewers: andrewkr

Reviewed By: andrewkr

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D56253
2016-04-01 16:19:12 -07:00
Islam AbdelRahman 8a1a603fdb Eliminate std::deque initialization while iterating over merge operands
Summary:
This patch is similar to D52563, When we iterate over a DB with merge operands we keep creating std::queue to store the operands, optimize this by reusing merge_operands_ data member

Before the patch

```
./db_bench --benchmarks="mergerandom,readseq,readseq,readseq,readseq" --db="/dev/shm/bench_merge_memcpy_on_the_fly/" --merge_operator="put" --merge_keys=10000 --num=10000

DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
mergerandom  :       3.757 micros/op 266141 ops/sec;   29.4 MB/s ( updates:10000)
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.413 micros/op 2423538 ops/sec;  268.1 MB/s
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.451 micros/op 2219071 ops/sec;  245.5 MB/s
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.420 micros/op 2382039 ops/sec;  263.5 MB/s
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.408 micros/op 2452017 ops/sec;  271.3 MB/s

DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
mergerandom  :       3.947 micros/op 253376 ops/sec;   28.0 MB/s ( updates:10000)
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.441 micros/op 2266473 ops/sec;  250.7 MB/s
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.471 micros/op 2122033 ops/sec;  234.8 MB/s
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.440 micros/op 2271407 ops/sec;  251.3 MB/s
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.429 micros/op 2331471 ops/sec;  257.9 MB/s
```

with the patch

```
./db_bench --benchmarks="mergerandom,readseq,readseq,readseq,readseq" --db="/dev/shm/bench_merge_memcpy_on_the_fly/" --merge_operator="put" --merge_keys=10000 --num=10000

DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
mergerandom  :       4.080 micros/op 245092 ops/sec;   27.1 MB/s ( updates:10000)
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.308 micros/op 3241843 ops/sec;  358.6 MB/s
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.312 micros/op 3200408 ops/sec;  354.0 MB/s
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.332 micros/op 3013962 ops/sec;  333.4 MB/s
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.300 micros/op 3328017 ops/sec;  368.2 MB/s

DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
mergerandom  :       3.973 micros/op 251705 ops/sec;   27.8 MB/s ( updates:10000)
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.320 micros/op 3123752 ops/sec;  345.6 MB/s
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.335 micros/op 2986641 ops/sec;  330.4 MB/s
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.339 micros/op 2950047 ops/sec;  326.4 MB/s
DB path: [/dev/shm/bench_merge_memcpy_on_the_fly/]
readseq      :       0.319 micros/op 3131565 ops/sec;  346.4 MB/s
```

Test Plan: make check -j64

Reviewers: yhchiang, andrewkr, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56031
2016-04-01 15:48:55 -07:00
Islam AbdelRahman f38540b12a WriteBatchWithIndex micro optimization
Summary:
  - Put key offset and key size in WriteBatchIndexEntry
  - Use vector for comparators in WriteBatchEntryComparator

I use a slightly modified version of @yoshinorim code to benchmark
https://gist.github.com/IslamAbdelRahman/b120f4fba8d6ff7d58d2

For Put I create a transaction that put a 1000000 keys and measure the time spent without commit.
For GetForUpdate I read the keys that I added in the Put transaction.

Original time:

```
 rm -rf /dev/shm/rocksdb-example/
 ./txn_bench put 1000000
 1000000 OK Ops | took      3.679 seconds
 ./txn_bench get_for_update 1000000
 1000000 OK Ops | took      3.940 seconds
```

New Time

```
  rm -rf /dev/shm/rocksdb-example/
 ./txn_bench put 1000000
 1000000 OK Ops | took      2.727 seconds
 ./txn_bench get_for_update 1000000
 1000000 OK Ops | took      3.880 seconds
```

It looks like there is no significant improvement in GetForUpdate() but we can see ~30% improvement in Put()

Test Plan: unittests

Reviewers: yhchiang, anthony, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, yoshinorim

Differential Revision: https://reviews.facebook.net/D55539
2016-04-01 15:23:46 -07:00
Adam Retter 200654067a Merge pull request #1053 from adamretter/benchmark-java-comparator
Benchmark Java comparator vs C++ comparator
2016-04-01 13:53:15 -07:00
Andrew Kryczka f2c43a4a27 Stderr info logger
Summary:
Adapted a stderr logger from the option tests. Moved it to a separate
header so we can reuse it, e.g., from ldb subcommands for faster debugging. This
is especially useful to make errors/warnings more visible when running
"ldb repair", which involves potential data loss.

Test Plan:
ran options_test and "ldb repair"

  $ ./ldb repair --db=./tmp/
  [WARN] **** Repaired rocksdb ./tmp/; recovered 1 files; 588bytes. Some data may have been lost. ****
  OK

Reviewers: IslamAbdelRahman, yhchiang, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D56151
2016-04-01 11:06:06 -07:00
Uddipta Maity b55e2165be Rocksdb backup can store optional application specific metadata
Summary:
Rocksdb backup engine maintains metadata about backups in separate files. But,
there was no way to add extra application specific data to it. Adding support
for that.
In some use cases, applications decide to restore a backup based on some
metadata. This will help those cases to cheaply decide whether to restore or
not.

Test Plan:
Added a unit test. Existing ones are passing

Sample meta file for BinaryMetadata test-

```

1459454043
0
metadata 6162630A64656600676869
2
private/1/MANIFEST-000001 crc32 1184723444
private/1/CURRENT crc32 3505765120

```

Reviewers: sdong, ldemailly, andrewkr

Reviewed By: andrewkr

Subscribers: andrewkr, dhruba, ldemailly

Differential Revision: https://reviews.facebook.net/D56007
2016-04-01 10:56:52 -07:00
Marton Trencseni 9b51987521 Adding pin_l0_filter_and_index_blocks_in_cache feature and related fixes.
Summary:
When a block based table file is opened, if prefetch_index_and_filter is true, it will prefetch the index and filter blocks, putting them into the block cache.
What this feature adds: when a L0 block based table file is opened, if pin_l0_filter_and_index_blocks_in_cache is true in the options (and prefetch_index_and_filter is true), then the filter and index blocks aren't released back to the block cache at the end of BlockBasedTableReader::Open(). Instead the table reader takes ownership of them, hence pinning them, ie. the LRU cache will never push them out. Meanwhile in the table reader, further accesses will not hit the block cache, thus avoiding lock contention.

Test Plan:
'export TEST_TMPDIR=/dev/shm/ && DISABLE_JEMALLOC=1 OPT=-g make all valgrind_check -j32' is OK.
I didn't run the Java tests, I don't have Java set up on my devserver.

Reviewers: sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56133
2016-04-01 10:42:39 -07:00
sdong 2feafa3db9 Change some RocksDB default options
Summary: Change some RocksDB default options to make it more friendly to server workloads.

Test Plan: Run all existing tests

Reviewers: yhchiang, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: sumeet, muthu, benj, MarkCallaghan, igor, leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55941
2016-03-31 17:12:18 -07:00
Yueh-Hsuan Chiang a558830f8f Fixed compile warnings in posix_logger.h and coding.h
Summary:
Fixed the following compile warnings:

/Users/yhchiang/rocksdb/util/posix_logger.h:32:11: error: unused variable 'kDebugLogChunkSize' [-Werror,-Wunused-const-variable]
const int kDebugLogChunkSize = 128 * 1024;
          ^
/Users/yhchiang/rocksdb/util/coding.h:24:20: error: unused variable 'kMaxVarint32Length' [-Werror,-Wunused-const-variable]
const unsigned int kMaxVarint32Length = 5;
                   ^
2 errors generated.

Test Plan: make clean rocksdb

Reviewers: igor, sdong, anthony, IslamAbdelRahman, rven, kradhakrishnan, adamretter

Reviewed By: adamretter

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D56223
2016-03-31 16:01:47 -07:00
Yueh-Hsuan Chiang 51c9464dfc Merge pull request #980 from adamretter/java-arm
ARM for the Java API
2016-03-31 13:56:45 -07:00
Igor Canadi 925b5d0025 Merge pull request #1054 from DCEngines/magic12
Remove the Magic number 12 used in record size checks
2016-03-30 21:38:19 -07:00
Sandeep Joshi 63e8f1b55b Formatted lines to adhere to 80 char limit 2016-03-31 08:26:55 +05:30
Gunnar Kudrjavets 994b3bd693 Add support for UBsan builds to RocksDB
Summary:
Undefined Behavior Sanitizer (ubsan) is //a good thing// which will help us to find sneaky bugs with low cost. Please see http://developerblog.redhat.com/2014/10/16/gcc-undefined-behavior-sanitizer-ubsan/ for more details and official GCC documentation for more context: https://gcc.gnu.org/onlinedocs/gcc/Instrumentation-Options.html.

Changes itself are quite simple and pretty much imitating whatever is implemented for ASan.

Hooking the UBsan validation build to Sandcastle is a separate step and will be dealt as separate diff because code is in internal repository.

Test Plan: Make sure that that there no regressions when it comes to builds and test pass rate.

Reviewers: leveldb, sdong, IslamAbdelRahman, yhchiang

Reviewed By: yhchiang

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56049
2016-03-30 15:59:24 -07:00
Laurent Demailly 21700a5106 to/from hex refactor
Summary:
Expose the inverse of ToString(hex=true) on Slice: Slice::DecodeHex
Refactor the other implementation of to/from hex in ldb_cmd.h to use the Slice
version
(Difference between the 2 is whether 0x is expected/produced in front of the hex
string or not)
Eliminated support for invalid odd length hex string - this is now invalid
instead of having 1/2 byte set
Added (inverse of HexToString) test for LDBCommand::StringToHex which also
indirectly tests Slice::ToString(true)

After moving the original implementation from ldb_cmd.h, updated it to much simpler/efficient version
(originally/inspired from https://github.com/facebook/wdt/blob/master/util/EncryptionUtils.cpp#L140-L169 )

Test Plan: run tests

Reviewers: uddipta, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56121
2016-03-30 14:36:48 -07:00
Sandeep Joshi 24420947d9 Replace kHeader by WriteBatchInternal::kHeader in few more places
kHeader was moved from write_batch.cc to header file because
it is being used wherever the number "12" was being used to
check for record size
2016-03-30 23:13:00 +05:30
Sandeep Joshi 3bdbe89614 Merge branch 'magic12' of https://github.com/dcengines/rocksdb into magic12 2016-03-30 23:09:39 +05:30
zensan 78711524b7 In all the places where log records are read, there was a check that
record.size() should not be less than 12.

This "magic number" seems to be the WriteBatch header (8 byte sequence
and 4 byte count).   Replaced all the places where "12" was used
by WriteBatchInternal::kHeader.
2016-03-30 23:05:22 +05:30
Islam AbdelRahman e7c64fb115 Imporve sst_file_manager comment
Summary: Improve the comment for sst_file_manager

Test Plan: none

Reviewers: yhchiang, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D56001
2016-03-29 13:20:26 -07:00
Islam AbdelRahman 99ffb3d533 Fix perf_context::merge_operator_time_nanos calculation
Summary: We were not measuring the time spent in merge_operator when called from Version::Get()

Test Plan: added a unittest

Reviewers: sdong, yhchiang

Reviewed By: yhchiang

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55905
2016-03-25 18:29:43 -07:00
sdong 07bb12d97d Update internal jemalloc and other versions
Summary: Update jemalloc to newer versions, as well as some other dependencies. Only effective for FB internal.

Test Plan: See all tests run

Reviewers: IslamAbdelRahman, MarkCallaghan

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55947
2016-03-25 10:42:20 -07:00
Yueh-Hsuan Chiang ad2fdaa823 Correct a typo in a comment
Summary: Correct a typo in a comment

Test Plan: No code change.

Reviewers: sdong, kradhakrishnan, IslamAbdelRahman

Reviewed By: kradhakrishnan, IslamAbdelRahman

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D55803
2016-03-24 19:39:13 -07:00
Yueh-Hsuan Chiang be9816b3d9 Fix data race issue when sub-compaction is used in CompactionJob
Summary:
When subcompaction is used, all subcompactions share the same Compaction
pointer in CompactionJob while each subcompaction all keeps their mutable
stats in SubcompactionState.  However, there're still some mutable part
that is currently store in the shared Compaction pointer.

This patch makes two changes:

1. Make the shared Compaction pointer const so that it can never be modified
   during the compaction.
2. Move necessary states from Compaction to SubcompactionState.
3. Make functions of Compaction const if the function does not modify
   its internal state.

Test Plan: rocksdb and MyRocks test

Reviewers: sdong, kradhakrishnan, andrewkr, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: andrewkr, dhruba, yoshinorim, gunnarku, leveldb

Differential Revision: https://reviews.facebook.net/D55923
2016-03-24 19:36:39 -07:00
Islam AbdelRahman e3802531f1 Merge pull request #1050 from yuslepukhin/support_db_test2
Add support for db_test2 for dev and CI runs
2016-03-23 17:59:30 -07:00
Dmitri Smirnov e7cc49cbdf Add support for db_test2 for dev and CI runs 2016-03-23 16:33:57 -07:00
sdong 3996770d0b Add comments to perf_context skip counters
Summary: Document the skipped counters in perf context more clearly.

Test Plan: Comment only.

Reviewers: IslamAbdelRahman, yhchiang, MarkCallaghan

Reviewed By: MarkCallaghan

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55833
2016-03-23 14:26:25 -07:00
Mike Kolupaev 4e85b74790 Make WritableFileWrapper not screw up preallocation
Summary:
Without this diff, this is what happens to compaction output file if it's a subclass of WritableFileWrapper:
- during compaction, all `PrepareWrite()` calls update `last_preallocated_block_` of the `WritableFileWrapper` itself, not of `target_`, since `PrepareWrite()` is not virtual,
- `PrepareWrite()` calls `Allocate()`, which is virtual; it does `fallocate()` on `target_`,
- after writing data, `target_->Close()` calls `GetPreallocationStatus()` of `target_`; it returns `last_preallocated_block_` of `target_`, which is zero because it was never touched before,
- `target_->Close()` doesn't call `ftruncate()`; file remains big.

This diff fixes it in a straightforward way, by making the methods virtual. `WritableFileWrapper` ends up having the useless fields `last_preallocated_block_` and `preallocation_block_size_`. I think ideally the preallocation logic should be outside `WritableFile`, the same way as `log_writer.h` and `file_reader_writer.h` moved some non-platform-specific logic out of Env, but that's probably not worth the effort now.

Test Plan: `make -j check`; I'm going to deploy it on our test tier and see if it fixes space reclamation problem there

Reviewers: yhchiang, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, sdong

Differential Revision: https://reviews.facebook.net/D54681
2016-03-23 09:14:56 -07:00
Siying Dong ec458dcdeb Merge pull request #1047 from PraveenSinghRao/wal_filter_ex
Avoid overloaded virtual function
2016-03-22 21:48:39 -07:00
Praveen Rao 583157f710 Avoid overloaded virtual function 2016-03-22 17:10:31 -07:00
sdong b9d4fa890b Options settable tests to use a different special charactor
Summary: Something changed and the special charactor seems to be conflict with an exisitng value. Change it to unblock the build.

Test Plan: Run the test and make sure it passes

Reviewers: kradhakrishnan

Reviewed By: kradhakrishnan

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55845
2016-03-22 16:21:53 -07:00
Siying Dong 60e34baef2 Merge pull request #1044 from PraveenSinghRao/wal_filter_ex
Allow per-column-family processing in wal_filter (publish log numbers for column family to wal_filter, and provide log number in the record callback)
2016-03-22 16:03:31 -07:00
Praveen Rao 136b8e0cad Merge from master 2016-03-22 12:38:44 -07:00
Praveen Rao 2dcbb3b4f3 Addressed review comments 2016-03-22 12:07:15 -07:00
sdong b1fafcaca6 Revert "Adding pin_l0_filter_and_index_blocks_in_cache feature."
This reverts commit 522de4f59e.

It has bug of index block cleaning up.
2016-03-21 11:50:42 -07:00
sdong 5f8741a69d Revert "Fix failing Java unit test."
This reverts commit d7ae42b0f8.

This is reverted as auto buld failure. This commit itself doesn't have any problem. Reverting as it depends on the commit to revert.
2016-03-21 11:46:57 -07:00
sdong 43bbb56198 tools/check_format_compatible.sh to use consistent version when testing backward and forward compatibility
Summary: Test seems to fail if we don't use consistent version between testing forward and backward compatibility.

Test Plan: Run the script (with some version removed manually to make it shorter)

Reviewers: IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55773
2016-03-21 11:13:26 -07:00
Marton Trencseni d7ae42b0f8 Fix failing Java unit test.
Test Plan: sent diff to sdong, passes :)

Reviewers: sdong

Reviewed By: sdong

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55749
2016-03-19 15:53:56 +00:00
agiardullo fbbb8a6144 Add test for Snapshot 0
Summary:
I ran into this assert when stress testing transactions.  It's pretty easy to repro.

Changing VersionSet::last_sequence_ to start at 1 seems pretty straightforward.  We would just need to change the 4 callers of SetLastSequence(), including recovery code.  I'd make this change myself, but I do not have enough time to test changes to recovery code-paths this week.  But checking in this test case (disabled) for future fixing.

Test Plan: n/a

Reviewers: yhchiang, kradhakrishnan, andrewkr, anthony, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D55311
2016-03-18 16:16:20 -07:00
Andrew Kryczka e182f03c1e Add unit tests for RepairDB
Summary:
Basic test cases:

- Manifest is lost or corrupt
- Manifest refers to too many or too few SST files
- SST file is corrupt
- Unflushed data is present when RepairDB is called

Depends on D55065 for its CreateFile() function in file_utils

Test Plan: Ran the tests.

Reviewers: IslamAbdelRahman, yhchiang, yoshinorim, sdong

Reviewed By: sdong

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55485
2016-03-18 15:18:42 -07:00
Praveen Rao 7d371863e5 travis build fixes 2016-03-18 14:43:22 -07:00
Karthikeyan Radhakrishnan fbea4dc660 Merge pull request #1042 from SherlockNoMad/HistFix
Fix in HistogramWindowingImpl
2016-03-18 14:27:55 -07:00
sdong 780d2b04cb Update format compatible checking tool
Summary: After introducing a less forward-compatible change, update the backward compatible checking tool.

Test Plan: Run it.

Reviewers: IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: IslamAbdelRahman, leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55695
2016-03-18 14:07:15 -07:00
Praveen Rao 4f1c74a46e merge from master 2016-03-18 12:48:01 -07:00
Praveen Rao f8c2189307 Publish log numbers for column family to wal_filter, and provide log number in the record callback 2016-03-18 12:32:15 -07:00
Marton Trencseni 44756260ae Reset block cache in failing unit test.
Test Plan: make -j40 check OPT=-g, on both /tmp and /dev/shm

Reviewers: sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55701
2016-03-18 06:13:54 +00:00
Marton Trencseni 522de4f59e Adding pin_l0_filter_and_index_blocks_in_cache feature.
Summary:
When a block based table file is opened, if prefetch_index_and_filter is true, it will prefetch the index and filter blocks, putting them into the block cache.
What this feature adds: when a L0 block based table file is opened, if pin_l0_filter_and_index_blocks_in_cache is true in the options (and prefetch_index_and_filter is true), then the filter and index blocks aren't released back to the block cache at the end of BlockBasedTableReader::Open(). Instead the table reader takes ownership of them, hence pinning them, ie. the LRU cache will never push them out. Meanwhile in the table reader, further accesses will not hit the block cache, thus avoiding lock contention.
When the table reader is destroyed, it releases the pinned blocks (if there were any). This has to happen before the cache is destroyed, so I had to introduce a TableReader::Close(), to guarantee the order of destruction.

Test Plan:
Added two unit tests for this. Existing unit tests run fine (default is pin_l0_filter_and_index_blocks_in_cache=false).

DISABLE_JEMALLOC=1 OPT=-g make all valgrind_check -j32
  Mac: OK.
  Linux: with D55287 patched in it's OK.

Reviewers: sdong

Reviewed By: sdong

Subscribers: andrewkr, leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54801
2016-03-17 22:40:01 +00:00
Islam AbdelRahman be22271265 Merge pull request #1041 from yuslepukhin/adjust_for_jemalloc
Latest versions of Jemalloc library do not require je_init()/je_unint()
2016-03-17 14:50:19 -07:00
SherlockNoMad 4ecc03c039 Fix in HistogramWindowingImpl 2016-03-17 14:28:41 -07:00
Dmitri Smirnov 2ca0994cf7 Latest versions of Jemalloc library do not require je_init()/je_unint()
calls. #ifdef in the source code and make this a default build option.
2016-03-17 11:25:20 -07:00
Gunnar Kudrjavets 90aff0c444 Update --max_write_buffer_number for compaction benchmarks
Summary: For compactions benchmarks (both level and universal) we'll use `--max_write_buffer_number=4`. For all the other benchmarks which don't customize the value of `--max_background_flushes` we'll continue using `--max_write_buffer_number=8`.

Test Plan:
To validate basic correctness and command-line options:

```
cd ~/rocksdb
NKEYS=10000000 ./tools/run_flash_bench.sh
```

Reviewers: MarkCallaghan

Reviewed By: MarkCallaghan

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55497
2016-03-17 10:14:23 -07:00
Andrew Kryczka 72224104d3 Forge current file for checkpoint
Summary:
This fixes a similar issue as D54711: "CURRENT" file can mutate between
GetLiveFiles() and copy to the tmp directory, in which case it would reference
the wrong manifest filename. To fix this, I forge the "CURRENT" file such that
it simply contains the filename for the manifest returned by GetLiveFiles().

- Changed CreateCheckpoint() to forge current file
- Added CreateFile() utility function
- Added test case that rolls manifest during checkpoint creation

Test Plan:
  $ ./checkpoint_test

Reviewers: sdong, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D55065
2016-03-17 10:07:21 -07:00
dhruba borthakur 33d568611d Merge pull request #1040 from bureau14/master
Fixes warnings and ensure correct int behavior on 32-bit platforms.
2016-03-17 03:23:30 -07:00
Edouard A 02e62ebbc8 Fixes warnings and ensure correct int behavior on 32-bit platforms. 2016-03-16 22:57:57 +01:00
Siying Dong 9cad56861a Merge pull request #1039 from bureau14/master
Improve documentation of the allow_os_buffer parameter.
2016-03-16 10:30:01 -07:00
Edouard Alligand 3d29f91466 Improve documentation of the allow_os_buffer parameter. 2016-03-16 15:37:55 +01:00
Islam AbdelRahman 3ff98bd209 Fix no compression test
Summary:
DBBlockCacheTest.TestWithCompressedBlockCache is depending on compression using snappy, so this test fail when snappy is not available
block this test when we don't have snappy

https://ci-builds.fb.com/view/rocksdb/job/rocksdb_no_compression/833/console

Test Plan: run the test when compression libraries are not avaliable

Reviewers: sdong, yiwu

Reviewed By: yiwu

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55413
2016-03-15 12:17:40 -07:00
Igor Canadi b9cc42a72c Merge pull request #1038 from SherlockNoMad/HistFix
Fix FB internal CI build failure
2016-03-15 11:43:06 -07:00
SherlockNoMad f76b260eff Fix FB internal CI build failure 2016-03-15 11:38:15 -07:00
Siying Dong 774922c680 Merge pull request #1026 from SherlockNoMad/Hist
Histogram Concurrency Improvement and Time-Windowing Support
2016-03-15 11:27:54 -07:00
Igor Canadi 17b879b91e Merge pull request #1037 from SherlockNoMad/BuildFix
Fix AppVeyor build error
2016-03-15 11:15:56 -07:00
SherlockNoMad f11b0df121 Fix AppVeyor build error 2016-03-15 10:57:33 -07:00
Gunnar Kudrjavets 6b03f93d4f Fix the build break on Ubuntu 15.10 when gcc 5.2.1 is used
Summary:
Using gcc 5.2.1 to compile RocksDB on Ubuntu 15.10 results in a warning about unused variable. Warning is treated as an error and all of it results in a build break. Fix the issue and make sure that everything compiles with that particular configuration.

P.S. Lint complains about a non-ASCII character used in the source code ("Wagner-Fischer"). Fix this as well.

Test Plan:
  # Everything compiler cleanly on the system which exhibited the problem.
  # `make clean && make -j 16` on CentOS.
  # `make clean && USE_CLANG=1 make -j 16` on CentOS.

Reviewers: sdong, igor

Reviewed By: igor

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55503
2016-03-15 10:30:10 -07:00
Gunnar Kudrjavets 697fab820a Updates to RocksDB subcompaction benchmarking script
Summary: Set of updates to the subcompaction benchmark script which are based on our internal discussions. The intent behind the changes is to make sure that the scripts will correctly reflect how we're doing the actual benchmarking.

Test Plan: Tested by exercising the full set of compaction benchmarks and validating the execution and consistency of results.

Reviewers: MarkCallaghan, sdong, yhchiang

Reviewed By: yhchiang

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55461
2016-03-14 23:09:04 -07:00
SherlockNoMad 58379bfb55 remove division from histogramwidowing impl 2016-03-14 21:41:13 -07:00
Dhruba Borthakur 1a2cc27e01 ColumnFamilyOptions SanitizeOptions is buggy on 32-bit platforms.
Summary:
The pre-existing code is trying to clamp between 65,536 and 0,
resulting in clamping to 65,536, resulting in very small buffers,
resulting in ShouldFlushNow() being true quite easily,
resulting in assertion failing and database performance
being "not what it should be".

https://github.com/facebook/rocksdb/issues/1018

Test Plan: make check

Reviewers: sdong, andrewkr, IslamAbdelRahman, yhchiang, igor

Reviewed By: igor

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55455
2016-03-14 16:21:54 -07:00
Igor Canadi e778c34e78 Merge pull request #1035 from bureau14/master
Added quasardb to the USERS.md file
2016-03-14 15:50:55 -07:00
Edouard A 5bd3da1c59 Added quasardb to the USERS.md file 2016-03-14 23:48:28 +01:00
sdong b2ae5950ba Index Reader should not be reused after DB restart
Summary:
In block based table reader, wow we put index reader to block cache, which can be retrieved after DB restart. However, index reader may reference internal comparator, which can be destroyed after DB restarts, causing problems.
Fix it by making cache key identical per table reader.

Test Plan: Add a new test which failed with out the commit but now pass.

Reviewers: IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: maro, yhchiang, kradhakrishnan, leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55287
2016-03-14 10:04:09 -07:00
Andrew Kryczka 0267655dad Update change log for 4.6 release
Summary: as titled

Test Plan: N/A

Reviewers: sdong, kradhakrishnan, anthony, yhchiang, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55323
2016-03-12 13:51:57 -08:00
Andrew Kryczka 08304c0867 Expose RepairDB as ldb command
Summary: This will make it easier for admins and devs to use RepairDB.

Test Plan:
Tried deleting the manifest and verified it recovers:

  $ ldb --create_if_missing --db=/tmp/test_db put ok ok
  $ rm -f /tmp/test_db/MANIFEST-000001
  $ ./ldb --db=/tmp/test_db repair
  $ ldb --db=/tmp/test_db get ok
  ok

Reviewers: yhchiang, sdong, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55359
2016-03-12 13:50:20 -08:00
SherlockNoMad fd664a27b5 Fix Build Error 2016-03-11 22:56:25 -08:00
Islam AbdelRahman 580fede347 Aggregate hot Iterator counters in LocalStatistics (DBIter::Next perf regression)
Summary:
This patch bump the counters in the frequent code path DBIter::Next() / DBIter::Prev() in a local data members and send them to Statistics when the iterator is destroyed
A better solution will be to have thread_local implementation for Statistics

New performance
```
readseq      :       0.035 micros/op 28597881 ops/sec; 3163.7 MB/s
     1,851,568,819      stalled-cycles-frontend   #   31.29% frontend cycles idle    [49.86%]
       884,929,823      stalled-cycles-backend    #   14.95% backend  cycles idle    [50.21%]
readreverse  :       0.071 micros/op 14077393 ops/sec; 1557.3 MB/s
     3,239,575,993      stalled-cycles-frontend   #   27.36% frontend cycles idle    [49.96%]
     1,558,253,983      stalled-cycles-backend    #   13.16% backend  cycles idle    [50.14%]

```

Existing performance

```
readreverse  :       0.174 micros/op 5732342 ops/sec;  634.1 MB/s
    20,570,209,389      stalled-cycles-frontend   #   70.71% frontend cycles idle    [50.01%]
    18,422,816,837      stalled-cycles-backend    #   63.33% backend  cycles idle    [50.04%]

readseq      :       0.119 micros/op 8400537 ops/sec;  929.3 MB/s
    15,634,225,844      stalled-cycles-frontend   #   79.07% frontend cycles idle    [49.96%]
    14,227,427,453      stalled-cycles-backend    #   71.95% backend  cycles idle    [50.09%]
```

Test Plan: unit tests

Reviewers: yhchiang, sdong, igor

Reviewed By: sdong

Subscribers: andrewkr, dhruba

Differential Revision: https://reviews.facebook.net/D55107
2016-03-11 19:01:12 -08:00
SherlockNoMad 54f6b9e162 Histogram Concurrency Improvement and Time-Windowing Support 2016-03-11 16:54:25 -08:00
agiardullo 790252805d Add multithreaded transaction test
Summary: Refactored db_bench transaction stress tests so that they can be called from unit tests as well.

Test Plan: run new unit test as well as db_bench

Reviewers: yhchiang, IslamAbdelRahman, sdong

Reviewed By: IslamAbdelRahman

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D55203
2016-03-11 15:16:52 -08:00
Baris Yazici e8e6cf0173 fix: handle_fatal_signal (sig=6) in std::vector<std::string, std::allocator<std::string> >::_M_range_check | c++/4.8.2/bits/stl_vector.h:794 #174
Summary:
Fix for https://github.com/facebook/mysql-5.6/issues/174

When there is no old files to purge, vector.at(i) function was crashing

if (old_info_log_file_count != 0 &&
      old_info_log_file_count >= db_options_.keep_log_file_num) {
    std::sort(old_info_log_files.begin(), old_info_log_files.end());
    size_t end = old_info_log_file_count - db_options_.keep_log_file_num;
    for (unsigned int i = 0; i <= end; i++) {
      std::string& to_delete = old_info_log_files.at(i);

Added check to old_info_log_file_count be non zero.

Test Plan: run existing tests

Reviewers: gunnarku, vasilep, sdong, yhchiang

Reviewed By: yhchiang

Subscribers: andrewkr, webscalesql-eng, dhruba

Differential Revision: https://reviews.facebook.net/D55245
2016-03-11 11:11:45 -08:00
Andrew Kryczka d9620239d2 Cleanup stale manifests outside of full purge
Summary:
- Keep track of obsolete manifests in VersionSet
- Updated FindObsoleteFiles() to put obsolete manifests in the JobContext for later use by PurgeObsoleteFiles()
- Added test case that verifies a stale manifest is deleted by a non-full purge

Test Plan:
  $ ./backupable_db_test --gtest_filter=BackupableDBTest.ChangeManifestDuringBackupCreation

Reviewers: IslamAbdelRahman, yoshinorim, sdong

Reviewed By: sdong

Subscribers: andrewkr, leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D55269
2016-03-10 18:16:21 -08:00
Yi Wu f71fc77b7c Cache to have an option to fail Cache::Insert() when full
Summary:
Cache to have an option to fail Cache::Insert() when full. Update call sites to check status and handle error.

I totally have no idea what's correct behavior of all the call sites when they encounter error. Please let me know if you see something wrong or more unit test is needed.

Test Plan: make check -j32, see tests pass.

Reviewers: anthony, yhchiang, andrewkr, IslamAbdelRahman, kradhakrishnan, sdong

Reviewed By: sdong

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54705
2016-03-10 17:35:19 -08:00
Igor Canadi ee8cc35201 Merge pull request #938 from alexander-fenster/master
added --no_value option to ldb scan to dump key only
2016-03-10 16:44:52 -08:00
Yueh-Hsuan Chiang 765597fa78 Update compaction score right after CompactFiles forms a compaction
Summary:
This is a follow-up patch of https://reviews.facebook.net/D54891.
As the information about files being compacted will also be used
when making compaction decision, it is necessary to update the compaction
score when a compaction plan has been made but not yet execute.

This patch adds a missing call to update the compaction score in
CompactFiles().

Test Plan: compact_files_test

Reviewers: sdong, IslamAbdelRahman, kradhakrishnan, yiwu, andrewkr

Reviewed By: andrewkr

Subscribers: andrewkr, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D55227
2016-03-10 14:34:28 -08:00
Alexander Fenster f0161c37b0 formatting fix 2016-03-10 13:34:42 -08:00
Yueh-Hsuan Chiang aa3f02d50c Improve comment in compaction.h and compaction_picker.h
Summary:
ReleaseCompactionFiles must be called when DB mutex is held,
but the documentation is mission.

Test Plan: no code change

Reviewers: anthony, IslamAbdelRahman, kradhakrishnan, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54987
2016-03-08 16:46:41 -08:00
agiardullo 2200295ee1 optimistic transactions support for reinitialization
Summary: Extend optimization in D53835 to optimistic transactions for completeness.

Test Plan: added test

Reviewers: sdong, IslamAbdelRahman, horuff, jkedgar

Reviewed By: horuff

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D55059
2016-03-07 19:03:09 -08:00
Islam AbdelRahman badd6b7846 Ignore db_test2
Summary: Add db_test2 to .gitignore

Test Plan: make sure db_test2 dont show in "git status"

Reviewers: sdong, yhchiang, andrewkr, anthony

Reviewed By: anthony

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D55191
2016-03-07 15:56:16 -08:00
agiardullo 200080ed72 Improve snapshot handling for Transaction reinitialization
Summary: Previously, reusing a transaction (by passing it as an argument to BeginTransaction) would not clear the transaction's snapshot.  This is not a clear, well-definited behavior.

Test Plan: improved test

Reviewers: sdong, IslamAbdelRahman, horuff, jkedgar

Reviewed By: jkedgar

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D55053
2016-03-07 13:28:11 -08:00
Islam AbdelRahman 171c8e80b1 Update dependencies / Fix Clang
Summary:
Update dependencies and fix broken clang symlink

version changes
```
zstd 0.4.7 => 0.5.1
jemalloc 4.0.3 => master
```

Test Plan:
make check -j64
USE_CLANG=1 make check -j64
ROCKSDB_FBCODE_BUILD_WITH_481=1 make check -j64

Reviewers: sdong, yhchiang, anthony, andrewkr

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D55179
2016-03-07 12:57:19 -08:00
sdong 294bdf9ee2 Change Property name from "rocksdb.current_version_number" to "rocksdb.current-super-version-number"
Summary: I realized I again is wrong about the naming convention. Let me change it to the correct one.

Test Plan: Run unit tests.

Reviewers: IslamAbdelRahman, kradhakrishnan, yhchiang, andrewkr

Reviewed By: andrewkr

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D55041
2016-03-04 18:15:29 -08:00
Yueh-Hsuan Chiang bf1c4089db Use pure if-then check instead of assert in EraseColumnFamilyInfo
Summary:
Use pure if-then check instead of assert in EraseColumnFamilyInfo
when the specified column family does not found in the cf_info_map_.
So the second deletion will be no op instead of crash.

Test Plan: existing test.

Reviewers: sdong, anthony, kradhakrishnan, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D55023
2016-03-04 16:03:31 -08:00
Yueh-Hsuan Chiang a7d4eb2f34 Fix a bug where flush does not happen when a manual compaction is running
Summary:
Currently, when rocksdb tries to run manual compaction to refit data into a level,
there's a ReFitLevel() process that requires no bg work is currently running.
When RocksDB plans to ReFitLevel(), it will do the following:

 1. pause scheduling new bg work.
 2. wait until all bg work finished
 3. do the ReFitLevel()
 4. unpause scheduling new bg work.

However, as it pause scheduling new bg work at step one and waiting for all bg work
finished in step 2, RocksDB will stop flushing until all bg work is done (which
could take a long time.)

This patch fix this issue by changing the way ReFitLevel() pause the background work:

1. pause scheduling compaction.
2. wait until all bg work finished.
3. pause scheduling flush
4. do ReFitLevel()
5. unpause both flush and compaction.

The major difference is that.  We only pause scheduling compaction in step 1 and wait
for all bg work finished in step 2.  This prevent flush being blocked for a long time.
Although there's a very rare case that ReFitLevel() might be in starvation in step 2,
but it's less likely the case as flush typically finish very fast.

Test Plan: existing test.

Reviewers: anthony, IslamAbdelRahman, kradhakrishnan, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D55029
2016-03-04 14:24:52 -08:00
Gunnar Kudrjavets 68189f7e1b Update benchmarks used to measure subcompaction performance
Summary: After closely working with Mark, Siying, and Yueh-Hsuan this set of changes reflects the updates needed to measure RocksDB subcompaction performance in a correct manner. The essence of the benchmark is executing `fillrandom` followed by `compact` with the correct set of options for various number of subcompactions specified.

Test Plan: Tested internally to verify correctness and reliability.

Reviewers: sdong, yhchiang, MarkCallaghan

Reviewed By: MarkCallaghan

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D55089
2016-03-04 12:32:11 -08:00
Islam AbdelRahman dfe96c72c3 Fix WriteLevel0TableForRecovery file delete protection
Summary:
The call to

```
CaptureCurrentFileNumberInPendingOutputs()
```

should be before

```
versions_->NewFileNumber()
```
Right now we are not actually protecting the file from being deleted

Test Plan: make check

Reviewers: sdong, anthony, yhchiang

Reviewed By: yhchiang

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D54645
2016-03-03 18:25:07 -08:00
Igor Canadi 451678c8c9 Merge pull request #1025 from SherlockNoMad/BuildFix
Fix Windows build
2016-03-03 15:24:25 -08:00
SherlockNoMad 58ecd91326 Fix Windows build 2016-03-03 15:08:24 -08:00
Andrew Kryczka 501927ffc4 [backupable db] Remove file size embedded in name workaround
Summary:
Now that we get sizes efficiently, we no longer need the workaround to
embed file size in filename.

Test Plan:
  $ ./backupable_db_test

Reviewers: sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D55035
2016-03-03 13:32:20 -08:00
sdong ef204df7ef Compaction always needs to be removed from level0_compactions_in_progress_ for universal compaction
Summary: We always put compaction to level0_compactions_in_progress_ for universal compaction, so we should also remove it. The bug causes assert failure when running manual compaction.

Test Plan:
TEST_TMPDIR=/dev/shm/ ./db_bench --benchmarks=fillrandom,compact --subcompactions=16 --compaction_style=1
always fails on my host. After the fix, it doesn't fail any more.

Reviewers: IslamAbdelRahman, andrewkr, kradhakrishnan, yhchiang

Reviewed By: yhchiang

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D55017
2016-03-02 21:23:28 -08:00
sdong e79ad9e184 Add Iterator Property rocksdb.iterator.version_number
Summary: We want to provide a way to detect whether an iterator is stale and needs to be recreated. Add a iterator property to return version number.

Test Plan: Add two unit tests for it.

Reviewers: IslamAbdelRahman, yhchiang, anthony, kradhakrishnan, andrewkr

Reviewed By: andrewkr

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54921
2016-03-02 16:23:59 -08:00
sdong 19ea40f8b6 Subcompaction boundary keys should not terminate after an empty level
Summary: Now we skip to add boundary keys to subcompaction candidates since we see an empty level. This makes subcompaction almost disabled for universal compaction. We should consider all files instead.

Test Plan: Run existing tests.

Reviewers: IslamAbdelRahman, andrewkr, yhchiang

Reviewed By: yhchiang

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D55005
2016-03-02 15:45:07 -08:00
Omar Sandoval deb08b8226 Add parsing of missing DB options
Summary: There are a few options in struct DBOptions that aren't handled by options_helper.cc. Add those missing options so they can be used by GetDBOptionsFromString() and friends.

Test Plan: Updated options_test.cc, reran all tests.

Reviewers: sdong, yhchiang

Reviewed By: yhchiang

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54603
2016-03-02 10:34:14 -08:00
Andrew Kryczka f8e90e8753 Get file attributes in bulk for VerifyBackup and CreateNewBackup
Summary:
For VerifyBackup(), backup files can be spread across "shared/",
"shared_checksum/", and "private/" subdirectories, so we have to
bulk get all three.

For CreateNewBackup(), we make two separate bulk calls: one for the
data files and one for WAL files.

There is also a new helper function, ExtendPathnameToSizeBytes(),
that translates the file attributes vector to a map. I decided to leave
GetChildrenFileAttributes()'s (from D53781) return type as vector to
keep it consistent with GetChildren().

Depends on D53781.

Test Plan:
verified relevant unit tests

  $ ./backupable_db_test

Reviewers: IslamAbdelRahman, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53919
2016-03-01 19:33:33 -08:00
sdong 12fd9b1868 Change BlockBasedTableOptions.format_version default to 2
Summary: BlockBasedTableOptions.format_version = 2 uses better encoding format. Now it's the time to make it default.

Test Plan: Run all existing tests.

Reviewers: igor, yhchiang, anthony, kradhakrishnan, andrewkr, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: MarkCallaghan, leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54879
2016-03-01 14:28:52 -08:00
sdong 4572a2d8c0 Update current version to 4.6
Summary: 4.5 is already cut, we can now increase the version in 4.6.

Test Plan: Not needed.

Reviewers: anthony, kradhakrishnan, andrewkr, yhchiang, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54963
2016-03-01 14:26:44 -08:00
sdong 74b660702e Rename iterator property "rocksdb.iterator.is.key.pinned" => "rocksdb.iterator.is-key-pinned"
Summary: Rename iterator property to folow property naming convention.

Test Plan: Run all existing tests.

Reviewers: andrewkr, anthony, yhchiang, kradhakrishnan, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54957
2016-03-01 13:47:12 -08:00
Islam AbdelRahman 6743135ea1 Fix DB::AddFile() issue when PurgeObsoleteFiles() is called
Summary:
In some situations the DB will scan all existing files in the DB path and delete the ones that are Obsolete.
If this happen during adding an external sst file. this could cause the file to be deleted while we are adding it.
This diff fix this issue

Test Plan:
unit test to reproduce the bug
existing unit tests

Reviewers: sdong, yhchiang, andrewkr

Reviewed By: andrewkr

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D54627
2016-03-01 12:05:29 -08:00
sdong 432f3adf2c Add DB Property "rocksdb.current_version_number"
Summary: Add a DB Property "rocksdb.current_version_number" for users to monitor version changes and stale iterators.

Test Plan: Add a unit test.

Reviewers: andrewkr, yhchiang, kradhakrishnan, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54927
2016-03-01 10:55:40 -08:00
Adam Retter 188bb2e7ad Fix formatting identified by arc lint 2016-03-01 16:00:45 +00:00
Adam Retter 0f2d2fcff6 Refactored tests to use try-with-resources 2016-03-01 16:00:45 +00:00
Adam Retter f8e02c7825 Deprecate org.rocksdb.AbstractNativeReference#dispose() and implement java.lang.AutoCloseable 2016-03-01 16:00:45 +00:00
Adam Retter 0f2fdfe23a Fix the javadoc and the formatting of the base-classes for objects with native references 2016-03-01 16:00:44 +00:00
sdong b5b1db167a Recompute compaction score after scheduling manual compaction
Summary: After we made manual compaction runnable concurrently with automaticallly compaction, we need to run ComputeCompactionScore() to prepare a coming compaction picking call before the compaction finishes.

Test Plan: Run existing tests.

Reviewers: yhchiang, IslamAbdelRahman, andrewkr, kradhakrishnan, anthony, igor

Reviewed By: igor

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54891
2016-02-29 17:17:51 -08:00
agiardullo 5ea9aa3c14 TransactionDB:ReinitializeTransaction
Summary: Add function to reinitialize a transaction object so that it can be reused.  This is an optimization so users can potentially avoid reallocating transaction objects.

Test Plan: added tests

Reviewers: yhchiang, kradhakrishnan, IslamAbdelRahman, sdong

Reviewed By: sdong

Subscribers: jkedgar, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53835
2016-02-29 16:27:32 -08:00
sdong 1f5954147b Introduce Iterator::GetProperty() and replace Iterator::IsKeyPinned()
Summary:
Add Iterator::GetProperty(), a way for users to communicate with iterator, and turn Iterator::IsKeyPinned() with it.
As a follow-up, I'll ask a property as the version number attached to the iterator

Test Plan: Rerun existing tests and add a negative test case.

Reviewers: yhchiang, andrewkr, kradhakrishnan, anthony, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54783
2016-02-29 14:01:31 -08:00
Igor Canadi 67789419fa Merge pull request #1020 from gongsu832/master
Modified Makefile and build_tools/build_detect_platform to compile on…
2016-02-29 13:12:57 -08:00
Andrew Kryczka 69c471bd9b Handle concurrent manifest update and backup creation
Summary:
Fixed two related race conditions in backup creation.

(1) CreateNewBackup() uses DB::DisableFileDeletions() to prevent table files
from being deleted while it is copying; however, the MANIFEST file could still
rotate during this time. The fix is to stop deleting the old manifest in the
rotation logic. It will be deleted safely later when PurgeObsoleteFiles() runs
(can only happen when file deletions are enabled).

(2) CreateNewBackup() did not account for the CURRENT file being mutable.
This is significant because the files returned by GetLiveFiles() contain a
particular manifest filename, but the manifest to which CURRENT refers can
change at any time. This causes problems when CURRENT changes between the call
to GetLiveFiles() and when it's copied to the backup directory. To workaround this, I
manually forge a CURRENT file referring to the manifest filename returned in
GetLiveFiles().

(2) also applies to the checkpointing code, so let me know if this approach is
good and I'll make the same change there.

Test Plan:
new test for roll manifest during backup creation.

running the test before this change:

  $ ./backupable_db_test --gtest_filter=BackupableDBTest.ChangeManifestDuringBackupCreation
  ...
  IO error: /tmp/rocksdbtest-9383/backupable_db/MANIFEST-000001: No such file or directory

running the test after this change:

  $ ./backupable_db_test --gtest_filter=BackupableDBTest.ChangeManifestDuringBackupCreation
  ...
  [ RUN      ] BackupableDBTest.ChangeManifestDuringBackupCreation
  [       OK ] BackupableDBTest.ChangeManifestDuringBackupCreation (2836 ms)

Reviewers: IslamAbdelRahman, anthony, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54711
2016-02-29 12:56:55 -08:00
root 3373c81fa8 Modify build_tools/build_detect_platform to detect and set -march=z10 on Linux s390x. 2016-02-29 15:02:52 -05:00
root 990509045f Merge branch 'master' of https://github.com/gongsu832/rocksdb
Rebase for -momit-leaf-frame-pointer fix.
2016-02-29 14:45:55 -05:00
Igor Canadi 3492889ab8 Merge pull request #1019 from javacruft/wip-omit-leaf-frame-pointer-archs
build: Improve -momit-leaf-frame-pointer usage
2016-02-29 09:32:16 -08:00
James Page 7ca731b12c build: Improve -momit-leaf-frame-pointer usage
-momit-leaf-frame-pointer is only supported on certain archs.
Detect this automatically based on which flags the compiler
understands.
2016-02-29 09:22:24 +00:00
root 21f17aaa60 Modified Makefile and build_tools/build_detect_platform to compile on Linux s390x. 2016-02-26 18:03:07 -05:00
sdong 8800975fb0 Make DBTestUniversalCompaction.IncreaseUniversalCompactionNumLevels more robust
Summary:
Based on thread scheduling, DBTestUniversalCompaction.IncreaseUniversalCompactionNumLevels can fail to flush enough files to trigger expected compactions. Fix it by waiting for flush after inserting each key.
There are failrue reported:

db/db_universal_compaction_test.cc:1134: Failure
Expected: (NumTableFilesAtLevel(options.num_levels - 1, 1)) > (0), actual: 0 vs 0

but I can't repro it. Try to fix the bug and see whether it goes away.

Test Plan: Run the test multiple time.

Reviewers: IslamAbdelRahman, anthony, andrewkr, kradhakrishnan, yhchiang

Reviewed By: yhchiang

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54747
2016-02-26 11:59:31 -08:00
agiardullo cd3fe675a9 Remove stale TODO
Summary: This was fixed by 0c2bd5cb

Test Plan: n/a

Reviewers: gabijs

Reviewed By: gabijs

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54753
2016-02-25 17:44:35 -08:00
Andrew Kryczka 69c98f0431 Reorder instance variables in backup test for proper destruction order
Summary:
As titled. This fixes the tsan error caused by logger_ being used in
backup_engine_'s destructor. It does not fix the transient unit test failure,
which is caused by MANIFEST file changing while backup is happening.

Test Plan:
verified the tsan error no longer happens on either success or
failure.

  $ COMPILE_WITH_TSAN=1 make -j32 backupable_db_test
  $ while ./backupable_db_test --gtest_filter=BackupableDBTest.CorruptionsTest ; do : ; done

Reviewers: sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54669
2016-02-24 10:32:11 -08:00
sdong 82f15fb15d Add test to make sure DropColumnFamily doesn't impact existing iterators
Summary: Add a test case in ColumnFamilyTest.ReadDroppedColumnFamily to make sure existing iterator is not impacted by column family dropping.

Test Plan: N/A

Reviewers: igor, yhchiang, anthony, andrewkr, kradhakrishnan, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54657
2016-02-24 10:25:38 -08:00
sdong 38201b3599 Fix assert failure when DBImpl::SyncWAL() conflicts with log rolling
Summary: DBImpl::SyncWAL() releases db mutex before calling DBImpl::MarkLogsSynced(), while inside DBImpl::MarkLogsSynced() we assert there is none or one outstanding log file. However, a memtable switch can happen in between and causing two or outstanding logs there, failing the assert. The diff adds a unit test that repros the issue and fix the assert so that the unit test passes.

Test Plan: Run the new tests.

Reviewers: anthony, kolmike, yhchiang, IslamAbdelRahman, kradhakrishnan, andrewkr

Reviewed By: andrewkr

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54621
2016-02-23 11:42:15 -08:00
Yi Wu 2568985ab3 IOStatsContext::ToString() add option to exclude zero counters
Summary: similar to D52809 add option to exclude zero counters.

Test Plan:
[yiwu@dev4504.prn1 ~/rocksdb] ./iostats_context_test
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from IOStatsContextTest
[ RUN      ] IOStatsContextTest.ToString
[       OK ] IOStatsContextTest.ToString (0 ms)
[----------] 1 test from IOStatsContextTest (0 ms total)

[----------] Global test environment tear-down
[==========] 1 test from 1 test case ran. (0 ms total)
[  PASSED  ] 1 test.

Reviewers: anthony, yhchiang, andrewkr, IslamAbdelRahman, kradhakrishnan, sdong

Reviewed By: sdong

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54591
2016-02-23 10:26:24 -08:00
Andrew Kryczka b046916656 Redo SyncPoints for flush while rolling test
Summary:
There was a race condition in the test where the rolling thread
acquired the mutex before the flush thread pinned the logger. Rather than add
more complicated synchronization to fix it, I followed Siying's suggestion to
use SyncPoint in the test code.

Comments in the LoadDependency() invocation explain the reason for each of the
sync points.

Test Plan:
Ran test 1000 times for tsan/asan. Will wait for all sandcastle tests
to finish before committing since this is a tricky test.

Reviewers: IslamAbdelRahman, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54615
2016-02-22 21:32:19 -08:00
Yueh-Hsuan Chiang 291ae4c206 Revert "Revert "Fixed the bug when both whole_key_filtering and prefix_extractor are set.""
Summary:
This reverts commit 73c31377bb, which mistakenly
reverts 73c31377bb that fixes a bug when both
whole_key_filtering and prefix_extractor are set

Test Plan: revert the patch

Reviewers: anthony, IslamAbdelRahman, rven, kradhakrishnan, sdong

Reviewed By: sdong

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D52707
2016-02-22 16:33:26 -08:00
Mike Kolupaev eef63ef807 Fixed CompactFiles() spuriously failing or corrupting DB
Summary:
We started getting two kinds of crashes since we started using `DB::CompactFiles()`:
(1) `CompactFiles()` fails saying something like "/data/logdevice/4440/shard12/012302.sst: No such file or directory", and presumably makes DB read-only,
(2) DB fails to open saying "Corruption: Can't access /267000.sst: IO error: /data/logdevice/4440/shard1/267000.sst: No such file or directory".

AFAICT, both can be explained by background thread deleting compaction output as "obsolete" while it's being written, before it's committed to manifest. If it ends up committed to the manifest, we get (2); if compaction notices the disappearance and fails, we get (1). The internal tasks t10068021 and t10134177 have some details about the investigation that led to this.

Test Plan: `make -j check`; the new test fails to reopen the DB without the fix

Reviewers: yhchiang

Reviewed By: yhchiang

Subscribers: dhruba, sdong

Differential Revision: https://reviews.facebook.net/D54561
2016-02-22 13:54:58 -08:00
Yueh-Hsuan Chiang 79ca039eb4 Relax the check condition of prefix_extractor in CheckOptionsCompatibility
Summary:
Relax the check condition of prefix_extractor in CheckOptionsCompatibility
by allowing changing value from non-nullptr to nullptr or nullptr to
non-nullptr.

Test Plan:
options_test
options_util_test

Reviewers: sdong, anthony, IslamAbdelRahman, kradhakrishnan, gunnarku

Reviewed By: gunnarku

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54477
2016-02-19 14:42:24 -08:00
Igor Canadi 4b1b4b8aec Merge pull request #1004 from yuslepukhin/child_attr
Implement ConsistentChildrenAttribute on Windows
2016-02-19 14:27:18 -08:00
Dmitri Smirnov 9ea2968d26 Implement ConsistentChildrenAttribute
by using default implementation for now as it works.
2016-02-19 14:20:34 -08:00
sdong c7f1a8a468 Fix LITE build thread_local_test
Summary: Recent change break thread_local_test by introducing exception, which is disabled in LITE build. Fix it by disabling exception handling in LITE build.

Test Plan: Build with both of LITE and non-LITE

Reviewers: anthony, IslamAbdelRahman, yhchiang, kradhakrishnan, andrewkr

Reviewed By: andrewkr

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54513
2016-02-19 13:57:18 -08:00
Igor Canadi 0914f0ca55 Merge pull request #1003 from yuslepukhin/fix_mutexlock_pthread_build
This addresses build issues on Windows
2016-02-19 13:12:36 -08:00
Dmitri Smirnov d37d348da8 This addresses build issues on Windows
https://github.com/facebook/rocksdb/issues/1002
2016-02-19 12:29:54 -08:00
Andrew Kryczka d825fc70d4 Use condition variable in log roller test
Summary:
Previously I just slept until the flush_thread was "probably" ready
since proper synchronization in test cases seemed like overkill. But then tsan
complained about it, so I did the synchronization (mostly) properly now.

Test Plan:
  $ COMPILE_WITH_TSAN=1 make -j32 auto_roll_logger_test
  $ ./auto_roll_logger_test

Reviewers: anthony, IslamAbdelRahman, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54399
2016-02-18 18:03:53 -08:00
Islam AbdelRahman 6b2a047dfc Fix SstFileManager uninitialized data member
Summary: I have introduced max_allowed_space_ but did not initialize it

Test Plan: make check

Reviewers: sdong, yhchiang, anthony

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D54357
2016-02-18 11:25:19 -08:00
Yueh-Hsuan Chiang a3db93c261 Remove the SyncPoint usage in the destructor of PosixEnv
Summary:
Remove the SyncPoint usage in the destructor of PosixEnv as none
of any active tests is using it.

SyncPoint is a test-only utility class, and it's a static varible.
As a result, using SyncPoint in the destructor of PosixEnv will
make default Env depends on SyncPoint.  Removing such dependency
could solve the problem crash issue only reproducable in Mac
environment.

Test Plan: OPT=-DTRAVIS V=1 make -j4 check on Mac environment

Reviewers: sdong, anthony

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54333
2016-02-17 23:32:14 -08:00
Islam AbdelRahman df9ba6df62 Introduce SstFileManager::SetMaxAllowedSpaceUsage() to cap disk space usage
Summary:
Introude SstFileManager::SetMaxAllowedSpaceUsage() that can be used to limit the maximum space usage allowed for RocksDB.
When this limit is exceeded WriteImpl() will fail and return Status::Aborted()

Test Plan: unit testing

Reviewers: yhchiang, anthony, andrewkr, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53763
2016-02-17 15:20:23 -08:00
Andrew Kryczka 3943d16780 Fix race conditions in auto-rolling logger
Summary:
For GetLogFileSize() and Flush(), they previously did not follow the
synchronization pattern for accessing logger_. This meant ResetLogger() could
cause logger_ destruction while the unsynchronized functions were accessing it,
causing a segfault.

Also made the mutex instance variable mutable so we can preserve
GetLogFileSize()'s const-ness.

Test Plan:
new test case, it's quite ugly because both threads need to access
one of the functions with SyncPoints (PosixLogger::Flush()), and also special
handling is needed to prevent the mutex and sync points from conflicting.

Reviewers: kradhakrishnan, IslamAbdelRahman, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54237
2016-02-17 12:06:45 -08:00
Andrew Kryczka d733dd5728 [build] Fix env_win.cc compiler errors
Summary: I broke it in D53781.

Test Plan: tried the same code in util/env_posix.cc and it compiled successfully

Reviewers: sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54303
2016-02-17 11:57:04 -08:00
Jonathan Wiepert cf38e56f2f Fix broken appveyor build caused by D53991
Test Plan:
pushed remote as D53991_appveyor_test branch. Appveyor run:
https://ci.appveyor.com/project/Facebook/rocksdb/build/1.0.651
shows the file is now found and the run is back to the state before the initial change:
https://ci.appveyor.com/project/Facebook/rocksdb/build/1.0.620

Reviewers: sdong, igor

Reviewed By: igor

Subscribers: igor, dhruba

Differential Revision: https://reviews.facebook.net/D54309
2016-02-17 11:49:52 -08:00
Igor Canadi 351252b497 Merge pull request #998 from fengjian0106/master
fix ios build error
2016-02-17 09:06:07 -08:00
fengjian 1336052497 fix ios build error 2016-02-17 20:22:40 +08:00
agiardullo d08d50295c Fix transaction locking
Summary: Broke transaction locking in 4.4 in D52197.  Will cherry-pick this change into 4.4 (which hasn't yet been fully released).  Repro'd using db_bench.

Test Plan: unit tests and db_Bench

Reviewers: sdong, yhchiang, kradhakrishnan, ngbronson

Reviewed By: ngbronson

Subscribers: ngbronson, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54021
2016-02-16 17:15:05 -08:00
Yueh-Hsuan Chiang 730a422c3a Improve the documentation of LoadLatestOptions
Summary: Improve the documentation of LoadLatestOptions

Test Plan: No code change

Reviewers: anthony, IslamAbdelRahman, kradhakrishnan, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54087
2016-02-16 14:55:24 -08:00
reid horuff a7b6f0748a Improve write_with_callback_test to sync WAL
Summary: Currently write_with_callback_test does not test with WAL syncing enabled. This addresses that.

Test Plan: write_with_callback_test

Reviewers: anthony

Reviewed By: anthony

Subscribers: leveldb, dhruba, hermanlee4

Differential Revision: https://reviews.facebook.net/D54255
2016-02-16 14:04:14 -08:00
reid horuff 5bcf952a87 Fix WriteImpl empty batch hanging issue
Summary: There is an issue in DBImpl::WriteImpl where if an empty writebatch comes in and sync=true then the logs will be marked as being synced yet the sync never actually happens because there is no data in the writebatch. This causes the next incoming batch to hang while waiting for the logs to complete syncing. This fix syncs logs even if the writebatch is empty.

Test Plan: DoubleEmptyBatch unit test in transaction_test.

Reviewers: yoshinorim, hermanlee4, sdong, ngbronson, anthony

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54057
2016-02-16 12:21:33 -08:00
Jonathan Wiepert 871cc5f987 fix build without gflags
Test Plan:
Built and ran with gflags:
% ./db_bench
LevelDB:    version 4.5
Date:       Tue Feb 16 12:04:23 2016
CPU:        40 * Intel(R) Xeon(R) CPU E5-2660 v2 @ 2.20GHz
...

And without gflags:
% ./db_bench
Please install gflags to run rocksdb tools
%

Reviewers: sdong, igor

Reviewed By: igor

Subscribers: igor, dhruba

Differential Revision: https://reviews.facebook.net/D54243
2016-02-16 12:16:47 -08:00
Igor Canadi c90d63a23d can_unlock set but not used
Test Plan: I couldn't repro, but I hope this fixes it. See the error here: https://evergreen.mongodb.com/task_log_raw/mongodb_mongo_master_ubuntu1404_rocksdb_compile_6e9fd902d5cb25aef992363efa128640affd5196_16_02_11_04_33_37/0?type=T

Reviewers: yhchiang, andrewkr, sdong, anthony

Reviewed By: anthony

Subscribers: meyering, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54123
2016-02-16 11:24:40 -08:00
Mike Kolupaev 44371501f0 Fixed a segfault when compaction fails
Summary: We've hit it today.

Test Plan: `make -j check`; didn't reproduce the issue

Reviewers: yhchiang

Reviewed By: yhchiang

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D54219
2016-02-16 11:11:16 -08:00
Igor Canadi 2f084d39b9 Merge pull request #992 from jofusa/jdennison/options-typo-fix
fixes typo in options logging
2016-02-16 10:26:25 -08:00
Jonathan Wiepert 7bd284c374 Separeate main from bench functionality to allow cusomizations
Summary: Isolate db_bench functionality from main so custom benchmark code can be written and managed

Test Plan:
Tested commands
./build_tools/regression_build_test.sh
./db_bench --db=/tmp/rocksdbtest-12321/dbbench --stats_interval_seconds=1 --num=1000
./db_bench --db=/tmp/rocksdbtest-12321/dbbench --stats_interval_seconds=1 --num=1000 --reads=500 --writes=500
./db_bench --db=/tmp/rocksdbtest-12321/dbbench --stats_interval_seconds=1 --num=1000 --merge_keys=100 --numdistinct=100 --num_column_families=3 --num_hot_column_families=1
./db_bench --stats_interval_seconds=1 --num=1000 --bloom_locality=1 --seed=5 --threads=5
./db_bench --duration=60 --value_size=50 --seek_nexts=10 --reverse_iterator=true --usee_uint64_comparator=true --batch-size=5
./db_bench --duration=60 --value_size=50 --seek_nexts=10 --reverse_iterator=true --use_uint64_comparator=true --batch_size=5
./db_bench --duration=60 --value_size=50 --seek_nexts=10 --reverse_iterator=true --usee_uint64_comparator=true --batch-size=5

Test Results - https://phabricator.fb.com/P56130387

Additional tests for:
./db_bench --duration=60 --value_size=50 --seek_nexts=10 --reverse_iterator=true --use_uint64_comparator=true --batch_size=5 --key_size=8 --merge_operator=put
./db_bench --stats_interval_seconds=1 --num=1000 --bloom_locality=1 --seed=5 --threads=5 --merge_operator=uint64add

Results: https://phabricator.fb.com/P56130607

Reviewers: yhchiang, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53991
2016-02-16 06:17:31 -08:00
John Dennison bd5f842bba fixes typo in options logging 2016-02-15 23:55:29 +00:00
Andrew Kryczka 1c868d6848 Fix includes for env_test
Summary:
Previously compilation failed when ROCKSDB_NO_FBCODE=1 because fcntl.h
wasn't included for open().

Related issue: https://github.com/facebook/rocksdb/issues/977

Test Plan:
verified below command works now:

  $ make clean && ROCKSDB_NO_FBCODE=1 ROCKSDB_DISABLE_FALLOCATE=1 make -j32 env_test

Reviewers: sdong, yhchiang, igor

Reviewed By: igor

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54135
2016-02-12 14:17:14 -08:00
krad 545a193952 Add J to commit_prereq so comilation/execution happens in parallel
Summary:
Users are confused on how to get the parallel compilation going. This
can help wire the parallelism.

Test Plan: Run manually

Reviewers: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53931
2016-02-12 12:09:07 -08:00
Andrew Kryczka 5bb7371ca6 [build] Evaluate test names only when db_test exists
Summary:
as titled, this will prevent the error that was printed because
test_names was evaluated before db_test was built.

Test Plan:
verified below command works and no longer prints errors:

  $ make release -j32

verified below command still finds the right tests:

  $ make J=32 parallel_check

Reviewers: igor

Reviewed By: igor

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54117
2016-02-12 10:49:35 -08:00
Andrew Kryczka 6a2b4fcb80 Add flag to forcibly disable fallocate
Summary:
see https://github.com/facebook/rocksdb/issues/977; there are issues
with fallocate() on certain filesystems/kernel versions that can lead it to pre-
allocating blocks but never freeing them, even if they're unused.

Test Plan:
verified build commands omit DROCKSDB_FALLOCATE_PRESENT when this env
variable is set.

without disabling it:

  $ ROCKSDB_NO_FBCODE=1 make -n env_test | grep -q DROCKSDB_FALLOCATE_PRESENT ; echo $?
  0

with disabling it:

  $ ROCKSDB_NO_FBCODE=1 DISABLE_FALLOCATE=1 make -n env_test | grep -q DROCKSDB_FALLOCATE_PRESENT ; echo $?
  1

Reviewers: kradhakrishnan, yhchiang, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54069
2016-02-11 17:00:01 -08:00
sdong 92a9ccf1a6 Add a new compaction priority that picks file whose overlapping ratio is smallest
Summary:
Add a new compaction priority as following:
For every file, we calculate total size of files overalapping with the file in the next level, over the file's size itself. The file with smallest ratio will be picked first.
My "db_bench --fillrandom" shows about 5% less compaction than kOldestSmallestSeqFirst if --hard_pending_compaction_bytes_limit value to keep LSM tree in shape. If not limiting hard_pending_compaction_bytes_limit, improvement is only 1% or 2%.

Test Plan: Add a unit test

Reviewers: andrewkr, kradhakrishnan, anthony, IslamAbdelRahman, yhchiang

Reviewed By: yhchiang

Subscribers: MarkCallaghan, leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D54075
2016-02-11 15:59:19 -08:00
Igor Canadi 3dc3d1c144 Merge pull request #984 from petermattis/pmattis/comparator-iterate-upper-bound
Use user_comparator when comparing against iterate_upper_bound.
2016-02-11 10:22:40 -08:00
Peter Mattis 239aaf2fc0 Use user_comparator when comparing against iterate_upper_bound.
Fixes #983.
2016-02-11 08:47:16 -05:00
Yueh-Hsuan Chiang 908100399c Fixed a dependency issue of ThreadLocalPtr
Summary:
When a child thread that uses ThreadLocalPtr, ThreadLocalPtr::OnThreadExit
will be called when that child thread is destroyed.  However,
OnThreadExit will try to access a static singleton of ThreadLocalPtr,
which will be destroyed when the main thread exit.  As a result,
when a child thread that uses ThreadLocalPtr exits AFTER the main thread
exits, illegal memory access will occur.

This diff includes a test that reproduce this legacy bug.

    ==2095206==ERROR: AddressSanitizer: heap-use-after-free on address
    0x608000007fa0 at pc 0x959b79 bp 0x7f5fa7426b60 sp 0x7f5fa7426b58
    READ of size 8 at 0x608000007fa0 thread T1

This patch fix this issue by having the thread local mutex never be deleted
(but will leak small piece of memory at the end.)   The patch also describe
a better solution (thread_local) in the comment that requires gcc 4.8.1 and
in latest clang as a future work once we agree to move toward gcc 4.8.

Test Plan:
COMPILE_WITH_ASAN=1 make thread_local_test -j32
./thread_local_test --gtest_filter="*MainThreadDiesFirst"

Reviewers: anthony, hermanlee4, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53013
2016-02-10 16:56:01 -08:00
Gunnar Kudrjavets 337671b688 Add universal compaction benchmarks to run_flash_bench.sh
Summary:
Implement a benchmark for universal compaction based on the feature description (see below), in-person discussions, and reading source code:

https://github.com/facebook/rocksdb/wiki/RocksDB-Tuning-Guide
https://github.com/facebook/rocksdb/wiki/Universal-Compaction
https://github.com/facebook/rocksdb/wiki/RocksDB-Tuning-Guide#universal-compaction

Universal compaction benchmark is based on `overwrite` benchmark, adding compaction specific options to it, and executing it for different values of subcompaction to understand the impact of scaling out subcompactions for a particular scenario.

Test Plan:
  - Execute the benchmark on various machines for multiple iterations to verify the reliability.
  - Observe the output to make sure that compaction is taking place.
  - Observe the execution to make sure that arguments passed to `db_bench` are correct.

Reviewers: sdong, MarkCallaghan

Reviewed By: MarkCallaghan

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54045
2016-02-10 15:30:47 -08:00
Yueh-Hsuan Chiang 3a67bffaa8 Fix an ASAN error in transaction_test.cc
Summary:
One test in transaction_test.cc forgets to call SyncPoint::DisableProcessing().
As a result, a program might to access the SyncPoint singleton after it
already goes out of scope.

This patch fix this error by calling SyncPoint::DisableProcessing().

Test Plan: transaction_test

Reviewers: sdong, IslamAbdelRahman, kradhakrishnan, anthony

Reviewed By: anthony

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54033
2016-02-10 12:06:59 -08:00
Adam Retter c5af85ecad Fix a memory leak of Slice objects from org.rocksdb.WBWIRocksIterator#entry1 2016-02-10 14:44:39 +00:00
Adam Retter e84137c8ae Remove unnessecary java.util.List expense in JNI 2016-02-10 14:44:39 +00:00
Adam Retter 76e8beeeb9 Pass by pointer from/to Java from JNI not by object 2016-02-10 14:44:39 +00:00
Adam Retter 0e7e6f6e4b Improve Javadoc 2016-02-10 14:44:39 +00:00
Adam Retter 18eb563058 Improve the speed and synchronization around the construction of Java/JNI objects 2016-02-10 14:44:39 +00:00
Yueh-Hsuan Chiang 2a04268be3 Temporarily disable unstable tests in memory_test.cc
Summary:
memory_test.cc has some tests that are not unstable but
hard to reproduce, and the cause is the test itself not
the code.  Temporarily disable the tests until
we have a good fix.

Test Plan: memory_test

Reviewers: sdong, anthony, IslamAbdelRahman, rven, kradhakrishnan

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D54009
2016-02-09 16:28:34 -08:00
Igor Canadi 08a78b6b44 Merge pull request #979 from facebook/update_licenses
Updated all copyright headers to the new format.
2016-02-09 15:46:18 -08:00
Baraa Hamodi 21e95811d1 Updated all copyright headers to the new format. 2016-02-09 15:12:00 -08:00
Andrew Kryczka 59b3ee658f Env function for bulk metadata retrieval
Summary:
Added this new function, which returns filename, size, and modified
timestamp for each file in the provided directory. The default implementation
retrieves the metadata sequentially using existing functions. In the next diff
I'll make HdfsEnv override this function to use libhdfs's bulk get function.

This won't work on windows due to the path separator.

Test Plan:
new unit test

  $ ./env_test --gtest_filter=EnvPosixTest.ConsistentChildrenMetadata

Reviewers: yhchiang, sdong

Reviewed By: sdong

Subscribers: IslamAbdelRahman, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53781
2016-02-09 14:54:32 -08:00
Yueh-Hsuan Chiang 4a8cbf4e31 Allows Get and MultiGet to read directly from SST files.
Summary:
Add kSstFileTier to ReadTier, which allows Get and MultiGet to
read only directly from SST files and skip mem-tables.

    kSstFileTier = 0x2      // data in SST files.
                          // Note that this ReadTier currently only supports
                          // Get and MultiGet and does not support iterators.

Test Plan: add new test in db_test.

Reviewers: anthony, IslamAbdelRahman, rven, kradhakrishnan, sdong

Reviewed By: sdong

Subscribers: igor, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53511
2016-02-09 11:20:22 -08:00
agiardullo fe93bf9b5d Transaction::UndoGetForUpdate
Summary: MyRocks wants to be able to un-lock a key that was just locked by GetForUpdate().  To do this safely, I am now keeping track of the number of reads(for update) and writes for each key in a transaction.  UndoGetForUpdate() will only unlock a key if it hasn't been written and the read count reaches 0.

Test Plan: more unit tests

Reviewers: igor, rven, yhchiang, spetrunia, sdong

Reviewed By: spetrunia, sdong

Subscribers: spetrunia, dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D47043
2016-02-09 10:46:11 -08:00
sdong 2608219cc9 crash_test: cover concurrent memtable insert in default crash test
Summary: Default crash test uses prefix hash memtable, which is not compatible to concurrent memtable. Allow prefix test run with skip list and use skip list memtable when concurrent insert is used.

Test Plan: Run "python -u tools/db_crashtest.py whitebox" and watch sometimes skip list is used.

Reviewers: anthony, yhchiang, kradhakrishnan, andrewkr, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53907
2016-02-09 08:02:38 -08:00
sdong a76e9093f0 Fix LITE db_test build broken by previous commit
Summary: Previous commit introduces a test that is not supported in LITE. Fix it.

Test Plan: Build the test with ROCKSDB_LITE.

Reviewers: kradhakrishnan, IslamAbdelRahman, anthony, yhchiang, andrewkr

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53901
2016-02-05 14:29:09 -08:00
sdong b1887c5dd9 Explictly fail when memtable doesn't support concurrent insert
Summary: If users turn on concurrent insert but the memtable doesn't support it, they might see unexcepted crash. Fix it by explicitly fail.

Test Plan:
Run different setting of stress_test and make sure it fails correctly.
Will add a unit test too.

Reviewers: anthony, kradhakrishnan, IslamAbdelRahman, yhchiang, andrewkr, ngbronson

Reviewed By: ngbronson

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53895
2016-02-05 14:15:50 -08:00
Gunnar Kudrjavets 8ed3438778 Add option to run fillseq with WAL enabled in addition to WAL disabled
Summary: This set of changes is part of the work to introduce benchmark for universal style compaction in RocksDB. It's conceptually separate from the compaction work, so sending it out as a separate diff to get it out of the way.

Test Plan:
  - Run `./tools/run_flash_bench.sh`.
  - Look at the contents of `report.txt` and `report2.txt` to make sure that data is reported and attributed correctly.
  - During `db_bench` execution time make sure that the correct flags are passed to `--disable_wal` depending on the benchmark being executed.

Reviewers: MarkCallaghan

Reviewed By: MarkCallaghan

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53865
2016-02-05 13:20:56 -08:00
sdong 73a9b0f4ba Update version to 4.5
Summary: Time to cut branch for release 4.5. Change the versions.

Test Plan: Not needed

Reviewers: IslamAbdelRahman, yhchiang, kradhakrishnan, andrewkr, anthony

Reviewed By: anthony

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53883
2016-02-05 13:05:54 -08:00
reid horuff 6f71d3b68b Improve perf of Pessimistic Transaction expirations (and optimistic transactions)
Summary:
copy from task 8196669:

1) Optimistic transactions do not support batching writes from different threads.
2) Pessimistic transactions do not support batching writes if an expiration time is set.

In these 2 cases, we currently do not do any write batching in DBImpl::WriteImpl() because there is a WriteCallback that could decide at the last minute to abort the write.  But we could support batching write operations with callbacks if we make sure to process the callbacks correctly.

To do this, we would first need to modify write_thread.cc to stop preventing writes with callbacks from being batched together.  Then we would need to change DBImpl::WriteImpl() to call all WriteCallback's in a batch, only write the batches that succeed, and correctly set the state of each batch's WriteThread::Writer.

Test Plan: Added test WriteWithCallbackTest to write_callback_test.cc which creates multiple client threads and verifies that writes are batched and executed properly.

Reviewers: hermanlee4, anthony, ngbronson

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D52863
2016-02-05 10:44:13 -08:00
Islam AbdelRahman 8e6172bc57 Add BlockBasedTableOptions::index_block_restart_interval
Summary: Add a new option to BlockBasedTableOptions that will allow us to change the restart interval for the index block

Test Plan: unit tests

Reviewers: yhchiang, anthony, andrewkr, sdong

Reviewed By: sdong

Subscribers: march, dhruba

Differential Revision: https://reviews.facebook.net/D53721
2016-02-05 10:22:37 -08:00
sdong 34a40bf911 Add --allow_concurrent_memtable_write in stress test and run it in crash_test
Summary: Add an option of --allow_concurrent_memtable_write in stress test and cover it in crash test

Test Plan: Run crash test and make sure three combinations of the two options show up randomly.

Reviewers: IslamAbdelRahman, yhchiang, andrewkr, anthony, kradhakrishnan

Reviewed By: kradhakrishnan

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53811
2016-02-04 16:15:18 -08:00
Siying Dong 73bf330c79 Merge pull request #973 from yuslepukhin/moveout_testcode
Move out of release build some test code
2016-02-04 15:50:21 -08:00
sdong f7c0f4e3ef perf_context.cc and iostats_context.cc use different output macro (fix unity build)
Summary: Unity build because perf_context.cc and iostats_context.cc define a different OUTPUT macro. Fix it.

Test Plan: Build unity

Reviewers: kradhakrishnan, anthony, yhchiang, andrewkr, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: divchenko, leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53769
2016-02-03 17:27:08 -08:00
Dmitri Smirnov 9656eab005 This partially addresses issue https://github.com/facebook/rocksdb/issues/935
testutil.cc and testharness.cc could not be moved out at this time
  as they are used by 4 benchmarks in release builds.
2016-02-03 17:06:10 -08:00
Jonathan Wiepert 14a322033f Remove references to files deleted in commit abb4052278
Summary:
Remove obolete references to files in src.mk
Fix incorrect path for reference in source.mk

Test Plan: Ran build to ensure changes do not break anything.

Reviewers: leveldb, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53733
2016-02-03 16:47:45 -08:00
krad 8445e53807 Add a mechanism to run all tests in sandcastle
Summary:
When making environment specific changes, it is better to run all CI
tests. This diff provides a mechanism to do that

Format is:

ROCKSDB_CHECK_ALL=1 arc diff

Test Plan: Submit request for diff

Reviewers: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53631
2016-02-03 16:34:08 -08:00
Igor Canadi 461cec4e8e Merge pull request #972 from adamretter/wb-threads
Threaded tests for WriteBatch
2016-02-03 12:32:46 -08:00
Adam Retter 9ab269ab39 Threaded tests for WriteBatch 2016-02-03 20:15:49 +00:00
krad bf767c6411 Minor fix to makefile
Summary: unit_481 is misspelt. Fixing it.

Test Plan: Running make commit_prereq

Reviewers: leveldb

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53757
2016-02-03 11:41:23 -08:00
Nathan Bronson 2c1db5ea51 always invalidate sequential-insertion cache for concurrent skiplist adds
Summary:
InlineSkipList::InsertConcurrently should invalidate the
sequential-insertion cache prev_[] for all inserts of multi-level nodes,
not just those that increase the height of the skip list.  The invariant
for prev_ is that prev_[i] (i > 0) is supposed to be the predecessor of
prev_[0] at level i.  Before this diff InsertConcurrently could violate
this constraint when inserting a multi-level node after prev_[i] but
before prev_[0].

This diff also reenables kConcurrentSkipList as db_test's
MultiThreaded/MultiThreadedDBTest.MultiThreaded/29.

Test Plan:
1. unit tests
2. temporarily hack kConcurrentSkipList timing so that it is fast but has a 1.5% failure rate on my dev box (1ms stagger on thread launch, 1s test duration, failure rate baseline over 1000 runs)
3. observe 1000 passes post-fix

Reviewers: igor, sdong

Reviewed By: sdong

Subscribers: MarkCallaghan, dhruba

Differential Revision: https://reviews.facebook.net/D53751
2016-02-03 11:08:16 -08:00
Siying Dong c12ff20ab2 Merge pull request #965 from koldat/jni_for_windows
Adding support for Windows JNI build
2016-02-03 10:41:57 -08:00
Gunnar Kudrjavets a09ce4fcd3 Skip some of the non-critical tests in ./tools/run_flash_bench.sh
Summary:
Some of the tests aren't considered to be critical when it comes to getting key benchmarking data for RocksDB. Therefore we'll introduce an environment variable `SKIP_LOW_PRI_TESTS` which enables skipping those test cases. By default all the tests will be run. If you want to optimize the test-case execution then do the following:

`
$ export SKIP_LOW_PRI_TESTS=1
$ ./tools/run_flash_bench.sh
`

Test Plan: Verified that when  `SKIP_LOW_PRI_TESTS` is not set then `benchmark.sh` is called for all the scenarios and when `SKIP_LOW_PRI_TESTS` is set to `1` then `benchmark.sh` is called only for the test-cases which are critical.

Reviewers: MarkCallaghan

Reviewed By: MarkCallaghan

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53739
2016-02-03 09:56:56 -08:00
Andrew Kryczka 284aa613a7 Eliminate duplicated property constants
Summary:
Before this diff, there were duplicated constants to refer to properties (user-
facing API had strings and InternalStats had an enum). I noticed these were
inconsistent in terms of which constants are provided, names of constants, and
documentation of constants. Overall it seemed annoying/error-prone to maintain
these duplicated constants.

So, this diff gets rid of InternalStats's constants and replaces them with a map
keyed on the user-facing constant. The value in that map contains a function
pointer to get the property value, so we don't need to do string matching while
holding db->mutex_. This approach has a side benefit of making many small
handler functions rather than a giant switch-statement.

Test Plan: db_properties_test passes, running "make commit-prereq -j32"

Reviewers: sdong, yhchiang, kradhakrishnan, IslamAbdelRahman, rven, anthony

Reviewed By: anthony

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53253
2016-02-02 19:14:56 -08:00
Gabriela Jacques da Silva 94be872eab Merge branch 'master' of github.com:facebook/rocksdb 2016-02-02 18:38:16 -08:00
Gabriela Jacques da Silva 0c2bd5cb4b Removing data race from expirable transactions
Summary:
Doing inline checking of transaction expiration instead of
using a callback.

Test Plan: To be added

Reviewers: anthony

Reviewed By: anthony

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53673
2016-02-02 18:37:44 -08:00
Nathan Bronson 5fcd1ba30a disable kConcurrentSkipList multithreaded test
Summary: Disable test that is intermittently failing

Test Plan: unit tests

Reviewers: igor, andrewkr, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53715
2016-02-02 18:24:47 -08:00
Andrew Kryczka 466c2c1bf7 Generate tags for *.c files
Summary:
db/c_test.c uses the functions in db/c.cc. If we have tags generated
for one but not the other, it's easy to make mistakes like updating a function
signature and missing a call site.

Test Plan:
  $ make tags

in vim:

  :cscope find s rocksdb_options_set_compression_options
  ...
  3    325  db/c_test.c <<main>>
            rocksdb_options_set_compression_options(options, -14, -1, 0);

Reviewers: sdong, yhchiang, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53685
2016-02-02 14:52:49 -08:00
Siying Dong 70c068c97a Merge pull request #960 from koldat/masterFixes2
Making use of GetSystemTimePreciseAsFileTime dynamic
2016-02-02 13:43:09 -08:00
Tomas Kolda a62c519bb6 RollLogFile tries to find non conflicting file until there is no conflict. 2016-02-02 10:33:49 +01:00
Tomas Kolda 57a95a7001 Making use of GetSystemTimePreciseAsFileTime dynamic - code review fixes 2016-02-02 10:23:56 +01:00
Tomas Kolda 502d41f150 Making use of GetSystemTimePreciseAsFileTime dynamic to not
break compatibility with Windows 7. The issue with rotated logs
was fixed other way.
2016-02-02 10:23:56 +01:00
Tomas Kolda 52153930d8 Adding support for Windows JNI build - fix Java unit test for release build of JNI DLL 2016-02-02 09:45:18 +01:00
Tomas Kolda e2972803a6 Adding support for Windows JNI build 2016-02-02 09:11:37 +01:00
Nathan Bronson 9c2cf9479b Fix for --allow_concurrent_memtable_write with batching
Summary:
Concurrent memtable adds were incorrectly computing
the last sequence number for a write batch group when the
write batches were not solitary.  This is the cause of
https://github.com/facebook/mysql-5.6/issues/155

Test Plan:
1. unit tests
2. new unit test
3. parallel db_bench stress tests with batch size of 10 and asserts enabled

Reviewers: igor, sdong

Reviewed By: sdong

Subscribers: IslamAbdelRahman, MarkCallaghan, dhruba

Differential Revision: https://reviews.facebook.net/D53595
2016-02-01 20:41:57 -08:00
sdong ac3fa9a6fe Travis CI to disable ROCKSDB_LITE tests
Summary: Travis CI fails most of the times because of timing out. To unblock it, disable LITE tests in Travis CI.

Test Plan: Will see.

Reviewers: anthony, yhchiang, kradhakrishnan, IslamAbdelRahman, andrewkr

Reviewed By: andrewkr

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53679
2016-02-01 18:42:01 -08:00
Siying Dong 7b943da1b2 Merge pull request #967 from SherlockNoMad/ValueSize
Add histogram for value size per operation
2016-02-01 17:59:08 -08:00
Siying Dong b5750790e0 Merge pull request #968 from yuslepukhin/one_shot_buffer
Enable per-request buffer allocation in RandomAccessFile
2016-02-01 16:39:55 -08:00
Islam AbdelRahman 1ad8182950 Fix WriteBatchTest.ManyUpdates, WriteBatchTest.LargeKeyValue under clang
Summary:
Fix current clang failure
https://ci-builds.fb.com/view/rocksdb/job/rocksdb_clang_build/1398/console

Test Plan:
make sure that both clang and g++ compilation succeed

USE_CLANG=1 make check -j64
make check -j64

Reviewers: anthony, yhchiang, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53667
2016-02-01 16:07:53 -08:00
sdong ad7ecca72d Add unit tests to verify large key/value
Summary:
Add unit tests:
(1) insert entries of 8MB key and 3GB value to DB
(2) insert entry of 3GB key and 3GB value into write batch and make sure we can read it.
(3) insert 3 billions of key-value pairs into write batch and make sure we can read it.
Disable them because not all platform can run it.

Test Plan: Run the tests

Reviewers: IslamAbdelRahman, yhchiang, kradhakrishnan, andrewkr, anthony

Reviewed By: anthony

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53619
2016-02-01 15:09:50 -08:00
Andrew Kryczka fdd70d1495 Skip filters for last L0 file if hit-optimized
Summary:
Following up on D53493, we can still enable the filter-skipping
optimization for last file in L0. It's correct to assume the key will be present
in the last L0 file when we're hit-optimized and L0 is deepest.

The FilePicker encapsulates the state for traversing each level's files, so I
needed to make it expose whether the returned file is last in its level.

Test Plan:
verified below test fails before this patch and passes afterwards.
The change to how the test memtable is populated is needed so file 1 has keys
(0, 30, 60), file 2 has keys (10, 40, 70), etc.

  $ ./db_universal_compaction_test --gtest_filter=UniversalCompactionNumLevels/DBTestUniversalCompaction.OptimizeFiltersForHits/*

Reviewers: sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53583
2016-02-01 14:58:46 -08:00
Dmytro Ivchenko aa5e3b7c04 PerfContext::ToString() add option to exclude zero counters
Test Plan: Added unit test to check w/ w/o zeros scenarios

Reviewers: yhchiang

Reviewed By: yhchiang

Subscribers: sdong, dhruba

Differential Revision: https://reviews.facebook.net/D52809
2016-02-01 13:41:13 -08:00
Dmitri Smirnov 36300fbbe3 Enable per-request buffer allocation in RandomAccessFile
This change impacts only non-buffered I/O on Windows.
 Currently, there is a buffer per RandomAccessFile
 instance that is protected by a lock. The reason we
 maintain the buffer is non-buffered I/O requires an aligned
 buffer to work.
 XPerf traces demonstrate that we accumulate a considerable
 wait time while waiting for that lock.
 This change enables to set random access buffer size to zero
 which would indicate a per request allocation.
 We are expecting that allocation expense would be much less than
 I/O costs plus wait time due to the fact that the memory heap
 would tend to re-use page aligned allocations especially with the
 use of Jemalloc.
 This change does not affect buffer use as a read_ahead_buffer for
 compaction purposes.
2016-02-01 13:14:37 -08:00
Yueh-Hsuan Chiang 1d854fa3d4 Fixed the asan error on column_family_test
Summary:
Fixed the asan error on column_family_test caused by not disabling
SyncPoint.

Test Plan: column_family_test

Reviewers: anthony, rven, kradhakrishnan, sdong, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53505
2016-02-01 12:45:45 -08:00
SherlockNoMad 37159a6448 Add histogram for value size per operation 2016-01-31 18:09:24 -08:00
Venkatesh Radhakrishnan 3b2a1ddd2e Add options.base_background_compactions as a number of compaction threads for low compaction debt
Summary:
If options.base_background_compactions is given, we try to schedule number of compactions not existing this number, only when L0 files increase to certain number, or pending compaction bytes more than certain threshold, we schedule compactions based on options.max_background_compactions.

The watermarks are calculated based on slowdown thresholds.

Test Plan:
Add new test cases in column_family_test.
Adding more unit tests.

Reviewers: IslamAbdelRahman, yhchiang, kradhakrishnan, rven, anthony

Reviewed By: anthony

Subscribers: leveldb, dhruba, yoshinorim

Differential Revision: https://reviews.facebook.net/D53409
2016-01-29 16:15:53 -08:00
sdong 6ee38bb15c Slowdown of writing to the last memtable should not override stopping
Summary: Now slowing down for the last mem table takes priority against some stopping conditions. This is logically confusing. Fix it.

Test Plan: Run all existing tests.

Reviewers: yhchiang, IslamAbdelRahman, kradhakrishnan, andrewkr, anthony

Reviewed By: anthony

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53529
2016-01-28 21:52:29 -08:00
Islam AbdelRahman d6c838f1e1 Add SstFileManager (component tracking all SST file in DBs and control the deletion rate)
Summary:
Add a new class SstFileTracker that will be notified whenever a DB add/delete/move and sst file, it will also replace DeleteScheduler
SstFileTracker can be used later to abort writes when we exceed a specific size

Test Plan: unit tests

Reviewers: rven, anthony, yhchiang, sdong

Reviewed By: sdong

Subscribers: igor, lovro, march, dhruba

Differential Revision: https://reviews.facebook.net/D50469
2016-01-28 18:35:01 -08:00
agiardullo 45768ade4f transaction allocation perf improvements
Summary: Removed a couple of memory allocations

Test Plan: changes covered by existing tests

Reviewers: rven, yhchiang, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53523
2016-01-28 17:32:28 -08:00
Igor Canadi 77926f93e5 Merge pull request #964 from benoitc/fix/pi2
fix build for raspberry 2
2016-01-28 10:24:11 -08:00
benoitc 03a5661a17 fix build for raspberry 2
fix #810
2016-01-28 11:43:58 +01:00
sdong 4b50f13540 Should not skip bloom filter for L0 during the query.
Summary: It's a regression bug caused by e089db40f9. With the change, if options.optimize_filters_for_hits=true and there are only L0 files (like single level universal compaction), we skip all the files in L0, which is more than necessary. Fix it by always trying to query bloom filter for files in level 0.

Test Plan: Add a unit test for it.

Reviewers: anthony, rven, yhchiang, IslamAbdelRahman, kradhakrishnan, andrewkr

Reviewed By: andrewkr

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53493
2016-01-27 16:16:39 -08:00
Siying Dong eadd221d3c Merge pull request #959 from koldat/master
Fixing generated GenerateBuildVersion.vcxproj on non English locale
2016-01-27 12:25:47 -08:00
Igor Canadi d209076fa0 Merge pull request #961 from wingify/master
Add Wingify to USERS.md
2016-01-27 11:53:47 -08:00
Ankit Jain 26c618004a Add Wingify to USERS.md
VWO, the flagship product of Wingify, uses RocksDB to populate the list of URLs where clients can run A/B test campaigns. It is also used to provide a way for the clients to see if VWO's Smart Code is installed on a specific URL of their account.
2016-01-28 01:04:49 +05:30
Tomas Kolda 4265f81e87 Remove util/auto_roll_logger.cc (it was moved to different directory) 2016-01-27 12:07:31 +01:00
Tomas Kolda d7f22b6d25 Fixing generated GenerateBuildVersion.vcxproj when one builds on different
locale than english. The problem is that date and time CLI utilities
generates different format so that REGEX in CMake does not work.
2016-01-27 10:51:20 +01:00
sdong d20915d52a Disable stats about mutex duration by default
Summary: Measuring mutex duration will measure time inside DB mutex, which breaks our best practice. Add a stat level in Statistics class. By default, disable to measure the mutex operations.

Test Plan: Add a unit test to make sure it is off by default.

Reviewers: rven, anthony, IslamAbdelRahman, kradhakrishnan, andrewkr, yhchiang

Reviewed By: yhchiang

Subscribers: MarkCallaghan, leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53367
2016-01-26 14:56:55 -08:00
Islam AbdelRahman 0c433cd1eb Fix issue in Iterator::Seek when using Block based filter block with prefix_extractor
Summary: Similar to D53385 we need to check InDomain before checking the filter block.

Test Plan: unit tests

Reviewers: yhchiang, rven, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53421
2016-01-26 14:47:42 -08:00
Andrew Kryczka 035857a312 Fix RocksDB lite build
Summary:
NewMemEnv() is defined in rocksdb lite but just returns nullptr --
would it be better to just not define it so we can catch issues like this at
compile-time?

Test Plan:
  $ make clean && OPT="-DTRAVIS -DROCKSDB_LITE" V=1 make -j32 db_test
  $ ./db_test --gtest_filter='DBTest.MemEnvTest'
  ...
  [  PASSED  ] 0 tests.

Reviewers: yhchiang, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53427
2016-01-26 13:15:36 -08:00
Islam AbdelRahman 77ef87ccb0 Update fbcode_config4.8.1.sh to use update_dependencies.sh
Summary:
This patch update fbcode_config4.8.1.sh to get it's dependencies the same way we updated fbcode_config.sh in D53037

as a result zstd is upgraded to 0.4.7 instead of 0.4.5

Test Plan:
make clean && ROCKSDB_FBCODE_BUILD_WITH_481=1 make check -j64
make clean && ROCKSDB_FBCODE_BUILD_WITH_481=1 USE_CLANG=1 make check -j64

Reviewers: yhchiang, andrewkr, rven, kradhakrishnan, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53355
2016-01-26 12:13:27 -08:00
Yueh-Hsuan Chiang 955ecf8b49 Fix an ASAN error in compact_files_test
Summary:
compact_files_test enables SyncPoint but never disable it before
the test terminates.  As a result, it might cause heap-use-after-free
error when some code path trying to access the static variable of
SyncPoint when it has already gone out of scope after the main thread
dies.

Test Plan:
COMPILE_WITH_ASAN=1 make compact_files_test -j32
./compact_files_test

Reviewers: sdong, anthony, kradhakrishnan, rven, andrewkr, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53379
2016-01-26 11:30:30 -08:00
Islam AbdelRahman b0afcdeeac Fix bug in block based tables with full filter block and prefix_extractor
Summary:
Right now when we are creating a BlockBasedTable with fill filter block
we add to the filter all the prefixes that are InDomain() based on the prefix_extractor

the problem is that when we read a key from the file, we check the filter block for the prefix whether or not it's InDomain()

Test Plan: unit tests

Reviewers: yhchiang, rven, anthony, kradhakrishnan, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53385
2016-01-26 11:07:08 -08:00
Andrew Kryczka 167bd8856d [directory includes cleanup] Finish removing util->db dependencies 2016-01-26 10:49:24 -08:00
Andrew Kryczka acd7d58695 [directory includes cleanup] Remove util->db dependency for ThreadStatusUtil
Summary:
We can avoid the dependency by forward-declaring ColumnFamilyData and
then treating it as a black box. That means callers of ThreadStatusUtil need to
explicitly provide more options, even if they can be derived from the
ColumnFamilyData, since ThreadStatusUtil doesn't include the definition.

This is part of a series of diffs to eliminate circular dependencies between
directories (e.g., db/* files depending on util/* files and vice-versa).

Test Plan:
  $ ./db_test --gtest_filter=DBTest.GetThreadStatus
  $ make -j32 commit-prereq

Reviewers: sdong, yhchiang, IslamAbdelRahman

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53361
2016-01-26 10:49:16 -08:00
Andrew Kryczka 46f9cd46af [directory includes cleanup] Move cross-function test points
Summary:
I split the db-specific test points out into a separate file under db/
directory. There were also a few bugs to fix in xfunc.{h,cc} that prevented it
from compiling previously; see https://reviews.facebook.net/D36825.

Test Plan:
compilation works now, below command works, will also run "make xfunc".

  $ make check ROCKSDB_XFUNC_TEST='managed_new' tests-regexp='DBTest' -j32

Reviewers: sdong, yhchiang

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53343
2016-01-26 10:49:05 -08:00
krad 22ecb752db Add valgrind to pre-commit sandcastle testing
Summary:
Initially I removed "valgrind" from the list since it take too much
time (3+hr) compared to tsan (40 min) when the tests are run in parallel. It is
not effective to run the tests in parallel in sandcastle and tsan takes about
3hrs as well.

Adding valgrind to the list.

Test Plan: Submit this diff and watch the run

Reviewers: sdong, rven

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53337
2016-01-26 10:40:57 -08:00
Venkatesh Radhakrishnan b7ecf3d214 Fix intermittent hang in ColumnFamilyTest.FlushAndDropRaceCondition
Summary:
ColumnFamilyTest.FlushAndDropRaceCondition sometimes
hangs because the sync point, "FlushJob::InstallResults", sleeps
holding the DB mutex. Fixing it by releasing the mutex before sleeping.

Test Plan:
seq 1000 |parallel --gnu --eta 't=/dev/shm/rdb-{}; rm -rf $t;
mkdir $t && export TEST_TMPDIR=$t; ./column_family_test
-gtest_filter=*FlushAndDropRaceCondition* > $t/log-{}'

Reviewers: IslamAbdelRahman, anthony, kradhakrishnan, yhchiang, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53349
2016-01-26 09:12:20 -08:00
sdong 38e1d7fea3 ldb to support --column_family option
Summary:
Add an option --column_family option, so that users can query or update specific column family.
Also add an create column family parameter to make unit test easier.
Still need to add unit tests.

Test Plan: Will add a test case in ldb python test.

Reviewers: yhchiang, rven, andrewkr, IslamAbdelRahman, kradhakrishnan, anthony

Reviewed By: anthony

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53265
2016-01-25 14:58:18 -08:00
sdong da33dfe188 Parameterize DBTest.Randomized
Summary: Break down DBTest.Randomized to multiple gtest tests based on config type

Test Plan: Run the test and all tests. Make sure configurations are correctly set

Reviewers: yhchiang, IslamAbdelRahman, rven, kradhakrishnan, andrewkr, anthony

Reviewed By: anthony

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53247
2016-01-25 14:49:08 -08:00
sdong fb9811ee9b Add a perf context level that doesn't measure time for mutex operations
Summary: Timing mutex operations can impact scalability of the system. Add a new perf context level that can measure time counters except for mutex.

Test Plan: Add a new unit test case to make sure it is not set.

Reviewers: IslamAbdelRahman, rven, kradhakrishnan, yhchiang, anthony

Reviewed By: anthony

Subscribers: MarkCallaghan, leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53199
2016-01-25 14:45:03 -08:00
Islam AbdelRahman f7ef1a6132 Include rest of dependencies in dependencies.sh
Summary:
This diff

  - Include the rest of the dependencies (kernel-headers, binutils, valgrind) in dependencies.sh
  - updtade zst to 0.4.7
  - It also fix a problem in clang scan build

Test Plan:
make check
USE_CLANG=1 make check
USE_CLANG=1 make analyze
coverage_test.sh

Reviewers: sdong, yhchiang, rven, andrewkr, kradhakrishnan

Reviewed By: kradhakrishnan

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53301
2016-01-25 14:35:04 -08:00
Andrew Kryczka 3e9209a078 Updated GetProperty documentation
Summary: As titled. Also added the kBaseLevel string, which was missing earlier.

Test Plan: built

Reviewers: yhchiang, anthony, rven, kradhakrishnan, IslamAbdelRahman, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53271
2016-01-25 14:04:55 -08:00
Venkatesh Radhakrishnan 40911e0b32 Run unit tests in parallel to find failing tests
Summary:
Added make targets parallel_test and parallel_dbtest to run
tests in parallel. Each test is run 32 times in parallel. There is a
timeout to catch hangs. The test continues after a failure and reports
non-zero status on failure

Test Plan: Run the two make targets

Reviewers: anthony, yhchiang, IslamAbdelRahman, kradhakrishnan, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53079
2016-01-25 14:04:27 -08:00
Islam AbdelRahman 2fbc59a348 Disallow SstFileWriter from creating empty sst files
Summary:
SstFileWriter may create an sst file with no entries
Right now this will fail when being ingested using DB::AddFile() saying that the keys are corrupted

Test Plan: make check

Reviewers: yhchiang, rven, anthony, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D52815
2016-01-25 13:47:07 -08:00
krad f53c95f81b Cosmetic fixes and comments for the reader
Summary:
Cosmetic fixes and some comments for the script. It is one big hack and
hopefully the comments will make it easy to maintain.

Test Plan: Run manual tests

Reviewers: sdong, rven

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53277
2016-01-25 10:45:29 -08:00
sdong f1ed170105 Add tests to make sure new DB or ColumnFamily options are settable through string
Summary: Add a test to fail if someone adds a DB options.

Test Plan: Run the test, run the test with valgrind. Add an option to DB option in the middle or in the end and make sure it fails.

Reviewers: yhchiang, anthony, IslamAbdelRahman, kradhakrishnan, rven, andrewkr

Reviewed By: andrewkr

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53097
2016-01-22 16:18:44 -08:00
krad f57596b0cb Improvements to pre-commit
Summary:
- UI is enhanced to lists the tests, status and the results
- We are using the same pre-commit tool as the make equivalent
- No more emails to user on failure
- Dropped valgrind from the list since it can be a time hogger (and can hurt
  scheduling for others)
- Patching bug fix
- Made the jobs run in parallel in sandcastle

Test Plan: Manual test

Reviewers: sdong, rven, igor

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53217

Making parallel requests to sandcastle

Test Plan: Run manual tests

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53259
2016-01-22 15:40:32 -08:00
Islam AbdelRahman 538eec0661 Update fbcode_config.sh to use latest versions automatically
Summary:
Update fbcode_config.sh so that It try to use the latest version for dependencies that we are using, after updating the code these libraries where updated

```
Snappy: 1.0.3 => 1.1.3
GFLAGS: 1.6 => 2.1.1
JEMALLOC: 3.6.0 => 4.0.3
```

I have also updated clang from 3.7 to 3.7.1

```
Clang 3.7 => 3.7.1
```

Another change is that we use the same tp2 directory as fbcode, so we dont need to keep changing commit hash every time we need to change a version of a compiler or a library

Test Plan:
make check -j64
USE_CLANG=1 make check -j64

DISABLE_JEMALLOC=1 OPT=-g make all valgrind_check -j32 (make sure it's running)

Reviewers: yhchiang, anthony, rven, kradhakrishnan, andrewkr, sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D53037
2016-01-22 11:56:32 -08:00
agiardullo 8019aa9b55 improve test for manifest write failure
Summary: Improve testing per discussion in D52989

Test Plan: ran test

Reviewers: sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53211
2016-01-22 11:47:59 -08:00
agiardullo bcd4ccbc33 Revert D7809
Summary: Revert the functionaility of D7809 (but I'm keeping the logging and test code).  We decided it was dangerous to ignore sync failures based on attempting to read the data written.  The read does not tell us whether the data was synced.

Test Plan: There was no test for the particular functionaility that was reverted.  Keeping the test code from D7809 that tests whether we set the DB to be readonly when paranoid checks are enabled.

Reviewers: rven, yhchiang, kradhakrishnan, IslamAbdelRahman, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D52989
2016-01-21 17:14:43 -08:00
krad b0a15e7fb9 Mechanism to run CI jobs on local branch via commit_prereq
Summary: This patch provides a mechanism to run pre commit tests on the local
branch before committing. This can help prevent frequent build breaks.

The tests can be run in parallel by specifying the J=<..> environment
variable.

Test Plan: Run manually

Reviewers: sdong rven tec

CC: leveldb@

Task ID: #9689218

Blame Rev:
2016-01-21 16:22:21 -08:00
Andrew Kryczka bb2888738c Cleanup property-related variable names
Summary:
I noticed these names were quite confusing while updating GetProperty
documentation.

Test Plan: running "make commit-prereq -j32"

Reviewers: yhchiang, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53073
2016-01-21 11:38:15 -08:00
Andrew Kryczka 29289333d0 Add named constants for remaining properties
Summary:
There were just these two properties that didn't have any named
constant.

Test Plan:
build and below test

  $ ./db_properties_test --gtest_filter=DBPropertiesTest.NumImmutableMemTable

Reviewers: yhchiang, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53103
2016-01-21 10:59:36 -08:00
sdong 2c2b72218d Disable OptionsParserTest.BlockBasedTableOptionsAllFieldsSettable under CLANG
Summary: OptionsParserTest.BlockBasedTableOptionsAllFieldsSettable is failiong under CLANG. Disable the test to unblock the build.

Test Plan: Run it both of CLANG and GCC

Reviewers: kradhakrishnan, rven, andrewkr, anthony, yhchiang, IslamAbdelRahman

Reviewed By: IslamAbdelRahman

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D53157
2016-01-21 10:56:34 -08:00
krad a300d9928e Added sandcastle pre-commit
Test Plan:
Lately we have been breaking our builds too often. This changes adds
the capability to schedule tests in sandcastle for every diff created. This will
help us increase the pre-commit testing bar.

This patch will dispatch signals to sandcastle to start running tests on the
diff. The test failures are reported to the user via email.

The user can also manually check the progress of test in sandcastle via the URL
provided.

Reviewers: sdong, rven

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D53001
2016-01-21 08:53:55 -08:00
sdong 202be23e46 Add test that verifies all options in BlockBasedTableOptions is settable through GetBlockBasedTableOptionsFromString()
Summary: Add a test OptionsParserTest.BlockBasedTableOptionsAdded, which will fail if a new option is added to BlockBasedTableOptions but is not settable through GetBlockBasedTableOptionsFromString().

Test Plan: Run the test. Also manually remove and add options and make sure it fails.

Reviewers: anthony, IslamAbdelRahman, kradhakrishnan, rven, yhchiang, andrewkr

Reviewed By: andrewkr

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D52953
2016-01-20 21:59:28 -08:00
Andrew Kryczka eceb5cb1b7 Split db_test.cc (part 1: properties)
Summary:
Moved all the tests that verify property correctness into a separate
file. The goal is to reduce compile time and complexity of db_test. I didn't
add parallelism for db_properties_test, even though these tests were
parallelized in db_test, since the file is small enough that it won't matter.

Some of these moves may be controversial since it's hard to say whether the
test is "verifying property correctness," or "using properties to verify
rocksdb's correctness." I'm interested in any opinions.

Test Plan: ran db_properties_test, also waiting on "make commit-prereq -j32"

Reviewers: yhchiang, IslamAbdelRahman, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D52995
2016-01-20 15:17:52 -08:00
sdong 94918ae84b db_bench: explicitly clear buffer in compress benchmark
Summary: It is reported that in compress benchmark in db_bench, zlib will cause an OOM. The suggestd fix was to clear the buffer.

Test Plan: Build and run compress benchmark.

Reviewers: IslamAbdelRahman, yhchiang, rven, andrewkr, kradhakrishnan, anthony

Reviewed By: anthony

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D52857
2016-01-19 18:11:46 -08:00
sdong fdbff42391 Crash test to make kill decision for every kill point
Summary:
In crash test, when coming to each kill point, we start a random class using seed as current second. With this approach, for every second, the random number used is the same. However, in each second, there are multiple kill points with different frequency. It makes it hard to reason about chance of kill point to trigger. With this commit, we use thread local random seed to generate the random number, so that it will take different values per second, hoping it makes chances of killing much easier to reason about.

Also significantly reduce the kill odd to make sure time before kiling is similar as before.

Test Plan: Run white box crash test and see the killing happens as expected and the run time time before killing reasonable.

Reviewers: kradhakrishnan, IslamAbdelRahman, rven, yhchiang, andrewkr, anthony

Reviewed By: anthony

Subscribers: leveldb, dhruba

Differential Revision: https://reviews.facebook.net/D52971
2016-01-19 18:11:24 -08:00
Siying Dong 39c3e94ff4 Merge pull request #954 from davidbernard/solaris_build
Solaris Build
2016-01-19 18:05:02 -08:00
David Bernard df7c2f3b59 As per google coding standard replace "using" in option_builder.cc and
geodb_impl.cc
2016-01-20 00:27:29 +00:00
David Bernard 12809b44b4 Revert "Change notification email for travis"
This reverts commit eaa5637568.
2016-01-19 22:46:39 +00:00
Mike Kolupaev 34704d5c7b [easy] Fixed a crash in LogAndApply() when CF creation failed
Summary: That line used to dereference `column_family_data`, which is nullptr if we're creating a column family.

Test Plan: `make -j check`

Reviewers: sdong

Reviewed By: sdong

Subscribers: dhruba

Differential Revision: https://reviews.facebook.net/D52881
2016-01-19 11:46:52 -08:00
Siying Dong 791dbafa93 Merge pull request #953 from sselva/master
Add Rakuten Marketing to USERS.md
2016-01-19 11:04:04 -08:00
Igor Canadi 594a5ef021 Merge pull request #955 from bcbrock/ppc64-build
Simple changes to support builds for ppc64[le] consistent with X86
2016-01-19 09:05:44 -08:00
bcbrock f423f05dcd Simple changes to support builds for ppc64[le] consistent with X86
These simple changes are required to allow builds on ppc64[le] systems
consistent with X86. The Makefile now recognizes both ppc64 and ppc64le, and
in the absence of PORTABLE=1, the code will be built analogously to the X86
-march=native.

Note that although GCC supports -mcpu=native -mtune=native on POWER, it
doesn't work correctly on all systems. This is why we need to get the actual
machine model from the AUX vector.
2016-01-19 09:08:19 -06:00
David Bernard 3f12e16f27 Make alloca.h optional 2016-01-19 06:17:31 +00:00
David Bernard eaa5637568 Change notification email for travis 2016-01-19 04:59:39 +00:00
David Bernard d78c6b28c4 Changes for build on solaris
Makefile adjust paths for solaris build
Makefile enable _GLIBCXX_USE_C99 so that std::to_string is available
db_compaction_test.cc Initialise a variable to avoid a compilation error
db_impl.cc Include <alloca.h>
db_test.cc Include <alloca.h>
Environment.java recognise solaris envrionment
options_bulder.cc Make log unambiguous
geodb_impl.cc Make log and floor unambiguous
2016-01-19 04:45:21 +00:00
Selva Sarangan 2e9fae3f2a Add Rakuten Marketing to USERS.md 2016-01-18 15:24:52 -08:00
agiardullo 83e1de92af move internal build to use zstd 0.4.5
Summary: update internal build scripts to use zstd 0.4.5.

Test Plan: built and ran tests with and without ROCKSDB_FBCODE_BUILD_WITH_481

Reviewers: yhchiang, rven, kradhakrishnan, sdong

Reviewed By: sdong

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D52833
2016-01-15 13:27:35 -08:00
Gunnar Kudrjavets aec10f734b Guard falloc.h inclusion to avoid build breaks
Summary: Depending on the order of include paths and versions of various headers we may end up in a situation where we'll encounter a build break caused by redefinition of constants. gcc-4.9-glibc-2.20 header update to include/bits/fcntl-linux.h introduced the definitions of FALLOC_FL_* constants. However, linux/falloc.h from kernel-headers also has FALLOC_FL_* constants defined. Therefore during the compilation we'll get "previously defined" errors.

Test Plan:
Both in the environment where the build break manifests (to make sure that the change fixed the problem) and in the environment where everything builds fine (to make sure that there are no regressions):

make clean
make -j 32

Reviewers: sdong, igor

Reviewed By: igor

Subscribers: dhruba, leveldb

Differential Revision: https://reviews.facebook.net/D52821
2016-01-14 22:47:15 -08:00
Alexander Fenster e16438bb86 fixing build warning 2016-01-11 11:23:33 -08:00
Alexander Fenster b73fbbaf64 added --no_value option to ldb scan to dump key only 2016-01-11 10:51:42 -08:00
714 changed files with 21198 additions and 13124 deletions
+1
View File
@@ -42,6 +42,7 @@ unity.a
tags
rocksdb_dump
rocksdb_undump
db_test2
java/out
java/target
+1 -1
View File
@@ -34,7 +34,7 @@ before_script:
# as EnvPosixTest::AllocateTest expects within the Travis OpenVZ environment.
script:
- if [[ "${TRAVIS_OS_NAME}" == 'linux' ]]; then OPT=-DTRAVIS CLANG_FORMAT_DIFF=/tmp/clang-format-diff.py make format || true; fi
- OPT=-DTRAVIS V=1 make -j4 check && OPT=-DTRAVIS V=1 make clean jclean rocksdbjava jtest && make clean && OPT="-DTRAVIS -DROCKSDB_LITE" V=1 make -j4 check
- OPT=-DTRAVIS V=1 make -j4 check && OPT=-DTRAVIS V=1 make clean jclean rocksdbjava jtest && make clean && OPT="-DTRAVIS -DROCKSDB_LITE" V=1 make -j4 static_lib
notifications:
email:
+41 -19
View File
@@ -13,8 +13,8 @@
# cd build
# 3. Run cmake to generate project files for Windows, add more options to enable required third-party libraries.
# See thirdparty.inc for more information.
# sample command: cmake -G "Visual Studio 12 Win64" -DGFLAGS=1 -DSNAPPY=1 -DJEMALLOC=1 ..
# OR for VS Studio 15 cmake -G "Visual Studio 14 Win64" -DGFLAGS=1 -DSNAPPY=1 -DJEMALLOC=1 ..
# sample command: cmake -G "Visual Studio 12 Win64" -DGFLAGS=1 -DSNAPPY=1 -DJEMALLOC=1 -DJNI=1 ..
# OR for VS Studio 15 cmake -G "Visual Studio 14 Win64" -DGFLAGS=1 -DSNAPPY=1 -DJEMALLOC=1 -DJNI=1 ..
# 4. Then build the project in debug mode (you may want to add /m[:<N>] flag to run msbuild in <N> parallel threads
# or simply /m ot use all avail cores)
# msbuild rocksdb.sln
@@ -30,10 +30,9 @@ cmake_minimum_required(VERSION 2.6)
project(rocksdb)
include(${CMAKE_CURRENT_SOURCE_DIR}/thirdparty.inc)
execute_process(COMMAND $ENV{COMSPEC} " /C date /T" OUTPUT_VARIABLE DATE)
execute_process(COMMAND $ENV{COMSPEC} " /C time /T" OUTPUT_VARIABLE TIME)
string(REGEX REPLACE "(..)/(..)/..(..).*" "\\1/\\2/\\3" DATE ${DATE})
execute_process(COMMAND powershell -Command "Get-Date -format MM_dd_yyyy" OUTPUT_VARIABLE DATE)
execute_process(COMMAND powershell -Command "Get-Date -format HH:mm:ss" OUTPUT_VARIABLE TIME)
string(REGEX REPLACE "(..)_(..)_..(..).*" "\\1/\\2/\\3" DATE ${DATE})
string(REGEX REPLACE "(..):(.....).*" " \\1:\\2" TIME ${TIME})
string(CONCAT GIT_DATE_TIME ${DATE} ${TIME})
@@ -99,6 +98,7 @@ add_subdirectory(third-party/gtest-1.7.0/fused-src/gtest)
# Main library source code
set(SOURCES
db/auto_roll_logger.cc
db/builder.cc
db/c.cc
db/column_family.cc
@@ -114,6 +114,7 @@ set(SOURCES
db/db_impl_debug.cc
db/db_impl_experimental.cc
db/db_impl_readonly.cc
db/db_info_dumper.cc
db/db_iter.cc
db/event_helpers.cc
db/experimental.cc
@@ -145,9 +146,12 @@ set(SOURCES
db/write_batch_base.cc
db/write_controller.cc
db/write_thread.cc
db/xfunc_test_points.cc
memtable/hash_cuckoo_rep.cc
memtable/hash_linklist_rep.cc
memtable/hash_skiplist_rep.cc
memtable/skiplistrep.cc
memtable/vectorrep.cc
port/stack_trace.cc
port/win/env_win.cc
port/win/port_win.cc
@@ -173,7 +177,6 @@ set(SOURCES
table/merger.cc
table/sst_file_writer.cc
table/meta_blocks.cc
table/mock_table.cc
table/plain_table_builder.cc
table/plain_table_factory.cc
table/plain_table_index.cc
@@ -182,9 +185,9 @@ set(SOURCES
table/table_properties.cc
table/two_level_iterator.cc
tools/sst_dump_tool.cc
tools/db_bench_tool.cc
tools/dump/db_dump_tool.cc
util/arena.cc
util/auto_roll_logger.cc
util/bloom.cc
util/build_version.cc
util/cache.cc
@@ -193,17 +196,18 @@ set(SOURCES
util/comparator.cc
util/concurrent_arena.cc
util/crc32c.cc
util/db_info_dumper.cc
util/delete_scheduler_impl.cc
util/delete_scheduler.cc
util/dynamic_bloom.cc
util/env.cc
util/env_hdfs.cc
util/event_logger.cc
util/file_util.cc
util/file_reader_writer.cc
util/sst_file_manager_impl.cc
util/filter_policy.cc
util/hash.cc
util/histogram.cc
util/histogram_windowing.cc
util/instrumented_mutex.cc
util/iostats_context.cc
tools/ldb_cmd.cc
@@ -211,7 +215,6 @@ set(SOURCES
util/logging.cc
util/log_buffer.cc
util/memenv.cc
util/mock_env.cc
util/murmurhash.cc
util/mutable_cf_options.cc
util/options.cc
@@ -223,7 +226,6 @@ set(SOURCES
util/perf_level.cc
util/random.cc
util/rate_limiter.cc
util/skiplistrep.cc
util/slice.cc
util/statistics.cc
util/status.cc
@@ -237,11 +239,12 @@ set(SOURCES
util/thread_status_updater.cc
util/thread_status_util.cc
util/thread_status_util_debug.cc
util/vectorrep.cc
util/transaction_test_util.cc
util/xfunc.cc
util/xxhash.cc
utilities/backupable/backupable_db.cc
utilities/checkpoint/checkpoint.cc
utilities/compaction_filters/remove_emptyvalue_compactionfilter.cc
utilities/document/document_db.cc
utilities/document/json_document.cc
utilities/document/json_document_builder.cc
@@ -275,6 +278,8 @@ set(SOURCES
# and linked to tests. Add test only code that is not #ifdefed for Release here.
set(TESTUTIL_SOURCE
db/db_test_util.cc
table/mock_table.cc
util/mock_env.cc
util/thread_status_updater_debug.cc
)
@@ -287,8 +292,19 @@ set_target_properties(rocksdb${ARTIFACT_SUFFIX} PROPERTIES COMPILE_FLAGS "-DROCK
add_dependencies(rocksdb${ARTIFACT_SUFFIX} GenerateBuildVersion)
target_link_libraries(rocksdb${ARTIFACT_SUFFIX} ${LIBS})
if (DEFINED JNI)
if (${JNI} EQUAL 1)
message(STATUS "JNI library is enabled")
add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/java)
else()
message(STATUS "JNI library is disabled")
endif()
else()
message(STATUS "JNI library is disabled")
endif()
set(APPS
db/db_bench.cc
tools/db_bench.cc
db/memtablerep_bench.cc
table/table_reader_bench.cc
tools/db_stress.cc
@@ -303,6 +319,7 @@ set(APPS
set(C_TESTS db/c_test.c)
set(TESTS
db/auto_roll_logger_test.cc
db/column_family_test.cc
db/compact_files_test.cc
db/compaction_iterator_test.cc
@@ -312,17 +329,20 @@ set(TESTS
db/comparator_db_test.cc
db/corruption_test.cc
db/cuckoo_table_db_test.cc
db/db_iter_test.cc
db/db_test.cc
db/db_compaction_filter_test.cc
db/db_compaction_test.cc
db/db_dynamic_level_test.cc
db/db_inplace_update_test.cc
db/db_iter_test.cc
db/db_log_iter_test.cc
db/db_properties_test.cc
db/db_table_properties_test.cc
db/db_tailing_iter_test.cc
db/db_test.cc
db/db_test2.cc
db/db_block_cache_test.cc
db/db_universal_compaction_test.cc
db/db_wal_test.cc
db/db_tailing_iter_test.cc
db/db_table_properties_test.cc
db/dbformat_test.cc
db/deletefile_test.cc
db/fault_injection_test.cc
@@ -340,6 +360,7 @@ set(TESTS
db/perf_context_test.cc
db/plain_table_db_test.cc
db/prefix_test.cc
db/repair_test.cc
db/skiplist_test.cc
db/table_properties_collector_test.cc
db/version_builder_test.cc
@@ -363,7 +384,6 @@ set(TESTS
tools/sst_dump_test.cc
util/arena_test.cc
util/autovector_test.cc
util/auto_roll_logger_test.cc
util/bloom_test.cc
util/cache_test.cc
util/coding_test.cc
@@ -376,8 +396,10 @@ set(TESTS
util/file_reader_writer_test.cc
util/heap_test.cc
util/histogram_test.cc
util/iostats_context_test.cc
util/memenv_test.cc
util/mock_env_test.cc
util/options_settable_test.cc
util/options_test.cc
util/rate_limiter_test.cc
util/slice_transform_test.cc
+9
View File
@@ -0,0 +1,9 @@
# RocksDB default options change log
## 4.7.0 (4/8/2016)
* options.write_buffer_size changes from 4MB to 64MB
* options.target_file_size_base changes from 2MB to 64MB
* options.max_bytes_for_level_base changes from 10MB to 256MB
* options.soft_pending_compaction_bytes_limit changes from 0 (disabled) to 64GB
* options.hard_pending_compaction_bytes_limit changes from 0 (disabled) to 256GB
* table_cache_numshardbits changes from 4 to 6
* max_file_opening_threads changes from 1 to 16
+25
View File
@@ -1,4 +1,29 @@
# Rocksdb Change Log
## 4.7.0 (4/8/2016)
### Public API Change
* rename options compaction_measure_io_stats to report_bg_io_stats and include flush too.
* Change some default options. Now default options will optimize for server-workloads. Also enable slowdown and full stop triggers for pending compaction bytes. These changes may cause sub-optimal performance or significant increase of resource usage. To avoid these risks, users can open existing RocksDB with options extracted from RocksDB option files. See https://github.com/facebook/rocksdb/wiki/RocksDB-Options-File for how to use RocksDB option files. Or you can call Options.OldDefaults() to recover old defaults. DEFAULT_OPTIONS_HISTORY.md will track change history of default options.
## 4.6.0 (3/10/2016)
### Public API Changes
* Change default of BlockBasedTableOptions.format_version to 2. It means default DB created by 4.6 or up cannot be opened by RocksDB version 3.9 or earlier.
* Added strict_capacity_limit option to NewLRUCache. If the flag is set to true, insert to cache will fail if no enough capacity can be free. Signiture of Cache::Insert() is updated accordingly.
* Tickers [NUMBER_DB_NEXT, NUMBER_DB_PREV, NUMBER_DB_NEXT_FOUND, NUMBER_DB_PREV_FOUND, ITER_BYTES_READ] are not updated immediately. The are updated when the Iterator is deleted.
* Add monotonically increasing counter (DB property "rocksdb.current-super-version-number") that increments upon any change to the LSM tree.
### New Features
* Add CompactionPri::kMinOverlappingRatio, a compaction picking mode friendly to write amplification.
* Deprecate Iterator::IsKeyPinned() and replace it with Iterator::GetProperty() with prop_name="rocksdb.iterator.is.key.pinned"
## 4.5.0 (2/5/2016)
### Public API Changes
* Add a new perf context level between kEnableCount and kEnableTime. Level 2 now does not include timers for mutexes.
* Statistics of mutex operation durations will not be measured by default. If you want to have them enabled, you need to set Statistics::stats_level_ to kAll.
* DBOptions::delete_scheduler and NewDeleteScheduler() are removed, please use DBOptions::sst_file_manager and NewSstFileManager() instead
### New Features
* ldb tool now supports operations to non-default column families.
* Add kPersistedTier to ReadTier. This option allows Get and MultiGet to read only the persited data and skip mem-tables if writes were done with disableWAL = true.
* Add DBOptions::sst_file_manager. Use NewSstFileManager() in include/rocksdb/sst_file_manager.h to create a SstFileManager that can be used to track the total size of SST files and control the SST files deletion rate.
## 4.4.0 (1/14/2016)
### Public API Changes
+1 -1
View File
@@ -21,7 +21,7 @@ depend on gflags. You will need to have gflags installed to run `make all`. This
use binaries compiled by `make all` in production.
* By default the binary we produce is optimized for the platform you're compiling on
(-march=native). If you want to build a portable binary, add 'PORTABLE=1' before
(-march=native or the equivalent). If you want to build a portable binary, add 'PORTABLE=1' before
your make commands, like this: `PORTABLE=1 make static_lib`
## Dependencies
+11
View File
@@ -0,0 +1,11 @@
This is the list of all known third-party language bindings for RocksDB. If something is missing, please open a pull request to add it.
* Java - https://github.com/facebook/rocksdb/tree/master/java
* Python - http://pyrocksdb.readthedocs.org/en/latest/
* Perl - https://metacpan.org/pod/RocksDB
* Node.js - https://npmjs.org/package/rocksdb
* Go - https://github.com/tecbot/gorocksdb
* Ruby - http://rubygems.org/gems/rocksdb-ruby
* Haskell - https://hackage.haskell.org/package/rocksdb-haskell
* PHP - https://github.com/Photonios/rocksdb-php
* C# - https://github.com/warrenfalk/rocksdb-sharp
+1 -1
View File
@@ -2,7 +2,7 @@ BSD License
For rocksdb software
Copyright (c) 2014, Facebook, Inc.
Copyright (c) 2011-present, Facebook, Inc.
All rights reserved.
---------------------------------------------------------------------
+111 -19
View File
@@ -84,7 +84,8 @@ endif
# compile with -O2 if debug level is not 2
ifneq ($(DEBUG_LEVEL), 2)
OPT += -O2 -fno-omit-frame-pointer
ifneq ($(MACHINE),ppc64) # ppc64 doesn't support -momit-leaf-frame-pointer
# Skip for archs that don't support -momit-leaf-frame-pointer
ifeq (,$(shell $(CXX) -fsyntax-only -momit-leaf-frame-pointer -xc /dev/null 2>&1))
OPT += -momit-leaf-frame-pointer
endif
endif
@@ -143,6 +144,9 @@ else
OPT += -DNDEBUG
endif
ifeq ($(PLATFORM), OS_SOLARIS)
PLATFORM_CXXFLAGS += -D _GLIBCXX_USE_C99
endif
ifneq ($(filter -DROCKSDB_LITE,$(OPT)),)
# found
CFLAGS += -fno-exceptions
@@ -170,6 +174,15 @@ else
pg = -pg
endif
# USAN doesn't work well with jemalloc. If we're compiling with USAN, we should use regular malloc.
ifdef COMPILE_WITH_UBSAN
DISABLE_JEMALLOC=1
EXEC_LDFLAGS += -fsanitize=undefined
PLATFORM_CCFLAGS += -fsanitize=undefined
PLATFORM_CXXFLAGS += -fsanitize=undefined
endif
ifndef DISABLE_JEMALLOC
EXEC_LDFLAGS := $(JEMALLOC_LIB) $(EXEC_LDFLAGS)
PLATFORM_CXXFLAGS += $(JEMALLOC_INCLUDE)
@@ -187,10 +200,6 @@ default: all
WARNING_FLAGS = -W -Wextra -Wall -Wsign-compare -Wshadow \
-Wno-unused-parameter
ifndef DISABLE_WARNING_AS_ERROR
WARNING_FLAGS += -Werror
endif
CFLAGS += $(WARNING_FLAGS) -I. -I./include $(PLATFORM_CCFLAGS) $(OPT)
CXXFLAGS += $(WARNING_FLAGS) -I. -I./include $(PLATFORM_CXXFLAGS) $(OPT) -Woverloaded-virtual -Wnon-virtual-dtor -Wno-missing-field-initializers
@@ -237,8 +246,12 @@ VALGRIND_VER := $(join $(VALGRIND_VER),valgrind)
VALGRIND_OPTS = --error-exitcode=$(VALGRIND_ERROR) --leak-check=full
BENCHTOOLOBJECTS = $(BENCH_SOURCES:.cc=.o) $(LIBOBJECTS) $(TESTUTIL)
TESTS = \
db_test \
db_test2 \
db_block_cache_test \
db_iter_test \
db_log_iter_test \
db_compaction_filter_test \
@@ -248,6 +261,7 @@ TESTS = \
db_tailing_iter_test \
db_universal_compaction_test \
db_wal_test \
db_properties_test \
db_table_properties_test \
block_hash_index_test \
autovector_test \
@@ -312,6 +326,7 @@ TESTS = \
rate_limiter_test \
delete_scheduler_test \
options_test \
options_settable_test \
options_util_test \
event_logger_test \
cuckoo_table_builder_test \
@@ -332,7 +347,8 @@ TESTS = \
compact_on_deletion_collector_test \
compaction_job_stats_test \
transaction_test \
ldb_cmd_test
ldb_cmd_test \
iostats_context_test
SUBSET := $(shell echo $(TESTS) |sed s/^.*$(ROCKSDBTESTS_START)/$(ROCKSDBTESTS_START)/)
@@ -460,7 +476,7 @@ test_names = \
-e '/^(\s*)(\S+)/; !$$1 and do {$$p=$$2; break};' \
-e 'print qq! $$p$$2!'
ifeq ($(MAKECMDGOALS),check)
ifneq (,$(filter check parallel_check,$(MAKECMDGOALS)),)
# Use /dev/shm if it has the sticky bit set (otherwise, /tmp),
# and create a randomly-named rocksdb.XXXX directory therein.
# We'll use that directory in the "make check" rules.
@@ -590,7 +606,7 @@ ldb_tests: ldb
crash_test: whitebox_crash_test blackbox_crash_test
blackbox_crash_test: db_stress
python -u tools/db_crashtest.py --simple blackbox
python -u tools/db_crashtest.py --simple blackbox
python -u tools/db_crashtest.py blackbox
whitebox_crash_test: db_stress
@@ -607,8 +623,18 @@ asan_crash_test:
COMPILE_WITH_ASAN=1 $(MAKE) crash_test
$(MAKE) clean
ubsan_check:
$(MAKE) clean
COMPILE_WITH_UBSAN=1 $(MAKE) check -j32
$(MAKE) clean
ubsan_crash_test:
$(MAKE) clean
COMPILE_WITH_UBSAN=1 $(MAKE) crash_test
$(MAKE) clean
valgrind_check: $(TESTS)
for t in $(filter-out %skiplist_test,$(TESTS)); do \
for t in $(filter-out %skiplist_test options_settable_test,$(TESTS)); do \
$(VALGRIND_VER) $(VALGRIND_OPTS) ./$$t; \
ret_code=$$?; \
if [ $$ret_code -ne 0 ]; then \
@@ -616,6 +642,46 @@ valgrind_check: $(TESTS)
fi; \
done
ifneq ($(PAR_TEST),)
parloop:
ret_bad=0; \
for t in $(PAR_TEST); do \
echo "===== Running $$t in parallel $(NUM_PAR)";\
if [ $(db_test) -eq 1 ]; then \
seq $(J) | v="$$t" parallel --gnu 's=$(TMPD)/rdb-{}; export TEST_TMPDIR=$$s;' \
'timeout 2m ./db_test --gtest_filter=$$v >> $$s/log-{} 2>1'; \
else\
seq $(J) | v="./$$t" parallel --gnu 's=$(TMPD)/rdb-{};' \
'export TEST_TMPDIR=$$s; timeout 10m $$v >> $$s/log-{} 2>1'; \
fi; \
ret_code=$$?; \
if [ $$ret_code -ne 0 ]; then \
ret_bad=$$ret_code; \
echo $$t exited with $$ret_code; \
fi; \
done; \
exit $$ret_bad;
endif
parallel_check: $(TESTS)
$(AM_V_GEN)if test "$(J)" > 1 \
&& (parallel --gnu --help 2>/dev/null) | \
grep -q 'GNU Parallel'; \
then \
echo Running in parallel $(J); \
else \
echo "Need to have GNU Parallel and J > 1"; exit 1; \
fi; \
ret_bad=0; \
echo $(J);\
echo Test Dir: $(TMPD); \
seq $(J) | parallel --gnu 's=$(TMPD)/rdb-{}; rm -rf $$s; mkdir $$s'; \
$(MAKE) PAR_TEST="$(shell $(test_names))" TMPD=$(TMPD) \
J=$(J) db_test=1 parloop; \
$(MAKE) PAR_TEST="$(filter-out db_test, $(TESTS))" \
TMPD=$(TMPD) J=$(J) db_test=0 parloop;
analyze: clean
$(CLANG_SCAN_BUILD) --use-analyzer=$(CLANG_ANALYZER) \
--use-c++=$(CXX) --use-cc=$(CC) --status-bugs \
@@ -652,7 +718,7 @@ clean:
tags:
ctags * -R
cscope -b `find . -name '*.cc'` `find . -name '*.h'`
cscope -b `find . -name '*.cc'` `find . -name '*.h'` `find . -name '*.c'`
format:
build_tools/format-diff.sh
@@ -667,7 +733,7 @@ $(LIBRARY): $(LIBOBJECTS)
$(AM_V_AR)rm -f $@
$(AM_V_at)$(AR) $(ARFLAGS) $@ $(LIBOBJECTS)
db_bench: db/db_bench.o $(LIBOBJECTS) $(TESTUTIL)
db_bench: tools/db_bench.o $(BENCHTOOLOBJECTS)
$(AM_LINK)
cache_bench: util/cache_bench.o $(LIBOBJECTS) $(TESTUTIL)
@@ -742,6 +808,12 @@ slice_transform_test: util/slice_transform_test.o $(LIBOBJECTS) $(TESTHARNESS)
db_test: db/db_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
db_test2: db/db_test2.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
db_block_cache_test: db/db_block_cache_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
db_log_iter_test: db/db_log_iter_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
@@ -769,6 +841,9 @@ db_universal_compaction_test: db/db_universal_compaction_test.o db/db_test_util.
db_wal_test: db/db_wal_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
db_properties_test: db/db_properties_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
db_table_properties_test: db/db_table_properties_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
@@ -946,6 +1021,9 @@ compact_files_test: db/compact_files_test.o $(LIBOBJECTS) $(TESTHARNESS)
options_test: util/options_test.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
options_settable_test: util/options_settable_test.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
options_util_test: utilities/options/options_util_test.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
@@ -970,7 +1048,7 @@ manual_compaction_test: db/manual_compaction_test.o $(LIBOBJECTS) $(TESTHARNESS)
filelock_test: util/filelock_test.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
auto_roll_logger_test: util/auto_roll_logger_test.o $(LIBOBJECTS) $(TESTHARNESS)
auto_roll_logger_test: db/auto_roll_logger_test.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
memtable_list_test: db/memtable_list_test.o $(LIBOBJECTS) $(TESTHARNESS)
@@ -988,12 +1066,18 @@ transaction_test: utilities/transactions/transaction_test.o $(LIBOBJECTS) $(TEST
sst_dump: tools/sst_dump.o $(LIBOBJECTS)
$(AM_LINK)
repair_test: db/repair_test.o db/db_test_util.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
ldb_cmd_test: tools/ldb_cmd_test.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_LINK)
ldb: tools/ldb.o $(LIBOBJECTS)
$(AM_LINK)
iostats_context_test: util/iostats_context_test.o $(LIBOBJECTS) $(TESTHARNESS)
$(AM_V_CCLD)$(CXX) $^ $(EXEC_LDFLAGS) -o $@ $(LDFLAGS)
#-------------------------------------------------
# make install related stuff
INSTALL_PATH ?= /usr/local
@@ -1036,7 +1120,11 @@ install: install-static
# ---------------------------------------------------------------------------
JAVA_INCLUDE = -I$(JAVA_HOME)/include/ -I$(JAVA_HOME)/include/linux
ARCH := $(shell getconf LONG_BIT)
ifeq ($(PLATFORM), OS_SOLARIS)
ARCH := $(shell isainfo -b)
else
ARCH := $(shell getconf LONG_BIT)
endif
ROCKSDBJNILIB = librocksdbjni-linux$(ARCH).so
ROCKSDB_JAR = rocksdbjni-$(ROCKSDB_MAJOR).$(ROCKSDB_MINOR).$(ROCKSDB_PATCH)-linux$(ARCH).jar
ROCKSDB_JAR_ALL = rocksdbjni-$(ROCKSDB_MAJOR).$(ROCKSDB_MINOR).$(ROCKSDB_PATCH).jar
@@ -1044,14 +1132,19 @@ ROCKSDB_JAVADOCS_JAR = rocksdbjni-$(ROCKSDB_MAJOR).$(ROCKSDB_MINOR).$(ROCKSDB_PA
ROCKSDB_SOURCES_JAR = rocksdbjni-$(ROCKSDB_MAJOR).$(ROCKSDB_MINOR).$(ROCKSDB_PATCH)-sources.jar
ifeq ($(PLATFORM), OS_MACOSX)
ROCKSDBJNILIB = librocksdbjni-osx.jnilib
ROCKSDB_JAR = rocksdbjni-$(ROCKSDB_MAJOR).$(ROCKSDB_MINOR).$(ROCKSDB_PATCH)-osx.jar
ROCKSDBJNILIB = librocksdbjni-osx.jnilib
ROCKSDB_JAR = rocksdbjni-$(ROCKSDB_MAJOR).$(ROCKSDB_MINOR).$(ROCKSDB_PATCH)-osx.jar
ifneq ("$(wildcard $(JAVA_HOME)/include/darwin)","")
JAVA_INCLUDE = -I$(JAVA_HOME)/include -I $(JAVA_HOME)/include/darwin
else
JAVA_INCLUDE = -I/System/Library/Frameworks/JavaVM.framework/Headers/
endif
endif
ifeq ($(PLATFORM), OS_SOLARIS)
ROCKSDBJNILIB = librocksdbjni-solaris$(ARCH).so
ROCKSDB_JAR = rocksdbjni-$(ROCKSDB_MAJOR).$(ROCKSDB_MINOR).$(ROCKSDB_PATCH)-solaris$(ARCH).jar
JAVA_INCLUDE = -I$(JAVA_HOME)/include/ -I$(JAVA_HOME)/include/solaris
endif
libz.a:
-rm -rf zlib-1.2.8
@@ -1145,11 +1238,10 @@ jtest: rocksdbjava
jdb_bench:
cd java;$(MAKE) db_bench;
commit-prereq:
$(MAKE) clean && $(MAKE) all check;
commit_prereq: build_tools/rocksdb-lego-determinator \
build_tools/precommit_checker.py
J=$(J) build_tools/precommit_checker.py unit unit_481 clang_unit tsan asan ubsan lite
$(MAKE) clean && $(MAKE) jclean && $(MAKE) rocksdbjava;
$(MAKE) clean && USE_CLANG=1 $(MAKE) all;
$(MAKE) clean && OPT=-DROCKSDB_LITE $(MAKE) static_lib;
xfunc:
for xftest in $(XFUNC_TESTS); do \
+11 -1
View File
@@ -45,4 +45,14 @@ Airbnb is using RocksDB as a storage engine for their personalized search servic
Pinterest's Object Retrieval System uses RocksDB for storage: https://www.youtube.com/watch?v=MtFEVEs_2Vo
## Smyte
[Smyte](https://www.smyte.com/) uses RocksDB as the storage layer for their core key-value storage, high-performance counters and time-windowed HyperLogLog services.
[Smyte](https://www.smyte.com/) uses RocksDB as the storage layer for their core key-value storage, high-performance counters and time-windowed HyperLogLog services.
## Rakuten Marketing
[Rakuten Marketing](https://marketing.rakuten.com/) uses RocksDB as the disk cache layer for the real-time bidding service in their Performance DSP.
## VWO, Wingify
[VWO's](https://vwo.com/) Smart Code checker and URL helper uses RocksDB to store all the URLs where VWO's Smart Code is installed.
## quasardb
[quasardb](https://www.quasardb.net) is a high-performance, distributed, transactional key-value database that integrates well with in-memory analytics engines such as Apache Spark.
quasardb uses a heavily tuned RocksDB as its persistence layer.
@@ -10,13 +10,184 @@ class FacebookArcanistConfiguration extends ArcanistConfiguration {
ArcanistBaseWorkflow $workflow,
$error_code) {
if ($command == 'diff' && !$workflow->isRawDiffSource()) {
$this->maybePushToJenkins($workflow);
$this->startTestsInJenkins($workflow);
$this->startTestsInSandcastle($workflow);
}
}
//////////////////////////////////////////////////////////////////////
/* Run tests in sandcastle */
function postURL($diffID, $url) {
$cmd = 'echo \'{"diff_id": "' . $diffID . '", '
. '"name":"click here for sandcastle tests for D' . $diffID . '", '
. '"link":"' . $url . '"}\' | '
. 'http_proxy=fwdproxy.any.facebook.com:8080 '
. 'https_proxy=fwdproxy.any.facebook.com:8080 arc call-conduit '
. 'differential.updateunitresults';
shell_exec($cmd);
}
function updateTestCommand($diffID, $test, $status) {
$cmd = 'echo \'{"diff_id": "' . $diffID . '", '
. '"name":"' . $test . '", '
. '"result":"' . $status . '"}\' | '
. 'http_proxy=fwdproxy.any.facebook.com:8080 '
. 'https_proxy=fwdproxy.any.facebook.com:8080 arc call-conduit '
. 'differential.updateunitresults';
return $cmd;
}
function updateTest($diffID, $test) {
shell_exec($this->updateTestCommand($diffID, $test, "waiting"));
}
function getSteps($diffID, $username, $test) {
$arcrc_content = exec("cat ~/.arcrc | gzip -f | base64 -w0");
// Sandcastle machines don't have arc setup. We copy the user certificate
// and authenticate using that in sandcastle
$setup = array(
"name" => "Setup arcrc",
"shell" => "echo " . $arcrc_content . " | base64 --decode"
. " | gzip -d > ~/.arcrc",
"user" => "root"
);
// arc demands certain permission on its config
$fix_permission = array(
"name" => "Fix environment",
"shell" => "chmod 600 ~/.arcrc",
"user" => "root"
);
// fbcode is a sub-repo. We cannot patch until we add it to ignore otherwise
// git thinks it is uncommited change
$fix_git_ignore = array(
"name" => "Fix git ignore",
"shell" => "echo fbcode >> .git/info/exclude",
"user" => "root"
);
// Patch the code (keep your fingures crossed)
$patch = array(
"name" => "Patch " . $diffID,
"shell" => "HTTPS_PROXY=fwdproxy:8080 arc --arcrc-file ~/.arcrc "
. "patch --diff " . $diffID,
"user" => "root"
);
// Clean up the user arc config we are using
$cleanup = array(
"name" => "Arc cleanup",
"shell" => "rm -f ~/.arcrc",
"user" => "root"
);
// Construct the steps in the order of execution
$steps[] = $setup;
$steps[] = $fix_permission;
$steps[] = $fix_git_ignore;
$steps[] = $patch;
// Run the actual command
$this->updateTest($diffID, $test);
$cmd = $this->updateTestCommand($diffID, $test, "running") . ";"
. "(./build_tools/precommit_checker.py " . $test
. "&& "
. $this->updateTestCommand($diffID, $test, "pass") . ")"
. "|| " . $this->updateTestCommand($diffID, $test, "fail")
. "; cat /tmp/precommit-check.log"
. "; for f in `ls t/log-*`; do echo \$f; cat \$f; done";
$run_test = array(
"name" => "Run " . $test,
"shell" => $cmd,
"user" => "root",
);
$steps[] = $run_test;
$steps[] = $cleanup;
return $steps;
}
function startTestsInSandcastle($workflow) {
// extract information we need from workflow or CLI
$diffID = $workflow->getDiffId();
$username = exec("whoami");
if ($diffID == null || $username == null) {
// there is no diff and we can't extract username
// we cannot schedule sandcasstle job
return;
}
if (strcmp(getenv("ROCKSDB_CHECK_ALL"), 1) == 0) {
// extract all tests from the CI definition
$output = file_get_contents("build_tools/rocksdb-lego-determinator");
preg_match_all('/[ ]{2}([a-zA-Z0-9_]+)[\)]{1}/', $output, $matches);
$tests = $matches[1];
} else {
// manually list of tests we want to run in sandcastle
$tests = array(
"unit", "unit_481", "clang_unit", "tsan", "asan", "lite", "valgrind"
);
}
// construct a job definition for each test and add it to the master plan
foreach ($tests as $test) {
$arg[] = array(
"name" => "RocksDB diff " . $diffID . " test " . $test,
"steps" => $this->getSteps($diffID, $username, $test)
);
}
// we cannot submit the parallel execution master plan to sandcastle
// we need supply the job plan as a determinator
// so we construct a small job that will spit out the master job plan
// which sandcastle will parse and execute
// Why compress ? Otherwise we run over the max string size.
$cmd = "echo " . base64_encode(json_encode($arg))
. " | gzip -f | base64 -w0";
$arg_encoded = shell_exec($cmd);
$command = array(
"name" => "Run diff " . $diffID . "for user " . $username,
"steps" => array()
);
$command["steps"][] = array(
"name" => "Generate determinator",
"shell" => "echo " . $arg_encoded . " | base64 --decode | gzip -d"
. " | base64 --decode",
"determinator" => true,
"user" => "root"
);
// submit to sandcastle
$url = 'https://interngraph.intern.facebook.com/sandcastle/generate?'
.'command=SandcastleUniversalCommand'
.'&vcs=rocksdb-git&revision=origin%2Fmaster&type=lego'
.'&user=krad&alias=rocksdb-precommit'
.'&command-args=' . urlencode(json_encode($command));
$cmd = 'https_proxy= HTTPS_PROXY= curl -s -k -F app=659387027470559 '
. '-F token=AeO_3f2Ya3TujjnxGD4 "' . $url . '"';
$output = shell_exec($cmd);
// extract sandcastle URL from the response
preg_match('/url": "(.+)"/', $output, $sandcastle_url);
echo "\nSandcastle URL: " . $sandcastle_url[1] . "\n";
// Ask phabricator to display it on the diff UI
$this->postURL($diffID, $sandcastle_url[1]);
}
//////////////////////////////////////////////////////////////////////
/* Send off builds to jenkins */
function maybePushToJenkins($workflow) {
function startTestsInJenkins($workflow) {
$diffID = $workflow->getDiffID();
if ($diffID === null) {
return;
@@ -31,5 +202,4 @@ class FacebookArcanistConfiguration extends ArcanistConfiguration {
."buildWithParameters?token=AUTH&DIFF_ID=$diffID";
system("curl --noproxy '*' \"$url\" > /dev/null 2>&1");
}
}
+1 -1
View File
@@ -1,5 +1,5 @@
#!/usr/bin/python
# Copyright (c) 2013, Facebook, Inc. All rights reserved.
# Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
# This source code is licensed under the BSD-style license found in the
# LICENSE file in the root directory of this source tree. An additional grant
# of patent rights can be found in the PATENTS file in the same directory.
+22 -17
View File
@@ -52,12 +52,7 @@ if [ -z "$ROCKSDB_NO_FBCODE" -a -d /mnt/gvfs/third-party ]; then
FBCODE_BUILD="true"
# If we're compiling with TSAN we need pic build
PIC_BUILD=$COMPILE_WITH_TSAN
if [ -z "$ROCKSDB_FBCODE_BUILD_WITH_481" ]; then
source "$PWD/build_tools/fbcode_config.sh"
else
# we need this to build with MySQL. Don't use for other purposes.
source "$PWD/build_tools/fbcode_config4.8.1.sh"
fi
source "$PWD/build_tools/fbcode_config.sh"
fi
# Delete existing output, if it exists
@@ -189,17 +184,19 @@ if [ "$CROSS_COMPILE" = "true" -o "$FBCODE_BUILD" = "true" ]; then
# Also don't need any compilation tests if compiling on fbcode
true
else
# Test whether fallocate is available
$CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null <<EOF
#include <fcntl.h>
#include <linux/falloc.h>
int main() {
int fd = open("/dev/null", 0);
fallocate(fd, FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE, 0, 1024);
}
if ! test $ROCKSDB_DISABLE_FALLOCATE; then
# Test whether fallocate is available
$CXX $CFLAGS -x c++ - -o /dev/null 2>/dev/null <<EOF
#include <fcntl.h>
#include <linux/falloc.h>
int main() {
int fd = open("/dev/null", 0);
fallocate(fd, FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE, 0, 1024);
}
EOF
if [ "$?" = 0 ]; then
COMMON_FLAGS="$COMMON_FLAGS -DROCKSDB_FALLOCATE_PRESENT"
if [ "$?" = 0 ]; then
COMMON_FLAGS="$COMMON_FLAGS -DROCKSDB_FALLOCATE_PRESENT"
fi
fi
# Test whether Snappy library is installed
@@ -359,7 +356,15 @@ if test "$USE_SSE"; then
# if Intel SSE instruction set is supported, set USE_SSE=1
COMMON_FLAGS="$COMMON_FLAGS -msse -msse4.2 "
elif test -z "$PORTABLE"; then
COMMON_FLAGS="$COMMON_FLAGS -march=native "
if test -n "`echo $TARGET_ARCHITECTURE | grep ^ppc64`"; then
# Tune for this POWER processor, treating '+' models as base models
POWER=`LD_SHOW_AUXV=1 /bin/true | grep AT_PLATFORM | grep -E -o power[0-9]+`
COMMON_FLAGS="$COMMON_FLAGS -mcpu=$POWER -mtune=$POWER "
elif test -n "`echo $TARGET_ARCHITECTURE | grep ^s390x`"; then
COMMON_FLAGS="$COMMON_FLAGS -march=z10 "
else
COMMON_FLAGS="$COMMON_FLAGS -march=native "
fi
fi
PLATFORM_CCFLAGS="$PLATFORM_CCFLAGS $COMMON_FLAGS"
+19
View File
@@ -0,0 +1,19 @@
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
GCC_BASE=/mnt/gvfs/third-party2/gcc/7331085db891a2ef4a88a48a751d834e8d68f4cb/7.x/centos7-native/b2ef2b6
CLANG_BASE=/mnt/gvfs/third-party2/llvm-fb/963d9aeda70cc4779885b1277484fe7544a04e3e/9.0.0/platform007/9e92d53/
LIBGCC_BASE=/mnt/gvfs/third-party2/libgcc/6ace84e956873d53638c738b6f65f3f469cca74c/7.x/platform007/5620abc
GLIBC_BASE=/mnt/gvfs/third-party2/glibc/192b0f42d63dcf6210d6ceae387b49af049e6e0c/2.26/platform007/f259413
SNAPPY_BASE=/mnt/gvfs/third-party2/snappy/7f9bdaada18f59bc27ec2b0871eb8a6144343aef/1.1.3/platform007/ca4da3d
ZLIB_BASE=/mnt/gvfs/third-party2/zlib/2d9f0b9a4274cc21f61272a9e89bdb859bce8f1f/1.2.8/platform007/ca4da3d
BZIP2_BASE=/mnt/gvfs/third-party2/bzip2/dc49a21c5fceec6456a7a28a94dcd16690af1337/1.0.6/platform007/ca4da3d
LZ4_BASE=/mnt/gvfs/third-party2/lz4/0f607f8fc442ea7d6b876931b1898bb573d5e5da/1.9.1/platform007/ca4da3d
ZSTD_BASE=/mnt/gvfs/third-party2/zstd/ca22bc441a4eb709e9e0b1f9fec9750fed7b31c5/1.4.x/platform007/15a3614
GFLAGS_BASE=/mnt/gvfs/third-party2/gflags/0b9929d2588991c65a57168bf88aff2db87c5d48/2.2.0/platform007/ca4da3d
JEMALLOC_BASE=/mnt/gvfs/third-party2/jemalloc/c26f08f47ac35fc31da2633b7da92d6b863246eb/master/platform007/c26c002
NUMA_BASE=/mnt/gvfs/third-party2/numa/3f3fb57a5ccc5fd21c66416c0b83e0aa76a05376/2.0.11/platform007/ca4da3d
LIBUNWIND_BASE=/mnt/gvfs/third-party2/libunwind/40c73d874898b386a71847f1b99115d93822d11f/1.4/platform007/6f3e0a9
TBB_BASE=/mnt/gvfs/third-party2/tbb/4ce8e8dba77cdbd81b75d6f0c32fd7a1b76a11ec/2018_U5/platform007/ca4da3d
KERNEL_HEADERS_BASE=/mnt/gvfs/third-party2/kernel-headers/fb251ecd2f5ae16f8671f7014c246e52a748fe0b/fb/platform007/da39a3e
BINUTILS_BASE=/mnt/gvfs/third-party2/binutils/ab9f09bba370e7066cafd4eb59752db93f2e8312/2.29.1/platform007/15a3614
VALGRIND_BASE=/mnt/gvfs/third-party2/valgrind/d42d152a15636529b0861ec493927200ebebca8e/3.15.0/platform007/ca4da3d
LUA_BASE=/mnt/gvfs/third-party2/lua/f0cd714433206d5139df61659eb7b28b1dea6683/5.3.4/platform007/5007832
+16
View File
@@ -0,0 +1,16 @@
GCC_BASE=/mnt/gvfs/third-party2/gcc/ebc96bc2fb751b5a0300b8d91a95bdf24ac1d88b/4.8.1/centos6-native/cc6c9dc
CLANG_BASE=/mnt/gvfs/third-party2/clang/fc904e50a9266b9d7b98cae1993afa0c5aae1440/3.7.1/centos6-native/9d9ecb9/
LIBGCC_BASE=/mnt/gvfs/third-party2/libgcc/f97108c12512b3b0789ac4515d836bdb1eae1142/4.8.1/gcc-4.8.1-glibc-2.17/8aac7fc
GLIBC_BASE=/mnt/gvfs/third-party2/glibc/f5484f168c0e4d19823d41df052c5870c6e575a4/2.17/gcc-4.8.1-glibc-2.17/99df8fc
SNAPPY_BASE=/mnt/gvfs/third-party2/snappy/cbf6f1f209e5bd160bdc5d971744e039f36b1566/1.1.3/gcc-4.8.1-glibc-2.17/c3f970a
ZLIB_BASE=/mnt/gvfs/third-party2/zlib/6d39cb54708049f527e713ad19f2aadb9d3667e8/1.2.8/gcc-4.8.1-glibc-2.17/c3f970a
BZIP2_BASE=/mnt/gvfs/third-party2/bzip2/2ddd45f0853bfc8bb1c27f0f447236a1a26c338a/1.0.6/gcc-4.8.1-glibc-2.17/c3f970a
LZ4_BASE=/mnt/gvfs/third-party2/lz4/6858fac689e0f92e584224d91bdb0e39f6c8320d/r131/gcc-4.8.1-glibc-2.17/c3f970a
ZSTD_BASE=/mnt/gvfs/third-party2/zstd/cb6c4880fcb4fee471574ba6af63a3882155a16a/0.5.1/gcc-4.8.1-glibc-2.17/c3f970a
GFLAGS_BASE=/mnt/gvfs/third-party2/gflags/c7275a4ceae0aca0929e56964a31dafc53c1ee96/2.1.1/gcc-4.8.1-glibc-2.17/c3f970a
JEMALLOC_BASE=/mnt/gvfs/third-party2/jemalloc/d9acac8a9a2c9378eb696e22ffa8bd0e97d9206b/master/gcc-4.8.1-glibc-2.17/8d31e51
NUMA_BASE=/mnt/gvfs/third-party2/numa/ae54a5ed22cdabb1c6446dce4e8ffae5b4446d73/2.0.8/gcc-4.8.1-glibc-2.17/c3f970a
LIBUNWIND_BASE=/mnt/gvfs/third-party2/libunwind/41bfa3759ce52c071f5fd547ec9ecd2522929f0a/trunk/gcc-4.8.1-glibc-2.17/675d945
KERNEL_HEADERS_BASE=/mnt/gvfs/third-party2/kernel-headers/1a48835975c66d30e47770ec419758ed3b9ba010/3.10.62-62_fbk17_03959_ge29cc63/gcc-4.8.1-glibc-2.17/da39a3e
BINUTILS_BASE=/mnt/gvfs/third-party2/binutils/71454c53fffcb716a0beb9a90047aff7fb5c984a/2.26/centos6-native/da39a3e
VALGRIND_BASE=/mnt/gvfs/third-party2/valgrind/af85c56f424cd5edfc2c97588299b44ecdec96bb/3.8.1/gcc-4.8.1-glibc-2.17/c3f970a
+72 -53
View File
@@ -6,128 +6,147 @@
# Environment variables that change the behavior of this script:
# PIC_BUILD -- if true, it will only take pic versions of libraries from fbcode. libraries that don't have pic variant will not be included
BASEDIR=`dirname $BASH_SOURCE`
source "$BASEDIR/dependencies.sh"
CFLAGS=""
# location of libgcc
LIBGCC_BASE="/mnt/gvfs/third-party2/libgcc/0473c80518a10d6efcbe24c5eeca3fb4ec9b519c/4.9.x/gcc-4.9-glibc-2.20/e1a7e4e"
LIBGCC_INCLUDE="$LIBGCC_BASE/include"
LIBGCC_LIBS=" -L $LIBGCC_BASE/libs"
# libgcc
LIBGCC_INCLUDE="$LIBGCC_BASE/include/c++/7.3.0"
LIBGCC_LIBS=" -L $LIBGCC_BASE/lib"
# location of glibc
GLIBC_REV=7397bed99280af5d9543439cdb7d018af7542720
GLIBC_INCLUDE="/mnt/gvfs/third-party2/glibc/$GLIBC_REV/2.20/gcc-4.9-glibc-2.20/99df8fc/include"
GLIBC_LIBS=" -L /mnt/gvfs/third-party2/glibc/$GLIBC_REV/2.20/gcc-4.9-glibc-2.20/99df8fc/lib"
SNAPPY_INCLUDE=" -I /mnt/gvfs/third-party2/snappy/b0f269b3ca47770121aa159b99e1d8d2ab260e1f/1.0.3/gcc-4.9-glibc-2.20/c32916f/include/"
# glibc
GLIBC_INCLUDE="$GLIBC_BASE/include"
GLIBC_LIBS=" -L $GLIBC_BASE/lib"
# snappy
SNAPPY_INCLUDE=" -I $SNAPPY_BASE/include/"
if test -z $PIC_BUILD; then
SNAPPY_LIBS=" /mnt/gvfs/third-party2/snappy/b0f269b3ca47770121aa159b99e1d8d2ab260e1f/1.0.3/gcc-4.9-glibc-2.20/c32916f/lib/libsnappy.a"
SNAPPY_LIBS=" $SNAPPY_BASE/lib/libsnappy.a"
else
SNAPPY_LIBS=" /mnt/gvfs/third-party2/snappy/b0f269b3ca47770121aa159b99e1d8d2ab260e1f/1.0.3/gcc-4.9-glibc-2.20/c32916f/lib/libsnappy_pic.a"
SNAPPY_LIBS=" $SNAPPY_BASE/lib/libsnappy_pic.a"
fi
CFLAGS+=" -DSNAPPY"
if test -z $PIC_BUILD; then
# location of zlib headers and libraries
ZLIB_INCLUDE=" -I /mnt/gvfs/third-party2/zlib/feb983d9667f4cf5e9da07ce75abc824764b67a1/1.2.8/gcc-4.9-glibc-2.20/4230243/include/"
ZLIB_LIBS=" /mnt/gvfs/third-party2/zlib/feb983d9667f4cf5e9da07ce75abc824764b67a1/1.2.8/gcc-4.9-glibc-2.20/4230243/lib/libz.a"
ZLIB_INCLUDE=" -I $ZLIB_BASE/include/"
ZLIB_LIBS=" $ZLIB_BASE/lib/libz.a"
CFLAGS+=" -DZLIB"
# location of bzip headers and libraries
BZIP_INCLUDE=" -I /mnt/gvfs/third-party2/bzip2/af004cceebb2dfd173ca29933ea5915e727aad2f/1.0.6/gcc-4.9-glibc-2.20/4230243/include/"
BZIP_LIBS=" /mnt/gvfs/third-party2/bzip2/af004cceebb2dfd173ca29933ea5915e727aad2f/1.0.6/gcc-4.9-glibc-2.20/4230243/lib/libbz2.a"
BZIP_INCLUDE=" -I $BZIP2_BASE/include/"
BZIP_LIBS=" $BZIP2_BASE/lib/libbz2.a"
CFLAGS+=" -DBZIP2"
LZ4_INCLUDE=" -I /mnt/gvfs/third-party2/lz4/6858fac689e0f92e584224d91bdb0e39f6c8320d/r131/gcc-4.9-glibc-2.20/e9936bf/include/"
LZ4_LIBS=" /mnt/gvfs/third-party2/lz4/6858fac689e0f92e584224d91bdb0e39f6c8320d/r131/gcc-4.9-glibc-2.20/e9936bf/lib/liblz4.a"
LZ4_INCLUDE=" -I $LZ4_BASE/include/"
LZ4_LIBS=" $LZ4_BASE/lib/liblz4.a"
CFLAGS+=" -DLZ4"
ZSTD_REV=810b81b4705def5243e998b54701f3c504e4009e
ZSTD_INCLUDE=" -I /mnt/gvfs/third-party2/zstd/$ZSTD_REV/0.4.2/gcc-4.8.1-glibc-2.17/c3f970a/include"
ZSTD_LIBS=" /mnt/gvfs/third-party2/zstd/$ZSTD_REV/0.4.2/gcc-4.8.1-glibc-2.17/c3f970a/lib/libzstd.a"
CFLAGS+=" -DZSTD"
fi
ZSTD_INCLUDE=" -I $ZSTD_BASE/include/"
if test -z $PIC_BUILD; then
ZSTD_LIBS=" $ZSTD_BASE/lib/libzstd.a"
else
ZSTD_LIBS=" $ZSTD_BASE/lib/libzstd_pic.a"
fi
CFLAGS+=" -DZSTD"
# location of gflags headers and libraries
GFLAGS_INCLUDE=" -I /mnt/gvfs/third-party2/gflags/0fa60e2b88de3e469db6c482d6e6dac72f5d65f9/1.6/gcc-4.9-glibc-2.20/4230243/include/"
GFLAGS_INCLUDE=" -I $GFLAGS_BASE/include/"
if test -z $PIC_BUILD; then
GFLAGS_LIBS=" /mnt/gvfs/third-party2/gflags/0fa60e2b88de3e469db6c482d6e6dac72f5d65f9/1.6/gcc-4.9-glibc-2.20/4230243/lib/libgflags.a"
GFLAGS_LIBS=" $GFLAGS_BASE/lib/libgflags.a"
else
GFLAGS_LIBS=" /mnt/gvfs/third-party2/gflags/0fa60e2b88de3e469db6c482d6e6dac72f5d65f9/1.6/gcc-4.9-glibc-2.20/4230243/lib/libgflags_pic.a"
GFLAGS_LIBS=" $GFLAGS_BASE/lib/libgflags_pic.a"
fi
CFLAGS+=" -DGFLAGS=google"
CFLAGS+=" -DGFLAGS=gflags"
# location of jemalloc
JEMALLOC_INCLUDE=" -I /mnt/gvfs/third-party2/jemalloc/bcd68e5e419efa4e61b9486d6854564d6d75a0b5/3.6.0/gcc-4.9-glibc-2.20/2aafc78/include/"
JEMALLOC_LIB=" /mnt/gvfs/third-party2/jemalloc/bcd68e5e419efa4e61b9486d6854564d6d75a0b5/3.6.0/gcc-4.9-glibc-2.20/2aafc78/lib/libjemalloc.a"
JEMALLOC_INCLUDE=" -I $JEMALLOC_BASE/include/"
JEMALLOC_LIB=" $JEMALLOC_BASE/lib/libjemalloc.a"
if test -z $PIC_BUILD; then
# location of numa
NUMA_INCLUDE=" -I /mnt/gvfs/third-party2/numa/bbefc39ecbf31d0ca184168eb613ef8d397790ee/2.0.8/gcc-4.9-glibc-2.20/4230243/include/"
NUMA_LIB=" /mnt/gvfs/third-party2/numa/bbefc39ecbf31d0ca184168eb613ef8d397790ee/2.0.8/gcc-4.9-glibc-2.20/4230243/lib/libnuma.a"
NUMA_INCLUDE=" -I $NUMA_BASE/include/"
NUMA_LIB=" $NUMA_BASE/lib/libnuma.a"
CFLAGS+=" -DNUMA"
# location of libunwind
LIBUNWIND="/mnt/gvfs/third-party2/libunwind/1de3b75e0afedfe5585b231bbb340ec7a1542335/1.1/gcc-4.9-glibc-2.20/34235e8/lib/libunwind.a"
LIBUNWIND="$LIBUNWIND_BASE/lib/libunwind.a"
fi
# location of TBB
TBB_INCLUDE=" -isystem $TBB_BASE/include/"
if test -z $PIC_BUILD; then
TBB_LIBS="$TBB_BASE/lib/libtbb.a"
else
TBB_LIBS="$TBB_BASE/lib/libtbb_pic.a"
fi
CFLAGS+=" -DTBB"
# use Intel SSE support for checksum calculations
export USE_SSE=1
BINUTILS="/mnt/gvfs/third-party2/binutils/0b6ad0c88ddd903333a48ae8bff134efac468e4a/2.25/centos6-native/da39a3e/bin"
BINUTILS="$BINUTILS_BASE/bin"
AR="$BINUTILS/ar"
DEPS_INCLUDE="$SNAPPY_INCLUDE $ZLIB_INCLUDE $BZIP_INCLUDE $LZ4_INCLUDE $ZSTD_INCLUDE $GFLAGS_INCLUDE $NUMA_INCLUDE"
DEPS_INCLUDE="$SNAPPY_INCLUDE $ZLIB_INCLUDE $BZIP_INCLUDE $LZ4_INCLUDE $ZSTD_INCLUDE $GFLAGS_INCLUDE $NUMA_INCLUDE $TBB_INCLUDE"
GCC_BASE="/mnt/gvfs/third-party2/gcc/1c67a0b88f64d4d9ced0382d141c76aaa7d62fba/4.9.x/centos6-native/1317bc4"
STDLIBS="-L $GCC_BASE/lib64"
CLANG_BASE="/mnt/gvfs/third-party2/clang/d81444dd214df3d2466734de45bb264a0486acc3/dev"
CLANG_BIN="$CLANG_BASE/centos6-native/af4b1a0/bin"
CLANG_BIN="$CLANG_BASE/bin"
CLANG_LIB="$CLANG_BASE/lib"
CLANG_SRC="$CLANG_BASE/../../src"
CLANG_ANALYZER="$CLANG_BIN/clang++"
CLANG_SCAN_BUILD="$CLANG_BASE/src/clang/tools/scan-build/scan-build"
CLANG_SCAN_BUILD="$CLANG_SRC/llvm/tools/clang/tools/scan-build/bin/scan-build"
if [ -z "$USE_CLANG" ]; then
# gcc
CC="$GCC_BASE/bin/gcc"
CXX="$GCC_BASE/bin/g++"
CFLAGS+=" -B$BINUTILS/gold"
CFLAGS+=" -isystem $GLIBC_INCLUDE"
CFLAGS+=" -isystem $LIBGCC_INCLUDE"
CFLAGS+=" -isystem $GLIBC_INCLUDE"
JEMALLOC=1
else
# clang
CLANG_INCLUDE="$CLANG_BASE/gcc-4.9-glibc-2.20/74c386f/lib/clang/dev/include/"
# clang
CLANG_INCLUDE="$CLANG_LIB/clang/stable/include"
CC="$CLANG_BIN/clang"
CXX="$CLANG_BIN/clang++"
KERNEL_HEADERS_INCLUDE="/mnt/gvfs/third-party2/kernel-headers/ffd14f660a43c4b92717986b1bba66722ef089d0/3.2.18_70_fbk11_00129_gc8882d0/gcc-4.9-glibc-2.20/da39a3e/include"
KERNEL_HEADERS_INCLUDE="$KERNEL_HEADERS_BASE/include"
CFLAGS+=" -B$BINUTILS/gold -nostdinc -nostdlib"
CFLAGS+=" -isystem $LIBGCC_BASE/include/c++/4.9.x "
CFLAGS+=" -isystem $LIBGCC_BASE/include/c++/4.9.x/x86_64-facebook-linux "
CFLAGS+=" -isystem $LIBGCC_BASE/include/c++/7.x "
CFLAGS+=" -isystem $LIBGCC_BASE/include/c++/7.x/x86_64-facebook-linux "
CFLAGS+=" -isystem $GLIBC_INCLUDE"
CFLAGS+=" -isystem $LIBGCC_INCLUDE"
CFLAGS+=" -isystem $CLANG_INCLUDE"
CFLAGS+=" -isystem $KERNEL_HEADERS_INCLUDE/linux "
CFLAGS+=" -isystem $KERNEL_HEADERS_INCLUDE "
CFLAGS+=" -Wno-expansion-to-defined "
CXXFLAGS="-nostdinc++"
fi
CFLAGS+=" $DEPS_INCLUDE"
CFLAGS+=" -DROCKSDB_PLATFORM_POSIX -DROCKSDB_LIB_IO_POSIX -DROCKSDB_FALLOCATE_PRESENT -DROCKSDB_MALLOC_USABLE_SIZE"
CFLAGS+=" -DROCKSDB_PLATFORM_POSIX -DROCKSDB_LIB_IO_POSIX -DROCKSDB_FALLOCATE_PRESENT -DROCKSDB_MALLOC_USABLE_SIZE -DROCKSDB_RANGESYNC_PRESENT -DROCKSDB_SCHED_GETCPU_PRESENT -DROCKSDB_SUPPORT_THREAD_LOCAL -DHAVE_SSE42"
CXXFLAGS+=" $CFLAGS"
EXEC_LDFLAGS=" $SNAPPY_LIBS $ZLIB_LIBS $BZIP_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS $NUMA_LIB"
EXEC_LDFLAGS+=" -Wl,--dynamic-linker,/usr/local/fbcode/gcc-4.9-glibc-2.20/lib/ld.so"
EXEC_LDFLAGS=" $SNAPPY_LIBS $ZLIB_LIBS $BZIP_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS $NUMA_LIB $TBB_LIBS"
EXEC_LDFLAGS+=" -B$BINUTILS/gold"
EXEC_LDFLAGS+=" -Wl,--dynamic-linker,/usr/local/fbcode/platform007/lib/ld.so"
EXEC_LDFLAGS+=" $LIBUNWIND"
EXEC_LDFLAGS+=" -Wl,-rpath=/usr/local/fbcode/gcc-4.9-glibc-2.20/lib"
EXEC_LDFLAGS+=" -Wl,-rpath=/usr/local/fbcode/platform007/lib"
# required by libtbb
EXEC_LDFLAGS+=" -ldl"
PLATFORM_LDFLAGS="$LIBGCC_LIBS $GLIBC_LIBS $STDLIBS -lgcc -lstdc++"
EXEC_LDFLAGS_SHARED="$SNAPPY_LIBS $ZLIB_LIBS $BZIP_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS"
EXEC_LDFLAGS_SHARED="$SNAPPY_LIBS $ZLIB_LIBS $BZIP_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS $TBB_LIBS"
VALGRIND_VER="/mnt/gvfs/third-party2/valgrind/6c45ef049cbf11c2df593addb712cd891049e737/3.10.0/gcc-4.9-glibc-2.20/4230243/bin/"
VALGRIND_VER="$VALGRIND_BASE/bin/"
export CC CXX AR CFLAGS CXXFLAGS EXEC_LDFLAGS EXEC_LDFLAGS_SHARED VALGRIND_VER JEMALLOC_LIB JEMALLOC_INCLUDE CLANG_ANALYZER CLANG_SCAN_BUILD
export CC CXX AR CFLAGS CXXFLAGS EXEC_LDFLAGS EXEC_LDFLAGS_SHARED VALGRIND_VER JEMALLOC_LIB JEMALLOC_INCLUDE CLANG_ANALYZER CLANG_SCAN_BUILD
+35 -44
View File
@@ -4,85 +4,77 @@
# fbcode settings. It uses the latest g++ compiler and also
# uses jemalloc
BASEDIR=`dirname $BASH_SOURCE`
source "$BASEDIR/dependencies_4.8.1.sh"
# location of libgcc
LIBGCC_BASE="/mnt/gvfs/third-party2/libgcc/d00277f4559e261ed0a81f30f23c0ce5564e359e/4.8.1/gcc-4.8.1-glibc-2.17/8aac7fc"
LIBGCC_INCLUDE="$LIBGCC_BASE/include"
LIBGCC_LIBS=" -L $LIBGCC_BASE/libs"
# location of glibc
GLIBC_REV=0600c95b31226b5e535614c590677d87c62d8016
GLIBC_INCLUDE="/mnt/gvfs/third-party2/glibc/$GLIBC_REV/2.17/gcc-4.8.1-glibc-2.17/99df8fc/include"
GLIBC_LIBS=" -L /mnt/gvfs/third-party2/glibc/$GLIBC_REV/2.17/gcc-4.8.1-glibc-2.17/99df8fc/lib"
GLIBC_INCLUDE="$GLIBC_BASE/include"
GLIBC_LIBS=" -L $GLIBC_BASE/lib"
# location of snappy headers and libraries
SNAPPY_REV=cbf6f1f209e5bd160bdc5d971744e039f36b1566
SNAPPY_INCLUDE=" -I /mnt/gvfs/third-party2/snappy/$SNAPPY_REV/1.1.3/gcc-4.8.1-glibc-2.17/c3f970a/include"
SNAPPY_LIBS=" /mnt/gvfs/third-party2/snappy/$SNAPPY_REV/1.1.3/gcc-4.8.1-glibc-2.17/c3f970a/lib/libsnappy.a"
SNAPPY_INCLUDE=" -I $SNAPPY_BASE/include"
SNAPPY_LIBS=" $SNAPPY_BASE/lib/libsnappy.a"
# location of zlib headers and libraries
ZLIB_REV=6d39cb54708049f527e713ad19f2aadb9d3667e8
ZLIB_INCLUDE=" -I /mnt/gvfs/third-party2/zlib/$ZLIB_REV/1.2.8/gcc-4.8.1-glibc-2.17/c3f970a/include"
ZLIB_LIBS=" /mnt/gvfs/third-party2/zlib/$ZLIB_REV/1.2.8/gcc-4.8.1-glibc-2.17/c3f970a/lib/libz.a"
ZLIB_INCLUDE=" -I $ZLIB_BASE/include"
ZLIB_LIBS=" $ZLIB_BASE/lib/libz.a"
# location of bzip headers and libraries
BZIP_REV=d6c789bfc2ec4c51a63d66df2878926b8158cde8
BZIP_INCLUDE=" -I /mnt/gvfs/third-party2/bzip2/$BZIP_REV/1.0.6/gcc-4.8.1-glibc-2.17/c3f970a/include/"
BZIP_LIBS=" /mnt/gvfs/third-party2/bzip2/$BZIP_REV/1.0.6/gcc-4.8.1-glibc-2.17/c3f970a/lib/libbz2.a"
BZIP2_INCLUDE=" -I $BZIP2_BASE/include/"
BZIP2_LIBS=" $BZIP2_BASE/lib/libbz2.a"
LZ4_REV=6858fac689e0f92e584224d91bdb0e39f6c8320d
LZ4_INCLUDE=" -I /mnt/gvfs/third-party2/lz4/$LZ4_REV/r131/gcc-4.8.1-glibc-2.17/c3f970a/include"
LZ4_LIBS=" /mnt/gvfs/third-party2/lz4/$LZ4_REV/r131/gcc-4.8.1-glibc-2.17/c3f970a/lib/liblz4.a"
LZ4_INCLUDE=" -I $LZ4_BASE/include"
LZ4_LIBS=" $LZ4_BASE/lib/liblz4.a"
ZSTD_REV=810b81b4705def5243e998b54701f3c504e4009e
ZSTD_INCLUDE=" -I /mnt/gvfs/third-party2/zstd/$ZSTD_REV/0.4.2/gcc-4.8.1-glibc-2.17/c3f970a/include"
ZSTD_LIBS=" /mnt/gvfs/third-party2/zstd/$ZSTD_REV/0.4.2/gcc-4.8.1-glibc-2.17/c3f970a/lib/libzstd.a"
ZSTD_INCLUDE=" -I $ZSTD_BASE/include"
ZSTD_LIBS=" $ZSTD_BASE/lib/libzstd.a"
# location of gflags headers and libraries
GFLAGS_REV=c7275a4ceae0aca0929e56964a31dafc53c1ee96
GFLAGS_INCLUDE=" -I /mnt/gvfs/third-party2/gflags/$GFLAGS_REV/2.1.1/gcc-4.8.1-glibc-2.17/c3f970a/include/"
GFLAGS_LIBS=" /mnt/gvfs/third-party2/gflags/$GFLAGS_REV/2.1.1/gcc-4.8.1-glibc-2.17/c3f970a/lib/libgflags.a"
GFLAGS_INCLUDE=" -I $GFLAGS_BASE/include/"
GFLAGS_LIBS=" $GFLAGS_BASE/lib/libgflags.a"
# location of jemalloc
JEMALLOC_REV=c370265e58c4b6602e798df23335a1e9913dae52
JEMALLOC_INCLUDE=" -I /mnt/gvfs/third-party2/jemalloc/$JEMALLOC_REV/4.0.3/gcc-4.8.1-glibc-2.17/8d31e51/include"
JEMALLOC_LIB="/mnt/gvfs/third-party2/jemalloc/$JEMALLOC_REV/4.0.3/gcc-4.8.1-glibc-2.17/8d31e51/lib/libjemalloc.a"
JEMALLOC_INCLUDE=" -I $JEMALLOC_BASE/include"
JEMALLOC_LIB="$JEMALLOC_BASE/lib/libjemalloc.a"
# location of numa
NUMA_REV=ae54a5ed22cdabb1c6446dce4e8ffae5b4446d73
NUMA_INCLUDE=" -I /mnt/gvfs/third-party2/numa/$NUMA_REV/2.0.8/gcc-4.8.1-glibc-2.17/c3f970a/include/"
NUMA_LIB=" /mnt/gvfs/third-party2/numa/$NUMA_REV/2.0.8/gcc-4.8.1-glibc-2.17/c3f970a/lib/libnuma.a"
NUMA_INCLUDE=" -I $NUMA_BASE/include/"
NUMA_LIB=" $NUMA_BASE/lib/libnuma.a"
# location of libunwind
LIBUNWIND_REV=121f1a75c4414683aea8c70b761bfaf187f7c1a3
LIBUNWIND="/mnt/gvfs/third-party2/libunwind/$LIBUNWIND_REV/trunk/gcc-4.8.1-glibc-2.17/675d945/lib/libunwind.a"
LIBUNWIND="$LIBUNWIND_BASE/lib/libunwind.a"
# use Intel SSE support for checksum calculations
export USE_SSE=1
BINUTILS="/mnt/gvfs/third-party2/binutils/75670d0d8ef4891fd1ec2a7513ef01cd002c823b/2.25/centos6-native/da39a3e/bin"
BINUTILS="$BINUTILS_BASE/bin"
AR="$BINUTILS/ar"
DEPS_INCLUDE="$SNAPPY_INCLUDE $ZLIB_INCLUDE $BZIP_INCLUDE $LZ4_INCLUDE $ZSTD_INCLUDE $GFLAGS_INCLUDE $NUMA_INCLUDE"
DEPS_INCLUDE="$SNAPPY_INCLUDE $ZLIB_INCLUDE $BZIP2_INCLUDE $LZ4_INCLUDE $ZSTD_INCLUDE $GFLAGS_INCLUDE $NUMA_INCLUDE"
GCC_BASE="/mnt/gvfs/third-party2/gcc/c0064002d2609ab649603f769f0bd110bbe48029/4.8.1/centos6-native/cc6c9dc"
STDLIBS="-L $GCC_BASE/lib64"
if [ -z "$USE_CLANG" ]; then
# gcc
CC="$GCC_BASE/bin/gcc"
CXX="$GCC_BASE/bin/g++"
CFLAGS="-B$BINUTILS/gold -m64 -mtune=generic"
CFLAGS+=" -isystem $GLIBC_INCLUDE"
CFLAGS+=" -isystem $LIBGCC_INCLUDE"
else
# clang
CLANG_BASE="/mnt/gvfs/third-party2/clang/ab054e9a490a8fd4537c0b6ec56e5c91c0f81c91/3.7"
CLANG_INCLUDE="$CLANG_BASE/gcc-4.8.1-glibc-2.17/ee9b060/lib/clang/3.7/include"
CC="$CLANG_BASE/centos6-native/b2feaee/bin/clang"
CXX="$CLANG_BASE/centos6-native/b2feaee/bin/clang++"
# clang
CLANG_BIN="$CLANG_BASE/bin"
CLANG_LIB="$CLANG_BASE/lib"
CLANG_INCLUDE="$CLANG_LIB/clang/*/include"
CC="$CLANG_BIN/clang"
CXX="$CLANG_BIN/clang++"
KERNEL_HEADERS_INCLUDE="/mnt/gvfs/third-party2/kernel-headers/1a48835975c66d30e47770ec419758ed3b9ba010/3.10.62-62_fbk17_03959_ge29cc63/gcc-4.8.1-glibc-2.17/da39a3e/include/"
KERNEL_HEADERS_INCLUDE="$KERNEL_HEADERS_BASE/include/"
CFLAGS="-B$BINUTILS/gold -nostdinc -nostdlib"
CFLAGS+=" -isystem $LIBGCC_BASE/include/c++/4.8.1 "
@@ -100,16 +92,15 @@ CFLAGS+=" -DROCKSDB_PLATFORM_POSIX -DROCKSDB_LIB_IO_POSIX -DROCKSDB_FALLOCATE_PR
CFLAGS+=" -DSNAPPY -DGFLAGS=google -DZLIB -DBZIP2 -DLZ4 -DZSTD -DNUMA"
CXXFLAGS+=" $CFLAGS"
EXEC_LDFLAGS=" $SNAPPY_LIBS $ZLIB_LIBS $BZIP_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS $NUMA_LIB"
EXEC_LDFLAGS=" $SNAPPY_LIBS $ZLIB_LIBS $BZIP2_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS $NUMA_LIB"
EXEC_LDFLAGS+=" -Wl,--dynamic-linker,/usr/local/fbcode/gcc-4.8.1-glibc-2.17/lib/ld.so"
EXEC_LDFLAGS+=" $LIBUNWIND"
EXEC_LDFLAGS+=" -Wl,-rpath=/usr/local/fbcode/gcc-4.8.1-glibc-2.17/lib"
PLATFORM_LDFLAGS="$LIBGCC_LIBS $GLIBC_LIBS $STDLIBS -lgcc -lstdc++"
EXEC_LDFLAGS_SHARED="$SNAPPY_LIBS $ZLIB_LIBS $BZIP_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS"
EXEC_LDFLAGS_SHARED="$SNAPPY_LIBS $ZLIB_LIBS $BZIP2_LIBS $LZ4_LIBS $ZSTD_LIBS $GFLAGS_LIBS"
VALGRIND_REV=af85c56f424cd5edfc2c97588299b44ecdec96bb
VALGRIND_VER="/mnt/gvfs/third-party2/valgrind/$VALGRIND_REV/3.8.1/gcc-4.8.1-glibc-2.17/c3f970a/bin/"
VALGRIND_VER="$VALGRIND_BASE/bin/"
export CC CXX AR CFLAGS CXXFLAGS EXEC_LDFLAGS EXEC_LDFLAGS_SHARED VALGRIND_VER JEMALLOC_LIB JEMALLOC_INCLUDE
+2 -2
View File
@@ -1,5 +1,5 @@
#!/bin/bash
# Copyright (c) 2013, Facebook, Inc. All rights reserved.
# Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
# This source code is licensed under the BSD-style license found in the
# LICENSE file in the root directory of this source tree. An additional grant
# of patent rights can be found in the PATENTS file in the same directory.
@@ -10,7 +10,7 @@ then
GIT="git"
fi
# Print out the colored progress info so that it can be brainlessly
# Print out the colored progress info so that it can be brainlessly
# distinguished by users.
function title() {
echo -e "\033[1;32m$*\033[0m"
+198
View File
@@ -0,0 +1,198 @@
#!/usr/local/fbcode/gcc-4.8.1-glibc-2.17-fb/bin/python2.7
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
from __future__ import unicode_literals
import argparse
import commands
import subprocess
import sys
import re
import os
import time
#
# Simple logger
#
class Log:
LOG_FILE = "/tmp/precommit-check.log"
def __init__(self):
self.filename = Log.LOG_FILE
self.f = open(self.filename, 'w+', 0)
def caption(self, str):
line = "\n##### %s #####\n" % str
if self.f:
self.f.write("%s \n" % line)
else:
print(line)
def error(self, str):
data = "\n\n##### ERROR ##### %s" % str
if self.f:
self.f.write("%s \n" % data)
else:
print(data)
def log(self, str):
if self.f:
self.f.write("%s \n" % str)
else:
print(str)
#
# Shell Environment
#
class Env(object):
def __init__(self, tests):
self.tests = tests
self.log = Log()
def shell(self, cmd, path=os.getcwd()):
if path:
os.chdir(path)
self.log.log("==== shell session ===========================")
self.log.log("%s> %s" % (path, cmd))
status = subprocess.call("cd %s; %s" % (path, cmd), shell=True,
stdout=self.log.f, stderr=self.log.f)
self.log.log("status = %s" % status)
self.log.log("============================================== \n\n")
return status
def GetOutput(self, cmd, path=os.getcwd()):
if path:
os.chdir(path)
self.log.log("==== shell session ===========================")
self.log.log("%s> %s" % (path, cmd))
status, out = commands.getstatusoutput(cmd)
self.log.log("status = %s" % status)
self.log.log("out = %s" % out)
self.log.log("============================================== \n\n")
return status, out
#
# Pre-commit checker
#
class PreCommitChecker(Env):
def __init__(self, tests):
Env.__init__(self, tests)
#
# Get commands for a given job from the determinator file
#
def get_commands(self, test):
status, out = self.GetOutput(
"build_tools/rocksdb-lego-determinator %s" % test, ".")
return status, out
#
# Run a specific CI job
#
def run_test(self, test):
self.log.caption("Running test %s locally" % test)
# get commands for the CI job determinator
status, cmds = self.get_commands(test)
if status != 0:
self.log.error("Error getting commands for test %s" % test)
return False
# Parse the JSON to extract the commands to run
cmds = re.findall("'shell':'([^\']*)'", cmds)
if len(cmds) == 0:
self.log.log("No commands found")
return False
# Run commands
for cmd in cmds:
# Replace J=<..> with the local environment variable
if "J" in os.environ:
cmd = cmd.replace("J=1", "J=%s" % os.environ["J"])
cmd = cmd.replace("make ", "make -j%s " % os.environ["J"])
# Run the command
status = self.shell(cmd, ".")
if status != 0:
self.log.error("Error running command %s for test %s"
% (cmd, test))
return False
return True
#
# Run specified CI jobs
#
def run_tests(self):
if not self.tests:
self.log.error("Invalid args. Please provide tests")
return False
self.print_separator()
self.print_row("TEST", "RESULT")
self.print_separator()
for test in self.tests:
start_time = time.time()
self.print_test(test)
result = self.run_test(test)
elapsed_min = (time.time() - start_time) / 60
if not result:
self.log.error("Error running test %s" % test)
self.print_result("FAIL (%dm)" % elapsed_min)
return False
self.print_result("PASS (%dm)" % elapsed_min)
self.print_separator()
return True
#
# Print a line
#
def print_separator(self):
print("".ljust(60, "-"))
#
# Print two colums
#
def print_row(self, c0, c1):
print("%s%s" % (c0.ljust(40), c1.ljust(20)))
def print_test(self, test):
print(test.ljust(40), end="")
sys.stdout.flush()
def print_result(self, result):
print(result.ljust(20))
#
# Main
#
parser = argparse.ArgumentParser(description='RocksDB pre-commit checker.')
# <test ....>
parser.add_argument('test', nargs='+',
help='CI test(s) to run. e.g: unit punit asan tsan ubsan')
print("Please follow log %s" % Log.LOG_FILE)
args = parser.parse_args()
checker = PreCommitChecker(args.test)
if not checker.run_tests():
print("Error running tests. Please check log file %s" % Log.LOG_FILE)
sys.exit(1)
sys.exit(0)
+51 -2
View File
@@ -67,6 +67,7 @@ ASAN="COMPILE_WITH_ASAN=1"
CLANG="USE_CLANG=1"
LITE="OPT=-DROCKSDB_LITE"
TSAN="COMPILE_WITH_TSAN=1"
UBSAN="COMPILE_WITH_UBSAN=1"
DISABLE_JEMALLOC="DISABLE_JEMALLOC=1"
PARSER="'parser':'egrep \'Failure|^#|Abort|Expected|Actual|GoogleTestFailure|^==\''"
@@ -376,6 +377,48 @@ ASAN_CRASH_TEST_COMMANDS="[
}
]"
#
# RocksDB test under undefined behavior sanitizer
#
UBSAN_TEST_COMMANDS="[
{
'name':'Rocksdb Unit Test under UBSAN',
'oncall':'$ONCALL',
'steps': [
$CLEANUP_ENV,
{
'name':'Test RocksDB debug under UBSAN',
'shell':'set -o pipefail && $SHM $UBSAN $DEBUG make J=1 ubsan_check',
'user':'root',
$PARSER
}
],
$REPORT
}
]"
#
# RocksDB crash testing under udnefined behavior sanitizer
#
UBSAN_CRASH_TEST_COMMANDS="[
{
'name':'Rocksdb crash test under UBSAN',
'oncall':'$ONCALL',
'timeout': 86400,
'steps': [
$CLEANUP_ENV,
{
'name':'Build and run RocksDB debug ubsan_crash_test',
'timeout': 86400,
'shell':'$SHM $DEBUG make J=1 ubsan_crash_test',
'user':'root',
$PARSER
},
],
$REPORT
}
]"
#
# RocksDB unit test under valgrind
#
@@ -462,9 +505,9 @@ run_format_compatible()
if [ -e "build_detect_platform" ]
then
sed "s/tcmalloc/nothingnothingnothing/g" build_detect_platform > $TEST_TMPDIR/temp_build_file
rm -rf build_detect_platform
rm -rf build_detect_platform
cp $TEST_TMPDIR/temp_build_file build_detect_platform
chmod +x build_detect_platform
chmod +x build_detect_platform
fi
make ldb -j32
@@ -632,6 +675,12 @@ case $1 in
asan_crash)
echo $ASAN_CRASH_TEST_COMMANDS
;;
ubsan)
echo $UBSAN_TEST_COMMANDS
;;
ubsan_crash)
echo $UBSAN_CRASH_TEST_COMMANDS
;;
valgrind)
echo $VALGRIND_TEST_COMMANDS
;;
+21 -9
View File
@@ -9,7 +9,7 @@ Param(
[string]$WorkFolder = "", # Direct tests to use that folder
[int]$Limit = -1, # -1 means run all otherwise limit for testing purposes
[string]$Exclude = "", # Expect a comma separated list, no spaces
[string]$Run = "db_test", # Run db_test|tests|testname1,testname2...
[string]$Run = "db_test", # Run db_test|db_test2|tests|testname1,testname2...
# Number of async tasks that would run concurrently. Recommend a number below 64.
# However, CPU utlization really depends on the storage media. Recommend ram based disk.
# a value of 1 will run everything serially
@@ -46,15 +46,25 @@ if($WorkFolder -eq "") {
$Env:TEST_TMPDIR = $WorkFolder
}
Write-Output "Root: $RootFolder, WorkFolder: $WorkFolder"
# Use JEMALLOC executables
if($EnableJE) {
$db_test = -Join ($BinariesFolder, "db_test_je.exe")
} else {
$db_test = -Join ($BinariesFolder, "db_test.exe")
if($Run -ceq "db_test" -or
$Run -ceq "db_test2" ) {
$file_name = $Run
if($EnableJE) {
$file_name += "_je"
}
$file_name += ".exe"
$db_test = -Join ($BinariesFolder, $file_name)
Write-Output "Binaries: $BinariesFolder db_test: $db_test"
}
Write-Output "Root: $RootFolder, WorkFolder: $WorkFolder"
Write-Output "Binaries: $BinariesFolder db_test: $db_test"
#Exclusions that we do not want to run
$ExcludeTests = New-Object System.Collections.Generic.HashSet[string]
@@ -153,7 +163,8 @@ function Discover-TestBinaries([string]$Pattern, $HashTable) {
$TestsToRun = [ordered]@{}
if($Run -ceq "db_test") {
if($Run -ceq "db_test" -or
$Run -ceq "db_test2") {
Normalize-DbTests -HashTable $TestsToRun
} elseif($Run -ceq "tests") {
if($EnableJE) {
@@ -225,7 +236,8 @@ function RunJobs($TestToLog, [int]$ConcurrencyVal, [bool]$AddForRerun)
Write-Host "Starting $k"
$log_path = ($TestToLog.$k)
if($Run -ceq "db_test") {
if($Run -ceq "db_test" -or
$Run -ceq "db_test2") {
$job = Start-Job -Name $k -ScriptBlock $InvokeTestCase -ArgumentList @($db_test,$k,$log_path)
} else {
[string]$Exe = -Join ($BinariesFolder, $k)
+127
View File
@@ -0,0 +1,127 @@
#!/bin/sh
#
# Update dependencies.sh file with the latest avaliable versions
BASEDIR=$(dirname $0)
OUTPUT=""
function log_variable()
{
echo "$1=${!1}" >> "$OUTPUT"
}
TP2_LATEST="/mnt/vol/engshare/fbcode/third-party2"
## $1 => lib name
## $2 => lib version (if not provided, will try to pick latest)
## $3 => platform (if not provided, will try to pick latest gcc)
##
## get_lib_base will set a variable named ${LIB_NAME}_BASE to the lib location
function get_lib_base()
{
local lib_name=$1
local lib_version=$2
local lib_platform=$3
local result="$TP2_LATEST/$lib_name/"
# Lib Version
if [ -z "$lib_version" ] || [ "$lib_version" = "LATEST" ]; then
# version is not provided, use latest
result=`ls -dr1v $result/*/ | head -n1`
else
result="$result/$lib_version/"
fi
# Lib Platform
if [ -z "$lib_platform" ]; then
# platform is not provided, use latest gcc
result=`ls -dr1v $result/gcc-*[^fb]/ | head -n1`
else
result="$result/$lib_platform/"
fi
result=`ls -1d $result/*/ | head -n1`
# lib_name => LIB_NAME_BASE
local __res_var=${lib_name^^}"_BASE"
__res_var=`echo $__res_var | tr - _`
# LIB_NAME_BASE=$result
eval $__res_var=`readlink -f $result`
log_variable $__res_var
}
###########################################################
# 4.9.x dependencies #
###########################################################
OUTPUT="$BASEDIR/dependencies.sh"
rm -f "$OUTPUT"
touch "$OUTPUT"
echo "Writing dependencies to $OUTPUT"
# Compilers locations
GCC_BASE=`readlink -f $TP2_LATEST/gcc/4.9.x/centos6-native/*/`
CLANG_BASE=`ls -d1 /mnt/gvfs/third-party2/clang/fc904e50a9266b9d7b98cae1993afa0c5aae1440/3.7.1/centos6-native/*/ | head -n1`
log_variable GCC_BASE
log_variable CLANG_BASE
# Libraries locations
get_lib_base libgcc 4.9.x
get_lib_base glibc 2.20
get_lib_base snappy LATEST
get_lib_base zlib LATEST
get_lib_base bzip2 LATEST
get_lib_base lz4 LATEST
get_lib_base zstd LATEST
get_lib_base gflags LATEST
get_lib_base jemalloc LATEST
get_lib_base numa LATEST
get_lib_base libunwind LATEST
get_lib_base kernel-headers LATEST
get_lib_base binutils LATEST centos6-native
get_lib_base valgrind LATEST
git diff $OUTPUT
###########################################################
# 4.8.1 dependencies #
###########################################################
OUTPUT="$BASEDIR/dependencies_4.8.1.sh"
rm -f "$OUTPUT"
touch "$OUTPUT"
echo "Writing 4.8.1 dependencies to $OUTPUT"
# Compilers locations
GCC_BASE=`readlink -f $TP2_LATEST/gcc/4.8.1/centos6-native/*/`
CLANG_BASE=`ls -d1 /mnt/gvfs/third-party2/clang/fc904e50a9266b9d7b98cae1993afa0c5aae1440/3.7.1/centos6-native/*/ | head -n1`
log_variable GCC_BASE
log_variable CLANG_BASE
# Libraries locations
get_lib_base libgcc 4.8.1 gcc-4.8.1-glibc-2.17
get_lib_base glibc 2.17 gcc-4.8.1-glibc-2.17
get_lib_base snappy LATEST gcc-4.8.1-glibc-2.17
get_lib_base zlib LATEST gcc-4.8.1-glibc-2.17
get_lib_base bzip2 LATEST gcc-4.8.1-glibc-2.17
get_lib_base lz4 LATEST gcc-4.8.1-glibc-2.17
get_lib_base zstd LATEST gcc-4.8.1-glibc-2.17
get_lib_base gflags LATEST gcc-4.8.1-glibc-2.17
get_lib_base jemalloc LATEST gcc-4.8.1-glibc-2.17
get_lib_base numa LATEST gcc-4.8.1-glibc-2.17
get_lib_base libunwind LATEST gcc-4.8.1-glibc-2.17
get_lib_base kernel-headers LATEST gcc-4.8.1-glibc-2.17
get_lib_base binutils LATEST centos6-native
get_lib_base valgrind 3.8.1 gcc-4.8.1-glibc-2.17
git diff $OUTPUT
@@ -1,9 +1,9 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
//
#include "util/auto_roll_logger.h"
#include "db/auto_roll_logger.h"
#include "util/mutexlock.h"
using namespace std;
@@ -12,7 +12,9 @@ namespace rocksdb {
// -- AutoRollLogger
Status AutoRollLogger::ResetLogger() {
TEST_SYNC_POINT("AutoRollLogger::ResetLogger:BeforeNewLogger");
status_ = env_->NewLogger(log_fname_, &logger_);
TEST_SYNC_POINT("AutoRollLogger::ResetLogger:AfterNewLogger");
if (!status_.ok()) {
return status_;
@@ -32,8 +34,16 @@ Status AutoRollLogger::ResetLogger() {
}
void AutoRollLogger::RollLogFile() {
std::string old_fname = OldInfoLogFileName(
dbname_, env_->NowMicros(), db_absolute_path_, db_log_dir_);
// This function is called when log is rotating. Two rotations
// can happen quickly (NowMicro returns same value). To not overwrite
// previous log file we increment by one micro second and try again.
uint64_t now = env_->NowMicros();
std::string old_fname;
do {
old_fname = OldInfoLogFileName(
dbname_, now, db_absolute_path_, db_log_dir_);
now++;
} while (env_->FileExists(old_fname).ok());
env_->RenameFile(log_fname_, old_fname);
}
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -13,6 +13,8 @@
#include "db/filename.h"
#include "port/port.h"
#include "port/util_logger.h"
#include "util/sync_point.h"
#include "util/mutexlock.h"
namespace rocksdb {
@@ -53,11 +55,26 @@ class AutoRollLogger : public Logger {
return status_;
}
size_t GetLogFileSize() const override { return logger_->GetLogFileSize(); }
size_t GetLogFileSize() const override {
std::shared_ptr<Logger> logger;
{
MutexLock l(&mutex_);
// pin down the current logger_ instance before releasing the mutex.
logger = logger_;
}
return logger->GetLogFileSize();
}
void Flush() override {
if (logger_) {
logger_->Flush();
std::shared_ptr<Logger> logger;
{
MutexLock l(&mutex_);
// pin down the current logger_ instance before releasing the mutex.
logger = logger_;
}
TEST_SYNC_POINT("AutoRollLogger::Flush:PinnedLogger");
if (logger) {
logger->Flush();
}
}
@@ -101,7 +118,7 @@ class AutoRollLogger : public Logger {
uint64_t ctime_;
uint64_t cached_now_access_count;
uint64_t call_NowMicros_every_N_records_;
port::Mutex mutex_;
mutable port::Mutex mutex_;
};
// Facade to craete logger automatically
@@ -1,17 +1,20 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
//
#include <string>
#include <thread>
#include <vector>
#include <cmath>
#include <iostream>
#include <fstream>
#include <iterator>
#include <algorithm>
#include "db/auto_roll_logger.h"
#include "port/port.h"
#include "util/sync_point.h"
#include "util/testharness.h"
#include "util/auto_roll_logger.h"
#include "rocksdb/db.h"
#include <sys/stat.h>
#include <errno.h>
@@ -260,7 +263,60 @@ TEST_F(AutoRollLoggerTest, CreateLoggerFromOptions) {
auto_roll_logger, options.log_file_time_to_roll,
kSampleMessage + ":CreateLoggerFromOptions - both");
}
#endif
TEST_F(AutoRollLoggerTest, LogFlushWhileRolling) {
DBOptions options;
shared_ptr<Logger> logger;
InitTestDb();
options.max_log_file_size = 1024 * 5;
ASSERT_OK(CreateLoggerFromOptions(kTestDir, options, &logger));
AutoRollLogger* auto_roll_logger =
dynamic_cast<AutoRollLogger*>(logger.get());
ASSERT_TRUE(auto_roll_logger);
std::thread flush_thread;
rocksdb::SyncPoint::GetInstance()->LoadDependency({
// Need to pin the old logger before beginning the roll, as rolling grabs
// the mutex, which would prevent us from accessing the old logger.
{"AutoRollLogger::Flush:PinnedLogger",
"AutoRollLoggerTest::LogFlushWhileRolling:PreRollAndPostThreadInit"},
// Need to finish the flush thread init before this callback because the
// callback accesses flush_thread.get_id() in order to apply certain sync
// points only to the flush thread.
{"AutoRollLoggerTest::LogFlushWhileRolling:PreRollAndPostThreadInit",
"AutoRollLoggerTest::LogFlushWhileRolling:FlushCallbackBegin"},
// Need to reset logger at this point in Flush() to exercise a race
// condition case, which is executing the flush with the pinned (old)
// logger after the roll has cut over to a new logger.
{"AutoRollLoggerTest::LogFlushWhileRolling:FlushCallback1",
"AutoRollLogger::ResetLogger:BeforeNewLogger"},
{"AutoRollLogger::ResetLogger:AfterNewLogger",
"AutoRollLoggerTest::LogFlushWhileRolling:FlushCallback2"},
});
rocksdb::SyncPoint::GetInstance()->SetCallBack(
"PosixLogger::Flush:BeginCallback", [&](void* arg) {
TEST_SYNC_POINT(
"AutoRollLoggerTest::LogFlushWhileRolling:FlushCallbackBegin");
if (std::this_thread::get_id() == flush_thread.get_id()) {
TEST_SYNC_POINT(
"AutoRollLoggerTest::LogFlushWhileRolling:FlushCallback1");
TEST_SYNC_POINT(
"AutoRollLoggerTest::LogFlushWhileRolling:FlushCallback2");
}
});
rocksdb::SyncPoint::GetInstance()->EnableProcessing();
flush_thread = std::thread([&]() { auto_roll_logger->Flush(); });
TEST_SYNC_POINT(
"AutoRollLoggerTest::LogFlushWhileRolling:PreRollAndPostThreadInit");
RollLogFileBySizeTest(auto_roll_logger, options.max_log_file_size,
kSampleMessage + ":LogFlushWhileRolling");
flush_thread.join();
rocksdb::SyncPoint::GetInstance()->DisableProcessing();
}
#endif // OS_WIN
TEST_F(AutoRollLoggerTest, InfoLogLevel) {
InitTestDb();
+18 -8
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -41,13 +41,17 @@ TableBuilder* NewTableBuilder(
const InternalKeyComparator& internal_comparator,
const std::vector<std::unique_ptr<IntTblPropCollectorFactory>>*
int_tbl_prop_collector_factories,
uint32_t column_family_id, WritableFileWriter* file,
const CompressionType compression_type,
uint32_t column_family_id, const std::string& column_family_name,
WritableFileWriter* file, const CompressionType compression_type,
const CompressionOptions& compression_opts, const bool skip_filters) {
assert((column_family_id ==
TablePropertiesCollectorFactory::Context::kUnknownColumnFamily) ==
column_family_name.empty());
return ioptions.table_factory->NewTableBuilder(
TableBuilderOptions(ioptions, internal_comparator,
int_tbl_prop_collector_factories, compression_type,
compression_opts, skip_filters),
compression_opts, skip_filters,
column_family_name),
column_family_id, file);
}
@@ -58,12 +62,16 @@ Status BuildTable(
const InternalKeyComparator& internal_comparator,
const std::vector<std::unique_ptr<IntTblPropCollectorFactory>>*
int_tbl_prop_collector_factories,
uint32_t column_family_id, std::vector<SequenceNumber> snapshots,
uint32_t column_family_id, const std::string& column_family_name,
std::vector<SequenceNumber> snapshots,
SequenceNumber earliest_write_conflict_snapshot,
const CompressionType compression,
const CompressionOptions& compression_opts, bool paranoid_file_checks,
InternalStats* internal_stats, const Env::IOPriority io_priority,
TableProperties* table_properties) {
TableProperties* table_properties, int level) {
assert((column_family_id ==
TablePropertiesCollectorFactory::Context::kUnknownColumnFamily) ==
column_family_name.empty());
// Reports the IOStats for flush for every following bytes.
const size_t kReportFlushIOStatsEvery = 1048576;
Status s;
@@ -87,7 +95,8 @@ Status BuildTable(
builder = NewTableBuilder(
ioptions, internal_comparator, int_tbl_prop_collector_factories,
column_family_id, file_writer.get(), compression, compression_opts);
column_family_id, column_family_name, file_writer.get(), compression,
compression_opts);
}
MergeHelper merge(env, internal_comparator.user_comparator(),
@@ -149,7 +158,8 @@ Status BuildTable(
ReadOptions(), env_options, internal_comparator, meta->fd, nullptr,
(internal_stats == nullptr) ? nullptr
: internal_stats->GetFileReadHist(0),
false));
false /* for_compaction */, nullptr /* arena */,
false /* skip_filter */, level));
s = it->status();
if (s.ok() && paranoid_file_checks) {
for (it->SeekToFirst(); it->Valid(); it->Next()) {
+12 -5
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -33,13 +33,16 @@ class WritableFileWriter;
class InternalStats;
class InternalIterator;
// @param column_family_name Name of the column family that is also identified
// by column_family_id, or empty string if unknown. It must outlive the
// TableBuilder returned by this function.
TableBuilder* NewTableBuilder(
const ImmutableCFOptions& options,
const InternalKeyComparator& internal_comparator,
const std::vector<std::unique_ptr<IntTblPropCollectorFactory>>*
int_tbl_prop_collector_factories,
uint32_t column_family_id, WritableFileWriter* file,
const CompressionType compression_type,
uint32_t column_family_id, const std::string& column_family_name,
WritableFileWriter* file, const CompressionType compression_type,
const CompressionOptions& compression_opts,
const bool skip_filters = false);
@@ -48,6 +51,9 @@ TableBuilder* NewTableBuilder(
// *meta will be filled with metadata about the generated table.
// If no data is present in *iter, meta->file_size will be set to
// zero, and no Table file will be produced.
//
// @param column_family_name Name of the column family that is also identified
// by column_family_id, or empty string if unknown.
extern Status BuildTable(
const std::string& dbname, Env* env, const ImmutableCFOptions& options,
const EnvOptions& env_options, TableCache* table_cache,
@@ -55,12 +61,13 @@ extern Status BuildTable(
const InternalKeyComparator& internal_comparator,
const std::vector<std::unique_ptr<IntTblPropCollectorFactory>>*
int_tbl_prop_collector_factories,
uint32_t column_family_id, std::vector<SequenceNumber> snapshots,
uint32_t column_family_id, const std::string& column_family_name,
std::vector<SequenceNumber> snapshots,
SequenceNumber earliest_write_conflict_snapshot,
const CompressionType compression,
const CompressionOptions& compression_opts, bool paranoid_file_checks,
InternalStats* internal_stats,
const Env::IOPriority io_priority = Env::IO_HIGH,
TableProperties* table_properties = nullptr);
TableProperties* table_properties = nullptr, int level = -1);
} // namespace rocksdb
+6 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -1288,6 +1288,11 @@ void rocksdb_block_based_options_set_cache_index_and_filter_blocks(
options->rep.cache_index_and_filter_blocks = v;
}
void rocksdb_block_based_options_set_pin_l0_filter_and_index_blocks_in_cache(
rocksdb_block_based_table_options_t* options, unsigned char v) {
options->rep.pin_l0_filter_and_index_blocks_in_cache = v;
}
void rocksdb_block_based_options_set_skip_table_builder_flush(
rocksdb_block_based_table_options_t* options, unsigned char v) {
options->rep.skip_table_builder_flush = v;
+73 -23
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -135,6 +135,10 @@ Status CheckConcurrentWritesSupported(const ColumnFamilyOptions& cf_options) {
"Delete filtering (filter_deletes) is not compatible with concurrent "
"memtable writes (allow_concurrent_memtable_writes)");
}
if (!cf_options.memtable_factory->IsInsertConcurrentlySupported()) {
return Status::InvalidArgument(
"Memtable doesn't concurrent writes (allow_concurrent_memtable_write)");
}
return Status::OK();
}
@@ -143,13 +147,10 @@ ColumnFamilyOptions SanitizeOptions(const DBOptions& db_options,
const ColumnFamilyOptions& src) {
ColumnFamilyOptions result = src;
result.comparator = icmp;
#ifdef OS_MACOSX
// TODO(icanadi) make write_buffer_size uint64_t instead of size_t
ClipToRange(&result.write_buffer_size, ((size_t)64) << 10, ((size_t)1) << 30);
#else
ClipToRange(&result.write_buffer_size,
((size_t)64) << 10, ((size_t)64) << 30);
#endif
size_t clamp_max = std::conditional<
sizeof(size_t) == 4, std::integral_constant<size_t, 0xffffffff>,
std::integral_constant<size_t, 64ull << 30>>::type::value;
ClipToRange(&result.write_buffer_size, ((size_t)64) << 10, clamp_max);
// if user sets arena_block_size, we trust user to use this value. Otherwise,
// calculate a proper value from writer_buffer_size;
if (result.arena_block_size <= 0) {
@@ -239,6 +240,17 @@ ColumnFamilyOptions SanitizeOptions(const DBOptions& db_options,
result.level0_slowdown_writes_trigger,
result.level0_file_num_compaction_trigger);
}
if (result.soft_pending_compaction_bytes_limit == 0) {
result.soft_pending_compaction_bytes_limit =
result.hard_pending_compaction_bytes_limit;
} else if (result.hard_pending_compaction_bytes_limit > 0 &&
result.soft_pending_compaction_bytes_limit >
result.hard_pending_compaction_bytes_limit) {
result.soft_pending_compaction_bytes_limit =
result.hard_pending_compaction_bytes_limit;
}
if (result.level_compaction_dynamic_level_bytes) {
if (result.compaction_style != kCompactionStyleLevel ||
db_options.db_paths.size() > 1U) {
@@ -513,6 +525,21 @@ std::unique_ptr<WriteControllerToken> SetupDelay(
}
return write_controller->GetDelayToken(write_rate);
}
int GetL0ThresholdSpeedupCompaction(int level0_file_num_compaction_trigger,
int level0_slowdown_writes_trigger) {
// SanitizeOptions() ensures it.
assert(level0_file_num_compaction_trigger <= level0_slowdown_writes_trigger);
// 1/4 of the way between L0 compaction trigger threshold and slowdown
// condition.
// Or twice as compaction trigger, if it is smaller.
return std::min(level0_file_num_compaction_trigger * 2,
level0_file_num_compaction_trigger +
(level0_slowdown_writes_trigger -
level0_file_num_compaction_trigger) /
4);
}
} // namespace
void ColumnFamilyData::RecalculateWriteStallConditions(
@@ -531,21 +558,6 @@ void ColumnFamilyData::RecalculateWriteStallConditions(
"(waiting for flush), max_write_buffer_number is set to %d",
name_.c_str(), imm()->NumNotFlushed(),
mutable_cf_options.max_write_buffer_number);
} else if (mutable_cf_options.max_write_buffer_number > 3 &&
imm()->NumNotFlushed() >=
mutable_cf_options.max_write_buffer_number - 1) {
write_controller_token_ =
SetupDelay(ioptions_.delayed_write_rate, write_controller,
compaction_needed_bytes, prev_compaction_needed_bytes_,
mutable_cf_options.disable_auto_compactions);
internal_stats_->AddCFStats(InternalStats::MEMTABLE_SLOWDOWN, 1);
Log(InfoLogLevel::WARN_LEVEL, ioptions_.info_log,
"[%s] Stalling writes because we have %d immutable memtables "
"(waiting for flush), max_write_buffer_number is set to %d "
"rate %" PRIu64,
name_.c_str(), imm()->NumNotFlushed(),
mutable_cf_options.max_write_buffer_number,
write_controller->delayed_write_rate());
} else if (vstorage->l0_delay_trigger_count() >=
mutable_cf_options.level0_stop_writes_trigger) {
write_controller_token_ = write_controller->GetStopToken();
@@ -567,6 +579,21 @@ void ColumnFamilyData::RecalculateWriteStallConditions(
"[%s] Stopping writes because of estimated pending compaction "
"bytes %" PRIu64,
name_.c_str(), compaction_needed_bytes);
} else if (mutable_cf_options.max_write_buffer_number > 3 &&
imm()->NumNotFlushed() >=
mutable_cf_options.max_write_buffer_number - 1) {
write_controller_token_ =
SetupDelay(ioptions_.delayed_write_rate, write_controller,
compaction_needed_bytes, prev_compaction_needed_bytes_,
mutable_cf_options.disable_auto_compactions);
internal_stats_->AddCFStats(InternalStats::MEMTABLE_SLOWDOWN, 1);
Log(InfoLogLevel::WARN_LEVEL, ioptions_.info_log,
"[%s] Stalling writes because we have %d immutable memtables "
"(waiting for flush), max_write_buffer_number is set to %d "
"rate %" PRIu64,
name_.c_str(), imm()->NumNotFlushed(),
mutable_cf_options.max_write_buffer_number,
write_controller->delayed_write_rate());
} else if (mutable_cf_options.level0_slowdown_writes_trigger >= 0 &&
vstorage->l0_delay_trigger_count() >=
mutable_cf_options.level0_slowdown_writes_trigger) {
@@ -598,6 +625,29 @@ void ColumnFamilyData::RecalculateWriteStallConditions(
"bytes %" PRIu64 " rate %" PRIu64,
name_.c_str(), vstorage->estimated_compaction_needed_bytes(),
write_controller->delayed_write_rate());
} else if (vstorage->l0_delay_trigger_count() >=
GetL0ThresholdSpeedupCompaction(
mutable_cf_options.level0_file_num_compaction_trigger,
mutable_cf_options.level0_slowdown_writes_trigger)) {
write_controller_token_ = write_controller->GetCompactionPressureToken();
Log(InfoLogLevel::WARN_LEVEL, ioptions_.info_log,
"[%s] Increasing compaction threads because we have %d level-0 "
"files ",
name_.c_str(), vstorage->l0_delay_trigger_count());
} else if (vstorage->estimated_compaction_needed_bytes() >=
mutable_cf_options.soft_pending_compaction_bytes_limit / 4) {
// Increase compaction threads if bytes needed for compaction exceeds
// 1/4 of threshold for slowing down.
// If soft pending compaction byte limit is not set, always speed up
// compaction.
write_controller_token_ = write_controller->GetCompactionPressureToken();
if (mutable_cf_options.soft_pending_compaction_bytes_limit > 0) {
Log(InfoLogLevel::WARN_LEVEL, ioptions_.info_log,
"[%s] Increasing compaction threads because of estimated pending "
"compaction "
"bytes %" PRIu64,
name_.c_str(), vstorage->estimated_compaction_needed_bytes());
}
} else {
write_controller_token_.reset();
}
+3 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -465,6 +465,8 @@ class ColumnFamilySet {
// Don't call while iterating over ColumnFamilySet
void FreeDeadColumnFamilies();
Cache* get_table_cache() { return table_cache_; }
private:
friend class ColumnFamilyData;
// helper function that gets called from cfd destructor
+198 -11
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -65,6 +65,7 @@ class ColumnFamilyTest : public testing::Test {
~ColumnFamilyTest() {
Close();
rocksdb::SyncPoint::GetInstance()->DisableProcessing();
Destroy();
delete env_;
}
@@ -1951,13 +1952,27 @@ TEST_F(ColumnFamilyTest, ReadDroppedColumnFamily) {
PutRandomData(1, kKeysNum, 100);
PutRandomData(2, kKeysNum, 100);
if (iter == 0) {
// Drop CF two
ASSERT_OK(db_->DropColumnFamily(handles_[2]));
} else {
// delete CF two
delete handles_[2];
handles_[2] = nullptr;
{
std::unique_ptr<Iterator> iterator(
db_->NewIterator(ReadOptions(), handles_[2]));
iterator->SeekToFirst();
if (iter == 0) {
// Drop CF two
ASSERT_OK(db_->DropColumnFamily(handles_[2]));
} else {
// delete CF two
delete handles_[2];
handles_[2] = nullptr;
}
// Make sure iterator created can still be used.
int count = 0;
for (; iterator->Valid(); iterator->Next()) {
ASSERT_OK(iterator->status());
++count;
}
ASSERT_OK(iterator->status());
ASSERT_EQ(count, kKeysNum);
}
// Add bunch more data to other CFs
@@ -1999,7 +2014,9 @@ TEST_F(ColumnFamilyTest, FlushAndDropRaceCondition) {
Reopen({options, options});
rocksdb::SyncPoint::GetInstance()->LoadDependency(
{{"VersionSet::LogAndApply::ColumnFamilyDrop:1",
{{"VersionSet::LogAndApply::ColumnFamilyDrop:0",
"FlushJob::WriteLevel0Table"},
{"VersionSet::LogAndApply::ColumnFamilyDrop:1",
"FlushJob::InstallResults"},
{"FlushJob::InstallResults",
"VersionSet::LogAndApply::ColumnFamilyDrop:2"}});
@@ -2045,7 +2062,6 @@ TEST_F(ColumnFamilyTest, FlushAndDropRaceCondition) {
Close();
Destroy();
rocksdb::SyncPoint::GetInstance()->DisableProcessing();
}
#ifndef ROCKSDB_LITE
@@ -2123,7 +2139,6 @@ TEST_F(ColumnFamilyTest, CreateAndDropRace) {
drop_cf_thread.join();
Close();
Destroy();
rocksdb::SyncPoint::GetInstance()->DisableProcessing();
for (auto* comparator : comparators) {
if (comparator) {
delete comparator;
@@ -2135,6 +2150,9 @@ TEST_F(ColumnFamilyTest, CreateAndDropRace) {
TEST_F(ColumnFamilyTest, WriteStallSingleColumnFamily) {
const uint64_t kBaseRate = 810000u;
db_options_.delayed_write_rate = kBaseRate;
db_options_.base_background_compactions = 2;
db_options_.max_background_compactions = 6;
Open({"default"});
ColumnFamilyData* cfd =
static_cast<ColumnFamilyHandleImpl*>(db_->DefaultColumnFamily())->cfd();
@@ -2160,6 +2178,7 @@ TEST_F(ColumnFamilyTest, WriteStallSingleColumnFamily) {
ASSERT_TRUE(!dbfull()->TEST_write_controler().IsStopped());
ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
ASSERT_EQ(kBaseRate, dbfull()->TEST_write_controler().delayed_write_rate());
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage->TEST_set_estimated_compaction_needed_bytes(400);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
@@ -2167,6 +2186,7 @@ TEST_F(ColumnFamilyTest, WriteStallSingleColumnFamily) {
ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
ASSERT_EQ(kBaseRate / 1.2,
dbfull()->TEST_write_controler().delayed_write_rate());
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage->TEST_set_estimated_compaction_needed_bytes(500);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
@@ -2222,6 +2242,7 @@ TEST_F(ColumnFamilyTest, WriteStallSingleColumnFamily) {
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_TRUE(dbfull()->TEST_write_controler().IsStopped());
ASSERT_TRUE(!dbfull()->TEST_write_controler().NeedsDelay());
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage->TEST_set_estimated_compaction_needed_bytes(3001);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
@@ -2246,6 +2267,7 @@ TEST_F(ColumnFamilyTest, WriteStallSingleColumnFamily) {
ASSERT_TRUE(dbfull()->TEST_write_controler().NeedsDelay());
ASSERT_EQ(kBaseRate / 1.2,
dbfull()->TEST_write_controler().delayed_write_rate());
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage->set_l0_delay_trigger_count(101);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
@@ -2318,6 +2340,73 @@ TEST_F(ColumnFamilyTest, WriteStallSingleColumnFamily) {
dbfull()->TEST_write_controler().delayed_write_rate());
}
TEST_F(ColumnFamilyTest, CompactionSpeedupSingleColumnFamily) {
db_options_.base_background_compactions = 2;
db_options_.max_background_compactions = 6;
Open({"default"});
ColumnFamilyData* cfd =
static_cast<ColumnFamilyHandleImpl*>(db_->DefaultColumnFamily())->cfd();
VersionStorageInfo* vstorage = cfd->current()->storage_info();
MutableCFOptions mutable_cf_options(
Options(db_options_, column_family_options_),
ImmutableCFOptions(Options(db_options_, column_family_options_)));
// Speed up threshold = min(4 * 2, 4 + (36 - 4)/4) = 8
mutable_cf_options.level0_file_num_compaction_trigger = 4;
mutable_cf_options.level0_slowdown_writes_trigger = 36;
mutable_cf_options.level0_stop_writes_trigger = 50;
// Speedup threshold = 200 / 4 = 50
mutable_cf_options.soft_pending_compaction_bytes_limit = 200;
mutable_cf_options.hard_pending_compaction_bytes_limit = 2000;
vstorage->TEST_set_estimated_compaction_needed_bytes(40);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(2, dbfull()->BGCompactionsAllowed());
vstorage->TEST_set_estimated_compaction_needed_bytes(50);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage->TEST_set_estimated_compaction_needed_bytes(300);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage->TEST_set_estimated_compaction_needed_bytes(45);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(2, dbfull()->BGCompactionsAllowed());
vstorage->set_l0_delay_trigger_count(7);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(2, dbfull()->BGCompactionsAllowed());
vstorage->set_l0_delay_trigger_count(9);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage->set_l0_delay_trigger_count(6);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(2, dbfull()->BGCompactionsAllowed());
// Speed up threshold = min(4 * 2, 4 + (12 - 4)/4) = 6
mutable_cf_options.level0_file_num_compaction_trigger = 4;
mutable_cf_options.level0_slowdown_writes_trigger = 16;
mutable_cf_options.level0_stop_writes_trigger = 30;
vstorage->set_l0_delay_trigger_count(5);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(2, dbfull()->BGCompactionsAllowed());
vstorage->set_l0_delay_trigger_count(7);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage->set_l0_delay_trigger_count(3);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(2, dbfull()->BGCompactionsAllowed());
}
TEST_F(ColumnFamilyTest, WriteStallTwoColumnFamilies) {
const uint64_t kBaseRate = 810000u;
db_options_.delayed_write_rate = kBaseRate;
@@ -2399,6 +2488,104 @@ TEST_F(ColumnFamilyTest, WriteStallTwoColumnFamilies) {
ASSERT_EQ(kBaseRate / 1.2,
dbfull()->TEST_write_controler().delayed_write_rate());
}
TEST_F(ColumnFamilyTest, CompactionSpeedupTwoColumnFamilies) {
db_options_.base_background_compactions = 2;
db_options_.max_background_compactions = 6;
column_family_options_.soft_pending_compaction_bytes_limit = 200;
column_family_options_.hard_pending_compaction_bytes_limit = 2000;
Open();
CreateColumnFamilies({"one"});
ColumnFamilyData* cfd =
static_cast<ColumnFamilyHandleImpl*>(db_->DefaultColumnFamily())->cfd();
VersionStorageInfo* vstorage = cfd->current()->storage_info();
ColumnFamilyData* cfd1 =
static_cast<ColumnFamilyHandleImpl*>(handles_[1])->cfd();
VersionStorageInfo* vstorage1 = cfd1->current()->storage_info();
MutableCFOptions mutable_cf_options(
Options(db_options_, column_family_options_),
ImmutableCFOptions(Options(db_options_, column_family_options_)));
// Speed up threshold = min(4 * 2, 4 + (36 - 4)/4) = 8
mutable_cf_options.level0_file_num_compaction_trigger = 4;
mutable_cf_options.level0_slowdown_writes_trigger = 36;
mutable_cf_options.level0_stop_writes_trigger = 30;
// Speedup threshold = 200 / 4 = 50
mutable_cf_options.soft_pending_compaction_bytes_limit = 200;
mutable_cf_options.hard_pending_compaction_bytes_limit = 2000;
MutableCFOptions mutable_cf_options1 = mutable_cf_options;
mutable_cf_options1.level0_slowdown_writes_trigger = 16;
vstorage->TEST_set_estimated_compaction_needed_bytes(40);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(2, dbfull()->BGCompactionsAllowed());
vstorage->TEST_set_estimated_compaction_needed_bytes(60);
cfd1->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(2, dbfull()->BGCompactionsAllowed());
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage1->TEST_set_estimated_compaction_needed_bytes(30);
cfd1->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage1->TEST_set_estimated_compaction_needed_bytes(70);
cfd1->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage->TEST_set_estimated_compaction_needed_bytes(20);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage1->TEST_set_estimated_compaction_needed_bytes(3);
cfd1->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(2, dbfull()->BGCompactionsAllowed());
vstorage->set_l0_delay_trigger_count(9);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage1->set_l0_delay_trigger_count(2);
cfd1->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(6, dbfull()->BGCompactionsAllowed());
vstorage->set_l0_delay_trigger_count(0);
cfd->RecalculateWriteStallConditions(mutable_cf_options);
ASSERT_EQ(2, dbfull()->BGCompactionsAllowed());
}
TEST_F(ColumnFamilyTest, LogSyncConflictFlush) {
Open();
CreateColumnFamiliesAndReopen({"one", "two"});
Put(0, "", "");
Put(1, "foo", "bar");
rocksdb::SyncPoint::GetInstance()->LoadDependency(
{{"DBImpl::SyncWAL:BeforeMarkLogsSynced:1",
"ColumnFamilyTest::LogSyncConflictFlush:1"},
{"ColumnFamilyTest::LogSyncConflictFlush:2",
"DBImpl::SyncWAL:BeforeMarkLogsSynced:2"}});
rocksdb::SyncPoint::GetInstance()->EnableProcessing();
std::thread thread([&] { db_->SyncWAL(); });
TEST_SYNC_POINT("ColumnFamilyTest::LogSyncConflictFlush:1");
Flush(1);
Put(1, "foo", "bar");
Flush(1);
TEST_SYNC_POINT("ColumnFamilyTest::LogSyncConflictFlush:2");
thread.join();
rocksdb::SyncPoint::GetInstance()->DisableProcessing();
Close();
}
} // namespace rocksdb
int main(int argc, char** argv) {
+59 -4
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2014, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -7,6 +7,7 @@
#include <mutex>
#include <string>
#include <thread>
#include <vector>
#include "rocksdb/db.h"
@@ -107,6 +108,7 @@ TEST_F(CompactFilesTest, L0ConflictsFiles) {
break;
}
}
rocksdb::SyncPoint::GetInstance()->DisableProcessing();
delete db;
}
@@ -141,9 +143,6 @@ TEST_F(CompactFilesTest, ObsoleteFiles) {
}
auto l0_files = collector->GetFlushedFiles();
CompactionOptions compact_opt;
compact_opt.compression = kNoCompression;
compact_opt.output_file_size_limit = kWriteBufferSize * 5;
ASSERT_OK(db->CompactFiles(CompactionOptions(), l0_files, 1));
// verify all compaction input files are deleted
@@ -153,6 +152,62 @@ TEST_F(CompactFilesTest, ObsoleteFiles) {
delete db;
}
TEST_F(CompactFilesTest, CapturingPendingFiles) {
Options options;
options.create_if_missing = true;
// Disable RocksDB background compaction.
options.compaction_style = kCompactionStyleNone;
// Always do full scans for obsolete files (needed to reproduce the issue).
options.delete_obsolete_files_period_micros = 0;
// Add listener.
FlushedFileCollector* collector = new FlushedFileCollector();
options.listeners.emplace_back(collector);
DB* db = nullptr;
DestroyDB(db_name_, options);
Status s = DB::Open(options, db_name_, &db);
assert(s.ok());
assert(db);
// Create 5 files.
for (int i = 0; i < 5; ++i) {
db->Put(WriteOptions(), "key" + ToString(i), "value");
db->Flush(FlushOptions());
}
auto l0_files = collector->GetFlushedFiles();
EXPECT_EQ(5, l0_files.size());
rocksdb::SyncPoint::GetInstance()->LoadDependency({
{"CompactFilesImpl:2", "CompactFilesTest.CapturingPendingFiles:0"},
{"CompactFilesTest.CapturingPendingFiles:1", "CompactFilesImpl:3"},
});
rocksdb::SyncPoint::GetInstance()->EnableProcessing();
// Start compacting files.
std::thread compaction_thread(
[&] { EXPECT_OK(db->CompactFiles(CompactionOptions(), l0_files, 1)); });
// In the meantime flush another file.
TEST_SYNC_POINT("CompactFilesTest.CapturingPendingFiles:0");
db->Put(WriteOptions(), "key5", "value");
db->Flush(FlushOptions());
TEST_SYNC_POINT("CompactFilesTest.CapturingPendingFiles:1");
compaction_thread.join();
rocksdb::SyncPoint::GetInstance()->DisableProcessing();
delete db;
// Make sure we can reopen the DB.
s = DB::Open(options, db_name_, &db);
ASSERT_TRUE(s.ok());
assert(db);
delete db;
}
} // namespace rocksdb
int main(int argc, char** argv) {
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2014, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2014, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+2 -31
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -162,9 +162,6 @@ Compaction::Compaction(VersionStorageInfo* vstorage,
deletion_compaction_(_deletion_compaction),
inputs_(std::move(_inputs)),
grandparents_(std::move(_grandparents)),
grandparent_index_(0),
seen_key_(false),
overlapped_bytes_(0),
score_(_score),
bottommost_level_(IsBottommostLevel(output_level_, vstorage, inputs_)),
is_full_compaction_(IsFullCompaction(vstorage, inputs_)),
@@ -290,32 +287,6 @@ bool Compaction::KeyNotExistsBeyondOutputLevel(
return true;
}
bool Compaction::ShouldStopBefore(const Slice& internal_key) {
// Scan to find earliest grandparent file that contains key.
const InternalKeyComparator* icmp = &cfd_->internal_comparator();
while (grandparent_index_ < grandparents_.size() &&
icmp->Compare(internal_key,
grandparents_[grandparent_index_]->largest.Encode()) > 0) {
if (seen_key_) {
overlapped_bytes_ += grandparents_[grandparent_index_]->fd.GetFileSize();
}
assert(grandparent_index_ + 1 >= grandparents_.size() ||
icmp->Compare(grandparents_[grandparent_index_]->largest.Encode(),
grandparents_[grandparent_index_+1]->smallest.Encode())
< 0);
grandparent_index_++;
}
seen_key_ = true;
if (overlapped_bytes_ > max_grandparent_overlap_bytes_) {
// Too much overlap for current output; start new output
overlapped_bytes_ = 0;
return true;
} else {
return false;
}
}
// Mark (or clear) each file that is being compacted
void Compaction::MarkFilesBeingCompacted(bool mark_as_compacted) {
for (size_t i = 0; i < num_input_levels(); i++) {
@@ -421,7 +392,7 @@ void Compaction::Summary(char* output, int len) {
snprintf(output + write, len - write, "]");
}
uint64_t Compaction::OutputFilePreallocationSize() {
uint64_t Compaction::OutputFilePreallocationSize() const {
uint64_t preallocation_size = 0;
if (cfd_->ioptions()->compaction_style == kCompactionStyleLevel ||
+20 -16
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -104,7 +104,7 @@ class Compaction {
}
// Returns the LevelFilesBrief of the specified compaction input level.
LevelFilesBrief* input_levels(size_t compaction_input_level) {
const LevelFilesBrief* input_levels(size_t compaction_input_level) const {
return &input_levels_[compaction_input_level];
}
@@ -132,12 +132,10 @@ class Compaction {
bool KeyNotExistsBeyondOutputLevel(const Slice& user_key,
std::vector<size_t>* level_ptrs) const;
// Returns true iff we should stop building the current output
// before processing "internal_key".
bool ShouldStopBefore(const Slice& internal_key);
// Clear all files to indicate that they are not being compacted
// Delete this compaction from the list of running compactions.
//
// Requirement: DB mutex held
void ReleaseCompactionFiles(Status status);
// Returns the summary of the compaction in "output" with maximum "len"
@@ -149,13 +147,13 @@ class Compaction {
double score() const { return score_; }
// Is this compaction creating a file in the bottom most level?
bool bottommost_level() { return bottommost_level_; }
bool bottommost_level() const { return bottommost_level_; }
// Does this compaction include all sst files?
bool is_full_compaction() { return is_full_compaction_; }
bool is_full_compaction() const { return is_full_compaction_; }
// Was this compaction triggered manually by the client?
bool is_manual_compaction() { return is_manual_compaction_; }
bool is_manual_compaction() const { return is_manual_compaction_; }
// Used when allow_trivial_move option is set in
// Universal compaction. If all the input files are
@@ -168,19 +166,21 @@ class Compaction {
// Used when allow_trivial_move option is set in
// Universal compaction. Returns true, if the input files
// are non-overlapping and can be trivially moved.
bool is_trivial_move() { return is_trivial_move_; }
bool is_trivial_move() const { return is_trivial_move_; }
// How many total levels are there?
int number_levels() const { return number_levels_; }
// Return the MutableCFOptions that should be used throughout the compaction
// procedure
const MutableCFOptions* mutable_cf_options() { return &mutable_cf_options_; }
const MutableCFOptions* mutable_cf_options() const {
return &mutable_cf_options_;
}
// Returns the size in bytes that the output file should be preallocated to.
// In level compaction, that is max_file_size_. In universal compaction, that
// is the sum of all input file sizes.
uint64_t OutputFilePreallocationSize();
uint64_t OutputFilePreallocationSize() const;
void SetInputVersion(Version* input_version);
@@ -223,6 +223,14 @@ class Compaction {
CompactionReason compaction_reason() { return compaction_reason_; }
const std::vector<FileMetaData*>& grandparents() const {
return grandparents_;
}
uint64_t max_grandparent_overlap_bytes() const {
return max_grandparent_overlap_bytes_;
}
private:
// mark (or clear) all files that are being compacted
void MarkFilesBeingCompacted(bool mark_as_compacted);
@@ -266,10 +274,6 @@ class Compaction {
// State used to check for number of of overlapping grandparent files
// (grandparent == "output_level_ + 1")
std::vector<FileMetaData*> grandparents_;
size_t grandparent_index_; // Index in grandparent_starts_
bool seen_key_; // Some output key has been seen
uint64_t overlapped_bytes_; // Bytes of overlap between current output
// and grandparent files
const double score_; // score that was used to pick this compaction.
// Is this compaction creating a file in the bottom most level?
+2 -2
View File
@@ -1,6 +1,6 @@
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file. See the AUTHORS file for names of contributors.
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -14,7 +14,7 @@ CompactionIterator::CompactionIterator(
InternalIterator* input, const Comparator* cmp, MergeHelper* merge_helper,
SequenceNumber last_sequence, std::vector<SequenceNumber>* snapshots,
SequenceNumber earliest_write_conflict_snapshot, Env* env,
bool expect_valid_internal_key, Compaction* compaction,
bool expect_valid_internal_key, const Compaction* compaction,
const CompactionFilter* compaction_filter, LogBuffer* log_buffer)
: input_(input),
cmp_(cmp),
+3 -3
View File
@@ -1,6 +1,6 @@
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file. See the AUTHORS file for names of contributors.
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -42,7 +42,7 @@ class CompactionIterator {
std::vector<SequenceNumber>* snapshots,
SequenceNumber earliest_write_conflict_snapshot, Env* env,
bool expect_valid_internal_key,
Compaction* compaction = nullptr,
const Compaction* compaction = nullptr,
const CompactionFilter* compaction_filter = nullptr,
LogBuffer* log_buffer = nullptr);
@@ -92,7 +92,7 @@ class CompactionIterator {
const SequenceNumber earliest_write_conflict_snapshot_;
Env* env_;
bool expect_valid_internal_key_;
Compaction* compaction_;
const Compaction* compaction_;
const CompactionFilter* compaction_filter_;
LogBuffer* log_buffer_;
bool bottommost_level_;
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+91 -22
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -51,6 +51,7 @@
#include "util/iostats_context_imp.h"
#include "util/log_buffer.h"
#include "util/logging.h"
#include "util/sst_file_manager_impl.h"
#include "util/mutexlock.h"
#include "util/perf_context_imp.h"
#include "util/stop_watch.h"
@@ -62,7 +63,7 @@ namespace rocksdb {
// Maintains state for each sub-compaction
struct CompactionJob::SubcompactionState {
Compaction* compaction;
const Compaction* compaction;
std::unique_ptr<CompactionIterator> c_iter;
// The boundaries of the key-range this compaction is interested in. No two
@@ -103,6 +104,14 @@ struct CompactionJob::SubcompactionState {
uint64_t num_output_records;
CompactionJobStats compaction_job_stats;
uint64_t approx_size;
// An index that used to speed up Compaction::ShouldStopBefore().
size_t grandparent_index = 0;
// The number of bytes overlapping between the current output and
// grandparent files used in Compaction::ShouldStopBefore().
uint64_t overlapped_bytes = 0;
// A flag determine whether the key has been seen in
// Compaction::ShouldStopBefore()
bool seen_key = false;
SubcompactionState(Compaction* c, Slice* _start, Slice* _end,
uint64_t size = 0)
@@ -114,7 +123,10 @@ struct CompactionJob::SubcompactionState {
total_bytes(0),
num_input_records(0),
num_output_records(0),
approx_size(size) {
approx_size(size),
grandparent_index(0),
overlapped_bytes(0),
seen_key(false) {
assert(compaction != nullptr);
}
@@ -133,6 +145,9 @@ struct CompactionJob::SubcompactionState {
num_output_records = std::move(o.num_output_records);
compaction_job_stats = std::move(o.compaction_job_stats);
approx_size = std::move(o.approx_size);
grandparent_index = std::move(o.grandparent_index);
overlapped_bytes = std::move(o.overlapped_bytes);
seen_key = std::move(o.seen_key);
return *this;
}
@@ -140,6 +155,38 @@ struct CompactionJob::SubcompactionState {
SubcompactionState(const SubcompactionState&) = delete;
SubcompactionState& operator=(const SubcompactionState&) = delete;
// Returns true iff we should stop building the current output
// before processing "internal_key".
bool ShouldStopBefore(const Slice& internal_key) {
const InternalKeyComparator* icmp =
&compaction->column_family_data()->internal_comparator();
const std::vector<FileMetaData*>& grandparents = compaction->grandparents();
// Scan to find earliest grandparent file that contains key.
while (grandparent_index < grandparents.size() &&
icmp->Compare(internal_key,
grandparents[grandparent_index]->largest.Encode()) >
0) {
if (seen_key) {
overlapped_bytes += grandparents[grandparent_index]->fd.GetFileSize();
}
assert(grandparent_index + 1 >= grandparents.size() ||
icmp->Compare(
grandparents[grandparent_index]->largest.Encode(),
grandparents[grandparent_index + 1]->smallest.Encode()) < 0);
grandparent_index++;
}
seen_key = true;
if (overlapped_bytes > compaction->max_grandparent_overlap_bytes()) {
// Too much overlap for current output; start new output
overlapped_bytes = 0;
return true;
}
return false;
}
};
// Maintains state for the entire compaction
@@ -211,6 +258,7 @@ CompactionJob::CompactionJob(
const EnvOptions& env_options, VersionSet* versions,
std::atomic<bool>* shutting_down, LogBuffer* log_buffer,
Directory* db_directory, Directory* output_directory, Statistics* stats,
InstrumentedMutex* db_mutex, Status* db_bg_error,
std::vector<SequenceNumber> existing_snapshots,
SequenceNumber earliest_write_conflict_snapshot,
std::shared_ptr<Cache> table_cache, EventLogger* event_logger,
@@ -230,6 +278,8 @@ CompactionJob::CompactionJob(
db_directory_(db_directory),
output_directory_(output_directory),
stats_(stats),
db_mutex_(db_mutex),
db_bg_error_(db_bg_error),
existing_snapshots_(std::move(existing_snapshots)),
earliest_write_conflict_snapshot_(earliest_write_conflict_snapshot),
table_cache_(std::move(table_cache)),
@@ -237,7 +287,9 @@ CompactionJob::CompactionJob(
paranoid_file_checks_(paranoid_file_checks),
measure_io_stats_(measure_io_stats) {
assert(log_buffer_ != nullptr);
ThreadStatusUtil::SetColumnFamily(compact_->compaction->column_family_data());
const auto* cfd = compact_->compaction->column_family_data();
ThreadStatusUtil::SetColumnFamily(cfd, cfd->ioptions()->env,
cfd->options()->enable_thread_tracking);
ThreadStatusUtil::SetThreadOperation(ThreadStatus::OP_COMPACTION);
ReportStartedCompaction(compaction);
}
@@ -249,8 +301,9 @@ CompactionJob::~CompactionJob() {
void CompactionJob::ReportStartedCompaction(
Compaction* compaction) {
ThreadStatusUtil::SetColumnFamily(
compact_->compaction->column_family_data());
const auto* cfd = compact_->compaction->column_family_data();
ThreadStatusUtil::SetColumnFamily(cfd, cfd->ioptions()->env,
cfd->options()->enable_thread_tracking);
ThreadStatusUtil::SetThreadOperationProperty(
ThreadStatus::COMPACTION_JOB_ID,
@@ -356,7 +409,7 @@ void CompactionJob::GenSubcompactionBoundaries() {
size_t num_files = flevel->num_files;
if (num_files == 0) {
break;
continue;
}
if (lvl == 0) {
@@ -415,12 +468,9 @@ void CompactionJob::GenSubcompactionBoundaries() {
// Group the ranges into subcompactions
const double min_file_fill_percent = 4.0 / 5;
uint64_t max_output_files =
static_cast<uint64_t>(
std::ceil(
sum / min_file_fill_percent /
cfd->GetCurrentMutableCFOptions()->MaxFileSizeForLevel(out_lvl))
);
uint64_t max_output_files = static_cast<uint64_t>(std::ceil(
sum / min_file_fill_percent /
cfd->GetCurrentMutableCFOptions()->MaxFileSizeForLevel(out_lvl)));
uint64_t subcompactions =
std::min({static_cast<uint64_t>(ranges.size()),
static_cast<uint64_t>(db_options_.max_subcompactions),
@@ -518,18 +568,17 @@ Status CompactionJob::Run() {
return status;
}
Status CompactionJob::Install(const MutableCFOptions& mutable_cf_options,
InstrumentedMutex* db_mutex) {
Status CompactionJob::Install(const MutableCFOptions& mutable_cf_options) {
AutoThreadOperationStageUpdater stage_updater(
ThreadStatus::STAGE_COMPACTION_INSTALL);
db_mutex->AssertHeld();
db_mutex_->AssertHeld();
Status status = compact_->status;
ColumnFamilyData* cfd = compact_->compaction->column_family_data();
cfd->internal_stats()->AddCompactionStats(
compact_->compaction->output_level(), compaction_stats_);
if (status.ok()) {
status = InstallCompactionResults(mutable_cf_options, db_mutex);
status = InstallCompactionResults(mutable_cf_options);
}
VersionStorageInfo::LevelSummaryStorage tmp;
auto vstorage = cfd->current()->storage_info();
@@ -666,7 +715,7 @@ void CompactionJob::ProcessKeyValueCompaction(SubcompactionState* sub_compact) {
if (end != nullptr &&
cfd->user_comparator()->Compare(c_iter->user_key(), *end) >= 0) {
break;
} else if (sub_compact->compaction->ShouldStopBefore(key) &&
} else if (sub_compact->ShouldStopBefore(key) &&
sub_compact->builder != nullptr) {
status = FinishCompactionOutputFile(input->status(), sub_compact);
if (!status.ok()) {
@@ -855,13 +904,33 @@ Status CompactionJob::FinishCompactionOutputFile(
event_logger_, cfd->ioptions()->listeners, meta->fd, info);
}
}
// Report new file to SstFileManagerImpl
auto sfm =
static_cast<SstFileManagerImpl*>(db_options_.sst_file_manager.get());
if (sfm && meta->fd.GetPathId() == 0) {
ColumnFamilyData* cfd = sub_compact->compaction->column_family_data();
auto fn = TableFileName(cfd->ioptions()->db_paths, meta->fd.GetNumber(),
meta->fd.GetPathId());
sfm->OnAddFile(fn);
if (sfm->IsMaxAllowedSpaceReached()) {
InstrumentedMutexLock l(db_mutex_);
if (db_bg_error_->ok()) {
s = Status::IOError("Max allowed space was reached");
*db_bg_error_ = s;
TEST_SYNC_POINT(
"CompactionJob::FinishCompactionOutputFile:MaxAllowedSpaceReached");
}
}
}
sub_compact->builder.reset();
return s;
}
Status CompactionJob::InstallCompactionResults(
const MutableCFOptions& mutable_cf_options, InstrumentedMutex* db_mutex) {
db_mutex->AssertHeld();
const MutableCFOptions& mutable_cf_options) {
db_mutex_->AssertHeld();
auto* compaction = compact_->compaction;
// paranoia: verify that the files that we started with
@@ -896,7 +965,7 @@ Status CompactionJob::InstallCompactionResults(
}
return versions_->LogAndApply(compaction->column_family_data(),
mutable_cf_options, compaction->edit(),
db_mutex, db_directory_);
db_mutex_, db_directory_);
}
void CompactionJob::RecordCompactionIOStats() {
@@ -950,7 +1019,7 @@ Status CompactionJob::OpenCompactionOutputFile(
cfd->ioptions()->optimize_filters_for_hits && bottommost_level_;
sub_compact->builder.reset(NewTableBuilder(
*cfd->ioptions(), cfd->internal_comparator(),
cfd->int_tbl_prop_collector_factories(), cfd->GetID(),
cfd->int_tbl_prop_collector_factories(), cfd->GetID(), cfd->GetName(),
sub_compact->outfile.get(), sub_compact->compaction->output_compression(),
cfd->ioptions()->compression_opts, skip_filters));
LogFlush(db_options_.info_log);
+7 -6
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -56,7 +56,8 @@ class CompactionJob {
const EnvOptions& env_options, VersionSet* versions,
std::atomic<bool>* shutting_down, LogBuffer* log_buffer,
Directory* db_directory, Directory* output_directory,
Statistics* stats,
Statistics* stats, InstrumentedMutex* db_mutex,
Status* db_bg_error,
std::vector<SequenceNumber> existing_snapshots,
SequenceNumber earliest_write_conflict_snapshot,
std::shared_ptr<Cache> table_cache, EventLogger* event_logger,
@@ -77,8 +78,7 @@ class CompactionJob {
Status Run();
// REQUIRED: mutex held
Status Install(const MutableCFOptions& mutable_cf_options,
InstrumentedMutex* db_mutex);
Status Install(const MutableCFOptions& mutable_cf_options);
private:
struct SubcompactionState;
@@ -95,8 +95,7 @@ class CompactionJob {
Status FinishCompactionOutputFile(const Status& input_status,
SubcompactionState* sub_compact);
Status InstallCompactionResults(const MutableCFOptions& mutable_cf_options,
InstrumentedMutex* db_mutex);
Status InstallCompactionResults(const MutableCFOptions& mutable_cf_options);
void RecordCompactionIOStats();
Status OpenCompactionOutputFile(SubcompactionState* sub_compact);
void CleanupCompaction();
@@ -130,6 +129,8 @@ class CompactionJob {
Directory* db_directory_;
Directory* output_directory_;
Statistics* stats_;
InstrumentedMutex* db_mutex_;
Status* db_bg_error_;
// If there were two snapshots with seq numbers s1 and
// s2 and s1 < s2, and if we find two instances of a key k1 then lies
// entirely within s1 and s2, then the earlier version of k1 can be safely
+2 -2
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -660,7 +660,7 @@ TEST_P(CompactionJobStatsTest, CompactionJobStatsTest) {
options.max_subcompactions = max_subcompactions_;
options.bytes_per_sync = 512 * 1024;
options.compaction_measure_io_stats = true;
options.report_bg_io_stats = true;
for (int test = 0; test < 2; ++test) {
DestroyAndReopen(options);
CreateAndReopenWithCF({"pikachu"}, options);
+6 -6
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -250,9 +250,9 @@ class CompactionJobTest : public testing::Test {
EventLogger event_logger(db_options_.info_log.get());
CompactionJob compaction_job(
0, &compaction, db_options_, env_options_, versions_.get(),
&shutting_down_, &log_buffer, nullptr, nullptr, nullptr, snapshots,
earliest_write_conflict_snapshot, table_cache_, &event_logger, false,
false, dbname_, &compaction_job_stats_);
&shutting_down_, &log_buffer, nullptr, nullptr, nullptr, &mutex_,
&bg_error_, snapshots, earliest_write_conflict_snapshot, table_cache_,
&event_logger, false, false, dbname_, &compaction_job_stats_);
VerifyInitializationOfCompactionJobStats(compaction_job_stats_);
@@ -262,8 +262,7 @@ class CompactionJobTest : public testing::Test {
s = compaction_job.Run();
ASSERT_OK(s);
mutex_.Lock();
ASSERT_OK(compaction_job.Install(*cfd->GetLatestMutableCFOptions(),
&mutex_));
ASSERT_OK(compaction_job.Install(*cfd->GetLatestMutableCFOptions()));
mutex_.Unlock();
if (expected_results.size() == 0) {
@@ -295,6 +294,7 @@ class CompactionJobTest : public testing::Test {
ColumnFamilyData* cfd_;
std::unique_ptr<CompactionFilter> compaction_filter_;
std::shared_ptr<MergeOperator> merge_op_;
Status bg_error_;
};
TEST_F(CompactionJobTest, Simple) {
+14 -2
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -141,7 +141,8 @@ CompactionPicker::~CompactionPicker() {}
// Delete this compaction from the list of running compactions.
void CompactionPicker::ReleaseCompactionFiles(Compaction* c, Status status) {
if (c->start_level() == 0) {
if (c->start_level() == 0 ||
ioptions_.compaction_style == kCompactionStyleUniversal) {
level0_compactions_in_progress_.erase(c);
}
if (!status.ok()) {
@@ -612,6 +613,17 @@ Compaction* CompactionPicker::CompactRange(
if (input_level == 0) {
level0_compactions_in_progress_.insert(compaction);
}
// Creating a compaction influences the compaction score because the score
// takes running compactions into account (by skipping files that are already
// being compacted). Since we just changed compaction score, we recalculate it
// here
{ // this piece of code recomputes compaction score
CompactionOptionsFIFO dummy_compaction_options_fifo;
vstorage->ComputeCompactionScore(mutable_cf_options,
dummy_compaction_options_fifo);
}
return compaction;
}
+3 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -83,6 +83,8 @@ class CompactionPicker {
#endif // ROCKSDB_LITE
// Free up the files that participated in a compaction
//
// Requirement: DB mutex held
void ReleaseCompactionFiles(Compaction* c, Status status);
// Returns true if any one of the specified files are being compacted
+86 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -195,6 +195,7 @@ TEST_F(CompactionPickerTest, LevelMaxScore) {
NewVersionStorage(6, kCompactionStyleLevel);
mutable_cf_options_.target_file_size_base = 10000000;
mutable_cf_options_.target_file_size_multiplier = 10;
mutable_cf_options_.max_bytes_for_level_base = 10 * 1024 * 1024;
Add(0, 1U, "150", "200", 1000000000U);
// Level 1 score 1.2
Add(1, 66U, "150", "200", 6000000U);
@@ -487,6 +488,90 @@ TEST_F(CompactionPickerTest, NeedsCompactionFIFO) {
}
#endif // ROCKSDB_LITE
TEST_F(CompactionPickerTest, CompactionPriMinOverlapping1) {
NewVersionStorage(6, kCompactionStyleLevel);
mutable_cf_options_.target_file_size_base = 10000000;
mutable_cf_options_.target_file_size_multiplier = 10;
mutable_cf_options_.max_bytes_for_level_base = 10 * 1024 * 1024;
mutable_cf_options_.compaction_pri = kMinOverlappingRatio;
Add(2, 6U, "150", "179", 50000000U);
Add(2, 7U, "180", "220", 50000000U);
Add(2, 8U, "321", "400", 50000000U); // File not overlapping
Add(2, 9U, "721", "800", 50000000U);
Add(3, 26U, "150", "170", 260000000U);
Add(3, 27U, "171", "179", 260000000U);
Add(3, 28U, "191", "220", 260000000U);
Add(3, 29U, "221", "300", 260000000U);
Add(3, 30U, "750", "900", 260000000U);
UpdateVersionStorageInfo();
std::unique_ptr<Compaction> compaction(level_compaction_picker.PickCompaction(
cf_name_, mutable_cf_options_, vstorage_.get(), &log_buffer_));
ASSERT_TRUE(compaction.get() != nullptr);
ASSERT_EQ(1U, compaction->num_input_files(0));
// Pick file 8 because it overlaps with 0 files on level 3.
ASSERT_EQ(8U, compaction->input(0, 0)->fd.GetNumber());
}
TEST_F(CompactionPickerTest, CompactionPriMinOverlapping2) {
NewVersionStorage(6, kCompactionStyleLevel);
mutable_cf_options_.target_file_size_base = 10000000;
mutable_cf_options_.target_file_size_multiplier = 10;
mutable_cf_options_.max_bytes_for_level_base = 10 * 1024 * 1024;
mutable_cf_options_.compaction_pri = kMinOverlappingRatio;
Add(2, 6U, "150", "175",
60000000U); // Overlaps with file 26, 27, total size 521M
Add(2, 7U, "176", "200", 60000000U); // Overlaps with file 27, 28, total size
// 520M, the smalelst overlapping
Add(2, 8U, "201", "300",
60000000U); // Overlaps with file 28, 29, total size 521M
Add(3, 26U, "100", "110", 261000000U);
Add(3, 26U, "150", "170", 261000000U);
Add(3, 27U, "171", "179", 260000000U);
Add(3, 28U, "191", "220", 260000000U);
Add(3, 29U, "221", "300", 261000000U);
Add(3, 30U, "321", "400", 261000000U);
UpdateVersionStorageInfo();
std::unique_ptr<Compaction> compaction(level_compaction_picker.PickCompaction(
cf_name_, mutable_cf_options_, vstorage_.get(), &log_buffer_));
ASSERT_TRUE(compaction.get() != nullptr);
ASSERT_EQ(1U, compaction->num_input_files(0));
// Picking file 7 because overlapping ratio is the biggest.
ASSERT_EQ(7U, compaction->input(0, 0)->fd.GetNumber());
}
TEST_F(CompactionPickerTest, CompactionPriMinOverlapping3) {
NewVersionStorage(6, kCompactionStyleLevel);
mutable_cf_options_.max_bytes_for_level_base = 10000000;
mutable_cf_options_.max_bytes_for_level_multiplier = 10;
mutable_cf_options_.compaction_pri = kMinOverlappingRatio;
// file 7 and 8 over lap with the same file, but file 8 is smaller so
// it will be picked.
Add(2, 6U, "150", "167", 60000000U); // Overlaps with file 26, 27
Add(2, 7U, "168", "169", 60000000U); // Overlaps with file 27
Add(2, 8U, "201", "300", 61000000U); // Overlaps with file 28, but the file
// itself is larger. Should be picked.
Add(3, 26U, "160", "165", 260000000U);
Add(3, 27U, "166", "170", 260000000U);
Add(3, 28U, "180", "400", 260000000U);
Add(3, 29U, "401", "500", 260000000U);
UpdateVersionStorageInfo();
std::unique_ptr<Compaction> compaction(level_compaction_picker.PickCompaction(
cf_name_, mutable_cf_options_, vstorage_.get(), &log_buffer_));
ASSERT_TRUE(compaction.get() != nullptr);
ASSERT_EQ(1U, compaction->num_input_files(0));
// Picking file 8 because overlapping ratio is the biggest.
ASSERT_EQ(8U, compaction->input(0, 0)->fd.GetNumber());
}
// This test exhibits the bug where we don't properly reset parent_index in
// PickCompaction()
TEST_F(CompactionPickerTest, ParentIndexResetBug) {
+1 -1
View File
@@ -1,6 +1,6 @@
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file. See the AUTHORS file for names of contributors.
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2014, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+240
View File
@@ -0,0 +1,240 @@
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
//
// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file. See the AUTHORS file for names of contributors.
#include <cstdlib>
#include "db/db_test_util.h"
#include "port/stack_trace.h"
namespace rocksdb {
static uint64_t TestGetTickerCount(const Options& options,
Tickers ticker_type) {
return options.statistics->getTickerCount(ticker_type);
}
class DBBlockCacheTest : public DBTestBase {
private:
size_t miss_count_ = 0;
size_t hit_count_ = 0;
size_t insert_count_ = 0;
size_t failure_count_ = 0;
size_t compressed_miss_count_ = 0;
size_t compressed_hit_count_ = 0;
size_t compressed_insert_count_ = 0;
size_t compressed_failure_count_ = 0;
public:
const size_t kNumBlocks = 10;
const size_t kValueSize = 100;
DBBlockCacheTest() : DBTestBase("/db_block_cache_test") {}
BlockBasedTableOptions GetTableOptions() {
BlockBasedTableOptions table_options;
// Set a small enough block size so that each key-value get its own block.
table_options.block_size = 1;
return table_options;
}
Options GetOptions(const BlockBasedTableOptions& table_options) {
Options options = CurrentOptions();
options.create_if_missing = true;
// options.compression = kNoCompression;
options.statistics = rocksdb::CreateDBStatistics();
options.table_factory.reset(new BlockBasedTableFactory(table_options));
return options;
}
void InitTable(const Options& options) {
std::string value(kValueSize, 'a');
for (size_t i = 0; i < kNumBlocks; i++) {
ASSERT_OK(Put(ToString(i), value.c_str()));
}
}
void RecordCacheCounters(const Options& options) {
miss_count_ = TestGetTickerCount(options, BLOCK_CACHE_MISS);
hit_count_ = TestGetTickerCount(options, BLOCK_CACHE_HIT);
insert_count_ = TestGetTickerCount(options, BLOCK_CACHE_ADD);
failure_count_ = TestGetTickerCount(options, BLOCK_CACHE_ADD_FAILURES);
compressed_miss_count_ =
TestGetTickerCount(options, BLOCK_CACHE_COMPRESSED_MISS);
compressed_hit_count_ =
TestGetTickerCount(options, BLOCK_CACHE_COMPRESSED_HIT);
compressed_insert_count_ =
TestGetTickerCount(options, BLOCK_CACHE_COMPRESSED_ADD);
compressed_failure_count_ =
TestGetTickerCount(options, BLOCK_CACHE_COMPRESSED_ADD_FAILURES);
}
void CheckCacheCounters(const Options& options, size_t expected_misses,
size_t expected_hits, size_t expected_inserts,
size_t expected_failures) {
size_t new_miss_count = TestGetTickerCount(options, BLOCK_CACHE_MISS);
size_t new_hit_count = TestGetTickerCount(options, BLOCK_CACHE_HIT);
size_t new_insert_count = TestGetTickerCount(options, BLOCK_CACHE_ADD);
size_t new_failure_count =
TestGetTickerCount(options, BLOCK_CACHE_ADD_FAILURES);
ASSERT_EQ(miss_count_ + expected_misses, new_miss_count);
ASSERT_EQ(hit_count_ + expected_hits, new_hit_count);
ASSERT_EQ(insert_count_ + expected_inserts, new_insert_count);
ASSERT_EQ(failure_count_ + expected_failures, new_failure_count);
miss_count_ = new_miss_count;
hit_count_ = new_hit_count;
insert_count_ = new_insert_count;
failure_count_ = new_failure_count;
}
void CheckCompressedCacheCounters(const Options& options,
size_t expected_misses,
size_t expected_hits,
size_t expected_inserts,
size_t expected_failures) {
size_t new_miss_count =
TestGetTickerCount(options, BLOCK_CACHE_COMPRESSED_MISS);
size_t new_hit_count =
TestGetTickerCount(options, BLOCK_CACHE_COMPRESSED_HIT);
size_t new_insert_count =
TestGetTickerCount(options, BLOCK_CACHE_COMPRESSED_ADD);
size_t new_failure_count =
TestGetTickerCount(options, BLOCK_CACHE_COMPRESSED_ADD_FAILURES);
ASSERT_EQ(compressed_miss_count_ + expected_misses, new_miss_count);
ASSERT_EQ(compressed_hit_count_ + expected_hits, new_hit_count);
ASSERT_EQ(compressed_insert_count_ + expected_inserts, new_insert_count);
ASSERT_EQ(compressed_failure_count_ + expected_failures, new_failure_count);
compressed_miss_count_ = new_miss_count;
compressed_hit_count_ = new_hit_count;
compressed_insert_count_ = new_insert_count;
compressed_failure_count_ = new_failure_count;
}
};
TEST_F(DBBlockCacheTest, TestWithoutCompressedBlockCache) {
ReadOptions read_options;
auto table_options = GetTableOptions();
auto options = GetOptions(table_options);
InitTable(options);
std::shared_ptr<Cache> cache = NewLRUCache(0, 0, false);
table_options.block_cache = cache;
options.table_factory.reset(new BlockBasedTableFactory(table_options));
Reopen(options);
RecordCacheCounters(options);
std::vector<std::unique_ptr<Iterator>> iterators(kNumBlocks - 1);
Iterator* iter = nullptr;
// Load blocks into cache.
for (size_t i = 0; i < kNumBlocks - 1; i++) {
iter = db_->NewIterator(read_options);
iter->Seek(ToString(i));
ASSERT_OK(iter->status());
CheckCacheCounters(options, 1, 0, 1, 0);
iterators[i].reset(iter);
}
size_t usage = cache->GetUsage();
ASSERT_LT(0, usage);
cache->SetCapacity(usage);
ASSERT_EQ(usage, cache->GetPinnedUsage());
// Test with strict capacity limit.
cache->SetStrictCapacityLimit(true);
iter = db_->NewIterator(read_options);
iter->Seek(ToString(kNumBlocks - 1));
ASSERT_TRUE(iter->status().IsIncomplete());
CheckCacheCounters(options, 1, 0, 0, 1);
delete iter;
iter = nullptr;
// Release interators and access cache again.
for (size_t i = 0; i < kNumBlocks - 1; i++) {
iterators[i].reset();
CheckCacheCounters(options, 0, 0, 0, 0);
}
ASSERT_EQ(0, cache->GetPinnedUsage());
for (size_t i = 0; i < kNumBlocks - 1; i++) {
iter = db_->NewIterator(read_options);
iter->Seek(ToString(i));
ASSERT_OK(iter->status());
CheckCacheCounters(options, 0, 1, 0, 0);
iterators[i].reset(iter);
}
}
#ifdef SNAPPY
TEST_F(DBBlockCacheTest, TestWithCompressedBlockCache) {
ReadOptions read_options;
auto table_options = GetTableOptions();
auto options = GetOptions(table_options);
options.compression = CompressionType::kSnappyCompression;
InitTable(options);
std::shared_ptr<Cache> cache = NewLRUCache(0, 0, false);
std::shared_ptr<Cache> compressed_cache = NewLRUCache(0, 0, false);
table_options.block_cache = cache;
table_options.block_cache_compressed = compressed_cache;
options.table_factory.reset(new BlockBasedTableFactory(table_options));
Reopen(options);
RecordCacheCounters(options);
std::vector<std::unique_ptr<Iterator>> iterators(kNumBlocks - 1);
Iterator* iter = nullptr;
// Load blocks into cache.
for (size_t i = 0; i < kNumBlocks - 1; i++) {
iter = db_->NewIterator(read_options);
iter->Seek(ToString(i));
ASSERT_OK(iter->status());
CheckCacheCounters(options, 1, 0, 1, 0);
CheckCompressedCacheCounters(options, 1, 0, 1, 0);
iterators[i].reset(iter);
}
size_t usage = cache->GetUsage();
ASSERT_LT(0, usage);
ASSERT_EQ(usage, cache->GetPinnedUsage());
size_t compressed_usage = compressed_cache->GetUsage();
ASSERT_LT(0, compressed_usage);
// Compressed block cache cannot be pinned.
ASSERT_EQ(0, compressed_cache->GetPinnedUsage());
// Set strict capacity limit flag. Now block will only load into compressed
// block cache.
cache->SetCapacity(usage);
cache->SetStrictCapacityLimit(true);
ASSERT_EQ(usage, cache->GetPinnedUsage());
// compressed_cache->SetCapacity(compressed_usage);
compressed_cache->SetCapacity(0);
// compressed_cache->SetStrictCapacityLimit(true);
iter = db_->NewIterator(read_options);
iter->Seek(ToString(kNumBlocks - 1));
ASSERT_TRUE(iter->status().IsIncomplete());
CheckCacheCounters(options, 1, 0, 0, 1);
CheckCompressedCacheCounters(options, 1, 0, 1, 0);
delete iter;
iter = nullptr;
// Clear strict capacity limit flag. This time we shall hit compressed block
// cache.
cache->SetStrictCapacityLimit(false);
iter = db_->NewIterator(read_options);
iter->Seek(ToString(kNumBlocks - 1));
ASSERT_OK(iter->status());
CheckCacheCounters(options, 1, 0, 1, 0);
CheckCompressedCacheCounters(options, 0, 1, 0, 0);
delete iter;
iter = nullptr;
}
#endif
} // namespace rocksdb
int main(int argc, char** argv) {
rocksdb::port::InstallStackTraceHandler();
::testing::InitGoogleTest(&argc, argv);
return RUN_ALL_TESTS();
}
+6 -11
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -368,11 +368,10 @@ TEST_F(DBTestCompactionFilter, CompactionFilter) {
// entries are deleted. The compaction should create bunch of 'DeleteFile'
// entries in VersionEdit, but none of the 'AddFile's.
TEST_F(DBTestCompactionFilter, CompactionFilterDeletesAll) {
Options options;
Options options = CurrentOptions();
options.compaction_filter_factory = std::make_shared<DeleteFilterFactory>();
options.disable_auto_compactions = true;
options.create_if_missing = true;
options = CurrentOptions(options);
DestroyAndReopen(options);
// put some data
@@ -400,11 +399,10 @@ TEST_F(DBTestCompactionFilter, CompactionFilterDeletesAll) {
TEST_F(DBTestCompactionFilter, CompactionFilterWithValueChange) {
do {
Options options;
Options options = CurrentOptions();
options.num_levels = 3;
options.compaction_filter_factory =
std::make_shared<ChangeFilterFactory>();
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
// Write 100K+1 keys, these are written to a few files
@@ -467,8 +465,7 @@ TEST_F(DBTestCompactionFilter, CompactionFilterWithMergeOperator) {
PutFixed64(&three, 3);
PutFixed64(&four, 4);
Options options;
options = CurrentOptions(options);
Options options = CurrentOptions();
options.create_if_missing = true;
options.merge_operator = MergeOperators::CreateUInt64AddOperator();
options.num_levels = 3;
@@ -621,11 +618,10 @@ TEST_F(DBTestCompactionFilter, CompactionFilterContextCfId) {
// Compaction filters should only be applied to records that are newer than the
// latest snapshot. This test inserts records and applies a delete filter.
TEST_F(DBTestCompactionFilter, CompactionFilterSnapshot) {
Options options;
Options options = CurrentOptions();
options.compaction_filter_factory = std::make_shared<DeleteFilterFactory>();
options.disable_auto_compactions = true;
options.create_if_missing = true;
options = CurrentOptions(options);
DestroyAndReopen(options);
// Put some data.
@@ -659,11 +655,10 @@ TEST_F(DBTestCompactionFilter, CompactionFilterSnapshot) {
// records newer than the snapshot will also be processed
TEST_F(DBTestCompactionFilter, CompactionFilterIgnoreSnapshot) {
std::string five = ToString(5);
Options options;
Options options = CurrentOptions();
options.compaction_filter_factory = std::make_shared<DeleteISFilterFactory>();
options.disable_auto_compactions = true;
options.create_if_missing = true;
options = CurrentOptions(options);
DestroyAndReopen(options);
// Put some data.
+136 -27
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -288,8 +288,7 @@ TEST_F(DBCompactionTest, SkipStatsUpdateTest) {
}
TEST_F(DBCompactionTest, TestTableReaderForCompaction) {
Options options;
options = CurrentOptions(options);
Options options = CurrentOptions();
options.env = env_;
options.new_table_reader_for_compaction_inputs = true;
options.max_open_files = 100;
@@ -493,14 +492,13 @@ TEST_F(DBCompactionTest, DisableStatsUpdateReopen) {
TEST_P(DBCompactionTestWithParam, CompactionTrigger) {
const int kNumKeysPerFile = 100;
Options options;
Options options = CurrentOptions();
options.write_buffer_size = 110 << 10; // 110KB
options.arena_block_size = 4 << 10;
options.num_levels = 3;
options.level0_file_num_compaction_trigger = 3;
options.max_subcompactions = max_subcompactions_;
options.memtable_factory.reset(new SpecialSkipListFactory(kNumKeysPerFile));
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
Random rnd(301);
@@ -533,11 +531,107 @@ TEST_P(DBCompactionTestWithParam, CompactionTrigger) {
ASSERT_EQ(NumTableFilesAtLevel(1, 1), 1);
}
TEST_F(DBCompactionTest, BGCompactionsAllowed) {
// Create several column families. Make compaction triggers in all of them
// and see number of compactions scheduled to be less than allowed.
const int kNumKeysPerFile = 100;
Options options = CurrentOptions();
options.write_buffer_size = 110 << 10; // 110KB
options.arena_block_size = 4 << 10;
options.num_levels = 3;
// Should speed up compaction when there are 4 files.
options.level0_file_num_compaction_trigger = 2;
options.level0_slowdown_writes_trigger = 20;
options.soft_pending_compaction_bytes_limit = 1 << 30; // Infinitely large
options.base_background_compactions = 1;
options.max_background_compactions = 3;
options.memtable_factory.reset(new SpecialSkipListFactory(kNumKeysPerFile));
// Block all threads in thread pool.
const size_t kTotalTasks = 4;
env_->SetBackgroundThreads(4, Env::LOW);
test::SleepingBackgroundTask sleeping_tasks[kTotalTasks];
for (size_t i = 0; i < kTotalTasks; i++) {
env_->Schedule(&test::SleepingBackgroundTask::DoSleepTask,
&sleeping_tasks[i], Env::Priority::LOW);
sleeping_tasks[i].WaitUntilSleeping();
}
CreateAndReopenWithCF({"one", "two", "three"}, options);
Random rnd(301);
for (int cf = 0; cf < 4; cf++) {
for (int num = 0; num < options.level0_file_num_compaction_trigger; num++) {
for (int i = 0; i < kNumKeysPerFile; i++) {
ASSERT_OK(Put(cf, Key(i), ""));
}
// put extra key to trigger flush
ASSERT_OK(Put(cf, "", ""));
dbfull()->TEST_WaitForFlushMemTable(handles_[cf]);
ASSERT_EQ(NumTableFilesAtLevel(0, cf), num + 1);
}
}
// Now all column families qualify compaction but only one should be
// scheduled, because no column family hits speed up condition.
ASSERT_EQ(1, env_->GetThreadPoolQueueLen(Env::Priority::LOW));
// Create two more files for one column family, which triggers speed up
// condition, three compactions will be scheduled.
for (int num = 0; num < options.level0_file_num_compaction_trigger; num++) {
for (int i = 0; i < kNumKeysPerFile; i++) {
ASSERT_OK(Put(2, Key(i), ""));
}
// put extra key to trigger flush
ASSERT_OK(Put(2, "", ""));
dbfull()->TEST_WaitForFlushMemTable(handles_[2]);
ASSERT_EQ(options.level0_file_num_compaction_trigger + num + 1,
NumTableFilesAtLevel(0, 2));
}
ASSERT_EQ(3, env_->GetThreadPoolQueueLen(Env::Priority::LOW));
// Unblock all threads to unblock all compactions.
for (size_t i = 0; i < kTotalTasks; i++) {
sleeping_tasks[i].WakeUp();
sleeping_tasks[i].WaitUntilDone();
}
dbfull()->TEST_WaitForCompact();
// Verify number of compactions allowed will come back to 1.
for (size_t i = 0; i < kTotalTasks; i++) {
sleeping_tasks[i].Reset();
env_->Schedule(&test::SleepingBackgroundTask::DoSleepTask,
&sleeping_tasks[i], Env::Priority::LOW);
sleeping_tasks[i].WaitUntilSleeping();
}
for (int cf = 0; cf < 4; cf++) {
for (int num = 0; num < options.level0_file_num_compaction_trigger; num++) {
for (int i = 0; i < kNumKeysPerFile; i++) {
ASSERT_OK(Put(cf, Key(i), ""));
}
// put extra key to trigger flush
ASSERT_OK(Put(cf, "", ""));
dbfull()->TEST_WaitForFlushMemTable(handles_[cf]);
ASSERT_EQ(NumTableFilesAtLevel(0, cf), num + 1);
}
}
// Now all column families qualify compaction but only one should be
// scheduled, because no column family hits speed up condition.
ASSERT_EQ(1, env_->GetThreadPoolQueueLen(Env::Priority::LOW));
for (size_t i = 0; i < kTotalTasks; i++) {
sleeping_tasks[i].WakeUp();
sleeping_tasks[i].WaitUntilDone();
}
}
TEST_P(DBCompactionTestWithParam, CompactionsGenerateMultipleFiles) {
Options options;
Options options = CurrentOptions();
options.write_buffer_size = 100000000; // Large write buffer
options.max_subcompactions = max_subcompactions_;
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
Random rnd(301);
@@ -564,9 +658,8 @@ TEST_P(DBCompactionTestWithParam, CompactionsGenerateMultipleFiles) {
TEST_F(DBCompactionTest, MinorCompactionsHappen) {
do {
Options options;
Options options = CurrentOptions();
options.write_buffer_size = 10000;
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
const int N = 500;
@@ -591,7 +684,7 @@ TEST_F(DBCompactionTest, MinorCompactionsHappen) {
}
TEST_F(DBCompactionTest, ZeroSeqIdCompaction) {
Options options;
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleLevel;
options.level0_file_num_compaction_trigger = 3;
@@ -605,7 +698,6 @@ TEST_F(DBCompactionTest, ZeroSeqIdCompaction) {
const size_t key_len =
static_cast<size_t>(compact_opt.output_file_size_limit) / 5;
options = CurrentOptions(options);
DestroyAndReopen(options);
std::vector<const Snapshot*> snaps;
@@ -649,10 +741,8 @@ TEST_F(DBCompactionTest, ZeroSeqIdCompaction) {
// if the database is shutdown during the memtable compaction.
TEST_F(DBCompactionTest, RecoverDuringMemtableCompaction) {
do {
Options options;
Options options = CurrentOptions();
options.env = env_;
options.write_buffer_size = 1000000;
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
// Trigger a long memtable compaction and reopen the database during it
@@ -676,10 +766,9 @@ TEST_P(DBCompactionTestWithParam, TrivialMoveOneFile) {
[&](void* arg) { trivial_move++; });
rocksdb::SyncPoint::GetInstance()->EnableProcessing();
Options options;
Options options = CurrentOptions();
options.write_buffer_size = 100000000;
options.max_subcompactions = max_subcompactions_;
options = CurrentOptions(options);
DestroyAndReopen(options);
int32_t num_keys = 80;
@@ -1271,10 +1360,9 @@ TEST_P(DBCompactionTestWithParam, TrivialMoveToLastLevelWithFiles) {
[&](void* arg) { non_trivial_move++; });
rocksdb::SyncPoint::GetInstance()->EnableProcessing();
Options options;
Options options = CurrentOptions();
options.write_buffer_size = 100000000;
options.max_subcompactions = max_subcompactions_;
options = CurrentOptions(options);
DestroyAndReopen(options);
int32_t value_size = 10 * 1024; // 10 KB
@@ -1563,7 +1651,7 @@ TEST_P(DBCompactionTestWithParam, ConvertCompactionStyle) {
int max_key_universal_insert = 600;
// Stage 1: generate a db with level compaction
Options options;
Options options = CurrentOptions();
options.write_buffer_size = 110 << 10; // 110KB
options.arena_block_size = 4 << 10;
options.num_levels = 4;
@@ -1573,7 +1661,6 @@ TEST_P(DBCompactionTestWithParam, ConvertCompactionStyle) {
options.target_file_size_base = 200 << 10; // 200KB
options.target_file_size_multiplier = 1;
options.max_subcompactions = max_subcompactions_;
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
for (int i = 0; i <= max_key_level_insert; i++) {
@@ -1898,7 +1985,7 @@ TEST_P(DBCompactionTestWithParam, DISABLED_CompactFilesOnLevelCompaction) {
std::set<std::string> overlapping_file_names;
std::vector<std::string> compaction_input_file_names;
for (int f = 0; f < file_picked; ++f) {
int level;
int level = 0;
auto file_meta = PickFileRandomly(cf_meta, &rnd, &level);
compaction_input_file_names.push_back(file_meta->name);
GetOverlappingFileNumbersForLevelCompaction(
@@ -2198,6 +2285,25 @@ TEST_P(DBCompactionTestWithParam, CompressLevelCompaction) {
Destroy(options);
}
TEST_F(DBCompactionTest, SanitizeCompactionOptionsTest) {
Options options = CurrentOptions();
options.max_background_compactions = 5;
options.soft_pending_compaction_bytes_limit = 0;
options.hard_pending_compaction_bytes_limit = 100;
options.create_if_missing = true;
DestroyAndReopen(options);
ASSERT_EQ(5, db_->GetOptions().base_background_compactions);
ASSERT_EQ(100, db_->GetOptions().soft_pending_compaction_bytes_limit);
options.base_background_compactions = 4;
options.max_background_compactions = 3;
options.soft_pending_compaction_bytes_limit = 200;
options.hard_pending_compaction_bytes_limit = 150;
DestroyAndReopen(options);
ASSERT_EQ(3, db_->GetOptions().base_background_compactions);
ASSERT_EQ(150, db_->GetOptions().soft_pending_compaction_bytes_limit);
}
// This tests for a bug that could cause two level0 compactions running
// concurrently
// TODO(aekmekji): Make sure that the reason this fails when run with
@@ -2269,10 +2375,9 @@ TEST_P(DBCompactionTestWithParam, ForceBottommostLevelCompaction) {
[&](void* arg) { non_trivial_move++; });
rocksdb::SyncPoint::GetInstance()->EnableProcessing();
Options options;
Options options = CurrentOptions();
options.write_buffer_size = 100000000;
options.max_subcompactions = max_subcompactions_;
options = CurrentOptions(options);
DestroyAndReopen(options);
int32_t value_size = 10 * 1024; // 10 KB
@@ -2362,14 +2467,14 @@ class CompactionPriTest : public DBTestBase,
};
TEST_P(CompactionPriTest, Test) {
Options options;
Options options = CurrentOptions();
options.write_buffer_size = 16 * 1024;
options.compaction_pri = static_cast<CompactionPri>(compaction_pri_);
options.hard_pending_compaction_bytes_limit = 256 * 1024;
options.max_bytes_for_level_base = 64 * 1024;
options.max_bytes_for_level_multiplier = 4;
options.compression = kNoCompression;
options = CurrentOptions(options);
DestroyAndReopen(options);
Random rnd(301);
@@ -2390,8 +2495,12 @@ TEST_P(CompactionPriTest, Test) {
}
}
INSTANTIATE_TEST_CASE_P(CompactionPriTest, CompactionPriTest,
::testing::Values(0, 1, 2));
INSTANTIATE_TEST_CASE_P(
CompactionPriTest, CompactionPriTest,
::testing::Values(CompactionPri::kByCompensatedSize,
CompactionPri::kOldestLargestSeqFirst,
CompactionPri::kOldestSmallestSeqFirst,
CompactionPri::kMinOverlappingRatio));
#endif // !defined(ROCKSDB_LITE)
} // namespace rocksdb
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+338 -175
View File
File diff suppressed because it is too large Load Diff
+19 -9
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -347,6 +347,10 @@ class DBImpl : public DB {
#endif // NDEBUG
// Return maximum background compaction alowed to be scheduled based on
// compaction status.
int BGCompactionsAllowed() const;
// Returns the list of live files in 'live' and the list
// of all files in the filesystem in 'candidate_files'.
// If force == false and the last call was less than
@@ -490,10 +494,11 @@ class DBImpl : public DB {
// Background process needs to call
// auto x = CaptureCurrentFileNumberInPendingOutputs()
// auto file_num = versions_->NewFileNumber();
// <do something>
// ReleaseFileNumberFromPendingOutputs(x)
// This will protect any temporary files created while <do something> is
// executing from being deleted.
// This will protect any file with number `file_num` or greater from being
// deleted while <do something> is running.
// -----------
// This function will capture current file number and append it to
// pending_outputs_. This will prevent any background process to delete any
@@ -518,7 +523,7 @@ class DBImpl : public DB {
SequenceNumber* max_sequence, bool read_only);
// The following two methods are used to flush a memtable to
// storage. The first one is used atdatabase RecoveryTime (when the
// storage. The first one is used at database RecoveryTime (when the
// database is opened) and is heavyweight because it holds the mutex
// for the entire period. The second method WriteLevel0Table supports
// concurrent flush memtables to storage.
@@ -818,7 +823,10 @@ class DBImpl : public DB {
// they're unique
std::atomic<int> next_job_id_;
bool flush_on_destroy_; // Used when disableWAL is true.
// A flag indicating whether the current rocksdb database has any
// data that is not yet persisted into either WAL or SST file.
// Used when disableWAL is true.
bool has_unpersisted_data_;
static const int KEEP_LOG_FILE_NUM = 1000;
// MSVC version 1800 still does not have constexpr for ::max()
@@ -836,9 +844,12 @@ class DBImpl : public DB {
// Unified interface for logging events
EventLogger event_logger_;
// A value of >0 temporarily disables scheduling of background work
// A value of > 0 temporarily disables scheduling of background work
int bg_work_paused_;
// A value of > 0 temporarily disables scheduling of background compaction
int bg_compaction_paused_;
// Guard against multiple concurrent refitting
bool refitting_level_;
@@ -889,9 +900,8 @@ class DBImpl : public DB {
bool* value_found = nullptr);
bool GetIntPropertyInternal(ColumnFamilyData* cfd,
DBPropertyType property_type,
bool need_out_of_mutex, bool is_locked,
uint64_t* value);
const DBPropertyInfo& property_info,
bool is_locked, uint64_t* value);
bool HasPendingManualCompaction();
bool HasExclusiveManualCompaction();
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+10 -8
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -54,10 +54,11 @@ Iterator* DBImplReadOnly::NewIterator(const ReadOptions& read_options,
auto db_iter = NewArenaWrappedDbIterator(
env_, *cfd->ioptions(), cfd->user_comparator(),
(read_options.snapshot != nullptr
? reinterpret_cast<const SnapshotImpl*>(
read_options.snapshot)->number_
? reinterpret_cast<const SnapshotImpl*>(read_options.snapshot)
->number_
: latest_snapshot),
super_version->mutable_cf_options.max_sequential_skip_in_iterations);
super_version->mutable_cf_options.max_sequential_skip_in_iterations,
super_version->version_number);
auto internal_iter = NewInternalIterator(
read_options, cfd, super_version, db_iter->GetArena());
db_iter->SetIterUnderDBIter(internal_iter);
@@ -81,10 +82,11 @@ Status DBImplReadOnly::NewIterators(
auto* db_iter = NewArenaWrappedDbIterator(
env_, *cfd->ioptions(), cfd->user_comparator(),
(read_options.snapshot != nullptr
? reinterpret_cast<const SnapshotImpl*>(
read_options.snapshot)->number_
: latest_snapshot),
sv->mutable_cf_options.max_sequential_skip_in_iterations);
? reinterpret_cast<const SnapshotImpl*>(read_options.snapshot)
->number_
: latest_snapshot),
sv->mutable_cf_options.max_sequential_skip_in_iterations,
sv->version_number);
auto* internal_iter = NewInternalIterator(
read_options, cfd, sv, db_iter->GetArena());
db_iter->SetIterUnderDBIter(internal_iter);
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -7,6 +7,8 @@
#define __STDC_FORMAT_MACROS
#endif
#include "db/db_info_dumper.h"
#include <inttypes.h>
#include <stdio.h>
#include <string>
@@ -16,7 +18,6 @@
#include "db/filename.h"
#include "rocksdb/options.h"
#include "rocksdb/env.h"
#include "util/db_info_dumper.h"
namespace rocksdb {
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+7 -13
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -18,12 +18,11 @@ class DBTestInPlaceUpdate : public DBTestBase {
TEST_F(DBTestInPlaceUpdate, InPlaceUpdate) {
do {
Options options;
Options options = CurrentOptions();
options.create_if_missing = true;
options.inplace_update_support = true;
options.env = env_;
options.write_buffer_size = 100000;
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
// Update key with values of smaller size
@@ -41,12 +40,11 @@ TEST_F(DBTestInPlaceUpdate, InPlaceUpdate) {
TEST_F(DBTestInPlaceUpdate, InPlaceUpdateLargeNewValue) {
do {
Options options;
Options options = CurrentOptions();
options.create_if_missing = true;
options.inplace_update_support = true;
options.env = env_;
options.write_buffer_size = 100000;
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
// Update key with values of larger size
@@ -64,7 +62,7 @@ TEST_F(DBTestInPlaceUpdate, InPlaceUpdateLargeNewValue) {
TEST_F(DBTestInPlaceUpdate, InPlaceUpdateCallbackSmallerSize) {
do {
Options options;
Options options = CurrentOptions();
options.create_if_missing = true;
options.inplace_update_support = true;
@@ -72,7 +70,6 @@ TEST_F(DBTestInPlaceUpdate, InPlaceUpdateCallbackSmallerSize) {
options.write_buffer_size = 100000;
options.inplace_callback =
rocksdb::DBTestInPlaceUpdate::updateInPlaceSmallerSize;
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
// Update key with values of smaller size
@@ -92,7 +89,7 @@ TEST_F(DBTestInPlaceUpdate, InPlaceUpdateCallbackSmallerSize) {
TEST_F(DBTestInPlaceUpdate, InPlaceUpdateCallbackSmallerVarintSize) {
do {
Options options;
Options options = CurrentOptions();
options.create_if_missing = true;
options.inplace_update_support = true;
@@ -100,7 +97,6 @@ TEST_F(DBTestInPlaceUpdate, InPlaceUpdateCallbackSmallerVarintSize) {
options.write_buffer_size = 100000;
options.inplace_callback =
rocksdb::DBTestInPlaceUpdate::updateInPlaceSmallerVarintSize;
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
// Update key with values of smaller varint size
@@ -120,7 +116,7 @@ TEST_F(DBTestInPlaceUpdate, InPlaceUpdateCallbackSmallerVarintSize) {
TEST_F(DBTestInPlaceUpdate, InPlaceUpdateCallbackLargeNewValue) {
do {
Options options;
Options options = CurrentOptions();
options.create_if_missing = true;
options.inplace_update_support = true;
@@ -128,7 +124,6 @@ TEST_F(DBTestInPlaceUpdate, InPlaceUpdateCallbackLargeNewValue) {
options.write_buffer_size = 100000;
options.inplace_callback =
rocksdb::DBTestInPlaceUpdate::updateInPlaceLargerSize;
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
// Update key with values of larger size
@@ -146,7 +141,7 @@ TEST_F(DBTestInPlaceUpdate, InPlaceUpdateCallbackLargeNewValue) {
TEST_F(DBTestInPlaceUpdate, InPlaceUpdateCallbackNoAction) {
do {
Options options;
Options options = CurrentOptions();
options.create_if_missing = true;
options.inplace_update_support = true;
@@ -154,7 +149,6 @@ TEST_F(DBTestInPlaceUpdate, InPlaceUpdateCallbackNoAction) {
options.write_buffer_size = 100000;
options.inplace_callback =
rocksdb::DBTestInPlaceUpdate::updateInPlaceNoAction;
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
// Callback function requests no actions from db
+113 -45
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -13,18 +13,20 @@
#include <string>
#include <limits>
#include "db/filename.h"
#include "db/dbformat.h"
#include "db/filename.h"
#include "db/merge_context.h"
#include "port/port.h"
#include "rocksdb/env.h"
#include "rocksdb/options.h"
#include "rocksdb/iterator.h"
#include "rocksdb/merge_operator.h"
#include "rocksdb/options.h"
#include "table/internal_iterator.h"
#include "util/arena.h"
#include "util/logging.h"
#include "util/mutexlock.h"
#include "util/perf_context_imp.h"
#include "util/string_util.h"
namespace rocksdb {
@@ -59,9 +61,47 @@ class DBIter: public Iterator {
kReverse
};
// LocalStatistics contain Statistics counters that will be aggregated per
// each iterator instance and then will be sent to the global statistics when
// the iterator is destroyed.
//
// The purpose of this approach is to avoid perf regression happening
// when multiple threads bump the atomic counters from a DBIter::Next().
struct LocalStatistics {
explicit LocalStatistics() { ResetCounters(); }
void ResetCounters() {
next_count_ = 0;
next_found_count_ = 0;
prev_count_ = 0;
prev_found_count_ = 0;
bytes_read_ = 0;
}
void BumpGlobalStatistics(Statistics* global_statistics) {
RecordTick(global_statistics, NUMBER_DB_NEXT, next_count_);
RecordTick(global_statistics, NUMBER_DB_NEXT_FOUND, next_found_count_);
RecordTick(global_statistics, NUMBER_DB_PREV, prev_count_);
RecordTick(global_statistics, NUMBER_DB_PREV_FOUND, prev_found_count_);
RecordTick(global_statistics, ITER_BYTES_READ, bytes_read_);
ResetCounters();
}
// Map to Tickers::NUMBER_DB_NEXT
uint64_t next_count_;
// Map to Tickers::NUMBER_DB_NEXT_FOUND
uint64_t next_found_count_;
// Map to Tickers::NUMBER_DB_PREV
uint64_t prev_count_;
// Map to Tickers::NUMBER_DB_PREV_FOUND
uint64_t prev_found_count_;
// Map to Tickers::ITER_BYTES_READ
uint64_t bytes_read_;
};
DBIter(Env* env, const ImmutableCFOptions& ioptions, const Comparator* cmp,
InternalIterator* iter, SequenceNumber s, bool arena_mode,
uint64_t max_sequential_skip_in_iterations,
uint64_t max_sequential_skip_in_iterations, uint64_t version_number,
const Slice* iterate_upper_bound = nullptr,
bool prefix_same_as_start = false)
: arena_mode_(arena_mode),
@@ -75,6 +115,7 @@ class DBIter: public Iterator {
valid_(false),
current_entry_is_merged_(false),
statistics_(ioptions.statistics),
version_number_(version_number),
iterate_upper_bound_(iterate_upper_bound),
prefix_same_as_start_(prefix_same_as_start),
iter_pinned_(false) {
@@ -84,6 +125,7 @@ class DBIter: public Iterator {
}
virtual ~DBIter() {
RecordTick(statistics_, NO_ITERATORS, -1);
local_stats_.BumpGlobalStatistics(statistics_);
if (!arena_mode_) {
delete iter_;
} else {
@@ -136,9 +178,27 @@ class DBIter: public Iterator {
}
return s;
}
virtual bool IsKeyPinned() const override {
assert(valid_);
return iter_pinned_ && saved_key_.IsKeyPinned();
virtual Status GetProperty(std::string prop_name,
std::string* prop) override {
if (prop == nullptr) {
return Status::InvalidArgument("prop is nullptr");
}
if (prop_name == "rocksdb.iterator.super-version-number") {
// First try to pass the value returned from inner iterator.
if (!iter_->GetProperty(prop_name, prop).ok()) {
*prop = ToString(version_number_);
}
return Status::OK();
} else if (prop_name == "rocksdb.iterator.is-key-pinned") {
if (valid_) {
*prop = (iter_pinned_ && saved_key_.IsKeyPinned()) ? "1" : "0";
} else {
*prop = "Iterator is not valid.";
}
return Status::OK();
}
return Status::InvalidArgument("Undentified property.");
}
virtual void Next() override;
@@ -186,12 +246,14 @@ class DBIter: public Iterator {
bool current_entry_is_merged_;
Statistics* statistics_;
uint64_t max_skip_;
uint64_t version_number_;
const Slice* iterate_upper_bound_;
IterKey prefix_start_;
bool prefix_same_as_start_;
bool iter_pinned_;
// List of operands for merge operator.
std::deque<std::string> merge_operands_;
MergeContext merge_context_;
LocalStatistics local_stats_;
// No copying allowed
DBIter(const DBIter&);
@@ -229,6 +291,9 @@ void DBIter::Next() {
PERF_COUNTER_ADD(internal_key_skipped_count, 1);
}
if (statistics_ != nullptr) {
local_stats_.next_count_++;
}
// Now we point to the next internal position, for both of merge and
// not merge cases.
if (!iter_->Valid()) {
@@ -236,18 +301,15 @@ void DBIter::Next() {
return;
}
FindNextUserEntry(true /* skipping the current user key */);
if (statistics_ != nullptr) {
RecordTick(statistics_, NUMBER_DB_NEXT);
if (valid_) {
RecordTick(statistics_, NUMBER_DB_NEXT_FOUND);
RecordTick(statistics_, ITER_BYTES_READ, key().size() + value().size());
}
}
if (valid_ && prefix_extractor_ && prefix_same_as_start_ &&
prefix_extractor_->Transform(saved_key_.GetKey())
.compare(prefix_start_.GetKey()) != 0) {
valid_ = false;
}
if (statistics_ != nullptr && valid_) {
local_stats_.next_found_count_++;
local_stats_.bytes_read_ += (key().size() + value().size());
}
}
// PRE: saved_key_ has the current user key if skipping
@@ -275,7 +337,7 @@ void DBIter::FindNextUserEntryInternal(bool skipping) {
if (ParseKey(&ikey)) {
if (iterate_upper_bound_ != nullptr &&
ikey.user_key.compare(*iterate_upper_bound_) >= 0) {
user_comparator_->Compare(ikey.user_key, *iterate_upper_bound_) >= 0) {
break;
}
@@ -350,9 +412,9 @@ void DBIter::MergeValuesNewToOld() {
return;
}
merge_context_.Clear();
// Start the merge process by pushing the first operand
std::deque<std::string> operands;
operands.push_front(iter_->value().ToString());
merge_context_.PushOperand(iter_->value());
ParsedInternalKey ikey;
for (iter_->Next(); iter_->Valid(); iter_->Next()) {
@@ -377,7 +439,8 @@ void DBIter::MergeValuesNewToOld() {
{
StopWatchNano timer(env_, statistics_ != nullptr);
PERF_TIMER_GUARD(merge_operator_time_nanos);
user_merge_operator_->FullMerge(ikey.user_key, &val, operands,
user_merge_operator_->FullMerge(ikey.user_key, &val,
merge_context_.GetOperands(),
&saved_value_, logger_);
RecordTick(statistics_, MERGE_OPERATION_TOTAL_TIME,
timer.ElapsedNanos());
@@ -389,7 +452,7 @@ void DBIter::MergeValuesNewToOld() {
// hit a merge, add the value as an operand and run associative merge.
// when complete, add result to operands and continue.
const Slice& val = iter_->value();
operands.push_front(val.ToString());
merge_context_.PushOperand(val);
} else {
assert(false);
}
@@ -402,8 +465,9 @@ void DBIter::MergeValuesNewToOld() {
// a deletion marker.
// feed null as the existing value to the merge operator, such that
// client can differentiate this scenario and do things accordingly.
user_merge_operator_->FullMerge(saved_key_.GetKey(), nullptr, operands,
&saved_value_, logger_);
user_merge_operator_->FullMerge(saved_key_.GetKey(), nullptr,
merge_context_.GetOperands(), &saved_value_,
logger_);
RecordTick(statistics_, MERGE_OPERATION_TOTAL_TIME, timer.ElapsedNanos());
}
}
@@ -415,10 +479,10 @@ void DBIter::Prev() {
}
PrevInternal();
if (statistics_ != nullptr) {
RecordTick(statistics_, NUMBER_DB_PREV);
local_stats_.prev_count_++;
if (valid_) {
RecordTick(statistics_, NUMBER_DB_PREV_FOUND);
RecordTick(statistics_, ITER_BYTES_READ, key().size() + value().size());
local_stats_.prev_found_count_++;
local_stats_.bytes_read_ += (key().size() + value().size());
}
}
if (valid_ && prefix_extractor_ && prefix_same_as_start_ &&
@@ -495,7 +559,7 @@ void DBIter::PrevInternal() {
// saved_value_
bool DBIter::FindValueForCurrentKey() {
assert(iter_->Valid());
merge_operands_.clear();
merge_context_.Clear();
// last entry before merge (could be kTypeDeletion, kTypeSingleDeletion or
// kTypeValue)
ValueType last_not_merge_type = kTypeDeletion;
@@ -515,19 +579,19 @@ bool DBIter::FindValueForCurrentKey() {
last_key_entry_type = ikey.type;
switch (last_key_entry_type) {
case kTypeValue:
merge_operands_.clear();
merge_context_.Clear();
saved_value_ = iter_->value().ToString();
last_not_merge_type = kTypeValue;
break;
case kTypeDeletion:
case kTypeSingleDeletion:
merge_operands_.clear();
merge_context_.Clear();
last_not_merge_type = last_key_entry_type;
PERF_COUNTER_ADD(internal_delete_skipped_count, 1);
break;
case kTypeMerge:
assert(user_merge_operator_ != nullptr);
merge_operands_.push_back(iter_->value().ToString());
merge_context_.PushOperandBack(iter_->value());
break;
default:
assert(false);
@@ -550,8 +614,8 @@ bool DBIter::FindValueForCurrentKey() {
StopWatchNano timer(env_, statistics_ != nullptr);
PERF_TIMER_GUARD(merge_operator_time_nanos);
user_merge_operator_->FullMerge(saved_key_.GetKey(), nullptr,
merge_operands_, &saved_value_,
logger_);
merge_context_.GetOperands(),
&saved_value_, logger_);
RecordTick(statistics_, MERGE_OPERATION_TOTAL_TIME,
timer.ElapsedNanos());
} else {
@@ -562,8 +626,8 @@ bool DBIter::FindValueForCurrentKey() {
StopWatchNano timer(env_, statistics_ != nullptr);
PERF_TIMER_GUARD(merge_operator_time_nanos);
user_merge_operator_->FullMerge(saved_key_.GetKey(), &temp_slice,
merge_operands_, &saved_value_,
logger_);
merge_context_.GetOperands(),
&saved_value_, logger_);
RecordTick(statistics_, MERGE_OPERATION_TOTAL_TIME,
timer.ElapsedNanos());
}
@@ -606,11 +670,11 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
// kTypeMerge. We need to collect all kTypeMerge values and save them
// in operands
std::deque<std::string> operands;
merge_context_.Clear();
while (iter_->Valid() &&
user_comparator_->Equal(ikey.user_key, saved_key_.GetKey()) &&
ikey.type == kTypeMerge) {
operands.push_front(iter_->value().ToString());
merge_context_.PushOperand(iter_->value());
iter_->Next();
FindParseableKey(&ikey, kForward);
}
@@ -621,7 +685,8 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
{
StopWatchNano timer(env_, statistics_ != nullptr);
PERF_TIMER_GUARD(merge_operator_time_nanos);
user_merge_operator_->FullMerge(saved_key_.GetKey(), nullptr, operands,
user_merge_operator_->FullMerge(saved_key_.GetKey(), nullptr,
merge_context_.GetOperands(),
&saved_value_, logger_);
RecordTick(statistics_, MERGE_OPERATION_TOTAL_TIME, timer.ElapsedNanos());
}
@@ -639,8 +704,9 @@ bool DBIter::FindValueForCurrentKeyUsingSeek() {
{
StopWatchNano timer(env_, statistics_ != nullptr);
PERF_TIMER_GUARD(merge_operator_time_nanos);
user_merge_operator_->FullMerge(saved_key_.GetKey(), &val, operands,
&saved_value_, logger_);
user_merge_operator_->FullMerge(saved_key_.GetKey(), &val,
merge_context_.GetOperands(), &saved_value_,
logger_);
RecordTick(statistics_, MERGE_OPERATION_TOTAL_TIME, timer.ElapsedNanos());
}
valid_ = true;
@@ -818,12 +884,13 @@ Iterator* NewDBIterator(Env* env, const ImmutableCFOptions& ioptions,
InternalIterator* internal_iter,
const SequenceNumber& sequence,
uint64_t max_sequential_skip_in_iterations,
uint64_t version_number,
const Slice* iterate_upper_bound,
bool prefix_same_as_start, bool pin_data) {
DBIter* db_iter =
new DBIter(env, ioptions, user_key_comparator, internal_iter, sequence,
false, max_sequential_skip_in_iterations, iterate_upper_bound,
prefix_same_as_start);
false, max_sequential_skip_in_iterations, version_number,
iterate_upper_bound, prefix_same_as_start);
if (pin_data) {
db_iter->PinData();
}
@@ -850,12 +917,13 @@ inline Slice ArenaWrappedDBIter::key() const { return db_iter_->key(); }
inline Slice ArenaWrappedDBIter::value() const { return db_iter_->value(); }
inline Status ArenaWrappedDBIter::status() const { return db_iter_->status(); }
inline Status ArenaWrappedDBIter::PinData() { return db_iter_->PinData(); }
inline Status ArenaWrappedDBIter::GetProperty(std::string prop_name,
std::string* prop) {
return db_iter_->GetProperty(prop_name, prop);
}
inline Status ArenaWrappedDBIter::ReleasePinnedData() {
return db_iter_->ReleasePinnedData();
}
inline bool ArenaWrappedDBIter::IsKeyPinned() const {
return db_iter_->IsKeyPinned();
}
void ArenaWrappedDBIter::RegisterCleanup(CleanupFunction function, void* arg1,
void* arg2) {
db_iter_->RegisterCleanup(function, arg1, arg2);
@@ -864,7 +932,7 @@ void ArenaWrappedDBIter::RegisterCleanup(CleanupFunction function, void* arg1,
ArenaWrappedDBIter* NewArenaWrappedDbIterator(
Env* env, const ImmutableCFOptions& ioptions,
const Comparator* user_key_comparator, const SequenceNumber& sequence,
uint64_t max_sequential_skip_in_iterations,
uint64_t max_sequential_skip_in_iterations, uint64_t version_number,
const Slice* iterate_upper_bound, bool prefix_same_as_start,
bool pin_data) {
ArenaWrappedDBIter* iter = new ArenaWrappedDBIter();
@@ -872,7 +940,7 @@ ArenaWrappedDBIter* NewArenaWrappedDbIterator(
auto mem = arena->AllocateAligned(sizeof(DBIter));
DBIter* db_iter =
new (mem) DBIter(env, ioptions, user_key_comparator, nullptr, sequence,
true, max_sequential_skip_in_iterations,
true, max_sequential_skip_in_iterations, version_number,
iterate_upper_bound, prefix_same_as_start);
iter->SetDBIter(db_iter);
+10 -11
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -9,6 +9,7 @@
#pragma once
#include <stdint.h>
#include <string>
#include "rocksdb/db.h"
#include "rocksdb/iterator.h"
#include "db/dbformat.h"
@@ -24,14 +25,12 @@ class InternalIterator;
// Return a new iterator that converts internal keys (yielded by
// "*internal_iter") that were live at the specified "sequence" number
// into appropriate user keys.
extern Iterator* NewDBIterator(Env* env, const ImmutableCFOptions& options,
const Comparator* user_key_comparator,
InternalIterator* internal_iter,
const SequenceNumber& sequence,
uint64_t max_sequential_skip_in_iterations,
const Slice* iterate_upper_bound = nullptr,
bool prefix_same_as_start = false,
bool pin_data = false);
extern Iterator* NewDBIterator(
Env* env, const ImmutableCFOptions& options,
const Comparator* user_key_comparator, InternalIterator* internal_iter,
const SequenceNumber& sequence, uint64_t max_sequential_skip_in_iterations,
uint64_t version_number, const Slice* iterate_upper_bound = nullptr,
bool prefix_same_as_start = false, bool pin_data = false);
// A wrapper iterator which wraps DB Iterator and the arena, with which the DB
// iterator is supposed be allocated. This class is used as an entry point of
@@ -66,7 +65,7 @@ class ArenaWrappedDBIter : public Iterator {
void RegisterCleanup(CleanupFunction function, void* arg1, void* arg2);
virtual Status PinData();
virtual Status ReleasePinnedData();
virtual bool IsKeyPinned() const override;
virtual Status GetProperty(std::string prop_name, std::string* prop) override;
private:
DBIter* db_iter_;
@@ -77,7 +76,7 @@ class ArenaWrappedDBIter : public Iterator {
extern ArenaWrappedDBIter* NewArenaWrappedDbIterator(
Env* env, const ImmutableCFOptions& options,
const Comparator* user_key_comparator, const SequenceNumber& sequence,
uint64_t max_sequential_skip_in_iterations,
uint64_t max_sequential_skip_in_iterations, uint64_t version_number,
const Slice* iterate_upper_bound = nullptr,
bool prefix_same_as_start = false, bool pin_data = false);
+91 -97
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2014, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -181,10 +181,9 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
internal_iter->AddPut("b", "val_b");
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(
NewDBIterator(env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, 10,
options.max_sequential_skip_in_iterations));
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -215,7 +214,7 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, options.max_sequential_skip_in_iterations));
10, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -241,7 +240,8 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, options.max_sequential_skip_in_iterations, ro.iterate_upper_bound));
10, options.max_sequential_skip_in_iterations, 0,
ro.iterate_upper_bound));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -273,7 +273,8 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, options.max_sequential_skip_in_iterations, ro.iterate_upper_bound));
10, options.max_sequential_skip_in_iterations, 0,
ro.iterate_upper_bound));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -308,7 +309,8 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, options.max_sequential_skip_in_iterations, ro.iterate_upper_bound));
10, options.max_sequential_skip_in_iterations, 0,
ro.iterate_upper_bound));
db_iter->SeekToLast();
ASSERT_TRUE(!db_iter->Valid());
@@ -337,7 +339,8 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
7, options.max_sequential_skip_in_iterations, ro.iterate_upper_bound));
7, options.max_sequential_skip_in_iterations, 0,
ro.iterate_upper_bound));
SetPerfLevel(kEnableCount);
ASSERT_TRUE(GetPerfLevel() == kEnableCount);
@@ -374,7 +377,8 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
4, options.max_sequential_skip_in_iterations, ro.iterate_upper_bound));
4, options.max_sequential_skip_in_iterations, 0,
ro.iterate_upper_bound));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -399,7 +403,8 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, options.max_sequential_skip_in_iterations, ro.iterate_upper_bound));
10, options.max_sequential_skip_in_iterations, 0,
ro.iterate_upper_bound));
db_iter->SeekToLast();
ASSERT_TRUE(!db_iter->Valid());
@@ -421,7 +426,8 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, options.max_sequential_skip_in_iterations, ro.iterate_upper_bound));
10, options.max_sequential_skip_in_iterations, 0,
ro.iterate_upper_bound));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -456,7 +462,8 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
7, options.max_sequential_skip_in_iterations, ro.iterate_upper_bound));
7, options.max_sequential_skip_in_iterations, 0,
ro.iterate_upper_bound));
SetPerfLevel(kEnableCount);
ASSERT_TRUE(GetPerfLevel() == kEnableCount);
@@ -482,10 +489,9 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
internal_iter->AddPut("b", "val_b");
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(
NewDBIterator(env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, 10,
options.max_sequential_skip_in_iterations));
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToFirst();
ASSERT_TRUE(db_iter->Valid());
@@ -524,10 +530,9 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
internal_iter->AddPut("b", "val_b");
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(
NewDBIterator(env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, 2,
options.max_sequential_skip_in_iterations));
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
2, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "b");
@@ -555,10 +560,9 @@ TEST_F(DBIteratorTest, DBIteratorPrevNext) {
internal_iter->AddPut("c", "val_c");
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(
NewDBIterator(env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, 10,
options.max_sequential_skip_in_iterations));
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "c");
@@ -583,10 +587,9 @@ TEST_F(DBIteratorTest, DBIteratorEmpty) {
TestIterator* internal_iter = new TestIterator(BytewiseComparator());
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(
NewDBIterator(env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, 0,
options.max_sequential_skip_in_iterations));
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
0, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(!db_iter->Valid());
}
@@ -595,10 +598,9 @@ TEST_F(DBIteratorTest, DBIteratorEmpty) {
TestIterator* internal_iter = new TestIterator(BytewiseComparator());
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(
NewDBIterator(env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, 0,
options.max_sequential_skip_in_iterations));
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
0, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToFirst();
ASSERT_TRUE(!db_iter->Valid());
}
@@ -617,10 +619,9 @@ TEST_F(DBIteratorTest, DBIteratorUseSkipCountSkips) {
}
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(
NewDBIterator(env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, 2,
options.max_sequential_skip_in_iterations));
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter, 2,
options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "c");
@@ -659,9 +660,8 @@ TEST_F(DBIteratorTest, DBIteratorUseSkip) {
options.statistics = rocksdb::CreateDBStatistics();
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, i + 2,
options.max_sequential_skip_in_iterations));
env_, ImmutableCFOptions(options), BytewiseComparator(),
internal_iter, i + 2, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -695,9 +695,8 @@ TEST_F(DBIteratorTest, DBIteratorUseSkip) {
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, i + 2,
options.max_sequential_skip_in_iterations));
env_, ImmutableCFOptions(options), BytewiseComparator(),
internal_iter, i + 2, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -724,9 +723,8 @@ TEST_F(DBIteratorTest, DBIteratorUseSkip) {
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, 202,
options.max_sequential_skip_in_iterations));
env_, ImmutableCFOptions(options), BytewiseComparator(),
internal_iter, 202, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -756,10 +754,9 @@ TEST_F(DBIteratorTest, DBIteratorUseSkip) {
}
internal_iter->AddPut("c", "200");
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(
NewDBIterator(env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, i,
options.max_sequential_skip_in_iterations));
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(),
internal_iter, i, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(!db_iter->Valid());
@@ -773,10 +770,9 @@ TEST_F(DBIteratorTest, DBIteratorUseSkip) {
}
internal_iter->AddPut("c", "200");
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(
NewDBIterator(env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, 200,
options.max_sequential_skip_in_iterations));
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
200, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "c");
@@ -809,9 +805,8 @@ TEST_F(DBIteratorTest, DBIteratorUseSkip) {
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, i + 2,
options.max_sequential_skip_in_iterations));
env_, ImmutableCFOptions(options), BytewiseComparator(),
internal_iter, i + 2, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -844,9 +839,8 @@ TEST_F(DBIteratorTest, DBIteratorUseSkip) {
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options),
BytewiseComparator(), internal_iter, i + 2,
options.max_sequential_skip_in_iterations));
env_, ImmutableCFOptions(options), BytewiseComparator(),
internal_iter, i + 2, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -887,7 +881,7 @@ TEST_F(DBIteratorTest, DBIterator1) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter, 1,
options.max_sequential_skip_in_iterations));
options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToFirst();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -913,7 +907,7 @@ TEST_F(DBIteratorTest, DBIterator2) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter, 0,
options.max_sequential_skip_in_iterations));
options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToFirst();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -936,7 +930,7 @@ TEST_F(DBIteratorTest, DBIterator3) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter, 2,
options.max_sequential_skip_in_iterations));
options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToFirst();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -958,7 +952,7 @@ TEST_F(DBIteratorTest, DBIterator4) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter, 4,
options.max_sequential_skip_in_iterations));
options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToFirst();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -987,7 +981,7 @@ TEST_F(DBIteratorTest, DBIterator5) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
0, options.max_sequential_skip_in_iterations));
0, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1009,7 +1003,7 @@ TEST_F(DBIteratorTest, DBIterator5) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
1, options.max_sequential_skip_in_iterations));
1, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1031,7 +1025,7 @@ TEST_F(DBIteratorTest, DBIterator5) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
2, options.max_sequential_skip_in_iterations));
2, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1053,7 +1047,7 @@ TEST_F(DBIteratorTest, DBIterator5) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
3, options.max_sequential_skip_in_iterations));
3, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1075,7 +1069,7 @@ TEST_F(DBIteratorTest, DBIterator5) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
4, options.max_sequential_skip_in_iterations));
4, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1097,7 +1091,7 @@ TEST_F(DBIteratorTest, DBIterator5) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
5, options.max_sequential_skip_in_iterations));
5, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1119,7 +1113,7 @@ TEST_F(DBIteratorTest, DBIterator5) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
6, options.max_sequential_skip_in_iterations));
6, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1145,7 +1139,7 @@ TEST_F(DBIteratorTest, DBIterator6) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
0, options.max_sequential_skip_in_iterations));
0, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1167,7 +1161,7 @@ TEST_F(DBIteratorTest, DBIterator6) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
1, options.max_sequential_skip_in_iterations));
1, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1189,7 +1183,7 @@ TEST_F(DBIteratorTest, DBIterator6) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
2, options.max_sequential_skip_in_iterations));
2, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1211,7 +1205,7 @@ TEST_F(DBIteratorTest, DBIterator6) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
3, options.max_sequential_skip_in_iterations));
3, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(!db_iter->Valid());
}
@@ -1229,7 +1223,7 @@ TEST_F(DBIteratorTest, DBIterator6) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
4, options.max_sequential_skip_in_iterations));
4, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1251,7 +1245,7 @@ TEST_F(DBIteratorTest, DBIterator6) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
5, options.max_sequential_skip_in_iterations));
5, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1273,7 +1267,7 @@ TEST_F(DBIteratorTest, DBIterator6) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
6, options.max_sequential_skip_in_iterations));
6, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1311,7 +1305,7 @@ TEST_F(DBIteratorTest, DBIterator7) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
0, options.max_sequential_skip_in_iterations));
0, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1345,7 +1339,7 @@ TEST_F(DBIteratorTest, DBIterator7) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
2, options.max_sequential_skip_in_iterations));
2, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -1385,7 +1379,7 @@ TEST_F(DBIteratorTest, DBIterator7) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
4, options.max_sequential_skip_in_iterations));
4, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -1425,7 +1419,7 @@ TEST_F(DBIteratorTest, DBIterator7) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
5, options.max_sequential_skip_in_iterations));
5, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -1470,7 +1464,7 @@ TEST_F(DBIteratorTest, DBIterator7) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
6, options.max_sequential_skip_in_iterations));
6, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -1516,7 +1510,7 @@ TEST_F(DBIteratorTest, DBIterator7) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
7, options.max_sequential_skip_in_iterations));
7, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -1556,7 +1550,7 @@ TEST_F(DBIteratorTest, DBIterator7) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
9, options.max_sequential_skip_in_iterations));
9, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -1602,7 +1596,7 @@ TEST_F(DBIteratorTest, DBIterator7) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
13, options.max_sequential_skip_in_iterations));
13, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -1649,7 +1643,7 @@ TEST_F(DBIteratorTest, DBIterator7) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
14, options.max_sequential_skip_in_iterations));
14, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -1678,7 +1672,7 @@ TEST_F(DBIteratorTest, DBIterator8) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, options.max_sequential_skip_in_iterations));
10, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "b");
@@ -1707,7 +1701,7 @@ TEST_F(DBIteratorTest, DBIterator9) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, options.max_sequential_skip_in_iterations));
10, options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
@@ -1754,7 +1748,7 @@ TEST_F(DBIteratorTest, DBIterator10) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, options.max_sequential_skip_in_iterations));
10, options.max_sequential_skip_in_iterations, 0));
db_iter->Seek("c");
ASSERT_TRUE(db_iter->Valid());
@@ -1778,9 +1772,9 @@ TEST_F(DBIteratorTest, SeekToLastOccurrenceSeq0) {
internal_iter->AddPut("b", "2");
internal_iter->Finish();
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter,
10, 0 /* force seek */));
std::unique_ptr<Iterator> db_iter(
NewDBIterator(env_, ImmutableCFOptions(options), BytewiseComparator(),
internal_iter, 10, 0 /* force seek */, 0));
db_iter->SeekToFirst();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1807,7 +1801,7 @@ TEST_F(DBIteratorTest, DBIterator11) {
std::unique_ptr<Iterator> db_iter(NewDBIterator(
env_, ImmutableCFOptions(options), BytewiseComparator(), internal_iter, 1,
options.max_sequential_skip_in_iterations));
options.max_sequential_skip_in_iterations, 0));
db_iter->SeekToFirst();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "a");
@@ -1832,7 +1826,7 @@ TEST_F(DBIteratorTest, DBIterator12) {
std::unique_ptr<Iterator> db_iter(
NewDBIterator(env_, ImmutableCFOptions(options), BytewiseComparator(),
internal_iter, 10, 0));
internal_iter, 10, 0, 0));
db_iter->SeekToLast();
ASSERT_TRUE(db_iter->Valid());
ASSERT_EQ(db_iter->key().ToString(), "c");
@@ -1874,7 +1868,7 @@ class DBIterWithMergeIterTest : public testing::Test {
db_iter_.reset(NewDBIterator(env_, ImmutableCFOptions(options_),
BytewiseComparator(), merge_iter,
8 /* read data earlier than seqId 8 */,
3 /* max iterators before reseek */));
3 /* max iterators before reseek */, 0));
}
Env* env_;
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
File diff suppressed because it is too large Load Diff
+29 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -215,6 +215,34 @@ TEST_F(DBTablePropertiesTest, GetPropertiesOfTablesInRange) {
TestGetPropertiesOfTablesInRange(std::move(ranges));
}
}
TEST_F(DBTablePropertiesTest, GetColumnFamilyNameProperty) {
std::string kExtraCfName = "pikachu";
CreateAndReopenWithCF({kExtraCfName}, CurrentOptions());
// Create one table per CF, then verify it was created with the column family
// name property.
for (int cf = 0; cf < 2; ++cf) {
Put(cf, "key", "val");
Flush(cf);
TablePropertiesCollection fname_to_props;
ASSERT_OK(db_->GetPropertiesOfAllTables(handles_[cf], &fname_to_props));
ASSERT_EQ(1U, fname_to_props.size());
std::string expected_cf_name;
if (cf > 0) {
expected_cf_name = kExtraCfName;
} else {
expected_cf_name = kDefaultColumnFamilyName;
}
ASSERT_EQ(expected_cf_name,
fname_to_props.begin()->second->column_family_name);
ASSERT_EQ(cf, static_cast<uint32_t>(
fname_to_props.begin()->second->column_family_id));
}
}
} // namespace rocksdb
#endif // ROCKSDB_LITE
+46 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -654,6 +654,51 @@ TEST_F(DBTestTailingIterator, ManagedTailingIteratorSeekToSame) {
ASSERT_EQ(found, iter->key().ToString());
}
TEST_F(DBTestTailingIterator, ForwardIteratorVersionProperty) {
Options options = CurrentOptions();
options.write_buffer_size = 1000;
ReadOptions read_options;
read_options.tailing = true;
Put("foo", "bar");
uint64_t v1, v2, v3, v4;
{
std::unique_ptr<Iterator> iter(db_->NewIterator(read_options));
iter->Seek("foo");
std::string prop_value;
ASSERT_OK(iter->GetProperty("rocksdb.iterator.super-version-number",
&prop_value));
v1 = static_cast<uint64_t>(std::atoi(prop_value.c_str()));
Put("foo1", "bar1");
Flush();
ASSERT_OK(iter->GetProperty("rocksdb.iterator.super-version-number",
&prop_value));
v2 = static_cast<uint64_t>(std::atoi(prop_value.c_str()));
iter->Seek("f");
ASSERT_OK(iter->GetProperty("rocksdb.iterator.super-version-number",
&prop_value));
v3 = static_cast<uint64_t>(std::atoi(prop_value.c_str()));
ASSERT_EQ(v1, v2);
ASSERT_GT(v3, v2);
}
{
std::unique_ptr<Iterator> iter(db_->NewIterator(read_options));
iter->Seek("foo");
std::string prop_value;
ASSERT_OK(iter->GetProperty("rocksdb.iterator.super-version-number",
&prop_value));
v4 = static_cast<uint64_t>(std::atoi(prop_value.c_str()));
}
ASSERT_EQ(v3, v4);
}
} // namespace rocksdb
#endif // !defined(ROCKSDB_LITE)
+793 -1575
View File
File diff suppressed because it is too large Load Diff
+864
View File
@@ -0,0 +1,864 @@
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
//
// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file. See the AUTHORS file for names of contributors.
#include <cstdlib>
#include "db/db_test_util.h"
#include "port/stack_trace.h"
#include "rocksdb/wal_filter.h"
namespace rocksdb {
static uint64_t TestGetTickerCount(const Options& options,
Tickers ticker_type) {
return options.statistics->getTickerCount(ticker_type);
}
class DBTest2 : public DBTestBase {
public:
DBTest2() : DBTestBase("/db_test2") {}
};
class PrefixFullBloomWithReverseComparator
: public DBTestBase,
public ::testing::WithParamInterface<bool> {
public:
PrefixFullBloomWithReverseComparator()
: DBTestBase("/prefix_bloom_reverse") {}
virtual void SetUp() override { if_cache_filter_ = GetParam(); }
bool if_cache_filter_;
};
TEST_P(PrefixFullBloomWithReverseComparator,
PrefixFullBloomWithReverseComparator) {
Options options = last_options_;
options.comparator = ReverseBytewiseComparator();
options.prefix_extractor.reset(NewCappedPrefixTransform(3));
options.statistics = rocksdb::CreateDBStatistics();
BlockBasedTableOptions bbto;
if (if_cache_filter_) {
bbto.no_block_cache = false;
bbto.cache_index_and_filter_blocks = true;
bbto.block_cache = NewLRUCache(1);
}
bbto.filter_policy.reset(NewBloomFilterPolicy(10, false));
bbto.whole_key_filtering = false;
options.table_factory.reset(NewBlockBasedTableFactory(bbto));
DestroyAndReopen(options);
ASSERT_OK(dbfull()->Put(WriteOptions(), "bar123", "foo"));
ASSERT_OK(dbfull()->Put(WriteOptions(), "bar234", "foo2"));
ASSERT_OK(dbfull()->Put(WriteOptions(), "foo123", "foo3"));
dbfull()->Flush(FlushOptions());
if (bbto.block_cache) {
bbto.block_cache->EraseUnRefEntries();
}
unique_ptr<Iterator> iter(db_->NewIterator(ReadOptions()));
iter->Seek("bar345");
ASSERT_OK(iter->status());
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("bar234", iter->key().ToString());
ASSERT_EQ("foo2", iter->value().ToString());
iter->Next();
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("bar123", iter->key().ToString());
ASSERT_EQ("foo", iter->value().ToString());
iter->Seek("foo234");
ASSERT_OK(iter->status());
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("foo123", iter->key().ToString());
ASSERT_EQ("foo3", iter->value().ToString());
iter->Seek("bar");
ASSERT_OK(iter->status());
ASSERT_TRUE(!iter->Valid());
}
INSTANTIATE_TEST_CASE_P(PrefixFullBloomWithReverseComparator,
PrefixFullBloomWithReverseComparator, testing::Bool());
TEST_F(DBTest2, IteratorPropertyVersionNumber) {
Put("", "");
Iterator* iter1 = db_->NewIterator(ReadOptions());
std::string prop_value;
ASSERT_OK(
iter1->GetProperty("rocksdb.iterator.super-version-number", &prop_value));
uint64_t version_number1 =
static_cast<uint64_t>(std::atoi(prop_value.c_str()));
Put("", "");
Flush();
Iterator* iter2 = db_->NewIterator(ReadOptions());
ASSERT_OK(
iter2->GetProperty("rocksdb.iterator.super-version-number", &prop_value));
uint64_t version_number2 =
static_cast<uint64_t>(std::atoi(prop_value.c_str()));
ASSERT_GT(version_number2, version_number1);
Put("", "");
Iterator* iter3 = db_->NewIterator(ReadOptions());
ASSERT_OK(
iter3->GetProperty("rocksdb.iterator.super-version-number", &prop_value));
uint64_t version_number3 =
static_cast<uint64_t>(std::atoi(prop_value.c_str()));
ASSERT_EQ(version_number2, version_number3);
iter1->SeekToFirst();
ASSERT_OK(
iter1->GetProperty("rocksdb.iterator.super-version-number", &prop_value));
uint64_t version_number1_new =
static_cast<uint64_t>(std::atoi(prop_value.c_str()));
ASSERT_EQ(version_number1, version_number1_new);
delete iter1;
delete iter2;
delete iter3;
}
TEST_F(DBTest2, CacheIndexAndFilterWithDBRestart) {
Options options = CurrentOptions();
options.create_if_missing = true;
options.statistics = rocksdb::CreateDBStatistics();
BlockBasedTableOptions table_options;
table_options.cache_index_and_filter_blocks = true;
table_options.filter_policy.reset(NewBloomFilterPolicy(20));
options.table_factory.reset(new BlockBasedTableFactory(table_options));
CreateAndReopenWithCF({"pikachu"}, options);
Put(1, "a", "begin");
Put(1, "z", "end");
ASSERT_OK(Flush(1));
TryReopenWithColumnFamilies({"default", "pikachu"}, options);
std::string value;
value = Get(1, "a");
}
#ifndef ROCKSDB_LITE
namespace {
void ValidateKeyExistence(DB* db, const std::vector<Slice>& keys_must_exist,
const std::vector<Slice>& keys_must_not_exist) {
// Ensure that expected keys exist
std::vector<std::string> values;
if (keys_must_exist.size() > 0) {
std::vector<Status> status_list =
db->MultiGet(ReadOptions(), keys_must_exist, &values);
for (size_t i = 0; i < keys_must_exist.size(); i++) {
ASSERT_OK(status_list[i]);
}
}
// Ensure that given keys don't exist
if (keys_must_not_exist.size() > 0) {
std::vector<Status> status_list =
db->MultiGet(ReadOptions(), keys_must_not_exist, &values);
for (size_t i = 0; i < keys_must_not_exist.size(); i++) {
ASSERT_TRUE(status_list[i].IsNotFound());
}
}
}
} // namespace
TEST_F(DBTest2, WalFilterTest) {
class TestWalFilter : public WalFilter {
private:
// Processing option that is requested to be applied at the given index
WalFilter::WalProcessingOption wal_processing_option_;
// Index at which to apply wal_processing_option_
// At other indexes default wal_processing_option::kContinueProcessing is
// returned.
size_t apply_option_at_record_index_;
// Current record index, incremented with each record encountered.
size_t current_record_index_;
public:
TestWalFilter(WalFilter::WalProcessingOption wal_processing_option,
size_t apply_option_for_record_index)
: wal_processing_option_(wal_processing_option),
apply_option_at_record_index_(apply_option_for_record_index),
current_record_index_(0) {}
virtual WalProcessingOption LogRecord(const WriteBatch& batch,
WriteBatch* new_batch,
bool* batch_changed) const override {
WalFilter::WalProcessingOption option_to_return;
if (current_record_index_ == apply_option_at_record_index_) {
option_to_return = wal_processing_option_;
}
else {
option_to_return = WalProcessingOption::kContinueProcessing;
}
// Filter is passed as a const object for RocksDB to not modify the
// object, however we modify it for our own purpose here and hence
// cast the constness away.
(const_cast<TestWalFilter*>(this)->current_record_index_)++;
return option_to_return;
}
virtual const char* Name() const override { return "TestWalFilter"; }
};
// Create 3 batches with two keys each
std::vector<std::vector<std::string>> batch_keys(3);
batch_keys[0].push_back("key1");
batch_keys[0].push_back("key2");
batch_keys[1].push_back("key3");
batch_keys[1].push_back("key4");
batch_keys[2].push_back("key5");
batch_keys[2].push_back("key6");
// Test with all WAL processing options
for (int option = 0;
option < static_cast<int>(
WalFilter::WalProcessingOption::kWalProcessingOptionMax);
option++) {
Options options = OptionsForLogIterTest();
DestroyAndReopen(options);
CreateAndReopenWithCF({ "pikachu" }, options);
// Write given keys in given batches
for (size_t i = 0; i < batch_keys.size(); i++) {
WriteBatch batch;
for (size_t j = 0; j < batch_keys[i].size(); j++) {
batch.Put(handles_[0], batch_keys[i][j], DummyString(1024));
}
dbfull()->Write(WriteOptions(), &batch);
}
WalFilter::WalProcessingOption wal_processing_option =
static_cast<WalFilter::WalProcessingOption>(option);
// Create a test filter that would apply wal_processing_option at the first
// record
size_t apply_option_for_record_index = 1;
TestWalFilter test_wal_filter(wal_processing_option,
apply_option_for_record_index);
// Reopen database with option to use WAL filter
options = OptionsForLogIterTest();
options.wal_filter = &test_wal_filter;
Status status =
TryReopenWithColumnFamilies({ "default", "pikachu" }, options);
if (wal_processing_option ==
WalFilter::WalProcessingOption::kCorruptedRecord) {
assert(!status.ok());
// In case of corruption we can turn off paranoid_checks to reopen
// databse
options.paranoid_checks = false;
ReopenWithColumnFamilies({ "default", "pikachu" }, options);
}
else {
assert(status.ok());
}
// Compute which keys we expect to be found
// and which we expect not to be found after recovery.
std::vector<Slice> keys_must_exist;
std::vector<Slice> keys_must_not_exist;
switch (wal_processing_option) {
case WalFilter::WalProcessingOption::kCorruptedRecord:
case WalFilter::WalProcessingOption::kContinueProcessing: {
fprintf(stderr, "Testing with complete WAL processing\n");
// we expect all records to be processed
for (size_t i = 0; i < batch_keys.size(); i++) {
for (size_t j = 0; j < batch_keys[i].size(); j++) {
keys_must_exist.push_back(Slice(batch_keys[i][j]));
}
}
break;
}
case WalFilter::WalProcessingOption::kIgnoreCurrentRecord: {
fprintf(stderr,
"Testing with ignoring record %" ROCKSDB_PRIszt " only\n",
apply_option_for_record_index);
// We expect the record with apply_option_for_record_index to be not
// found.
for (size_t i = 0; i < batch_keys.size(); i++) {
for (size_t j = 0; j < batch_keys[i].size(); j++) {
if (i == apply_option_for_record_index) {
keys_must_not_exist.push_back(Slice(batch_keys[i][j]));
}
else {
keys_must_exist.push_back(Slice(batch_keys[i][j]));
}
}
}
break;
}
case WalFilter::WalProcessingOption::kStopReplay: {
fprintf(stderr,
"Testing with stopping replay from record %" ROCKSDB_PRIszt
"\n",
apply_option_for_record_index);
// We expect records beyond apply_option_for_record_index to be not
// found.
for (size_t i = 0; i < batch_keys.size(); i++) {
for (size_t j = 0; j < batch_keys[i].size(); j++) {
if (i >= apply_option_for_record_index) {
keys_must_not_exist.push_back(Slice(batch_keys[i][j]));
}
else {
keys_must_exist.push_back(Slice(batch_keys[i][j]));
}
}
}
break;
}
default:
assert(false); // unhandled case
}
bool checked_after_reopen = false;
while (true) {
// Ensure that expected keys exists
// and not expected keys don't exist after recovery
ValidateKeyExistence(db_, keys_must_exist, keys_must_not_exist);
if (checked_after_reopen) {
break;
}
// reopen database again to make sure previous log(s) are not used
//(even if they were skipped)
// reopn database with option to use WAL filter
options = OptionsForLogIterTest();
ReopenWithColumnFamilies({ "default", "pikachu" }, options);
checked_after_reopen = true;
}
}
}
TEST_F(DBTest2, WalFilterTestWithChangeBatch) {
class ChangeBatchHandler : public WriteBatch::Handler {
private:
// Batch to insert keys in
WriteBatch* new_write_batch_;
// Number of keys to add in the new batch
size_t num_keys_to_add_in_new_batch_;
// Number of keys added to new batch
size_t num_keys_added_;
public:
ChangeBatchHandler(WriteBatch* new_write_batch,
size_t num_keys_to_add_in_new_batch)
: new_write_batch_(new_write_batch),
num_keys_to_add_in_new_batch_(num_keys_to_add_in_new_batch),
num_keys_added_(0) {}
virtual void Put(const Slice& key, const Slice& value) override {
if (num_keys_added_ < num_keys_to_add_in_new_batch_) {
new_write_batch_->Put(key, value);
++num_keys_added_;
}
}
};
class TestWalFilterWithChangeBatch : public WalFilter {
private:
// Index at which to start changing records
size_t change_records_from_index_;
// Number of keys to add in the new batch
size_t num_keys_to_add_in_new_batch_;
// Current record index, incremented with each record encountered.
size_t current_record_index_;
public:
TestWalFilterWithChangeBatch(size_t change_records_from_index,
size_t num_keys_to_add_in_new_batch)
: change_records_from_index_(change_records_from_index),
num_keys_to_add_in_new_batch_(num_keys_to_add_in_new_batch),
current_record_index_(0) {}
virtual WalProcessingOption LogRecord(const WriteBatch& batch,
WriteBatch* new_batch,
bool* batch_changed) const override {
if (current_record_index_ >= change_records_from_index_) {
ChangeBatchHandler handler(new_batch, num_keys_to_add_in_new_batch_);
batch.Iterate(&handler);
*batch_changed = true;
}
// Filter is passed as a const object for RocksDB to not modify the
// object, however we modify it for our own purpose here and hence
// cast the constness away.
(const_cast<TestWalFilterWithChangeBatch*>(this)
->current_record_index_)++;
return WalProcessingOption::kContinueProcessing;
}
virtual const char* Name() const override {
return "TestWalFilterWithChangeBatch";
}
};
std::vector<std::vector<std::string>> batch_keys(3);
batch_keys[0].push_back("key1");
batch_keys[0].push_back("key2");
batch_keys[1].push_back("key3");
batch_keys[1].push_back("key4");
batch_keys[2].push_back("key5");
batch_keys[2].push_back("key6");
Options options = OptionsForLogIterTest();
DestroyAndReopen(options);
CreateAndReopenWithCF({ "pikachu" }, options);
// Write given keys in given batches
for (size_t i = 0; i < batch_keys.size(); i++) {
WriteBatch batch;
for (size_t j = 0; j < batch_keys[i].size(); j++) {
batch.Put(handles_[0], batch_keys[i][j], DummyString(1024));
}
dbfull()->Write(WriteOptions(), &batch);
}
// Create a test filter that would apply wal_processing_option at the first
// record
size_t change_records_from_index = 1;
size_t num_keys_to_add_in_new_batch = 1;
TestWalFilterWithChangeBatch test_wal_filter_with_change_batch(
change_records_from_index, num_keys_to_add_in_new_batch);
// Reopen database with option to use WAL filter
options = OptionsForLogIterTest();
options.wal_filter = &test_wal_filter_with_change_batch;
ReopenWithColumnFamilies({ "default", "pikachu" }, options);
// Ensure that all keys exist before change_records_from_index_
// And after that index only single key exists
// as our filter adds only single key for each batch
std::vector<Slice> keys_must_exist;
std::vector<Slice> keys_must_not_exist;
for (size_t i = 0; i < batch_keys.size(); i++) {
for (size_t j = 0; j < batch_keys[i].size(); j++) {
if (i >= change_records_from_index && j >=
num_keys_to_add_in_new_batch) {
keys_must_not_exist.push_back(Slice(batch_keys[i][j]));
}
else {
keys_must_exist.push_back(Slice(batch_keys[i][j]));
}
}
}
bool checked_after_reopen = false;
while (true) {
// Ensure that expected keys exists
// and not expected keys don't exist after recovery
ValidateKeyExistence(db_, keys_must_exist, keys_must_not_exist);
if (checked_after_reopen) {
break;
}
// reopen database again to make sure previous log(s) are not used
//(even if they were skipped)
// reopn database with option to use WAL filter
options = OptionsForLogIterTest();
ReopenWithColumnFamilies({ "default", "pikachu" }, options);
checked_after_reopen = true;
}
}
TEST_F(DBTest2, WalFilterTestWithChangeBatchExtraKeys) {
class TestWalFilterWithChangeBatchAddExtraKeys : public WalFilter {
public:
virtual WalProcessingOption LogRecord(const WriteBatch& batch,
WriteBatch* new_batch,
bool* batch_changed) const override {
*new_batch = batch;
new_batch->Put("key_extra", "value_extra");
*batch_changed = true;
return WalProcessingOption::kContinueProcessing;
}
virtual const char* Name() const override {
return "WalFilterTestWithChangeBatchExtraKeys";
}
};
std::vector<std::vector<std::string>> batch_keys(3);
batch_keys[0].push_back("key1");
batch_keys[0].push_back("key2");
batch_keys[1].push_back("key3");
batch_keys[1].push_back("key4");
batch_keys[2].push_back("key5");
batch_keys[2].push_back("key6");
Options options = OptionsForLogIterTest();
DestroyAndReopen(options);
CreateAndReopenWithCF({ "pikachu" }, options);
// Write given keys in given batches
for (size_t i = 0; i < batch_keys.size(); i++) {
WriteBatch batch;
for (size_t j = 0; j < batch_keys[i].size(); j++) {
batch.Put(handles_[0], batch_keys[i][j], DummyString(1024));
}
dbfull()->Write(WriteOptions(), &batch);
}
// Create a test filter that would add extra keys
TestWalFilterWithChangeBatchAddExtraKeys test_wal_filter_extra_keys;
// Reopen database with option to use WAL filter
options = OptionsForLogIterTest();
options.wal_filter = &test_wal_filter_extra_keys;
Status status = TryReopenWithColumnFamilies({ "default", "pikachu" },
options);
ASSERT_TRUE(status.IsNotSupported());
// Reopen without filter, now reopen should succeed - previous
// attempt to open must not have altered the db.
options = OptionsForLogIterTest();
ReopenWithColumnFamilies({ "default", "pikachu" }, options);
std::vector<Slice> keys_must_exist;
std::vector<Slice> keys_must_not_exist; // empty vector
for (size_t i = 0; i < batch_keys.size(); i++) {
for (size_t j = 0; j < batch_keys[i].size(); j++) {
keys_must_exist.push_back(Slice(batch_keys[i][j]));
}
}
ValidateKeyExistence(db_, keys_must_exist, keys_must_not_exist);
}
TEST_F(DBTest2, WalFilterTestWithColumnFamilies) {
class TestWalFilterWithColumnFamilies : public WalFilter {
private:
// column_family_id -> log_number map (provided to WALFilter)
std::map<uint32_t, uint64_t> cf_log_number_map_;
// column_family_name -> column_family_id map (provided to WALFilter)
std::map<std::string, uint32_t> cf_name_id_map_;
// column_family_name -> keys_found_in_wal map
// We store keys that are applicable to the column_family
// during recovery (i.e. aren't already flushed to SST file(s))
// for verification against the keys we expect.
std::map<uint32_t, std::vector<std::string>> cf_wal_keys_;
public:
virtual void ColumnFamilyLogNumberMap(
const std::map<uint32_t, uint64_t>& cf_lognumber_map,
const std::map<std::string, uint32_t>& cf_name_id_map) override {
cf_log_number_map_ = cf_lognumber_map;
cf_name_id_map_ = cf_name_id_map;
}
virtual WalProcessingOption LogRecordFound(unsigned long long log_number,
const std::string& log_file_name,
const WriteBatch& batch,
WriteBatch* new_batch,
bool* batch_changed) override {
class LogRecordBatchHandler : public WriteBatch::Handler {
private:
const std::map<uint32_t, uint64_t> & cf_log_number_map_;
std::map<uint32_t, std::vector<std::string>> & cf_wal_keys_;
unsigned long long log_number_;
public:
LogRecordBatchHandler(unsigned long long current_log_number,
const std::map<uint32_t, uint64_t> & cf_log_number_map,
std::map<uint32_t, std::vector<std::string>> & cf_wal_keys) :
cf_log_number_map_(cf_log_number_map),
cf_wal_keys_(cf_wal_keys),
log_number_(current_log_number){}
virtual Status PutCF(uint32_t column_family_id, const Slice& key,
const Slice& /*value*/) override {
auto it = cf_log_number_map_.find(column_family_id);
assert(it != cf_log_number_map_.end());
unsigned long long log_number_for_cf = it->second;
// If the current record is applicable for column_family_id
// (i.e. isn't flushed to SST file(s) for column_family_id)
// add it to the cf_wal_keys_ map for verification.
if (log_number_ >= log_number_for_cf) {
cf_wal_keys_[column_family_id].push_back(std::string(key.data(),
key.size()));
}
return Status::OK();
}
} handler(log_number, cf_log_number_map_, cf_wal_keys_);
batch.Iterate(&handler);
return WalProcessingOption::kContinueProcessing;
}
virtual const char* Name() const override {
return "WalFilterTestWithColumnFamilies";
}
const std::map<uint32_t, std::vector<std::string>> &
GetColumnFamilyKeys() {
return cf_wal_keys_;
}
const std::map<std::string, uint32_t> & GetColumnFamilyNameIdMap() {
return cf_name_id_map_;
}
};
std::vector<std::vector<std::string>> batch_keys_pre_flush(3);
batch_keys_pre_flush[0].push_back("key1");
batch_keys_pre_flush[0].push_back("key2");
batch_keys_pre_flush[1].push_back("key3");
batch_keys_pre_flush[1].push_back("key4");
batch_keys_pre_flush[2].push_back("key5");
batch_keys_pre_flush[2].push_back("key6");
Options options = OptionsForLogIterTest();
DestroyAndReopen(options);
CreateAndReopenWithCF({ "pikachu" }, options);
// Write given keys in given batches
for (size_t i = 0; i < batch_keys_pre_flush.size(); i++) {
WriteBatch batch;
for (size_t j = 0; j < batch_keys_pre_flush[i].size(); j++) {
batch.Put(handles_[0], batch_keys_pre_flush[i][j], DummyString(1024));
batch.Put(handles_[1], batch_keys_pre_flush[i][j], DummyString(1024));
}
dbfull()->Write(WriteOptions(), &batch);
}
//Flush default column-family
db_->Flush(FlushOptions(), handles_[0]);
// Do some more writes
std::vector<std::vector<std::string>> batch_keys_post_flush(3);
batch_keys_post_flush[0].push_back("key7");
batch_keys_post_flush[0].push_back("key8");
batch_keys_post_flush[1].push_back("key9");
batch_keys_post_flush[1].push_back("key10");
batch_keys_post_flush[2].push_back("key11");
batch_keys_post_flush[2].push_back("key12");
// Write given keys in given batches
for (size_t i = 0; i < batch_keys_post_flush.size(); i++) {
WriteBatch batch;
for (size_t j = 0; j < batch_keys_post_flush[i].size(); j++) {
batch.Put(handles_[0], batch_keys_post_flush[i][j], DummyString(1024));
batch.Put(handles_[1], batch_keys_post_flush[i][j], DummyString(1024));
}
dbfull()->Write(WriteOptions(), &batch);
}
// On Recovery we should only find the second batch applicable to default CF
// But both batches applicable to pikachu CF
// Create a test filter that would add extra keys
TestWalFilterWithColumnFamilies test_wal_filter_column_families;
// Reopen database with option to use WAL filter
options = OptionsForLogIterTest();
options.wal_filter = &test_wal_filter_column_families;
Status status =
TryReopenWithColumnFamilies({ "default", "pikachu" }, options);
ASSERT_TRUE(status.ok());
// verify that handles_[0] only has post_flush keys
// while handles_[1] has pre and post flush keys
auto cf_wal_keys = test_wal_filter_column_families.GetColumnFamilyKeys();
auto name_id_map =
test_wal_filter_column_families.GetColumnFamilyNameIdMap();
size_t index = 0;
auto keys_cf = cf_wal_keys[name_id_map[kDefaultColumnFamilyName]];
//default column-family, only post_flush keys are expected
for (size_t i = 0; i < batch_keys_post_flush.size(); i++) {
for (size_t j = 0; j < batch_keys_post_flush[i].size(); j++) {
Slice key_from_the_log(keys_cf[index++]);
Slice batch_key(batch_keys_post_flush[i][j]);
ASSERT_TRUE(key_from_the_log.compare(batch_key) == 0);
}
}
ASSERT_TRUE(index == keys_cf.size());
index = 0;
keys_cf = cf_wal_keys[name_id_map["pikachu"]];
//pikachu column-family, all keys are expected
for (size_t i = 0; i < batch_keys_pre_flush.size(); i++) {
for (size_t j = 0; j < batch_keys_pre_flush[i].size(); j++) {
Slice key_from_the_log(keys_cf[index++]);
Slice batch_key(batch_keys_pre_flush[i][j]);
ASSERT_TRUE(key_from_the_log.compare(batch_key) == 0);
}
}
for (size_t i = 0; i < batch_keys_post_flush.size(); i++) {
for (size_t j = 0; j < batch_keys_post_flush[i].size(); j++) {
Slice key_from_the_log(keys_cf[index++]);
Slice batch_key(batch_keys_post_flush[i][j]);
ASSERT_TRUE(key_from_the_log.compare(batch_key) == 0);
}
}
ASSERT_TRUE(index == keys_cf.size());
}
#endif // ROCKSDB_LITE
TEST_F(DBTest2, DISABLED_FirstSnapshotTest) {
Options options;
options.write_buffer_size = 100000; // Small write buffer
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
// This snapshot will have sequence number 0. When compaction encounters
// this snapshot, CompactionIterator::findEarliestVisibleSnapshot() will
// assert as it expects non-zero snapshots.
//
// One fix would be to simply remove this assert. However, a better fix
// might
// be to always have db sequence numbers start from 1 so that no code is
// ever
// confused by 0.
const Snapshot* s1 = db_->GetSnapshot();
Put(1, "k1", std::string(100000, 'x')); // Fill memtable
Put(1, "k2", std::string(100000, 'y')); // Trigger flush
db_->ReleaseSnapshot(s1);
}
class PinL0IndexAndFilterBlocksTest : public DBTestBase,
public testing::WithParamInterface<bool> {
public:
PinL0IndexAndFilterBlocksTest() : DBTestBase("/db_pin_l0_index_bloom_test") {}
virtual void SetUp() override { infinite_max_files_ = GetParam(); }
bool infinite_max_files_;
};
TEST_P(PinL0IndexAndFilterBlocksTest,
IndexAndFilterBlocksOfNewTableAddedToCacheWithPinning) {
Options options = CurrentOptions();
if (infinite_max_files_) {
options.max_open_files = -1;
}
options.create_if_missing = true;
options.statistics = rocksdb::CreateDBStatistics();
BlockBasedTableOptions table_options;
table_options.cache_index_and_filter_blocks = true;
table_options.pin_l0_filter_and_index_blocks_in_cache = true;
table_options.filter_policy.reset(NewBloomFilterPolicy(20));
options.table_factory.reset(new BlockBasedTableFactory(table_options));
CreateAndReopenWithCF({"pikachu"}, options);
ASSERT_OK(Put(1, "key", "val"));
// Create a new table.
ASSERT_OK(Flush(1));
// index/filter blocks added to block cache right after table creation.
ASSERT_EQ(1, TestGetTickerCount(options, BLOCK_CACHE_FILTER_MISS));
ASSERT_EQ(0, TestGetTickerCount(options, BLOCK_CACHE_FILTER_HIT));
ASSERT_EQ(1, TestGetTickerCount(options, BLOCK_CACHE_INDEX_MISS));
ASSERT_EQ(0, TestGetTickerCount(options, BLOCK_CACHE_INDEX_HIT));
// only index/filter were added
ASSERT_EQ(2, TestGetTickerCount(options, BLOCK_CACHE_ADD));
ASSERT_EQ(0, TestGetTickerCount(options, BLOCK_CACHE_DATA_MISS));
std::string value;
// Miss and hit count should remain the same, they're all pinned.
db_->KeyMayExist(ReadOptions(), handles_[1], "key", &value);
ASSERT_EQ(1, TestGetTickerCount(options, BLOCK_CACHE_FILTER_MISS));
ASSERT_EQ(0, TestGetTickerCount(options, BLOCK_CACHE_FILTER_HIT));
ASSERT_EQ(1, TestGetTickerCount(options, BLOCK_CACHE_INDEX_MISS));
ASSERT_EQ(0, TestGetTickerCount(options, BLOCK_CACHE_INDEX_HIT));
// Miss and hit count should remain the same, they're all pinned.
value = Get(1, "key");
ASSERT_EQ(1, TestGetTickerCount(options, BLOCK_CACHE_FILTER_MISS));
ASSERT_EQ(0, TestGetTickerCount(options, BLOCK_CACHE_FILTER_HIT));
ASSERT_EQ(1, TestGetTickerCount(options, BLOCK_CACHE_INDEX_MISS));
ASSERT_EQ(0, TestGetTickerCount(options, BLOCK_CACHE_INDEX_HIT));
}
TEST_P(PinL0IndexAndFilterBlocksTest,
MultiLevelIndexAndFilterBlocksCachedWithPinning) {
Options options = CurrentOptions();
if (infinite_max_files_) {
options.max_open_files = -1;
}
options.create_if_missing = true;
options.statistics = rocksdb::CreateDBStatistics();
BlockBasedTableOptions table_options;
table_options.cache_index_and_filter_blocks = true;
table_options.pin_l0_filter_and_index_blocks_in_cache = true;
table_options.filter_policy.reset(NewBloomFilterPolicy(20));
options.table_factory.reset(new BlockBasedTableFactory(table_options));
CreateAndReopenWithCF({"pikachu"}, options);
Put(1, "a", "begin");
Put(1, "z", "end");
ASSERT_OK(Flush(1));
// move this table to L1
dbfull()->TEST_CompactRange(0, nullptr, nullptr, handles_[1]);
// reset block cache
table_options.block_cache = NewLRUCache(64 * 1024);
options.table_factory.reset(NewBlockBasedTableFactory(table_options));
TryReopenWithColumnFamilies({"default", "pikachu"}, options);
// create new table at L0
Put(1, "a2", "begin2");
Put(1, "z2", "end2");
ASSERT_OK(Flush(1));
table_options.block_cache->EraseUnRefEntries();
// get base cache values
uint64_t fm = TestGetTickerCount(options, BLOCK_CACHE_FILTER_MISS);
uint64_t fh = TestGetTickerCount(options, BLOCK_CACHE_FILTER_HIT);
uint64_t im = TestGetTickerCount(options, BLOCK_CACHE_INDEX_MISS);
uint64_t ih = TestGetTickerCount(options, BLOCK_CACHE_INDEX_HIT);
std::string value;
// this should be read from L0
// so cache values don't change
value = Get(1, "a2");
ASSERT_EQ(fm, TestGetTickerCount(options, BLOCK_CACHE_FILTER_MISS));
ASSERT_EQ(fh, TestGetTickerCount(options, BLOCK_CACHE_FILTER_HIT));
ASSERT_EQ(im, TestGetTickerCount(options, BLOCK_CACHE_INDEX_MISS));
ASSERT_EQ(ih, TestGetTickerCount(options, BLOCK_CACHE_INDEX_HIT));
// this should be read from L1
// the file is opened, prefetching results in a cache filter miss
// the block is loaded and added to the cache,
// then the get results in a cache hit for L1
value = Get(1, "a");
ASSERT_EQ(fm + 1, TestGetTickerCount(options, BLOCK_CACHE_FILTER_MISS));
ASSERT_EQ(im + 1, TestGetTickerCount(options, BLOCK_CACHE_INDEX_MISS));
}
INSTANTIATE_TEST_CASE_P(PinL0IndexAndFilterBlocksTest,
PinL0IndexAndFilterBlocksTest, ::testing::Bool());
} // namespace rocksdb
int main(int argc, char** argv) {
rocksdb::port::InstallStackTraceHandler();
::testing::InitGoogleTest(&argc, argv);
return RUN_ALL_TESTS();
}
+83 -40
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -73,63 +73,69 @@ DBTestBase::~DBTestBase() {
delete env_;
}
// Switch to a fresh database with the next option configuration to
// test. Return false if there are no more configurations to test.
bool DBTestBase::ChangeOptions(int skip_mask) {
for (option_config_++; option_config_ < kEnd; option_config_++) {
bool DBTestBase::ShouldSkipOptions(int option_config, int skip_mask) {
#ifdef ROCKSDB_LITE
// These options are not supported in ROCKSDB_LITE
if (option_config_ == kHashSkipList ||
option_config_ == kPlainTableFirstBytePrefix ||
option_config_ == kPlainTableCappedPrefix ||
option_config_ == kPlainTableCappedPrefixNonMmap ||
option_config_ == kPlainTableAllBytesPrefix ||
option_config_ == kVectorRep || option_config_ == kHashLinkList ||
option_config_ == kHashCuckoo ||
option_config_ == kUniversalCompaction ||
option_config_ == kUniversalCompactionMultiLevel ||
option_config_ == kUniversalSubcompactions ||
option_config_ == kFIFOCompaction) {
continue;
if (option_config == kHashSkipList ||
option_config == kPlainTableFirstBytePrefix ||
option_config == kPlainTableCappedPrefix ||
option_config == kPlainTableCappedPrefixNonMmap ||
option_config == kPlainTableAllBytesPrefix ||
option_config == kVectorRep || option_config == kHashLinkList ||
option_config == kHashCuckoo || option_config == kUniversalCompaction ||
option_config == kUniversalCompactionMultiLevel ||
option_config == kUniversalSubcompactions ||
option_config == kFIFOCompaction ||
option_config == kConcurrentSkipList) {
return true;
}
#endif
if ((skip_mask & kSkipDeletesFilterFirst) &&
option_config_ == kDeletesFilterFirst) {
continue;
option_config == kDeletesFilterFirst) {
return true;
}
if ((skip_mask & kSkipUniversalCompaction) &&
(option_config_ == kUniversalCompaction ||
option_config_ == kUniversalCompactionMultiLevel)) {
continue;
(option_config == kUniversalCompaction ||
option_config == kUniversalCompactionMultiLevel)) {
return true;
}
if ((skip_mask & kSkipMergePut) && option_config_ == kMergePut) {
continue;
if ((skip_mask & kSkipMergePut) && option_config == kMergePut) {
return true;
}
if ((skip_mask & kSkipNoSeekToLast) &&
(option_config_ == kHashLinkList || option_config_ == kHashSkipList)) {
continue;
(option_config == kHashLinkList || option_config == kHashSkipList)) {
return true;
}
if ((skip_mask & kSkipPlainTable) &&
(option_config_ == kPlainTableAllBytesPrefix ||
option_config_ == kPlainTableFirstBytePrefix ||
option_config_ == kPlainTableCappedPrefix ||
option_config_ == kPlainTableCappedPrefixNonMmap)) {
continue;
(option_config == kPlainTableAllBytesPrefix ||
option_config == kPlainTableFirstBytePrefix ||
option_config == kPlainTableCappedPrefix ||
option_config == kPlainTableCappedPrefixNonMmap)) {
return true;
}
if ((skip_mask & kSkipHashIndex) &&
(option_config_ == kBlockBasedTableWithPrefixHashIndex ||
option_config_ == kBlockBasedTableWithWholeKeyHashIndex)) {
continue;
(option_config == kBlockBasedTableWithPrefixHashIndex ||
option_config == kBlockBasedTableWithWholeKeyHashIndex)) {
return true;
}
if ((skip_mask & kSkipHashCuckoo) && (option_config_ == kHashCuckoo)) {
continue;
if ((skip_mask & kSkipHashCuckoo) && (option_config == kHashCuckoo)) {
return true;
}
if ((skip_mask & kSkipFIFOCompaction) &&
option_config_ == kFIFOCompaction) {
continue;
if ((skip_mask & kSkipFIFOCompaction) && option_config == kFIFOCompaction) {
return true;
}
if ((skip_mask & kSkipMmapReads) && option_config_ == kWalDirAndMmapReads) {
if ((skip_mask & kSkipMmapReads) && option_config == kWalDirAndMmapReads) {
return true;
}
return false;
}
// Switch to a fresh database with the next option configuration to
// test. Return false if there are no more configurations to test.
bool DBTestBase::ChangeOptions(int skip_mask) {
for (option_config_++; option_config_ < kEnd; option_config_++) {
if (ShouldSkipOptions(option_config_, skip_mask)) {
continue;
}
break;
@@ -204,6 +210,8 @@ Options DBTestBase::CurrentOptions(
const anon::OptionsOverride& options_override) {
Options options;
options.write_buffer_size = 4090 * 4096;
options.target_file_size_base = 2 * 1024 * 1024;
options.max_bytes_for_level_base = 10 * 1024 * 1024;
return CurrentOptions(options, options_override);
}
@@ -295,6 +303,7 @@ Options DBTestBase::CurrentOptions(
case kPerfOptions:
options.soft_rate_limit = 2.0;
options.delayed_write_rate = 8 * 1024 * 1024;
options.report_bg_io_stats = true;
// TODO(3.13) -- test more options
break;
case kDeletesFilterFirst:
@@ -333,6 +342,10 @@ Options DBTestBase::CurrentOptions(
options.prefix_extractor.reset(NewNoopTransform());
break;
}
case kBlockBasedTableWithIndexRestartInterval: {
table_options.index_block_restart_interval = 8;
break;
}
case kOptimizeFiltersForHits: {
options.optimize_filters_for_hits = true;
set_block_based_table_factory = true;
@@ -356,6 +369,11 @@ Options DBTestBase::CurrentOptions(
options.max_subcompactions = 4;
break;
}
case kConcurrentSkipList: {
options.allow_concurrent_memtable_write = true;
options.enable_write_thread_adaptive_yield = true;
break;
}
default:
break;
@@ -1000,4 +1018,29 @@ void DBTestBase::CopyFile(const std::string& source,
ASSERT_OK(destfile->Close());
}
std::unordered_map<std::string, uint64_t> DBTestBase::GetAllSSTFiles(
uint64_t* total_size) {
std::unordered_map<std::string, uint64_t> res;
if (total_size) {
*total_size = 0;
}
std::vector<std::string> files;
env_->GetChildren(dbname_, &files);
for (auto& file_name : files) {
uint64_t number;
FileType type;
std::string file_path = dbname_ + "/" + file_name;
if (ParseFileName(file_name, &number, &type) && type == kTableFile) {
uint64_t file_size = 0;
env_->GetFileSize(file_path, &file_size);
res[file_path] = file_size;
if (total_size) {
*total_size += file_size;
}
}
}
return res;
}
} // namespace rocksdb
+12 -5
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -19,6 +19,7 @@
#endif
#include <algorithm>
#include <map>
#include <set>
#include <string>
#include <thread>
@@ -27,7 +28,6 @@
#include <vector>
#include "db/db_impl.h"
#include "db/db_test_util.h"
#include "db/dbformat.h"
#include "db/filename.h"
#include "memtable/hash_linklist_rep.h"
@@ -525,9 +525,11 @@ class DBTestBase : public testing::Test {
kOptimizeFiltersForHits = 27,
kRowCache = 28,
kRecycleLogFiles = 29,
kLevelSubcompactions = 30,
kUniversalSubcompactions = 31,
kEnd = 30
kConcurrentSkipList = 30,
kEnd = 31,
kLevelSubcompactions = 31,
kUniversalSubcompactions = 32,
kBlockBasedTableWithIndexRestartInterval = 33,
};
int option_config_;
@@ -573,6 +575,8 @@ class DBTestBase : public testing::Test {
return std::string(buf);
}
static bool ShouldSkipOptions(int option_config, int skip_mask = kNoSkip);
// Switch to a fresh database with the next option configuration to
// test. Return false if there are no more configurations to test.
bool ChangeOptions(int skip_mask = kNoSkip);
@@ -749,6 +753,9 @@ class DBTestBase : public testing::Test {
void CopyFile(const std::string& source, const std::string& destination,
uint64_t size = 0);
std::unordered_map<std::string, uint64_t> GetAllSSTFiles(
uint64_t* total_size = nullptr);
};
} // namespace rocksdb
+88 -34
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -14,6 +14,11 @@
namespace rocksdb {
static uint64_t TestGetTickerCount(const Options& options,
Tickers ticker_type) {
return options.statistics->getTickerCount(ticker_type);
}
static std::string CompressibleString(Random* rnd, int len) {
std::string r;
test::CompressibleString(rnd, 0.8, len, &r);
@@ -119,8 +124,7 @@ class DelayFilterFactory : public CompactionFilterFactory {
// Make sure we don't trigger a problem if the trigger conditon is given
// to be 0, which is invalid.
TEST_P(DBTestUniversalCompaction, UniversalCompactionSingleSortedRun) {
Options options;
options = CurrentOptions(options);
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleUniversal;
options.num_levels = num_levels_;
@@ -154,6 +158,71 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionSingleSortedRun) {
}
}
TEST_P(DBTestUniversalCompaction, OptimizeFiltersForHits) {
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleUniversal;
options.compaction_options_universal.size_ratio = 5;
options.num_levels = num_levels_;
options.write_buffer_size = 105 << 10; // 105KB
options.arena_block_size = 4 << 10;
options.target_file_size_base = 32 << 10; // 32KB
// trigger compaction if there are >= 4 files
options.level0_file_num_compaction_trigger = 4;
BlockBasedTableOptions bbto;
bbto.cache_index_and_filter_blocks = true;
bbto.filter_policy.reset(NewBloomFilterPolicy(10, false));
bbto.whole_key_filtering = true;
options.table_factory.reset(NewBlockBasedTableFactory(bbto));
options.optimize_filters_for_hits = true;
options.statistics = rocksdb::CreateDBStatistics();
options.memtable_factory.reset(new SpecialSkipListFactory(3));
DestroyAndReopen(options);
// block compaction from happening
env_->SetBackgroundThreads(1, Env::LOW);
test::SleepingBackgroundTask sleeping_task_low;
env_->Schedule(&test::SleepingBackgroundTask::DoSleepTask, &sleeping_task_low,
Env::Priority::LOW);
for (int num = 0; num < options.level0_file_num_compaction_trigger; num++) {
Put(Key(num * 10), "val");
if (num) {
dbfull()->TEST_WaitForFlushMemTable();
}
Put(Key(30 + num * 10), "val");
Put(Key(60 + num * 10), "val");
}
Put("", "");
dbfull()->TEST_WaitForFlushMemTable();
// Query set of non existing keys
for (int i = 5; i < 90; i += 10) {
ASSERT_EQ(Get(Key(i)), "NOT_FOUND");
}
// Make sure bloom filter is used at least once.
ASSERT_GT(TestGetTickerCount(options, BLOOM_FILTER_USEFUL), 0);
auto prev_counter = TestGetTickerCount(options, BLOOM_FILTER_USEFUL);
// Make sure bloom filter is used for all but the last L0 file when looking
// up a non-existent key that's in the range of all L0 files.
ASSERT_EQ(Get(Key(35)), "NOT_FOUND");
ASSERT_EQ(prev_counter + NumTableFilesAtLevel(0) - 1,
TestGetTickerCount(options, BLOOM_FILTER_USEFUL));
prev_counter = TestGetTickerCount(options, BLOOM_FILTER_USEFUL);
// Unblock compaction and wait it for happening.
sleeping_task_low.WakeUp();
dbfull()->TEST_WaitForCompact();
// The same queries will not trigger bloom filter
for (int i = 5; i < 90; i += 10) {
ASSERT_EQ(Get(Key(i)), "NOT_FOUND");
}
ASSERT_EQ(prev_counter, TestGetTickerCount(options, BLOOM_FILTER_USEFUL));
}
// TODO(kailiu) The tests on UniversalCompaction has some issues:
// 1. A lot of magic numbers ("11" or "12").
// 2. Made assumption on the memtable flush conditions, which may change from
@@ -266,13 +335,12 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionTrigger) {
}
TEST_P(DBTestUniversalCompaction, UniversalCompactionSizeAmplification) {
Options options;
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleUniversal;
options.num_levels = num_levels_;
options.write_buffer_size = 100 << 10; // 100KB
options.target_file_size_base = 32 << 10; // 32KB
options.level0_file_num_compaction_trigger = 3;
options = CurrentOptions(options);
DestroyAndReopen(options);
CreateAndReopenWithCF({"pikachu"}, options);
@@ -317,12 +385,12 @@ TEST_P(DBTestUniversalCompaction, CompactFilesOnUniversalCompaction) {
ChangeCompactOptions();
Options options;
options.create_if_missing = true;
options.write_buffer_size = kEntrySize * kEntriesPerBuffer;
options.compaction_style = kCompactionStyleLevel;
options.num_levels = 1;
options.target_file_size_base = options.write_buffer_size;
options.compression = kNoCompression;
options = CurrentOptions(options);
options.write_buffer_size = kEntrySize * kEntriesPerBuffer;
CreateAndReopenWithCF({"pikachu"}, options);
ASSERT_EQ(options.compaction_style, kCompactionStyleUniversal);
Random rnd(301);
@@ -380,12 +448,11 @@ TEST_P(DBTestUniversalCompaction, CompactFilesOnUniversalCompaction) {
}
TEST_P(DBTestUniversalCompaction, UniversalCompactionTargetLevel) {
Options options;
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleUniversal;
options.write_buffer_size = 100 << 10; // 100KB
options.num_levels = 7;
options.disable_auto_compactions = true;
options = CurrentOptions(options);
DestroyAndReopen(options);
// Generate 3 overlapping files
@@ -425,14 +492,13 @@ class DBTestUniversalCompactionMultiLevels
};
TEST_P(DBTestUniversalCompactionMultiLevels, UniversalCompactionMultiLevels) {
Options options;
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleUniversal;
options.num_levels = num_levels_;
options.write_buffer_size = 100 << 10; // 100KB
options.level0_file_num_compaction_trigger = 8;
options.max_background_compactions = 3;
options.target_file_size_base = 32 * 1024;
options = CurrentOptions(options);
CreateAndReopenWithCF({"pikachu"}, options);
// Trigger compaction if size amplification exceeds 110%
@@ -468,7 +534,7 @@ TEST_P(DBTestUniversalCompactionMultiLevels, UniversalCompactionTrivialMove) {
});
rocksdb::SyncPoint::GetInstance()->EnableProcessing();
Options options;
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleUniversal;
options.compaction_options_universal.allow_trivial_move = true;
options.num_levels = 3;
@@ -476,7 +542,6 @@ TEST_P(DBTestUniversalCompactionMultiLevels, UniversalCompactionTrivialMove) {
options.level0_file_num_compaction_trigger = 3;
options.max_background_compactions = 2;
options.target_file_size_base = 32 * 1024;
options = CurrentOptions(options);
DestroyAndReopen(options);
CreateAndReopenWithCF({"pikachu"}, options);
@@ -515,7 +580,7 @@ class DBTestUniversalCompactionParallel :
};
TEST_P(DBTestUniversalCompactionParallel, UniversalCompactionParallel) {
Options options;
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleUniversal;
options.num_levels = num_levels_;
options.write_buffer_size = 1 << 10; // 1KB
@@ -524,7 +589,6 @@ TEST_P(DBTestUniversalCompactionParallel, UniversalCompactionParallel) {
options.max_background_flushes = 3;
options.target_file_size_base = 1 * 1024;
options.compaction_options_universal.max_size_amplification_percent = 110;
options = CurrentOptions(options);
DestroyAndReopen(options);
CreateAndReopenWithCF({"pikachu"}, options);
@@ -581,7 +645,7 @@ INSTANTIATE_TEST_CASE_P(DBTestUniversalCompactionParallel,
::testing::Bool()));
TEST_P(DBTestUniversalCompaction, UniversalCompactionOptions) {
Options options;
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleUniversal;
options.write_buffer_size = 105 << 10; // 105KB
options.arena_block_size = 4 << 10; // 4KB
@@ -589,7 +653,6 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionOptions) {
options.level0_file_num_compaction_trigger = 4;
options.num_levels = num_levels_;
options.compaction_options_universal.compression_size_percent = -1;
options = CurrentOptions(options);
DestroyAndReopen(options);
CreateAndReopenWithCF({"pikachu"}, options);
@@ -702,14 +765,13 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionCompressRatio1) {
return;
}
Options options;
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleUniversal;
options.write_buffer_size = 100 << 10; // 100KB
options.target_file_size_base = 32 << 10; // 32KB
options.level0_file_num_compaction_trigger = 2;
options.num_levels = num_levels_;
options.compaction_options_universal.compression_size_percent = 70;
options = CurrentOptions(options);
DestroyAndReopen(options);
Random rnd(301);
@@ -770,14 +832,13 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionCompressRatio2) {
if (!Snappy_Supported()) {
return;
}
Options options;
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleUniversal;
options.write_buffer_size = 100 << 10; // 100KB
options.target_file_size_base = 32 << 10; // 32KB
options.level0_file_num_compaction_trigger = 2;
options.num_levels = num_levels_;
options.compaction_options_universal.compression_size_percent = 95;
options = CurrentOptions(options);
DestroyAndReopen(options);
Random rnd(301);
@@ -813,7 +874,7 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionTrivialMoveTest1) {
});
rocksdb::SyncPoint::GetInstance()->EnableProcessing();
Options options;
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleUniversal;
options.compaction_options_universal.allow_trivial_move = true;
options.num_levels = 2;
@@ -821,7 +882,6 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionTrivialMoveTest1) {
options.level0_file_num_compaction_trigger = 3;
options.max_background_compactions = 1;
options.target_file_size_base = 32 * 1024;
options = CurrentOptions(options);
DestroyAndReopen(options);
CreateAndReopenWithCF({"pikachu"}, options);
@@ -858,7 +918,7 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionTrivialMoveTest2) {
rocksdb::SyncPoint::GetInstance()->EnableProcessing();
Options options;
Options options = CurrentOptions();
options.compaction_style = kCompactionStyleUniversal;
options.compaction_options_universal.allow_trivial_move = true;
options.num_levels = 15;
@@ -866,7 +926,6 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionTrivialMoveTest2) {
options.level0_file_num_compaction_trigger = 8;
options.max_background_compactions = 4;
options.target_file_size_base = 64 * 1024;
options = CurrentOptions(options);
DestroyAndReopen(options);
CreateAndReopenWithCF({"pikachu"}, options);
@@ -892,7 +951,7 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionTrivialMoveTest2) {
}
TEST_P(DBTestUniversalCompaction, UniversalCompactionFourPaths) {
Options options;
Options options = CurrentOptions();
options.db_paths.emplace_back(dbname_, 300 * 1024);
options.db_paths.emplace_back(dbname_ + "_2", 300 * 1024);
options.db_paths.emplace_back(dbname_ + "_3", 500 * 1024);
@@ -905,7 +964,6 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionFourPaths) {
options.arena_block_size = 4 << 10;
options.level0_file_num_compaction_trigger = 2;
options.num_levels = 1;
options = CurrentOptions(options);
std::vector<std::string> filenames;
env_->GetChildren(options.db_paths[1].path, &filenames);
@@ -1032,16 +1090,11 @@ TEST_P(DBTestUniversalCompaction, IncreaseUniversalCompactionNumLevels) {
for (int i = 0; i <= max_key1; i++) {
// each value is 10K
ASSERT_OK(Put(1, Key(i), RandomString(&rnd, 10000)));
dbfull()->TEST_WaitForFlushMemTable(handles_[1]);
}
ASSERT_OK(Flush(1));
dbfull()->TEST_WaitForCompact();
int non_level0_num_files = 0;
for (int i = 1; i < options.num_levels; i++) {
non_level0_num_files += NumTableFilesAtLevel(i, 1);
}
ASSERT_EQ(non_level0_num_files, 0);
// Stage 2: reopen with universal compaction, num_levels=4
options.compaction_style = kCompactionStyleUniversal;
options.num_levels = 4;
@@ -1054,6 +1107,7 @@ TEST_P(DBTestUniversalCompaction, IncreaseUniversalCompactionNumLevels) {
for (int i = max_key1 + 1; i <= max_key2; i++) {
// each value is 10K
ASSERT_OK(Put(1, Key(i), RandomString(&rnd, 10000)));
dbfull()->TEST_WaitForFlushMemTable(handles_[1]);
}
ASSERT_OK(Flush(1));
dbfull()->TEST_WaitForCompact();
@@ -1084,6 +1138,7 @@ TEST_P(DBTestUniversalCompaction, IncreaseUniversalCompactionNumLevels) {
for (int i = max_key2 + 1; i <= max_key3; i++) {
// each value is 10K
ASSERT_OK(Put(1, Key(i), RandomString(&rnd, 10000)));
dbfull()->TEST_WaitForFlushMemTable(handles_[1]);
}
ASSERT_OK(Flush(1));
dbfull()->TEST_WaitForCompact();
@@ -1095,7 +1150,7 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionSecondPathRatio) {
if (!Snappy_Supported()) {
return;
}
Options options;
Options options = CurrentOptions();
options.db_paths.emplace_back(dbname_, 500 * 1024);
options.db_paths.emplace_back(dbname_ + "_2", 1024 * 1024 * 1024);
options.compaction_style = kCompactionStyleUniversal;
@@ -1107,7 +1162,6 @@ TEST_P(DBTestUniversalCompaction, UniversalCompactionSecondPathRatio) {
options.num_levels = 1;
options.memtable_factory.reset(
new SpecialSkipListFactory(KNumKeysByGenerateNewFile - 1));
options = CurrentOptions(options);
std::vector<std::string> filenames;
env_->GetChildren(options.db_paths[1].path, &filenames);
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+7 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -465,6 +465,12 @@ class InternalKeySliceTransform : public SliceTransform {
const SliceTransform* const transform_;
};
// Read the key of a record from a write batch.
// if this record represent the default column family then cf_record
// must be passed as false, otherwise it must be passed as true.
extern bool ReadKeyFromWriteBatchEntry(Slice* input, Slice* key,
bool cf_record);
// Read record from a write batch piece from input.
// tag, column_family, key, value and blob are return values. Callers own the
// Slice they point to.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+2 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -37,6 +37,7 @@ class DeleteFileTest : public testing::Test {
DeleteFileTest() {
db_ = nullptr;
env_ = Env::Default();
options_.delete_obsolete_files_period_micros = 0; // always do full purge
options_.enable_thread_tracking = true;
options_.write_buffer_size = 1024*1024*1000;
options_.target_file_size_base = 1024*1024*1000;
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2014, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2015, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+43 -12
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -66,7 +66,7 @@ FlushJob::FlushJob(const std::string& dbname, ColumnFamilyData* cfd,
JobContext* job_context, LogBuffer* log_buffer,
Directory* db_directory, Directory* output_file_directory,
CompressionType output_compression, Statistics* stats,
EventLogger* event_logger)
EventLogger* event_logger, bool measure_io_stats)
: dbname_(dbname),
cfd_(cfd),
db_options_(db_options),
@@ -83,7 +83,8 @@ FlushJob::FlushJob(const std::string& dbname, ColumnFamilyData* cfd,
output_file_directory_(output_file_directory),
output_compression_(output_compression),
stats_(stats),
event_logger_(event_logger) {
event_logger_(event_logger),
measure_io_stats_(measure_io_stats) {
// Update the thread status to indicate flush.
ReportStartedFlush();
TEST_SYNC_POINT("FlushJob::FlushJob()");
@@ -94,7 +95,8 @@ FlushJob::~FlushJob() {
}
void FlushJob::ReportStartedFlush() {
ThreadStatusUtil::SetColumnFamily(cfd_);
ThreadStatusUtil::SetColumnFamily(cfd_, cfd_->ioptions()->env,
cfd_->options()->enable_thread_tracking);
ThreadStatusUtil::SetThreadOperation(ThreadStatus::OP_FLUSH);
ThreadStatusUtil::SetThreadOperationProperty(
ThreadStatus::COMPACTION_JOB_ID,
@@ -120,6 +122,21 @@ void FlushJob::RecordFlushIOStats() {
Status FlushJob::Run(FileMetaData* file_meta) {
AutoThreadOperationStageUpdater stage_run(
ThreadStatus::STAGE_FLUSH_RUN);
// I/O measurement variables
PerfLevel prev_perf_level = PerfLevel::kEnableTime;
uint64_t prev_write_nanos = 0;
uint64_t prev_fsync_nanos = 0;
uint64_t prev_range_sync_nanos = 0;
uint64_t prev_prepare_write_nanos = 0;
if (measure_io_stats_) {
prev_perf_level = GetPerfLevel();
SetPerfLevel(PerfLevel::kEnableTime);
prev_write_nanos = IOSTATS(write_nanos);
prev_fsync_nanos = IOSTATS(fsync_nanos);
prev_range_sync_nanos = IOSTATS(range_sync_nanos);
prev_prepare_write_nanos = IOSTATS(prepare_write_nanos);
}
// Save the contents of the earliest memtable as a new Table
FileMetaData meta;
autovector<MemTable*> mems;
@@ -179,6 +196,18 @@ Status FlushJob::Run(FileMetaData* file_meta) {
}
stream.EndArray();
if (measure_io_stats_) {
if (prev_perf_level != PerfLevel::kEnableTime) {
SetPerfLevel(prev_perf_level);
}
stream << "file_write_nanos" << (IOSTATS(write_nanos) - prev_write_nanos);
stream << "file_range_sync_nanos"
<< (IOSTATS(range_sync_nanos) - prev_range_sync_nanos);
stream << "file_fsync_nanos" << (IOSTATS(fsync_nanos) - prev_fsync_nanos);
stream << "file_prepare_write_nanos"
<< (IOSTATS(prepare_write_nanos) - prev_prepare_write_nanos);
}
return s;
}
@@ -233,14 +262,15 @@ Status FlushJob::WriteLevel0Table(const autovector<MemTable*>& mems,
TEST_SYNC_POINT_CALLBACK("FlushJob::WriteLevel0Table:output_compression",
&output_compression_);
s = BuildTable(dbname_, db_options_.env, *cfd_->ioptions(), env_options_,
cfd_->table_cache(), iter.get(), meta,
cfd_->internal_comparator(),
cfd_->int_tbl_prop_collector_factories(), cfd_->GetID(),
existing_snapshots_, earliest_write_conflict_snapshot_,
output_compression_, cfd_->ioptions()->compression_opts,
mutable_cf_options_.paranoid_file_checks,
cfd_->internal_stats(), Env::IO_HIGH, &table_properties_);
s = BuildTable(
dbname_, db_options_.env, *cfd_->ioptions(), env_options_,
cfd_->table_cache(), iter.get(), meta, cfd_->internal_comparator(),
cfd_->int_tbl_prop_collector_factories(), cfd_->GetID(),
cfd_->GetName(), existing_snapshots_,
earliest_write_conflict_snapshot_, output_compression_,
cfd_->ioptions()->compression_opts,
mutable_cf_options_.paranoid_file_checks, cfd_->internal_stats(),
Env::IO_HIGH, &table_properties_, 0 /* level */);
info.table_properties = table_properties_;
LogFlush(db_options_.info_log);
}
@@ -270,6 +300,7 @@ Status FlushJob::WriteLevel0Table(const autovector<MemTable*>& mems,
if (!db_options_.disableDataSync && output_file_directory_ != nullptr) {
output_file_directory_->Fsync();
}
TEST_SYNC_POINT("FlushJob::WriteLevel0Table");
db_mutex_->Lock();
}
base->Unref();
+3 -2
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -62,7 +62,7 @@ class FlushJob {
JobContext* job_context, LogBuffer* log_buffer,
Directory* db_directory, Directory* output_file_directory,
CompressionType output_compression, Statistics* stats,
EventLogger* event_logger);
EventLogger* event_logger, bool measure_io_stats);
~FlushJob();
@@ -93,6 +93,7 @@ class FlushJob {
Statistics* stats_;
EventLogger* event_logger_;
TableProperties table_properties_;
bool measure_io_stats_;
};
} // namespace rocksdb
+8 -8
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -93,7 +93,7 @@ TEST_F(FlushJobTest, Empty) {
db_options_, *cfd->GetLatestMutableCFOptions(),
env_options_, versions_.get(), &mutex_, &shutting_down_,
{}, kMaxSequenceNumber, &job_context, nullptr, nullptr,
nullptr, kNoCompression, nullptr, &event_logger);
nullptr, kNoCompression, nullptr, &event_logger, false);
ASSERT_OK(flush_job.Run());
job_context.Clean();
}
@@ -132,7 +132,7 @@ TEST_F(FlushJobTest, NonEmpty) {
db_options_, *cfd->GetLatestMutableCFOptions(),
env_options_, versions_.get(), &mutex_, &shutting_down_,
{}, kMaxSequenceNumber, &job_context, nullptr, nullptr,
nullptr, kNoCompression, nullptr, &event_logger);
nullptr, kNoCompression, nullptr, &event_logger, true);
FileMetaData fd;
mutex_.Lock();
ASSERT_OK(flush_job.Run(&fd));
@@ -192,11 +192,11 @@ TEST_F(FlushJobTest, Snapshots) {
}
EventLogger event_logger(db_options_.info_log.get());
FlushJob flush_job(dbname_, versions_->GetColumnFamilySet()->GetDefault(),
db_options_, *cfd->GetLatestMutableCFOptions(),
env_options_, versions_.get(), &mutex_, &shutting_down_,
snapshots, kMaxSequenceNumber, &job_context, nullptr,
nullptr, nullptr, kNoCompression, nullptr, &event_logger);
FlushJob flush_job(
dbname_, versions_->GetColumnFamilySet()->GetDefault(), db_options_,
*cfd->GetLatestMutableCFOptions(), env_options_, versions_.get(), &mutex_,
&shutting_down_, snapshots, kMaxSequenceNumber, &job_context, nullptr,
nullptr, nullptr, kNoCompression, nullptr, &event_logger, true);
mutex_.Lock();
ASSERT_OK(flush_job.Run());
mutex_.Unlock();
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+14 -4
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -10,15 +10,16 @@
#include <string>
#include <utility>
#include "db/job_context.h"
#include "db/column_family.h"
#include "db/db_impl.h"
#include "db/db_iter.h"
#include "db/column_family.h"
#include "db/dbformat.h"
#include "db/job_context.h"
#include "rocksdb/env.h"
#include "rocksdb/slice.h"
#include "rocksdb/slice_transform.h"
#include "table/merger.h"
#include "db/dbformat.h"
#include "util/string_util.h"
#include "util/sync_point.h"
namespace rocksdb {
@@ -471,6 +472,15 @@ Status ForwardIterator::status() const {
return immutable_status_;
}
Status ForwardIterator::GetProperty(std::string prop_name, std::string* prop) {
assert(prop != nullptr);
if (prop_name == "rocksdb.iterator.super-version-number") {
*prop = ToString(sv_->version_number);
return Status::OK();
}
return Status::InvalidArgument();
}
void ForwardIterator::RebuildIterators(bool refresh_sv) {
// Clean up
Cleanup(refresh_sv);
+3 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
@@ -71,6 +71,8 @@ class ForwardIterator : public InternalIterator {
virtual Slice key() const override;
virtual Slice value() const override;
virtual Status status() const override;
virtual Status GetProperty(std::string prop_name, std::string* prop) override;
bool TEST_CheckDeletedIters(int* deleted_iters, int* num_iters);
private:
+1 -1
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
+21 -12
View File
@@ -1,4 +1,4 @@
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional
// grant of patent rights can be found in the PATENTS file in the same
@@ -147,8 +147,9 @@ class InlineSkipList {
// values are ok.
std::atomic<int> max_height_; // Height of the entire list
// Used for optimizing sequential insert patterns. Tricky. prev_[i] for
// i up to max_height_ - 1 (inclusive) is the predecessor of prev_[0].
// Used for optimizing sequential insert patterns. Tricky. prev_height_
// of zero means prev_ is undefined. Otherwise: prev_[i] for i up
// to max_height_ - 1 (inclusive) is the predecessor of prev_[0], and
// prev_height_ is the height of prev_[0]. prev_[0] can only be equal
// to head when max_height_ and prev_height_ are both 1.
Node** prev_;
@@ -510,11 +511,10 @@ InlineSkipList<Comparator>::AllocateNode(size_t key_size, int height) {
template <class Comparator>
void InlineSkipList<Comparator>::Insert(const char* key) {
// InsertConcurrently can't maintain the prev_ invariants when it needs
// to increase max_height_. In that case it sets prev_height_ to zero,
// letting us know that we should ignore it. A relaxed load suffices
// here because write thread synchronization separates Insert calls
// from InsertConcurrently calls.
// InsertConcurrently often can't maintain the prev_ invariants, so
// it just sets prev_height_ to zero, letting us know that we should
// ignore it. A relaxed load suffices here because write thread
// synchronization separates Insert calls from InsertConcurrently calls.
auto prev_height = prev_height_.load(std::memory_order_relaxed);
// fast path for sequential insertion
@@ -595,15 +595,24 @@ void InlineSkipList<Comparator>::InsertConcurrently(const char* key) {
int height = x->UnstashHeight();
assert(height >= 1 && height <= kMaxHeight_);
// We don't have a lock-free algorithm for updating prev_, but we do have
// the option of invalidating the entire sequential-insertion cache.
// prev_'s invariant is that prev_[i] (i > 0) is the predecessor of
// prev_[0] at that level. We're only going to violate that if height
// > 1 and key lands after prev_[height - 1] but before prev_[0].
// Comparisons are pretty expensive, so an easier version is to just
// clear the cache if height > 1. We only write to prev_height_ if the
// nobody else has, to avoid invalidating the root of the skip list in
// all of the other CPU caches.
if (height > 1 && prev_height_.load(std::memory_order_relaxed) != 0) {
prev_height_.store(0, std::memory_order_relaxed);
}
int max_height = max_height_.load(std::memory_order_relaxed);
while (height > max_height) {
if (max_height_.compare_exchange_strong(max_height, height)) {
// successfully updated it
max_height = height;
// we dont have a lock-free algorithm for fixing up prev_, so just
// mark it invalid
prev_height_.store(0, std::memory_order_relaxed);
break;
}
// else retry, possibly exiting the loop because somebody else

Some files were not shown because too many files have changed in this diff Show More