Skip to content

Commit

Permalink
Merge branch 'main' into 701-expose-enable-speedb-features-via-the-ja…
Browse files Browse the repository at this point in the history
…va-api
  • Loading branch information
ofriedma authored Dec 28, 2023
2 parents b129e16 + d4d2a0a commit a90d918
Show file tree
Hide file tree
Showing 5 changed files with 11 additions and 10 deletions.
9 changes: 3 additions & 6 deletions .github/workflows/check_license_and_history.yml
Original file line number Diff line number Diff line change
Expand Up @@ -50,8 +50,7 @@ jobs:
echo $file does not have the Apache 2.0 license header && exit_code=222
fi
done
#exit $exit_code
exit 0
exit $exit_code
- name: Check HISTORY PR
run: |
set +e
Expand All @@ -64,8 +63,7 @@ jobs:
else
history_not_in=0
fi
#exit $history_not_in
exit 0
exit $history_not_in
fi
echo "No files were added"
exit 0
Expand All @@ -82,8 +80,7 @@ jobs:
else
history_not_in=0
fi
#exit $history_not_in
exit 0
exit $history_not_in
fi
echo "No files were added"
exit 0
4 changes: 4 additions & 0 deletions HISTORY.md
Original file line number Diff line number Diff line change
@@ -1,9 +1,11 @@
# Speedb Change Log

## Unreleased
Based on RocksDB 8.6.7

### New Features
* Added ConfigOptions::compare_to. When set, this value causes only values that have been changed to be part of the serialized output (#648).
* Rebase on RocksDB 8.6.7

### Enhancements
* Added a kUseBaseAddress flag and GetBaseOffset flag to OptionTypeInfo. If this flag is set and a function is used for processing options, the function is passed the base address of the struct rather than the specific field (#397)
Expand All @@ -12,9 +14,11 @@
### Bug Fixes
* Stall deadlock consists small cfs (#637).
* Proactive Flushes: Fix a race in the ShouldInitiateAnotherFlushMemOnly that may cause the method to return an incorrect answer (#758).
* Fix CI failure after changing compation_readahead_size default to 0 (#794).

### Miscellaneous
* Remove leftover references to ROCKSDB_LITE (#755).
* Options: Set compaction_readahead_size default to 0. The current default of 2Mb is not optimal for most of our use cases. Having a value of 0 means that the FS will use its default size for prefetching (true only with https://github.com/speedb-io/speedb/pull/788).

## Hazlenut 2.7.0 (27/10/2023)
Based on RocksDB 8.1.1
Expand Down
3 changes: 1 addition & 2 deletions db/db_options_test.cc
Original file line number Diff line number Diff line change
Expand Up @@ -1069,8 +1069,7 @@ TEST_F(DBOptionsTest, CompactionReadaheadSizeChange) {
ASSERT_OK(s);
}

ASSERT_EQ(1024 * 1024 * 2,
dbfull()->GetDBOptions().compaction_readahead_size);
ASSERT_EQ(0, dbfull()->GetDBOptions().compaction_readahead_size);
ASSERT_OK(dbfull()->SetDBOptions({{"compaction_readahead_size", "256"}}));
ASSERT_EQ(256, dbfull()->GetDBOptions().compaction_readahead_size);
for (int i = 0; i < 1024; i++) {
Expand Down
1 change: 1 addition & 0 deletions file/prefetch_test.cc
Original file line number Diff line number Diff line change
Expand Up @@ -171,6 +171,7 @@ TEST_P(PrefetchTest, Basic) {
SyncPoint::GetInstance()->SetCallBack("FilePrefetchBuffer::Prefetch:Start",
[&](void*) { buff_prefetch_count++; });
SyncPoint::GetInstance()->EnableProcessing();
options.compaction_readahead_size = 2 * 1024 * 1024;

Status s = TryReopen(options);
if (use_direct_io && (s.IsNotSupported() || s.IsInvalidArgument())) {
Expand Down
4 changes: 2 additions & 2 deletions include/rocksdb/options.h
Original file line number Diff line number Diff line change
Expand Up @@ -1005,10 +1005,10 @@ struct DBOptions {
// running RocksDB on spinning disks, you should set this to at least 2MB.
// That way RocksDB's compaction is doing sequential instead of random reads.
//
// Default: 2MB
// Default: 0
//
// Dynamically changeable through SetDBOptions() API.
size_t compaction_readahead_size = 2 * 1024 * 1024;
size_t compaction_readahead_size = 0;

// This is a maximum buffer size that is used by WinMmapReadableFile in
// unbuffered disk I/O mode. We need to maintain an aligned buffer for
Expand Down

0 comments on commit a90d918

Please sign in to comment.