Summary:
This is an effort to club all string related utility functions into one common place, in string_util, so that it is easier for everyone to know what string processing functions are available. Right now they seem to be spread out across multiple modules, like logging and options_helper.
Check the sub-commits for easier reviewing.
Closes https://github.com/facebook/rocksdb/pull/2094
Differential Revision: D4837730
Pulled By: sagar0
fbshipit-source-id: 344278a
Summary:
Move some files under util/ to new directories env/, monitoring/ options/ and cache/
Closes https://github.com/facebook/rocksdb/pull/2090
Differential Revision: D4833681
Pulled By: siying
fbshipit-source-id: 2fd8bef
Summary:
Level-based L0->L0 compaction operates on spans of files that aren't currently being compacted. It reduces the number of L0 files, thus making write stall conditions harder to reach.
- L0->L0 is triggered when base level is unavailable due to pending compactions
- L0->L0 always outputs one file of at most `max_level0_burst_file_size` bytes.
- Subcompactions are disabled for L0->L0 since we want to output one file.
- Input files are chosen as the longest span of available files that will fit within the size limit. This minimizes number of files in L0.
Closes https://github.com/facebook/rocksdb/pull/2027
Differential Revision: D4760318
Pulled By: ajkr
fbshipit-source-id: 9d07183
Summary:
I've needed Env timing measurements a few times now, so finally built something for it.
Closes https://github.com/facebook/rocksdb/pull/2073
Differential Revision: D4811231
Pulled By: ajkr
fbshipit-source-id: 218a249
Summary:
Refactor WriteImpl() so when I plug-in the pipeline write code (which is
an alternative approach for WriteThread), some of the logic can be
reuse. I split out the following methods from WriteImpl():
* PreprocessWrite()
* HandleWALFull() (previous MaybeFlushColumnFamilies())
* HandleWriteBufferFull()
* WriteToWAL()
Also adding a constructor to WriteThread::Writer, and move WriteContext into db_impl.h.
No real logic change in this patch.
Closes https://github.com/facebook/rocksdb/pull/2042
Differential Revision: D4781014
Pulled By: yiwu-arbug
fbshipit-source-id: d45ca18
Summary:
It is confusing to have auto_roll_logger to stay under db/, which has nothing to do with database. Move filename together as it is a dependency.
Closes https://github.com/facebook/rocksdb/pull/2080
Differential Revision: D4821141
Pulled By: siying
fbshipit-source-id: ca7d768
Summary:
some fbcode services override it, we need to keep it virtual.
original change: #1756
Closes https://github.com/facebook/rocksdb/pull/2065
Differential Revision: D4808123
Pulled By: ajkr
fbshipit-source-id: 5eaeea7
Summary:
There still are many warnings (most of them about invalid printf format
for long long), but it builds if FAIL_ON_WARNINGS is disabled.
Closes https://github.com/facebook/rocksdb/pull/2052
Differential Revision: D4807355
Pulled By: siying
fbshipit-source-id: ef03786
Summary:
previously we only cleaned up .tmp files under "shared/" and "private/" directories in case the previous backup failed. we need to do the same for "shared_checksum/"; otherwise, the subsequent backup will fail if it tries to backup at least one of the same files.
Closes https://github.com/facebook/rocksdb/pull/2062
Differential Revision: D4805599
Pulled By: ajkr
fbshipit-source-id: eaa6088
Summary:
Add a parameter to Checkpoint::CreateCheckpoint() so that flush can be skipped if total log file size is within a threshold.
Closes https://github.com/facebook/rocksdb/pull/1993
Differential Revision: D4719842
Pulled By: siying
fbshipit-source-id: 4f9d9e1
Summary:
PinnableSlice
Summary:
Currently the point lookup values are copied to a string provided by the
user. This incures an extra memcpy cost. This patch allows doing point lookup
via a PinnableSlice which pins the source memory location (instead of
copying their content) and releases them after the content is consumed
by the user. The old API of Get(string) is translated to the new API
underneath.
Here is the summary for improvements:
value 100 byte: 1.8% regular, 1.2% merge values
value 1k byte: 11.5% regular, 7.5% merge values
value 10k byte: 26% regular, 29.9% merge values
The improvement for merge could be more if we extend this approach to
pin the merge output and delay the full merge operation until the user
actually needs it. We have put that for future work.
PS:
Sometimes we observe a small decrease in performance when switching from
t5452014 to this patch but with the old Get(string) API. The d
Closes https://github.com/facebook/rocksdb/pull/1756
Differential Revision: D4391738
Pulled By: maysamyabandeh
fbshipit-source-id: 6f3edd3
Summary:
This is the metric I plan to use for adaptive rate limiting. The statistics are updated only if the rate limiter is drained by flush or compaction. I believe (but am not certain) that this is the normal case.
The Statistics object is passed in RateLimiter::Request() to avoid requiring changes to client code, which would've been necessary if we passed it in the RateLimiter constructor.
Closes https://github.com/facebook/rocksdb/pull/1946
Differential Revision: D4646489
Pulled By: ajkr
fbshipit-source-id: d8e0161
Summary:
InsertPathnameToSizeBytes() is called on shared/ and shared_checksum/ directories, which only exist for certain configurations. If we try to list a non-existent directory's contents, some Envs will dump an error message. Let's avoid this by checking whether the directory exists before listing its contents.
Closes https://github.com/facebook/rocksdb/pull/1895
Differential Revision: D4596301
Pulled By: ajkr
fbshipit-source-id: c809679
Summary:
`WriteBatchWithIndex` has an incorrect implicitly-generated move constructor (it will copy the pointer causing a double-free on destruction). Just switch to `unique_ptr` so we get correct move semantics for free.
Closes https://github.com/facebook/rocksdb/pull/1899
Differential Revision: D4598896
Pulled By: ajkr
fbshipit-source-id: 2373d47
Summary:
As the last step in backup creation, the .tmp directory is renamed omitting the .tmp suffix. In case the process terminates before this, the .tmp directory will be left behind. Even if this happens, we want future backups to succeed, so I added some checks/cleanup for this case.
Closes https://github.com/facebook/rocksdb/pull/1896
Differential Revision: D4597323
Pulled By: ajkr
fbshipit-source-id: 48900d8
Summary:
Remove disableDataSync, and another similarly named disable_data_sync options.
This is being done to simplify options, and also because the performance gains of this feature can be achieved by other methods.
Closes https://github.com/facebook/rocksdb/pull/1859
Differential Revision: D4541292
Pulled By: sagar0
fbshipit-source-id: 5b3a6ca
Summary:
introduce new methods into a public threadpool interface,
- allow submission of std::functions as they allow greater flexibility.
- add Joining methods to the implementation to join scheduled and submitted jobs with
an option to cancel jobs that did not start executing.
- Remove ugly `#ifdefs` between pthread and std implementation, make it uniform.
- introduce pimpl for a drop in replacement of the implementation
- Introduce rocksdb::port::Thread typedef which is a replacement for std::thread. On Posix Thread defaults as before std::thread.
- Implement WindowsThread that allocates memory in a more controllable manner than windows std::thread with a replaceable implementation.
- should be no functionality changes.
Closes https://github.com/facebook/rocksdb/pull/1823
Differential Revision: D4492902
Pulled By: siying
fbshipit-source-id: c74cb11
Summary:
merger.h was always a confusing name for me, simply give the file a better name
Closes https://github.com/facebook/rocksdb/pull/1836
Differential Revision: D4505357
Pulled By: IslamAbdelRahman
fbshipit-source-id: 07b28d8
Summary:
The Env registration framework supports registering client Envs and selecting which one to instantiate according to a text field. This enabled things like adding the -env_uri argument to db_bench, so the same binary could be reused with different Envs just by changing CLI config.
Now this problem has come up again in a non-Env context, as I want to instantiate a client Statistics implementation from db_bench, which is configured entirely via text parameters. Also, in the future we may wish to use it for deserializing client objects when loading OPTIONS file.
This diff generalizes the Env registration logic to work with arbitrary types.
- Generalized registration and instantiation code by templating them
- The entire implementation is in a header file as that's Google style guide's recommendation for template definitions
- Pattern match with std::regex_match rather than checking prefix, which was the previous behavior
- Rename functions/files to be non-Env-specific
Closes https://github.com/facebook/rocksdb/pull/1776
Differential Revision: D4421933
Pulled By: ajkr
fbshipit-source-id: 34647d1
Summary:
using ~0UL for mask uses a uint32_t at least in MSVC, but a uint64_t is required for it to work properly
Closes https://github.com/facebook/rocksdb/pull/1777
Differential Revision: D4444004
Pulled By: yiwu-arbug
fbshipit-source-id: 057cc42
Summary:
Consider the following single column family scenario:
prepare in log A
commit in log B
*WAL is too large, flush all CFs to releast log A*
*CFA is on log B so we do not see CFA is depending on log A so no flush is requested*
To fix this we must also consider the log containing the prepare section when determining what log a CF is dependent on.
Closes https://github.com/facebook/rocksdb/pull/1768
Differential Revision: D4403265
Pulled By: reidHoruff
fbshipit-source-id: ce800ff
Summary:
also change variable name `direct_io_` to `use_direct_io_` in WritableFile to make it consistent with read path.
Closes https://github.com/facebook/rocksdb/pull/1770
Differential Revision: D4416435
Pulled By: lightmark
fbshipit-source-id: 4143c53
Summary:
Previously the only way to increment a handle's refcount was to invoke Lookup(), which (1) did hash table lookup to get cache handle, (2) incremented that handle's refcount. For a future DeleteRange optimization, I added a function, Ref(), for when the caller already has a cache handle and only needs to do (2).
Closes https://github.com/facebook/rocksdb/pull/1761
Differential Revision: D4397114
Pulled By: ajkr
fbshipit-source-id: 9addbe5
Summary:
Enable directIO on WritableFileImpl::Append
with offset being current length of the file.
Enable UniqueID tests on Windows, disable others but
leeting them to compile. Unique tests are valuable to
detect failures on different filesystems and upcoming
ReFS.
Clear output in WinEnv Getchildren.This is different from
previous strategy, do not touch output on failure.
Make sure DBTest.OpenWhenOpen works with windows error message
Closes https://github.com/facebook/rocksdb/pull/1746
Differential Revision: D4385681
Pulled By: IslamAbdelRahman
fbshipit-source-id: c07b702
Summary:
In the test the last change to AAAZZZ in handles[1] is deleting it. The
result of the get must be NotFound then. Previosuly the test did not
check for the return value of Get and assumed that the status is ok. It
then move ahead asserting the returned value. The passed-by-reference
string value however was not changed (since the key was not found) and
the asserted value is what it contained before doing the Get.
Closes https://github.com/facebook/rocksdb/pull/1753
Differential Revision: D4390982
Pulled By: maysamyabandeh
fbshipit-source-id: dd55a34
Summary:
Fix some memory leaks in the test. Also rename the test class name from DBTest to CheckpointTest to avoid confusion.
Closes https://github.com/facebook/rocksdb/pull/1752
Differential Revision: D4390355
Pulled By: siying
fbshipit-source-id: 0fa388a
Summary:
Currently the point lookup values are copied to a string provided by the user.
This incures an extra memcpy cost. This patch allows doing point lookup
via a PinnableSlice which pins the source memory location (instead of
copying their content) and releases them after the content is consumed
by the user. The old API of Get(string) is translated to the new API
underneath.
Here is the summary for improvements:
1. value 100 byte: 1.8% regular, 1.2% merge values
2. value 1k byte: 11.5% regular, 7.5% merge values
3. value 10k byte: 26% regular, 29.9% merge values
The improvement for merge could be more if we extend this approach to
pin the merge output and delay the full merge operation until the user
actually needs it. We have put that for future work.
PS:
Sometimes we observe a small decrease in performance when switching from
t5452014 to this patch but with the old Get(string) API. The difference
is a little and could be noise. More importantly it is safely
cancelled
Closes https://github.com/facebook/rocksdb/pull/1732
Differential Revision: D4374613
Pulled By: maysamyabandeh
fbshipit-source-id: a077f1a
Summary:
#1733 started using SizeFileBytes(), so our dummy log file implementation should stop asserting that this function isn't called.
Closes https://github.com/facebook/rocksdb/pull/1740
Differential Revision: D4376055
Pulled By: ajkr
fbshipit-source-id: 2854d89
Summary:
Since the backup work as snapshot, we should only copy
the bytes of the wal while we get the alive files.
Closes https://github.com/facebook/rocksdb/pull/1733
Differential Revision: D4373457
Pulled By: ajkr
fbshipit-source-id: 389318f
Summary:
File copying happens when creating checkpoints and bulkloading files from different FS partition. We should fsync the files when copying them to guarantee durability. A side effect will be that the dirty pages in file system buffers won't grow too large.
Closes https://github.com/facebook/rocksdb/pull/1728
Differential Revision: D4371083
Pulled By: siying
fbshipit-source-id: 579e14c
Summary:
If 2PC is enabled, checkpoint may not copy previous log files that contain uncommitted prepare records. In this diff we keep those files.
Closes https://github.com/facebook/rocksdb/pull/1724
Differential Revision: D4368319
Pulled By: siying
fbshipit-source-id: cc2c746
Summary:
Improve cache options logging to info log.
Also print the value of
cache_index_and_filter_blocks_with_high_priority.
Closes https://github.com/facebook/rocksdb/pull/1709
Differential Revision: D4358776
Pulled By: yiwu-arbug
fbshipit-source-id: 8f030a0
Summary:
In persistent_cache/block_cache_tier.cc, timers are never restarted, so the latency measured is not correct.
Closes https://github.com/facebook/rocksdb/pull/1707
Differential Revision: D4355828
Pulled By: siying
fbshipit-source-id: cd5f9e1
Summary:
Fixes compile error:
In file included from ./util/statistics.h:17:0,
from ./util/stop_watch.h:8,
from ./util/perf_step_timer.h:9,
from ./util/iostats_context_imp.h:8,
from ./util/posix_logger.h:27,
from ./port/util_logger.h:18,
from ./db/auto_roll_logger.h:15,
from db/auto_roll_logger.cc:6:
./util/thread_local.h:65:16: error: 'function' in namespace 'std' does not name a template type
typedef std::function<void(void*, void*)> FoldFunc;
Closes https://github.com/facebook/rocksdb/pull/1656
Differential Revision: D4318702
Pulled By: yiwu-arbug
fbshipit-source-id: 8c5d17a
Summary:
hopefully the last of the gcc-7 compile errors
Closes https://github.com/facebook/rocksdb/pull/1675
Differential Revision: D4332106
Pulled By: IslamAbdelRahman
fbshipit-source-id: 139448c
Summary:
We used to treat any failure to read a backup's meta-file as if the backup were corrupted; however, we should distinguish corruption errors from errors in the backup Env. This fixes an issue where callers would get inconsistent results from GetBackupInfo() if they called it on an engine that encountered Env error during initialization. Now we fail Initialize() in this case so callers cannot invoke GetBackupInfo() on such engines.
Closes https://github.com/facebook/rocksdb/pull/1654
Differential Revision: D4318573
Pulled By: ajkr
fbshipit-source-id: f7a7c54
Summary:
The two tests keep failing in travis. Disable them and will fix later.
Closes https://github.com/facebook/rocksdb/pull/1648
Differential Revision: D4316389
Pulled By: yiwu-arbug
fbshipit-source-id: 0a370e7
Summary:
Some users are assuming NotFound means the backup does not
exist at the provided path, which is a reasonable assumption. We need to
stop returning NotFound for system errors.
Depends on #1644
Closes https://github.com/facebook/rocksdb/pull/1645
Differential Revision: D4312233
Pulled By: ajkr
fbshipit-source-id: 5343c10
Sagar Vemuri