- 22 11月, 2022 6 次提交
-
-
由 anand76 提交于
Summary: Update HISTORY.md, version.h, and check_format_compatible.sh Pull Request resolved: https://github.com/facebook/rocksdb/pull/10974 Reviewed By: akankshamahajan15 Differential Revision: D41455289 Pulled By: anand1976 fbshipit-source-id: 99888ebcb9109e5ced80584a66b20123f8783c0b
-
由 Changyu Bi 提交于
Summary: before this PR, if there is a range tombstone-only file generated in penultimate level, it is marked the `last_level_temperature`. This PR fixes this issue. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10972 Test Plan: added unit test for this scenario. Reviewed By: ajkr Differential Revision: D41449215 Pulled By: cbi42 fbshipit-source-id: 1e06b5ae3bc0183db2991a45965a9807a7e8be0c
-
由 anand76 提交于
Summary: Update HISTORY.md for 7.9.0 release. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10973 Reviewed By: pdillinger Differential Revision: D41453720 Pulled By: anand1976 fbshipit-source-id: 47a23d4b6539ec6a9a09c9e69c026f7c8b10afa7
-
由 Peter Dillinger 提交于
Summary: Can simplify some ugly code in cache_dump_load_impl.cc by having an API in SecondaryCache that can directly consume persisted data. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10945 Test Plan: existing tests for CacheDumper, added basic unit test Reviewed By: anand1976 Differential Revision: D41231497 Pulled By: pdillinger fbshipit-source-id: b8ec993ef7d3e7efd68aae8602fd3f858da58068
-
由 Andrew Kryczka 提交于
Summary: We were not resetting it in non-debug mode so it could be true once and then stay true for future keys where it should be false. This PR adds the reset logic. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10967 Test Plan: - built `db_bench` with DEBUG_LEVEL=0 - ran benchmark: `TEST_TMPDIR=/dev/shm/prefix ./db_bench -benchmarks=fillrandom -compaction_style=1 -preserve_internal_time_seconds=100 -preclude_last_level_data_seconds=10 -write_buffer_size=1048576 -target_file_size_base=1048576 -subcompactions=8 -duration=120` - compared "output_to_penultimate_level: X bytes + last: Y bytes" lines in LOG output - Before this fix, Y was always zero - After this fix, Y gradually increased throughout the benchmark Reviewed By: riversand963 Differential Revision: D41417726 Pulled By: ajkr fbshipit-source-id: ace1e9a289e751a5b0c2fbaa8addd4eda5525329
-
由 Peter Dillinger 提交于
Summary: Background. One of the core risks of chosing HyperClockCache is ending up with degraded performance if estimated_entry_charge is very significantly wrong. Too low leads to under-utilized hash table, which wastes a bit of (tracked) memory and likely increases access times due to larger working set size (more TLB misses). Too high leads to fully populated hash table (at some limit with reasonable lookup performance) and not being able to cache as many objects as the memory limit would allow. In either case, performance degradation is graceful/continuous but can be quite significant. For example, cutting block size in half without updating estimated_entry_charge could lead to a large portion of configured block cache memory (up to roughly 1/3) going unused. Fix. This change adds a mechanism through which the DB periodically probes the block cache(s) for "problems" to report, and adds diagnostics to the HyperClockCache for bad estimated_entry_charge. The periodic probing is currently done with DumpStats / stats_dump_period_sec, and diagnostics reported to info_log (normally LOG file). Pull Request resolved: https://github.com/facebook/rocksdb/pull/10965 Test Plan: unit test included. Doesn't cover all the implemented subtleties of reporting, but ensures basics of when to report or not. Also manual testing with db_bench. Create db with ``` ./db_bench --benchmarks=fillrandom,flush --num=3000000 --disable_wal=1 ``` Use and check LOG file for HyperClockCache for various block sizes (used as estimated_entry_charge) ``` ./db_bench --use_existing_db --benchmarks=readrandom --num=3000000 --duration=20 --stats_dump_period_sec=8 --cache_type=hyper_clock_cache -block_size=XXXX ``` Seeing warnings / errors or not as expected. Reviewed By: anand1976 Differential Revision: D41406932 Pulled By: pdillinger fbshipit-source-id: 4ca56162b73017e4b9cec2cad74466f49c27a0a7
-
- 18 11月, 2022 2 次提交
-
-
由 Yanqin Jin 提交于
Summary: Pull Request resolved: https://github.com/facebook/rocksdb/pull/10948 Test Plan: make crash_test_with_ts Reviewed By: ltamasi Differential Revision: D41390854 Pulled By: riversand963 fbshipit-source-id: 599e114da8e2b2bbff5628fb8c67fa0393a31c05
-
由 Peter Dillinger 提交于
Summary: After a couple minor bug fixes and successful productions roll-outs in a few places, I think we can mark this as production-ready. It has a clear value proposition for many workloads, even if we don't have clear advice for every workload yet. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10963 Test Plan: existing tests, comment changes only Reviewed By: siying Differential Revision: D41384083 Pulled By: pdillinger fbshipit-source-id: 56359f01a57bb28de8697666b342382fac72ce6d
-
- 17 11月, 2022 2 次提交
-
-
由 Levi Tamasi 提交于
Summary: Pull Request resolved: https://github.com/facebook/rocksdb/pull/10959 Reviewed By: akankshamahajan15 Differential Revision: D41348198 Pulled By: ltamasi fbshipit-source-id: 51e89d03c1fe87f576a766f609a7f233a519c83d
-
由 Peter Dillinger 提交于
Summary: This was just a stepping stone to what eventually became HyperClockCache, and is now just more code to maintain. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10954 Test Plan: tests updated Reviewed By: akankshamahajan15 Differential Revision: D41310123 Pulled By: pdillinger fbshipit-source-id: 618ee148a1a0a29ee756ba8fe28359617b7cd67c
-
- 16 11月, 2022 2 次提交
-
-
由 Peter Dillinger 提交于
Summary: No material changes to code or comments, just re-arranging things to prepare for a big refactoring, making it easier to what changed. Some specifics: * This groups things together in Cache in anticipation of secondary cache features being marked production-ready (vs. experimental). * CacheEntryRole will be needed in definition of class Cache, so that has been moved above it. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10942 Test Plan: existing tests Reviewed By: anand1976 Differential Revision: D41205509 Pulled By: pdillinger fbshipit-source-id: 3f2559ab1651c758918dc97056951fa2b5eb0348
-
由 Levi Tamasi 提交于
Summary: With the recent changes, `GetMergeOperands` is now supported for wide-column entities as well, so we can use it for verification purposes in the non-batched stress tests. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10952 Test Plan: Ran a simple non-batched ops blackbox crash test. Reviewed By: riversand963 Differential Revision: D41292114 Pulled By: ltamasi fbshipit-source-id: 70b4c756a4a1fecb445c16c7096aad805a51203c
-
- 15 11月, 2022 3 次提交
-
-
由 Akanksha Mahajan 提交于
Summary: Fix db_stress failure in async_io in FilePrefetchBuffer. From the logs, assertion was caused when - prev_offset_ = offset but somehow prev_len != 0 and explicit_prefetch_submitted_ = true. That scenario is when we send async request to prefetch buffer during seek but in second seek that data is found in cache. prev_offset_ and prev_len_ get updated but we were not setting explicit_prefetch_submitted_ = false because of which buffers were getting out of sync. It's possible a read by another thread might have loaded the block into the cache in the meantime. Particular assertion example: ``` prev_offset: 0, prev_len_: 8097 , offset: 0, length: 8097, actual_length: 8097 , actual_offset: 0 , curr_: 0, bufs_[curr_].offset_: 4096 ,bufs_[curr_].CurrentSize(): 48541 , async_len_to_read: 278528, bufs_[curr_].async_in_progress_: false second: 1, bufs_[second].offset_: 282624 ,bufs_[second].CurrentSize(): 0, async_len_to_read: 262144 ,bufs_[second].async_in_progress_: true , explicit_prefetch_submitted_: true , copy_to_third_buffer: false ``` As we can see curr_ was expected to read 278528 but it read 48541. Also buffers are out of sync. Also `explicit_prefetch_submitted_` is set true but prev_len not 0. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10949 Test Plan: - Ran db_bench for regression to make sure there is no regression; - Ran db_stress failing without this fix, - Ran build-linux-mini-crashtest 7- 8 times locally + CircleCI Reviewed By: anand1976 Differential Revision: D41257786 Pulled By: akankshamahajan15 fbshipit-source-id: 1d100f94f8c06bbbe4cc76ca27f1bbc820c2494f
-
由 xiaochenfan 提交于
Summary: zlib(https://zlib.net/) has released v1.2.13. 1.2.12 is no longer available for downloading and Makefile for rocksdb will be broken due to can't find the source .tar.gz. https://nvd.nist.gov/vuln/detail/CVE-2022-37434 This pr update the version number and the shasum of new .tar.gz file. (1.2.13) Fixes https://github.com/facebook/rocksdb/issues/10876 Pull Request resolved: https://github.com/facebook/rocksdb/pull/10833 Reviewed By: hx235 Differential Revision: D40575954 Pulled By: ajkr fbshipit-source-id: 3e560e453ddf58d045214fc4e64f83bef91f22e5
-
由 akankshamahajan 提交于
Summary: Update unit test to avoid timeout Pull Request resolved: https://github.com/facebook/rocksdb/pull/10950 Reviewed By: hx235 Differential Revision: D41258892 Pulled By: akankshamahajan15 fbshipit-source-id: cbfe94da63e9e54544a307845deb79ba42458301
-
- 14 11月, 2022 1 次提交
-
-
由 anand76 提交于
Summary: Add stats for time spent in the ReadAsync call, and async read errors. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10947 Test Plan: Run db_bench and look at stats Reviewed By: akankshamahajan15 Differential Revision: D41236637 Pulled By: anand1976 fbshipit-source-id: 70539b69a28491d57acead449436a761f7108acf
-
- 12 11月, 2022 3 次提交
-
-
由 Peter Dillinger 提交于
Summary: Compressed block cache depends on reading the block compression marker beyond the payload block size. Only the payload bytes were being saved and loaded from SecondaryCache -> boom! This removes some unnecessary code attempting to combine these two competing features. Note that BlockContents was previously used for block-based filter in block cache, but that support has been removed. Also marking block_cache_compressed as deprecated in this commit as we expect it to be replaced with SecondaryCache. This problem was discovered during refactoring but didn't want to combine bug fix with that refactoring. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10944 Test Plan: test added that fails on base revision (at least with ASAN) Reviewed By: akankshamahajan15 Differential Revision: D41205578 Pulled By: pdillinger fbshipit-source-id: 1b29d36c7a6552355ac6511fcdc67038ef4af29f
-
由 Levi Tamasi 提交于
Summary: The patch extends the compaction logic to handle `Merge`s in conjunction with wide-column entities. As usual, the merge operation is applied to the anonymous default column, and any other columns are unaffected. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10946 Test Plan: `make check` Reviewed By: riversand963 Differential Revision: D41233722 Pulled By: ltamasi fbshipit-source-id: dfd9b1362222f01bafcecb139eb48480eb279fed
-
由 akankshamahajan 提交于
Summary: Fix async_io regression in scans due to incorrect check which was causing the valid data in buffer to be cleared during seek. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10939 Test Plan: - stress tests export CRASH_TEST_EXT_ARGS="--async_io=1" make crash_test -j32 - Ran db_bench command which was caught the regression: ./db_bench --db=/rocksdb_async_io_testing/prefix_scan --disable_wal=1 --use_existing_db=true --benchmarks="seekrandom" -key_size=32 -value_size=512 -num=50000000 -use_direct_reads=false -seek_nexts=963 -duration=30 -ops_between_duration_checks=1 --async_io=true --compaction_readahead_size=4194304 --log_readahead_size=0 --blob_compaction_readahead_size=0 --initial_auto_readahead_size=65536 --num_file_reads_for_auto_readahead=0 --max_auto_readahead_size=524288 seekrandom : 3777.415 micros/op 264 ops/sec 30.000 seconds 7942 operations; 132.3 MB/s (7942 of 7942 found) Reviewed By: anand1976 Differential Revision: D41173899 Pulled By: akankshamahajan15 fbshipit-source-id: 2d75b06457d65b1851c92382565d9c3fac329dfe
-
- 11 11月, 2022 2 次提交
-
-
由 Levi Tamasi 提交于
Summary: The patch adds `Merge` support for wide-column entities in `DBIter`. As before, the `Merge` operation is applied to the default column of the entity; any other columns are unchanged. As a small cleanup, the PR also changes the signature of `DBIter::Merge` to simply return a boolean instead of the `Merge` operation's `Status` since the actual `Status` is already stored in a member variable. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10941 Test Plan: `make check` Reviewed By: riversand963 Differential Revision: D41195471 Pulled By: ltamasi fbshipit-source-id: 362cf555897296e252c3de5ddfbd569ef34f85ef
-
由 Levi Tamasi 提交于
Summary: The patch untangles some nested ifs in `MergeHelper::MergeUntil`. This will come in handy when extending the compaction logic to support `Merge` for wide-column entities, and also enables us to eliminate some repeated branching on value type and to decrease the scope of some variables. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10943 Test Plan: `make check` Reviewed By: riversand963 Differential Revision: D41201946 Pulled By: ltamasi fbshipit-source-id: 890bd3d4e31cdccadca614489a94686d76485ba9
-
- 10 11月, 2022 1 次提交
-
-
由 Levi Tamasi 提交于
Summary: The patch refines/reworks `MergeHelper::TimedFullMerge(WithEntity)` a bit in two ways. First, it eliminates the recently introduced `TimedFullMerge` overload, which makes the responsibilities clearer by making sure the query result (`value` for `Get`, `columns` for `GetEntity`) is set uniformly in `SaveValue` and `GetContext`. Second, it changes the interface of `TimedFullMergeWithEntity` so it exposes its result in a serialized form; this is a more decoupled design which will come in handy when adding support for `Merge` with wide-column entities to `DBIter`. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10932 Test Plan: `make check` Reviewed By: akankshamahajan15 Differential Revision: D41129399 Pulled By: ltamasi fbshipit-source-id: 69d8da358c77d4fc7e8c40f4dafc2c129a710677
-
- 09 11月, 2022 2 次提交
-
-
由 Levi Tamasi 提交于
Summary: `DBIter::saved_value_` stores the result of any `Merge` that was performed to compute the iterator's current value. This value can be ditched whenever the iterator's position is changed, and is already cleared in `Seek`, `SeekForPrev`, `SeekToFirst`, and `SeekToLast`. With the patch, it is also cleared in `Next` and `Prev`. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10934 Test Plan: `make check` Reviewed By: akankshamahajan15 Differential Revision: D41133473 Pulled By: ltamasi fbshipit-source-id: cf9e936f48151e64e455cc1664d6e9f4a03aa308
-
由 Daniel Engel 提交于
Summary: Hello, As discussed previously in this [discussion](https://github.com/facebook/rocksdb/pull/9680#discussion_r853105163), the mentioned PR introduced a regression in portable versions that compile with MSVC - crc_3way optimization won't be used even in cases where it is supported. This PR aims to fix just that. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10667 Reviewed By: akankshamahajan15 Differential Revision: D40644592 Pulled By: ajkr fbshipit-source-id: dadbeb10d57c19800e74288258ec3b96095557dd
-
- 08 11月, 2022 3 次提交
-
-
由 Jay Zhuang 提交于
Summary: Pull Request resolved: https://github.com/facebook/rocksdb/pull/10917 Reviewed By: ajkr Differential Revision: D41070371 Pulled By: jay-zhuang fbshipit-source-id: f211aa4f9931d06a38b32042f73a5e207d996caa
-
由 Levi Tamasi 提交于
Summary: The patch fixes a bug where `GetContext::Merge` (and `MergeEntity`) does not update the ticker `READ_NUM_MERGE_OPERANDS` because it implicitly uses the default parameter value of `update_num_ops_stats=false` when calling `MergeHelper::TimedFullMerge`. Also, to prevent such issues going forward, the PR removes the default parameter values from the `TimedFullMerge` methods. In addition, it removes an unused/unnecessary parameter from `TimedFullMergeWithEntity`, and does some cleanup at the call sites of these methods. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10925 Test Plan: `make check` Reviewed By: riversand963 Differential Revision: D41096453 Pulled By: ltamasi fbshipit-source-id: fc60646d32b4d516b8fe81e265c3f020a32fd7f8
-
由 Yanqin Jin 提交于
Summary: In transaction unit tests, replace a few member variable lambdas with non-static methods. It's easier for gdb to work with variables in methods than in lambdas. (Seen similar things to https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86675). Pull Request resolved: https://github.com/facebook/rocksdb/pull/10924 Test Plan: make check Reviewed By: jay-zhuang Differential Revision: D41072241 Pulled By: riversand963 fbshipit-source-id: e4fa491de573c4656225a86a75af926c1df827f6
-
- 05 11月, 2022 2 次提交
-
-
由 Andrew Kryczka 提交于
Summary: Prevents `MemTableList::PickMemtablesToFlush()` from picking non-consecutive memtables. It leads to wrong ordering in L0 if the files are committed, or an error like below if force_consistency_checks=true catches it: ``` Corruption: force_consistency_checks: VersionBuilder: L0 file https://github.com/facebook/rocksdb/issues/25 with seqno 320416 368066 vs. file https://github.com/facebook/rocksdb/issues/24 with seqno 336037 352068 ``` Pull Request resolved: https://github.com/facebook/rocksdb/pull/10921 Test Plan: fix the expectation in the existing test of this behavior Reviewed By: riversand963 Differential Revision: D41046935 Pulled By: ajkr fbshipit-source-id: 783696bff56115063d5dc5856dfaed6a9881d1ab
-
由 anand76 提交于
Summary: Add the new benchmark Pull Request resolved: https://github.com/facebook/rocksdb/pull/10919 Reviewed By: akankshamahajan15 Differential Revision: D41017025 Pulled By: anand1976 fbshipit-source-id: 5220815d66de1f689b7f09d9c5266cebf4e345d1
-
- 04 11月, 2022 1 次提交
-
-
由 Yanqin Jin 提交于
Summary: When performing Merge during range scan, iterator should understand value types of kDeletionWithTimestamp. Also add an additional check in debug mode to MergeHelper, and account for the presence of compaction filter. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10915 Test Plan: make check Reviewed By: ltamasi Differential Revision: D40960039 Pulled By: riversand963 fbshipit-source-id: dd79d86d7c79d05755bb939a3d94e0c53ddd7f59
-
- 03 11月, 2022 8 次提交
-
-
由 Levi Tamasi 提交于
Summary: The patch adds `Merge` support for wide-column entities to the point lookup APIs, i.e. `Get`, `MultiGet`, `GetEntity`, and `GetMergeOperands`. (I plan to update the iterator and compaction logic in separate PRs.) In terms of semantics, the `Merge` operation is applied to the default (anonymous) column; any other columns in the entity are unaffected. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10916 Test Plan: `make check` Reviewed By: riversand963 Differential Revision: D40962311 Pulled By: ltamasi fbshipit-source-id: 244bc9d172be1af2f204796b2f89104e4d2fa373
-
由 Peter Dillinger 提交于
Summary: For clean-up and in preparation for some other anticipated changes, including * A new dynamically-scaling variant of HyperClockCache * SecondaryCache support for HyperClockCache This change does some refactoring for current and future code sharing and reusability. (Including follow-up on https://github.com/facebook/rocksdb/issues/10843) ## clock_cache.h * TBD whether new variant will be a HyperClockCache or use some other name, so namespace is just clock_cache for the family of structures. * A number of helper functions introduced and used. * Pre-emptively split ClockHandle (shared among lock-free clock cache variants) and HandleImpl (specific to a kind of Table), and introduce template to plug new Table implementation into ClockCacheShard. ## clock_cache.cc * Mostly using helper functions. Some things like `Rollback()` and `FreeDataMarkEmpty()` were not combined because `Rollback()` is Table-specific while `FreeDataMarkEmpty()` can be used with different table implementations. * Performance testing indicated that despite more opportunities for parallelism, making a local copy of handle data for processing after marking an entry empty was slower than doing that processing before marking the entry empty (but after marking it "under construction"), thus avoiding a few words of copying data. At least for now, this answers the "TODO? Delay freeing?" questions (no). Pull Request resolved: https://github.com/facebook/rocksdb/pull/10887 Test Plan: fixed a unit testing gap; other minor test updates for refactoring No functionality change ## Performance Same setup as https://github.com/facebook/rocksdb/issues/10801: Before: `readrandom [AVG 81 runs] : 627992 (± 5124) ops/sec` After: `readrandom [AVG 81 runs] : 637512 (± 4866) ops/sec` I've been getting some inconsistent results on restarts like the system is not being fair to the two processes, so I'm not sure there's such a real difference. Reviewed By: anand1976 Differential Revision: D40959240 Pulled By: pdillinger fbshipit-source-id: 0a8f3646b3bdb5bc7aaad60b26790b0779189949
-
由 Tal Zussman 提交于
Summary: Pull Request resolved: https://github.com/facebook/rocksdb/pull/10825 Reviewed By: akankshamahajan15 Differential Revision: D40419234 Pulled By: ajkr fbshipit-source-id: 2d700154eb5b2943d10a0f944f2b414ece353e4a
-
由 Yanqin Jin 提交于
Summary: With TSAN build on CircleCI (see mini-tsan in .circleci/config). Sometimes `SeqAdvanceConcurrentTest.SeqAdvanceConcurrent` will get stuck when an experimental feature called "unordered write" is enabled. Stack trace will be the following ``` Thread 7 (Thread 0x7f2284a1c700 (LWP 481523) "write_prepared_"): #0 0x00000000004fa3f5 in __tsan_atomic64_load () at ./db/merge_context.h:15 https://github.com/facebook/rocksdb/issues/1 0x00000000005e5942 in std::__atomic_base<unsigned long>::load (this=0x7b74000012f8, __m=std::memory_order_seq_cst) at /usr/bin/../lib/gcc/x86_64-linux-gnu/11/../../../../include/c++/11/bits/atomic_base.h:481 https://github.com/facebook/rocksdb/issues/2 std::__atomic_base<unsigned long>::operator unsigned long (this=0x7b74000012f8) at /usr/bin/../lib/gcc/x86_64-linux-gnu/11/../../../../include/c++/11/bits/atomic_base.h:341 https://github.com/facebook/rocksdb/issues/3 0x00000000005bf001 in rocksdb::SeqAdvanceConcurrentTest_SeqAdvanceConcurrent_Test::TestBody()::$_9::operator()(void*) const (this=0x7b14000085e8) at utilities/transactions/write_prepared_transaction_test.cc:1702 Thread 6 (Thread 0x7f228421b700 (LWP 481521) "write_prepared_"): #0 0x000000000052178c in __tsan::MetaMap::GetAndLock(__tsan::ThreadState*, unsigned long, unsigned long, bool, bool) () at ./db/merge_context.h:15 https://github.com/facebook/rocksdb/issues/1 0x00000000004fa48e in __tsan_atomic64_load () at ./db/merge_context.h:15 https://github.com/facebook/rocksdb/issues/2 0x00000000005e5942 in std::__atomic_base<unsigned long>::load (this=0x7b74000012f8, __m=std::memory_order_seq_cst) at /usr/bin/../lib/gcc/x86_64-linux-gnu/11/../../../../include/c++/11/bits/atomic_base.h:481 https://github.com/facebook/rocksdb/issues/3 std::__atomic_base<unsigned long>::operator unsigned long (this=0x7b74000012f8) at /usr/bin/../lib/gcc/x86_64-linux-gnu/11/../../../../include/c++/11/bits/atomic_base.h:341 https://github.com/facebook/rocksdb/issues/4 0x00000000005bf001 in rocksdb::SeqAdvanceConcurrentTest_SeqAdvanceConcurrent_Test::TestBody()::$_9::operator()(void*) const (this=0x7b14000085e8) at utilities/transactions/write_prepared_transaction_test.cc:1702 ``` This is problematic and suspicious. Two threads will get stuck in the same place trying to load from an atomic variable. https://github.com/facebook/rocksdb/blob/7.8.fb/utilities/transactions/write_prepared_transaction_test.cc#L1694:L1707. Not sure why two threads can reach the same point. The stack trace shows that there may be a deadlock, since the two threads are on the same write thread (one is doing Prepare, while the other is trying to commit). Pull Request resolved: https://github.com/facebook/rocksdb/pull/10909 Test Plan: On CircleCI mini-tsan, apply a patch first so that we have a higher chance of hitting the same problematic situation, ``` diff --git a/utilities/transactions/write_prepared_transaction_test.cc b/utilities/transactions/write_prepared_transaction_test.cc index 4bc1f3744..bd5dc4924 100644 --- a/utilities/transactions/write_prepared_transaction_test.cc +++ b/utilities/transactions/write_prepared_transaction_test.cc @@ -1714,13 +1714,13 @@ TEST_P(SeqAdvanceConcurrentTest, SeqAdvanceConcurrent) { size_t d = (n % base[bi + 1]) / base[bi]; switch (d) { case 0: - threads.emplace_back(txn_t0, bi); + threads.emplace_back(txn_t3, bi); break; case 1: - threads.emplace_back(txn_t1, bi); + threads.emplace_back(txn_t3, bi); break; case 2: - threads.emplace_back(txn_t2, bi); + threads.emplace_back(txn_t3, bi); break; case 3: threads.emplace_back(txn_t3, bi); ``` then build and run tests ``` COMPILE_WITH_TSAN=1 CC=clang-13 CXX=clang++-13 ROCKSDB_DISABLE_ALIGNED_NEW=1 USE_CLANG=1 make V=1 -j32 check gtest-parallel -r 100 ./write_prepared_transaction_test --gtest_filter=TwoWriteQueues/SeqAdvanceConcurrentTest.SeqAdvanceConcurrent/19 ``` In the above, `SeqAdvanceConcurrent/19`. The tests 10 to 19 correspond to unordered write in which Prepare() and Commit() can both enter the same write thread. Before this PR, there is a high chance of hitting the deadlock. With this PR, no deadlock has been encountered so far. Reviewed By: ltamasi Differential Revision: D40869387 Pulled By: riversand963 fbshipit-source-id: 81e82a70c263e4f3417597a201b081ee54f1deab
-
由 Brord van Wierst 提交于
Summary: Cross compiling rocksdb with rust bindings to android leads to an error since 7.4.0 (Incusion of madvise) This is due to missing placeholders for non-linux platforms. This PR adds the missing placeholders. See https://github.com/rust-rocksdb/rust-rocksdb/issues/697 for the specific error thrown. I have just completed the CLA :) Pull Request resolved: https://github.com/facebook/rocksdb/pull/10881 Reviewed By: akankshamahajan15 Differential Revision: D40726103 Pulled By: ajkr fbshipit-source-id: 6b391636a74ef7e20d0daf47d332ddf0c14d5c34
-
由 Adam Retter 提交于
Summary: The user may override the detection of whether to use GNU libc (the default) or musl libc by setting the environment variable: `ROCKSDB_MUSL_LIBC=true`. Builds upon and supersedes: https://github.com/facebook/rocksdb/pull/9977 Pull Request resolved: https://github.com/facebook/rocksdb/pull/10889 Reviewed By: akankshamahajan15 Differential Revision: D40788431 Pulled By: ajkr fbshipit-source-id: ef594d973fc14cbadf28bfb38434231a18a2107c
-
由 Brad Smith 提交于
Summary: Pull Request resolved: https://github.com/facebook/rocksdb/pull/10902 Reviewed By: akankshamahajan15 Differential Revision: D40839659 Pulled By: ajkr fbshipit-source-id: 06be5919622f8cce1fce1097c5e654900bf7f8fb
-
由 Andrew Kryczka 提交于
Summary: Ran `find ./db/ -type f | xargs clang-format -i`. Excluded minor changes it tried to make on db/db_impl/. Everything else it changed was directly under db/ directory. Included minor manual touchups mentioned in PR commit history. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10910 Reviewed By: riversand963 Differential Revision: D40880683 Pulled By: ajkr fbshipit-source-id: cfe26cda05b3fb9a72e3cb82c286e21d8c5c4174
-
- 02 11月, 2022 1 次提交
-
-
由 akankshamahajan 提交于
Summary: Fix memory corruption error in scans if async_io is enabled. Memory corruption happened if data is overlapping between two buffers. If there is IOError while reading the data, it leads to empty buffer and other buffer already in progress of async read goes again for reading causing the error. Fix: Added check to abort IO in second buffer if curr_ got empty. This PR also fixes db_stress failures which happened when buffers are not aligned. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10890 Test Plan: - Ran make crash_test -j32 with async_io enabled. - Ran benchmarks to make sure there is no regression. Reviewed By: anand1976 Differential Revision: D40881731 Pulled By: akankshamahajan15 fbshipit-source-id: 39fcf2134c7b1bbb08415ede3e1ef261ac2dbc58
-
- 01 11月, 2022 1 次提交
-
-
由 Yanqin Jin 提交于
Summary: This PR implements the originally disabled `Merge()` APIs when user-defined timestamp is enabled. Simplest usage: ```cpp // assume string append merge op is used with '.' as delimiter. // ts1 < ts2 db->Put(WriteOptions(), "key", ts1, "v0"); db->Merge(WriteOptions(), "key", ts2, "1"); ReadOptions ro; ro.timestamp = &ts2; db->Get(ro, "key", &value); ASSERT_EQ("v0.1", value); ``` Some code comments are added for clarity. Note: support for timestamp in `DB::GetMergeOperands()` will be done in a follow-up PR. Pull Request resolved: https://github.com/facebook/rocksdb/pull/10819 Test Plan: make check Reviewed By: ltamasi Differential Revision: D40603195 Pulled By: riversand963 fbshipit-source-id: f96d6f183258f3392d80377025529f7660503013
-