Summary:
Most popular versions of GCC can't identify platform on ARM if "-march=native" is specified. Remove it to unblock most people.
Closes https://github.com/facebook/rocksdb/pull/3346
Differential Revision: D6690544
Pulled By: siying
fbshipit-source-id: bbaba9fe2645b6b37144b36ea75beeff88992b49
Summary:
**# Summary**
RocksDB uses SSE crc32 intrinsics to calculate the crc32 values but it does it in single way fashion (not pipelined on single CPU core). Intel's whitepaper () published an algorithm that uses 3-way pipelining for the crc32 intrinsics, then use pclmulqdq intrinsic to combine the values. Because pclmulqdq has overhead on its own, this algorithm will show perf gains on buffers larger than 216 bytes, which makes RocksDB a perfect user, since most of the buffers RocksDB call crc32c on is over 4KB. Initial db_bench show tremendous CPU gain.
This change uses the 3-way SSE algorithm by default. The old SSE algorithm is now behind a compiler tag NO_THREEWAY_CRC32C. If user compiles the code with NO_THREEWAY_CRC32C=1 then the old SSE Crc32c algorithm would be used. If the server does not have SSE4.2 at the run time the slow way (Non SSE) will be used.
**# Performance Test Results**
We ran the FillRandom and ReadRandom benchmarks in db_bench. ReadRandom is the point of interest here since it calculates the CRC32 for the in-mem buffers. We did 3 runs for each algorithm.
Before this change the CRC32 value computation takes about 11.5% of total CPU cost, and with the new 3-way algorithm it reduced to around 4.5%. The overall throughput also improved from 25.53MB/s to 27.63MB/s.
1) ReadRandom in db_bench overall metrics
PER RUN
Algorithm | run | micros/op | ops/sec |Throughput (MB/s)
3-way | 1 | 4.143 | 241387 | 26.7
3-way | 2 | 3.775 | 264872 | 29.3
3-way | 3 | 4.116 | 242929 | 26.9
FastCrc32c|1 | 4.037 | 247727 | 27.4
FastCrc32c|2 | 4.648 | 215166 | 23.8
FastCrc32c|3 | 4.352 | 229799 | 25.4
AVG
Algorithm | Average of micros/op | Average of ops/sec | Average of Throughput (MB/s)
3-way | 4.01 | 249,729 | 27.63
FastCrc32c | 4.35 | 230,897 | 25.53
2) Crc32c computation CPU cost (inclusive samples percentage)
PER RUN
Implementation | run | TotalSamples | Crc32c percentage
3-way | 1 | 4,572,250,000 | 4.37%
3-way | 2 | 3,779,250,000 | 4.62%
3-way | 3 | 4,129,500,000 | 4.48%
FastCrc32c | 1 | 4,663,500,000 | 11.24%
FastCrc32c | 2 | 4,047,500,000 | 12.34%
FastCrc32c | 3 | 4,366,750,000 | 11.68%
**# Test Plan**
make -j64 corruption_test && ./corruption_test
By default it uses 3-way SSE algorithm
NO_THREEWAY_CRC32C=1 make -j64 corruption_test && ./corruption_test
make clean && DEBUG_LEVEL=0 make -j64 db_bench
make clean && DEBUG_LEVEL=0 NO_THREEWAY_CRC32C=1 make -j64 db_bench
Closes https://github.com/facebook/rocksdb/pull/3173
Differential Revision: D6330882
Pulled By: yingsu00
fbshipit-source-id: 8ec3d89719533b63b536a736663ca6f0dd4482e9
Summary:
I started adding gflags support for cmake on linux and got frustrated that I'd need to duplicate the build_detect_platform logic, which determines namespace based on attempting compilation. We can do it differently -- use the GFLAGS_NAMESPACE macro if available, and if not, that indicates it's an old gflags version without configurable namespace so we can simply hardcode "google".
Closes https://github.com/facebook/rocksdb/pull/3212
Differential Revision: D6456973
Pulled By: ajkr
fbshipit-source-id: 3e6d5bde3ca00d4496a120a7caf4687399f5d656
Summary:
If possible, use -march or -mcpu to get enable all features available on the local CPU or architecture. Only if this is impossible, we will manually set -msse4.2. It should be safe as there'll be a warning printed if `USE_SSE` is set and the provided flags are insufficient to support SSE4.2.
Closes https://github.com/facebook/rocksdb/pull/3156
Differential Revision: D6304703
Pulled By: ajkr
fbshipit-source-id: 030a53491263300cae7fafb429114d87acc828ef
Summary:
Add per-exe execution capability
Add fix parsing of groups/tests
Add timer test exclusion
Fix unit tests
Ifdef threadpool specific tests that do not pass on Vista threadpool.
Remove spurious outout from prefix_test so test case listing works
properly.
Fix not using standard test directories results in file creation errors
in sst_dump_test.
BlobDb fixes:
In C++ end() iterators can not be dereferenced. They are not valid.
When deleting blob_db_ set it to nullptr before any other code executes.
Not fixed:. On Windows you can not delete a file while it is open.
[ RUN ] BlobDBTest.ReadWhileGC
d:\dev\rocksdb\rocksdb\utilities\blob_db\blob_db_test.cc(75): error: DestroyBlobDB(dbname_, options, bdb_options)
IO error: Failed to delete: d:/mnt/db\testrocksdb-17444/blob_db_test/blob_dir/000001.blob: Permission denied
d:\dev\rocksdb\rocksdb\utilities\blob_db\blob_db_test.cc(75): error: DestroyBlobDB(dbname_, options, bdb_options)
IO error: Failed to delete: d:/mnt/db\testrocksdb-17444/blob_db_test/blob_dir/000001.blob: Permission denied
write_batch
Should not call front() if there is a chance the container is empty
Closes https://github.com/facebook/rocksdb/pull/3152
Differential Revision: D6293274
Pulled By: sagar0
fbshipit-source-id: 318c3717c22087fae13b18715dffb24565dbd956
Summary:
* make `checksum_type_string_map` available for lite
* comment out `FilesPerLevel` in lite mode.
* travis and legocastle lite build also build `all` target and run tests
Closes https://github.com/facebook/rocksdb/pull/3015
Differential Revision: D6069822
Pulled By: yiwu-arbug
fbshipit-source-id: 9fe92ac220e711e9e6ed4e921bd25ef4314796a0
Summary:
Update dependencies.sh. Also update tbb to 4.3, which is the latest available in TP2.
Closes https://github.com/facebook/rocksdb/pull/2812
Differential Revision: D5741394
Pulled By: yiwu-arbug
fbshipit-source-id: cafa0b7179f9a44669e5ccace818a02b42336781
Summary:
Most of the data used here in shell commands is not generated directly from user input but some data (ie: from environment variables) may have been external influenced. It is a good practice to escape this data before using it in a shell command.
Originally D4800264 but we never quite got it merged.
Reviewed By: yiwu-arbug
Differential Revision: D5595052
fbshipit-source-id: c09d8b47fe35fc6a47afb4933ccad9d56ca8d7be
Summary:
With some compilers, `-std=c++11` is necessary for <cstdint> to be
available. Pass this flag via $PLATFORM_CXXFLAGS. Fixes#2488.
Closes https://github.com/facebook/rocksdb/pull/2545
Differential Revision: D5620610
Pulled By: yiwu-arbug
fbshipit-source-id: 2f975b8c1ad52e283e677d9a33543abd064f13ce
Summary:
TSAN shows error when we grab too many locks at the same time. In TSAN crash test, make one shard key cover 2^22 keys so that no many keys will be hold at the same time.
Closes https://github.com/facebook/rocksdb/pull/2719
Differential Revision: D5609035
Pulled By: siying
fbshipit-source-id: 930e5d63fff92dbc193dc154c4c615efbdf06c6a
Summary:
Seems the only function of the script is to create a new branch, which can be done easily. I'm removing it.
Closes https://github.com/facebook/rocksdb/pull/2623
Differential Revision: D5468681
Pulled By: yiwu-arbug
fbshipit-source-id: 87dea5ecc4c85e06941ccbc36993f7f589063878
Summary:
We don't need this format compatibility hack anymore. We should remove it to make things simpler.
Closes https://github.com/facebook/rocksdb/pull/2607
Differential Revision: D5444107
Pulled By: siying
fbshipit-source-id: 7ef587dd0cacfc15a4083a137adba8e6bfddac7e
Summary:
We forgot to add the new flag in internal build script. Add it.
Closes https://github.com/facebook/rocksdb/pull/2360
Differential Revision: D5121428
Pulled By: siying
fbshipit-source-id: af72d48cd855b37df1ce3c1fbb00c80377ba6e4f
Summary:
We've had a couple CockroachDB users fail to build RocksDB on exotic platforms, so I figured I'd try my hand at solving these issues upstream. The problems stem from a) `USE_SSE=1` being too aggressive about turning on SSE4.2, even on toolchains that don't support SSE4.2 and b) RocksDB attempting to detect support for thread-local storage based on OS, even though it can vary by compiler on the same OS.
See the individual commit messages for details. Regarding SSE support, this PR should change virtually nothing for non-CMake based builds. `make`, `PORTABLE=1 make`, `USE_SSE=1 make`, and `PORTABLE=1 USE_SSE=1 make` function exactly as before, except that SSE support will be automatically disabled when a simple SSE4.2-using test program fails to compile, as it does on OpenBSD. (OpenBSD's ports GCC supports SSE4.2, but its binutils do not, so `__SSE_4_2__` is defined but an SSE4.2-using program will fail to assemble.) A warning is emitted in this case. The CMake build is modified to support the same set of options, except that `USE_SSE` is spelled `FORCE_SSE42` because `USE_SSE` is rather useless now that we can automatically detect SSE support, and I figure changing options in the CMake build is less disruptive than changing the non-CMake build.
I've tested these changes on all the platforms I can get my hands on (macOS, Windows MSVC, Windows MinGW, and OpenBSD) and it all works splendidly. Let me know if there's anything you object to—I obviously don't mean to break any of your build pipelines in the process of fixing ours downstream.
Closes https://github.com/facebook/rocksdb/pull/2199
Differential Revision: D5054042
Pulled By: yiwu-arbug
fbshipit-source-id: 938e1fc665c049c02ae15698e1409155b8e72171
Summary:
Currently, the RPM package will install the lib and header files into `/usr/package/lib` and `/usr/package/include` which is not in the default search paths. It is reasonable to install them under `/usr/lib` and `/usr/include` so that no extra configuration is required.
Closes https://github.com/facebook/rocksdb/pull/2221
Differential Revision: D5054030
Pulled By: yiwu-arbug
fbshipit-source-id: 1d23de5ff21f07e6738c9dfa04429acd7a839143
Summary:
- added a feature test in build_detect_platform to check whether sched_getcpu() is available. glibc offers it only on some platforms (e.g., linux but not mac); this way should be easier than maintaining a list of platforms on which it's available.
- refactored PhysicalCoreID() to be simpler / less repetitive. ordered the conditional compilation clauses from most-to-least preferred
Closes https://github.com/facebook/rocksdb/pull/2272
Differential Revision: D5038093
Pulled By: ajkr
fbshipit-source-id: 81d7db3cc620250de220bdeb3194b2b3d7673de7
Summary:
VALGRIND_VER was left empty after moving the environment to GCC-5. Set it back.
Closes https://github.com/facebook/rocksdb/pull/2234
Differential Revision: D4978534
Pulled By: siying
fbshipit-source-id: f0640d58e8f575f75fb3f8b92e686c9e0b6a59bb
Summary:
Replacement of #2147
The change was squashed due to a lot of conflicts.
Closes https://github.com/facebook/rocksdb/pull/2194
Differential Revision: D4929799
Pulled By: siying
fbshipit-source-id: 5cd49c254737a1d5ac13f3c035f128e86524c581
Summary:
When people are working off of a rocksdb fork, i.e. when their 'origin'
points to github.com/<username>/rocksdb, the script creates a new branch
and pushes to their origin. The new branch created by this script should
instead be pushed to github.com/facebook/rocksdb. Many people might
have named facebook/rocksdb remote as 'upstream' (or something else).
This fix provides an option to specify the remote to push the branch to.
The default is still 'origin'
More context:
When I created 5.4 branch using this script, it got pushed to sagar0/rocksdb instead of facebook/rocksdb, as I was working off of a fork. My 'origin' was pointing to sagar0/rocksdb. My 'upstream' was set to 'facebook/rocksdb'. So, I had to manually push the branch to my 'upstream'.
Closes https://github.com/facebook/rocksdb/pull/2156
Differential Revision: D4885333
Pulled By: sagar0
fbshipit-source-id: 9410eab5bd9bbefc340059800bd6b8434406729d
Summary:
Rework test running script.
New options SuiteRun - runs specified executables as google suite
test cases in parallel.
Run - this option now runs executables in parallel the same as 'tests'
RunAll - scans for test executables and attempts to run them all
as suites except those mentiones in RunOnly (hardcoded in the script)
or specified either in $ExcludeTestCases $ExcludeTestExe
Closes https://github.com/facebook/rocksdb/pull/2089
Differential Revision: D4832212
Pulled By: yiwu-arbug
fbshipit-source-id: 954990c
Summary:
Since non_shn CI was made to run in parallel, /dev/shm is automatically used. It defeated the purpose of the test to cover a non-ramfs file system.
Closes https://github.com/facebook/rocksdb/pull/2031
Differential Revision: D4764804
Pulled By: siying
fbshipit-source-id: 5666bda
Summary:
This fixes an issue when the most recent readers assume that alignment is always set even if direct io is off.
Also adjust slightly appveyor script to run db_basic_test cases concurrently.
Closes https://github.com/facebook/rocksdb/pull/1959
Differential Revision: D4671972
Pulled By: IslamAbdelRahman
fbshipit-source-id: 1886620
Summary:
The PATH update should put the Java path in the beginning, rather than the end. Otherwise, it will be overwritten. Also upgrade the Java version.
Closes https://github.com/facebook/rocksdb/pull/1912
Differential Revision: D4609854
Pulled By: siying
fbshipit-source-id: 3dc04f2
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:
The gcc-4.8.1-glibc-2.17 platform is deprecated and will be removed
soon.
Closes https://github.com/facebook/rocksdb/pull/1839
Differential Revision: D4509684
Pulled By: siying
fbshipit-source-id: 3efe296
Summary:
Update clang-format script to format diff since last commit from master,
instead of just last commit. In our common workflow we usually endup
with multiple commits for a single PR. This change make it easier to
format all stacking changes.
Closes https://github.com/facebook/rocksdb/pull/1684
Differential Revision: D4340597
Pulled By: yiwu-arbug
fbshipit-source-id: c18949e
Summary:
Previously:
$ make format
Makefile:104: Warning: Compiling in debug mode. Don't use the resulting binary in production
build_tools/format-diff.sh
You didn't have clang-format-diff.py available in your computer!
You can download it by running:
curl http://goo.gl/iUW1u2
Makefile:868: recipe for target 'format' failed
make: *** [format] Error 128
$ curl http://goo.gl/iUW1u2 > ~/bin/clang-format-diff.py
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 276 0 276 0 0 148 0 --:--:-- 0:00:01 --:--:-- 148m
$ more ~/bin/clang-format-diff.py
<HTML>
<HEAD>
<TITLE>Moved Permanently</TITLE>
</HEAD>
<BODY BGCOLOR="#FFFFFF" TEXT="#000000">
<H1>Moved Permanently</H1>
The document has moved <A HREF="https://raw.github.com/leaningtech/duetto-clang/master/tools/clang-format/clang-format-diff.py">here</A>.
</BODY>
</HTML>
Closes https://github.com/facebook/rocksdb/pull/1680
Differential Revision: D4338495
Pulled By: yiwu-arbug
fbshipit-source-id: e2b24d8
Summary:
This diff includes an implementation of CompactionFilter that allows
users to write CompactionFilter in Lua. With this ability, users can
dynamically change compaction filter logic without requiring building
the rocksdb binary and restarting the database.
To compile, WITH_LUA_PATH must be specified to the base directory
of lua.
Closes https://github.com/facebook/rocksdb/pull/1478
Differential Revision: D4150138
Pulled By: yhchiang
fbshipit-source-id: ed84222
Summary:
Rocksdb currently has many references to std::map.emplace_back()
which is not implemented in gcc 4.7, but valid in gcc 4.8. Confirmed that
it did not build with gcc 4.7, but builds fine with gcc 4.8
Closes https://github.com/facebook/rocksdb/pull/1272
Differential Revision: D4101385
Pulled By: IslamAbdelRahman
fbshipit-source-id: f6af453
Summary:
valgrind_test is the correct way to run valgrind tests.
this is becasue we need to force DISABLE_JEMALLOC
Test Plan: Running sandcastle and contrun
Reviewers: IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D65451
Summary:
ZSTD 1.0.0 is coming. We can finally add a support of ZSTD without worrying about compatibility.
Still keep ZSTDNotFinal for compatibility reason.
Test Plan: Run all tests. Run db_bench with ZSTD version with RocksDB built with ZSTD 1.0 and older.
Reviewers: andrewkr, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: cyan, igor, IslamAbdelRahman, leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D63141
