5b2bbacb6f
Summary: New experimental option BBTO::optimize_filters_for_memory builds filters that maximize their use of "usable size" from malloc_usable_size, which is also used to compute block cache charges. Rather than always "rounding up," we track state in the BloomFilterPolicy object to mix essentially "rounding down" and "rounding up" so that the average FP rate of all generated filters is the same as without the option. (YMMV as heavily accessed filters might be unluckily lower accuracy.) Thus, the option near-minimizes what the block cache considers as "memory used" for a given target Bloom filter false positive rate and Bloom filter implementation. There are no forward or backward compatibility issues with this change, though it only works on the format_version=5 Bloom filter. With Jemalloc, we see about 10% reduction in memory footprint (and block cache charge) for Bloom filters, but 1-2% increase in storage footprint, due to encoding efficiency losses (FP rate is non-linear with bits/key). Why not weighted random round up/down rather than state tracking? By only requiring malloc_usable_size, we don't actually know what the next larger and next smaller usable sizes for the allocator are. We pick a requested size, accept and use whatever usable size it has, and use the difference to inform our next choice. This allows us to narrow in on the right balance without tracking/predicting usable sizes. Why not weight history of generated filter false positive rates by number of keys? This could lead to excess skew in small filters after generating a large filter. Results from filter_bench with jemalloc (irrelevant details omitted): (normal keys/filter, but high variance) $ ./filter_bench -quick -impl=2 -average_keys_per_filter=30000 -vary_key_count_ratio=0.9 Build avg ns/key: 29.6278 Number of filters: 5516 Total size (MB): 200.046 Reported total allocated memory (MB): 220.597 Reported internal fragmentation: 10.2732% Bits/key stored: 10.0097 Average FP rate %: 0.965228 $ ./filter_bench -quick -impl=2 -average_keys_per_filter=30000 -vary_key_count_ratio=0.9 -optimize_filters_for_memory Build avg ns/key: 30.5104 Number of filters: 5464 Total size (MB): 200.015 Reported total allocated memory (MB): 200.322 Reported internal fragmentation: 0.153709% Bits/key stored: 10.1011 Average FP rate %: 0.966313 (very few keys / filter, optimization not as effective due to ~59 byte internal fragmentation in blocked Bloom filter representation) $ ./filter_bench -quick -impl=2 -average_keys_per_filter=1000 -vary_key_count_ratio=0.9 Build avg ns/key: 29.5649 Number of filters: 162950 Total size (MB): 200.001 Reported total allocated memory (MB): 224.624 Reported internal fragmentation: 12.3117% Bits/key stored: 10.2951 Average FP rate %: 0.821534 $ ./filter_bench -quick -impl=2 -average_keys_per_filter=1000 -vary_key_count_ratio=0.9 -optimize_filters_for_memory Build avg ns/key: 31.8057 Number of filters: 159849 Total size (MB): 200 Reported total allocated memory (MB): 208.846 Reported internal fragmentation: 4.42297% Bits/key stored: 10.4948 Average FP rate %: 0.811006 (high keys/filter) $ ./filter_bench -quick -impl=2 -average_keys_per_filter=1000000 -vary_key_count_ratio=0.9 Build avg ns/key: 29.7017 Number of filters: 164 Total size (MB): 200.352 Reported total allocated memory (MB): 221.5 Reported internal fragmentation: 10.5552% Bits/key stored: 10.0003 Average FP rate %: 0.969358 $ ./filter_bench -quick -impl=2 -average_keys_per_filter=1000000 -vary_key_count_ratio=0.9 -optimize_filters_for_memory Build avg ns/key: 30.7131 Number of filters: 160 Total size (MB): 200.928 Reported total allocated memory (MB): 200.938 Reported internal fragmentation: 0.00448054% Bits/key stored: 10.1852 Average FP rate %: 0.963387 And from db_bench (block cache) with jemalloc: $ ./db_bench -db=/dev/shm/dbbench.no_optimize -benchmarks=fillrandom -format_version=5 -value_size=90 -bloom_bits=10 -num=2000000 -threads=8 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=false $ ./db_bench -db=/dev/shm/dbbench -benchmarks=fillrandom -format_version=5 -value_size=90 -bloom_bits=10 -num=2000000 -threads=8 -optimize_filters_for_memory -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=false $ (for FILE in /dev/shm/dbbench.no_optimize/*.sst; do ./sst_dump --file=$FILE --show_properties | grep 'filter block' ; done) | awk '{ t += $4; } END { print t; }' 17063835 $ (for FILE in /dev/shm/dbbench/*.sst; do ./sst_dump --file=$FILE --show_properties | grep 'filter block' ; done) | awk '{ t += $4; } END { print t; }' 17430747 $ #^ 2.1% additional filter storage $ ./db_bench -db=/dev/shm/dbbench.no_optimize -use_existing_db -benchmarks=readrandom,stats -statistics -bloom_bits=10 -num=2000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=false -duration=10 -cache_index_and_filter_blocks -cache_size=1000000000 rocksdb.block.cache.index.add COUNT : 33 rocksdb.block.cache.index.bytes.insert COUNT : 8440400 rocksdb.block.cache.filter.add COUNT : 33 rocksdb.block.cache.filter.bytes.insert COUNT : 21087528 rocksdb.bloom.filter.useful COUNT : 4963889 rocksdb.bloom.filter.full.positive COUNT : 1214081 rocksdb.bloom.filter.full.true.positive COUNT : 1161999 $ #^ 1.04 % observed FP rate $ ./db_bench -db=/dev/shm/dbbench -use_existing_db -benchmarks=readrandom,stats -statistics -bloom_bits=10 -num=2000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=false -optimize_filters_for_memory -duration=10 -cache_index_and_filter_blocks -cache_size=1000000000 rocksdb.block.cache.index.add COUNT : 33 rocksdb.block.cache.index.bytes.insert COUNT : 8448592 rocksdb.block.cache.filter.add COUNT : 33 rocksdb.block.cache.filter.bytes.insert COUNT : 18220328 rocksdb.bloom.filter.useful COUNT : 5360933 rocksdb.bloom.filter.full.positive COUNT : 1321315 rocksdb.bloom.filter.full.true.positive COUNT : 1262999 $ #^ 1.08 % observed FP rate, 13.6% less memory usage for filters (Due to specific key density, this example tends to generate filters that are "worse than average" for internal fragmentation. "Better than average" cases can show little or no improvement.) Pull Request resolved: https://github.com/facebook/rocksdb/pull/6427 Test Plan: unit test added, 'make check' with gcc, clang and valgrind Reviewed By: siying Differential Revision: D22124374 Pulled By: pdillinger fbshipit-source-id: f3e3aa152f9043ddf4fae25799e76341d0d8714e
562 lines
25 KiB
C++
562 lines
25 KiB
C++
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
|
|
// This source code is licensed under both the GPLv2 (found in the
|
|
// COPYING file in the root directory) and Apache 2.0 License
|
|
// (found in the LICENSE.Apache file in the root 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 <cstring>
|
|
|
|
#include "options/options_helper.h"
|
|
#include "rocksdb/convenience.h"
|
|
#include "test_util/testharness.h"
|
|
|
|
#ifndef GFLAGS
|
|
bool FLAGS_enable_print = false;
|
|
#else
|
|
#include "util/gflags_compat.h"
|
|
using GFLAGS_NAMESPACE::ParseCommandLineFlags;
|
|
DEFINE_bool(enable_print, false, "Print options generated to console.");
|
|
#endif // GFLAGS
|
|
|
|
namespace ROCKSDB_NAMESPACE {
|
|
|
|
// Verify options are settable from options strings.
|
|
// We take the approach that depends on compiler behavior that copy constructor
|
|
// won't touch implicit padding bytes, so that the test is fragile.
|
|
// As a result, we only run the tests to verify new fields in options are
|
|
// settable through string on limited platforms as it depends on behavior of
|
|
// compilers.
|
|
#ifndef ROCKSDB_LITE
|
|
#if defined OS_LINUX || defined OS_WIN
|
|
#ifndef __clang__
|
|
|
|
class OptionsSettableTest : public testing::Test {
|
|
public:
|
|
OptionsSettableTest() {}
|
|
};
|
|
|
|
const char kSpecialChar = 'z';
|
|
typedef std::vector<std::pair<size_t, size_t>> OffsetGap;
|
|
|
|
void FillWithSpecialChar(char* start_ptr, size_t total_size,
|
|
const OffsetGap& excluded,
|
|
char special_char = kSpecialChar) {
|
|
size_t offset = 0;
|
|
for (auto& pair : excluded) {
|
|
std::memset(start_ptr + offset, special_char, pair.first - offset);
|
|
offset = pair.first + pair.second;
|
|
}
|
|
std::memset(start_ptr + offset, special_char, total_size - offset);
|
|
}
|
|
|
|
int NumUnsetBytes(char* start_ptr, size_t total_size,
|
|
const OffsetGap& excluded) {
|
|
int total_unset_bytes_base = 0;
|
|
size_t offset = 0;
|
|
for (auto& pair : excluded) {
|
|
for (char* ptr = start_ptr + offset; ptr < start_ptr + pair.first; ptr++) {
|
|
if (*ptr == kSpecialChar) {
|
|
total_unset_bytes_base++;
|
|
}
|
|
}
|
|
offset = pair.first + pair.second;
|
|
}
|
|
for (char* ptr = start_ptr + offset; ptr < start_ptr + total_size; ptr++) {
|
|
if (*ptr == kSpecialChar) {
|
|
total_unset_bytes_base++;
|
|
}
|
|
}
|
|
return total_unset_bytes_base;
|
|
}
|
|
|
|
// Return true iff two structs are the same except excluded fields.
|
|
bool CompareBytes(char* start_ptr1, char* start_ptr2, size_t total_size,
|
|
const OffsetGap& excluded) {
|
|
size_t offset = 0;
|
|
for (auto& pair : excluded) {
|
|
for (; offset < pair.first; offset++) {
|
|
if (*(start_ptr1 + offset) != *(start_ptr2 + offset)) {
|
|
return false;
|
|
}
|
|
}
|
|
offset = pair.first + pair.second;
|
|
}
|
|
for (; offset < total_size; offset++) {
|
|
if (*(start_ptr1 + offset) != *(start_ptr2 + offset)) {
|
|
return false;
|
|
}
|
|
}
|
|
return true;
|
|
}
|
|
|
|
// If the test fails, likely a new option is added to BlockBasedTableOptions
|
|
// but it cannot be set through GetBlockBasedTableOptionsFromString(), or the
|
|
// test is not updated accordingly.
|
|
// After adding an option, we need to make sure it is settable by
|
|
// GetBlockBasedTableOptionsFromString() and add the option to the input string
|
|
// passed to the GetBlockBasedTableOptionsFromString() in this test.
|
|
// If it is a complicated type, you also need to add the field to
|
|
// kBbtoExcluded, and maybe add customized verification for it.
|
|
TEST_F(OptionsSettableTest, BlockBasedTableOptionsAllFieldsSettable) {
|
|
// Items in the form of <offset, size>. Need to be in ascending order
|
|
// and not overlapping. Need to updated if new pointer-option is added.
|
|
const OffsetGap kBbtoExcluded = {
|
|
{offsetof(struct BlockBasedTableOptions, flush_block_policy_factory),
|
|
sizeof(std::shared_ptr<FlushBlockPolicyFactory>)},
|
|
{offsetof(struct BlockBasedTableOptions, block_cache),
|
|
sizeof(std::shared_ptr<Cache>)},
|
|
{offsetof(struct BlockBasedTableOptions, persistent_cache),
|
|
sizeof(std::shared_ptr<PersistentCache>)},
|
|
{offsetof(struct BlockBasedTableOptions, block_cache_compressed),
|
|
sizeof(std::shared_ptr<Cache>)},
|
|
{offsetof(struct BlockBasedTableOptions, filter_policy),
|
|
sizeof(std::shared_ptr<const FilterPolicy>)},
|
|
};
|
|
|
|
// In this test, we catch a new option of BlockBasedTableOptions that is not
|
|
// settable through GetBlockBasedTableOptionsFromString().
|
|
// We count padding bytes of the option struct, and assert it to be the same
|
|
// as unset bytes of an option struct initialized by
|
|
// GetBlockBasedTableOptionsFromString().
|
|
|
|
char* bbto_ptr = new char[sizeof(BlockBasedTableOptions)];
|
|
|
|
// Count padding bytes by setting all bytes in the memory to a special char,
|
|
// copy a well constructed struct to this memory and see how many special
|
|
// bytes left.
|
|
BlockBasedTableOptions* bbto = new (bbto_ptr) BlockBasedTableOptions();
|
|
FillWithSpecialChar(bbto_ptr, sizeof(BlockBasedTableOptions), kBbtoExcluded);
|
|
// It based on the behavior of compiler that padding bytes are not changed
|
|
// when copying the struct. It's prone to failure when compiler behavior
|
|
// changes. We verify there is unset bytes to detect the case.
|
|
*bbto = BlockBasedTableOptions();
|
|
int unset_bytes_base =
|
|
NumUnsetBytes(bbto_ptr, sizeof(BlockBasedTableOptions), kBbtoExcluded);
|
|
ASSERT_GT(unset_bytes_base, 0);
|
|
bbto->~BlockBasedTableOptions();
|
|
|
|
// Construct the base option passed into
|
|
// GetBlockBasedTableOptionsFromString().
|
|
bbto = new (bbto_ptr) BlockBasedTableOptions();
|
|
FillWithSpecialChar(bbto_ptr, sizeof(BlockBasedTableOptions), kBbtoExcluded);
|
|
// This option is not setable:
|
|
bbto->use_delta_encoding = true;
|
|
|
|
char* new_bbto_ptr = new char[sizeof(BlockBasedTableOptions)];
|
|
BlockBasedTableOptions* new_bbto =
|
|
new (new_bbto_ptr) BlockBasedTableOptions();
|
|
FillWithSpecialChar(new_bbto_ptr, sizeof(BlockBasedTableOptions),
|
|
kBbtoExcluded);
|
|
|
|
// Need to update the option string if a new option is added.
|
|
ASSERT_OK(GetBlockBasedTableOptionsFromString(
|
|
*bbto,
|
|
"cache_index_and_filter_blocks=1;"
|
|
"cache_index_and_filter_blocks_with_high_priority=true;"
|
|
"pin_l0_filter_and_index_blocks_in_cache=1;"
|
|
"pin_top_level_index_and_filter=1;"
|
|
"index_type=kHashSearch;"
|
|
"data_block_index_type=kDataBlockBinaryAndHash;"
|
|
"index_shortening=kNoShortening;"
|
|
"data_block_hash_table_util_ratio=0.75;"
|
|
"checksum=kxxHash;hash_index_allow_collision=1;no_block_cache=1;"
|
|
"block_cache=1M;block_cache_compressed=1k;block_size=1024;"
|
|
"block_size_deviation=8;block_restart_interval=4; "
|
|
"metadata_block_size=1024;"
|
|
"partition_filters=false;"
|
|
"optimize_filters_for_memory=true;"
|
|
"index_block_restart_interval=4;"
|
|
"filter_policy=bloomfilter:4:true;whole_key_filtering=1;"
|
|
"format_version=1;"
|
|
"hash_index_allow_collision=false;"
|
|
"verify_compression=true;read_amp_bytes_per_bit=0;"
|
|
"enable_index_compression=false;"
|
|
"block_align=true",
|
|
new_bbto));
|
|
|
|
ASSERT_EQ(unset_bytes_base,
|
|
NumUnsetBytes(new_bbto_ptr, sizeof(BlockBasedTableOptions),
|
|
kBbtoExcluded));
|
|
|
|
ASSERT_TRUE(new_bbto->block_cache.get() != nullptr);
|
|
ASSERT_TRUE(new_bbto->block_cache_compressed.get() != nullptr);
|
|
ASSERT_TRUE(new_bbto->filter_policy.get() != nullptr);
|
|
|
|
bbto->~BlockBasedTableOptions();
|
|
new_bbto->~BlockBasedTableOptions();
|
|
|
|
delete[] bbto_ptr;
|
|
delete[] new_bbto_ptr;
|
|
}
|
|
|
|
// If the test fails, likely a new option is added to DBOptions
|
|
// but it cannot be set through GetDBOptionsFromString(), or the test is not
|
|
// updated accordingly.
|
|
// After adding an option, we need to make sure it is settable by
|
|
// GetDBOptionsFromString() and add the option to the input string passed to
|
|
// DBOptionsFromString()in this test.
|
|
// If it is a complicated type, you also need to add the field to
|
|
// kDBOptionsExcluded, and maybe add customized verification for it.
|
|
TEST_F(OptionsSettableTest, DBOptionsAllFieldsSettable) {
|
|
const OffsetGap kDBOptionsExcluded = {
|
|
{offsetof(struct DBOptions, env), sizeof(Env*)},
|
|
{offsetof(struct DBOptions, rate_limiter),
|
|
sizeof(std::shared_ptr<RateLimiter>)},
|
|
{offsetof(struct DBOptions, sst_file_manager),
|
|
sizeof(std::shared_ptr<SstFileManager>)},
|
|
{offsetof(struct DBOptions, info_log), sizeof(std::shared_ptr<Logger>)},
|
|
{offsetof(struct DBOptions, statistics),
|
|
sizeof(std::shared_ptr<Statistics>)},
|
|
{offsetof(struct DBOptions, db_paths), sizeof(std::vector<DbPath>)},
|
|
{offsetof(struct DBOptions, db_log_dir), sizeof(std::string)},
|
|
{offsetof(struct DBOptions, wal_dir), sizeof(std::string)},
|
|
{offsetof(struct DBOptions, write_buffer_manager),
|
|
sizeof(std::shared_ptr<WriteBufferManager>)},
|
|
{offsetof(struct DBOptions, listeners),
|
|
sizeof(std::vector<std::shared_ptr<EventListener>>)},
|
|
{offsetof(struct DBOptions, row_cache), sizeof(std::shared_ptr<Cache>)},
|
|
{offsetof(struct DBOptions, wal_filter), sizeof(const WalFilter*)},
|
|
{offsetof(struct DBOptions, file_checksum_gen_factory),
|
|
sizeof(std::shared_ptr<FileChecksumGenFactory>)},
|
|
};
|
|
|
|
char* options_ptr = new char[sizeof(DBOptions)];
|
|
|
|
// Count padding bytes by setting all bytes in the memory to a special char,
|
|
// copy a well constructed struct to this memory and see how many special
|
|
// bytes left.
|
|
DBOptions* options = new (options_ptr) DBOptions();
|
|
FillWithSpecialChar(options_ptr, sizeof(DBOptions), kDBOptionsExcluded);
|
|
// It based on the behavior of compiler that padding bytes are not changed
|
|
// when copying the struct. It's prone to failure when compiler behavior
|
|
// changes. We verify there is unset bytes to detect the case.
|
|
*options = DBOptions();
|
|
int unset_bytes_base =
|
|
NumUnsetBytes(options_ptr, sizeof(DBOptions), kDBOptionsExcluded);
|
|
ASSERT_GT(unset_bytes_base, 0);
|
|
options->~DBOptions();
|
|
|
|
options = new (options_ptr) DBOptions();
|
|
FillWithSpecialChar(options_ptr, sizeof(DBOptions), kDBOptionsExcluded);
|
|
|
|
char* new_options_ptr = new char[sizeof(DBOptions)];
|
|
DBOptions* new_options = new (new_options_ptr) DBOptions();
|
|
FillWithSpecialChar(new_options_ptr, sizeof(DBOptions), kDBOptionsExcluded);
|
|
|
|
// Need to update the option string if a new option is added.
|
|
ASSERT_OK(
|
|
GetDBOptionsFromString(*options,
|
|
"wal_bytes_per_sync=4295048118;"
|
|
"delete_obsolete_files_period_micros=4294967758;"
|
|
"WAL_ttl_seconds=4295008036;"
|
|
"WAL_size_limit_MB=4295036161;"
|
|
"max_write_batch_group_size_bytes=1048576;"
|
|
"wal_dir=path/to/wal_dir;"
|
|
"db_write_buffer_size=2587;"
|
|
"max_subcompactions=64330;"
|
|
"table_cache_numshardbits=28;"
|
|
"max_open_files=72;"
|
|
"max_file_opening_threads=35;"
|
|
"max_background_jobs=8;"
|
|
"base_background_compactions=3;"
|
|
"max_background_compactions=33;"
|
|
"use_fsync=true;"
|
|
"use_adaptive_mutex=false;"
|
|
"max_total_wal_size=4295005604;"
|
|
"compaction_readahead_size=0;"
|
|
"new_table_reader_for_compaction_inputs=false;"
|
|
"keep_log_file_num=4890;"
|
|
"skip_stats_update_on_db_open=false;"
|
|
"skip_checking_sst_file_sizes_on_db_open=false;"
|
|
"max_manifest_file_size=4295009941;"
|
|
"db_log_dir=path/to/db_log_dir;"
|
|
"skip_log_error_on_recovery=true;"
|
|
"writable_file_max_buffer_size=1048576;"
|
|
"paranoid_checks=true;"
|
|
"is_fd_close_on_exec=false;"
|
|
"bytes_per_sync=4295013613;"
|
|
"strict_bytes_per_sync=true;"
|
|
"enable_thread_tracking=false;"
|
|
"recycle_log_file_num=0;"
|
|
"create_missing_column_families=true;"
|
|
"log_file_time_to_roll=3097;"
|
|
"max_background_flushes=35;"
|
|
"create_if_missing=false;"
|
|
"error_if_exists=true;"
|
|
"delayed_write_rate=4294976214;"
|
|
"manifest_preallocation_size=1222;"
|
|
"allow_mmap_writes=false;"
|
|
"stats_dump_period_sec=70127;"
|
|
"stats_persist_period_sec=54321;"
|
|
"persist_stats_to_disk=true;"
|
|
"stats_history_buffer_size=14159;"
|
|
"allow_fallocate=true;"
|
|
"allow_mmap_reads=false;"
|
|
"use_direct_reads=false;"
|
|
"use_direct_io_for_flush_and_compaction=false;"
|
|
"max_log_file_size=4607;"
|
|
"random_access_max_buffer_size=1048576;"
|
|
"advise_random_on_open=true;"
|
|
"fail_if_options_file_error=false;"
|
|
"enable_pipelined_write=false;"
|
|
"unordered_write=false;"
|
|
"allow_concurrent_memtable_write=true;"
|
|
"wal_recovery_mode=kPointInTimeRecovery;"
|
|
"enable_write_thread_adaptive_yield=true;"
|
|
"write_thread_slow_yield_usec=5;"
|
|
"write_thread_max_yield_usec=1000;"
|
|
"access_hint_on_compaction_start=NONE;"
|
|
"info_log_level=DEBUG_LEVEL;"
|
|
"dump_malloc_stats=false;"
|
|
"allow_2pc=false;"
|
|
"avoid_flush_during_recovery=false;"
|
|
"avoid_flush_during_shutdown=false;"
|
|
"allow_ingest_behind=false;"
|
|
"preserve_deletes=false;"
|
|
"concurrent_prepare=false;"
|
|
"two_write_queues=false;"
|
|
"manual_wal_flush=false;"
|
|
"seq_per_batch=false;"
|
|
"atomic_flush=false;"
|
|
"avoid_unnecessary_blocking_io=false;"
|
|
"log_readahead_size=0;"
|
|
"write_dbid_to_manifest=false;"
|
|
"best_efforts_recovery=false",
|
|
new_options));
|
|
|
|
ASSERT_EQ(unset_bytes_base, NumUnsetBytes(new_options_ptr, sizeof(DBOptions),
|
|
kDBOptionsExcluded));
|
|
|
|
options->~DBOptions();
|
|
new_options->~DBOptions();
|
|
|
|
delete[] options_ptr;
|
|
delete[] new_options_ptr;
|
|
}
|
|
|
|
template <typename T1, typename T2>
|
|
inline int offset_of(T1 T2::*member) {
|
|
static T2 obj;
|
|
return int(size_t(&(obj.*member)) - size_t(&obj));
|
|
}
|
|
|
|
// If the test fails, likely a new option is added to ColumnFamilyOptions
|
|
// but it cannot be set through GetColumnFamilyOptionsFromString(), or the
|
|
// test is not updated accordingly.
|
|
// After adding an option, we need to make sure it is settable by
|
|
// GetColumnFamilyOptionsFromString() and add the option to the input
|
|
// string passed to GetColumnFamilyOptionsFromString()in this test.
|
|
// If it is a complicated type, you also need to add the field to
|
|
// kColumnFamilyOptionsExcluded, and maybe add customized verification
|
|
// for it.
|
|
TEST_F(OptionsSettableTest, ColumnFamilyOptionsAllFieldsSettable) {
|
|
// options in the excluded set need to appear in the same order as in
|
|
// ColumnFamilyOptions.
|
|
const OffsetGap kColumnFamilyOptionsExcluded = {
|
|
{offset_of(&ColumnFamilyOptions::inplace_callback),
|
|
sizeof(UpdateStatus(*)(char*, uint32_t*, Slice, std::string*))},
|
|
{offset_of(
|
|
&ColumnFamilyOptions::memtable_insert_with_hint_prefix_extractor),
|
|
sizeof(std::shared_ptr<const SliceTransform>)},
|
|
{offset_of(&ColumnFamilyOptions::compression_per_level),
|
|
sizeof(std::vector<CompressionType>)},
|
|
{offset_of(
|
|
&ColumnFamilyOptions::max_bytes_for_level_multiplier_additional),
|
|
sizeof(std::vector<int>)},
|
|
{offset_of(&ColumnFamilyOptions::memtable_factory),
|
|
sizeof(std::shared_ptr<MemTableRepFactory>)},
|
|
{offset_of(&ColumnFamilyOptions::table_properties_collector_factories),
|
|
sizeof(ColumnFamilyOptions::TablePropertiesCollectorFactories)},
|
|
{offset_of(&ColumnFamilyOptions::comparator), sizeof(Comparator*)},
|
|
{offset_of(&ColumnFamilyOptions::merge_operator),
|
|
sizeof(std::shared_ptr<MergeOperator>)},
|
|
{offset_of(&ColumnFamilyOptions::compaction_filter),
|
|
sizeof(const CompactionFilter*)},
|
|
{offset_of(&ColumnFamilyOptions::compaction_filter_factory),
|
|
sizeof(std::shared_ptr<CompactionFilterFactory>)},
|
|
{offset_of(&ColumnFamilyOptions::prefix_extractor),
|
|
sizeof(std::shared_ptr<const SliceTransform>)},
|
|
{offset_of(&ColumnFamilyOptions::snap_refresh_nanos), sizeof(uint64_t)},
|
|
{offset_of(&ColumnFamilyOptions::table_factory),
|
|
sizeof(std::shared_ptr<TableFactory>)},
|
|
{offset_of(&ColumnFamilyOptions::cf_paths), sizeof(std::vector<DbPath>)},
|
|
{offset_of(&ColumnFamilyOptions::compaction_thread_limiter),
|
|
sizeof(std::shared_ptr<ConcurrentTaskLimiter>)},
|
|
};
|
|
|
|
char* options_ptr = new char[sizeof(ColumnFamilyOptions)];
|
|
|
|
// Count padding bytes by setting all bytes in the memory to a special char,
|
|
// copy a well constructed struct to this memory and see how many special
|
|
// bytes left.
|
|
ColumnFamilyOptions* options = new (options_ptr) ColumnFamilyOptions();
|
|
FillWithSpecialChar(options_ptr, sizeof(ColumnFamilyOptions),
|
|
kColumnFamilyOptionsExcluded);
|
|
|
|
// It based on the behavior of compiler that padding bytes are not changed
|
|
// when copying the struct. It's prone to failure when compiler behavior
|
|
// changes. We verify there is unset bytes to detect the case.
|
|
*options = ColumnFamilyOptions();
|
|
|
|
// Deprecatd option which is not initialized. Need to set it to avoid
|
|
// Valgrind error
|
|
options->max_mem_compaction_level = 0;
|
|
|
|
int unset_bytes_base = NumUnsetBytes(options_ptr, sizeof(ColumnFamilyOptions),
|
|
kColumnFamilyOptionsExcluded);
|
|
ASSERT_GT(unset_bytes_base, 0);
|
|
options->~ColumnFamilyOptions();
|
|
|
|
options = new (options_ptr) ColumnFamilyOptions();
|
|
FillWithSpecialChar(options_ptr, sizeof(ColumnFamilyOptions),
|
|
kColumnFamilyOptionsExcluded);
|
|
|
|
// Following options are not settable through
|
|
// GetColumnFamilyOptionsFromString():
|
|
options->rate_limit_delay_max_milliseconds = 33;
|
|
options->compaction_options_universal = CompactionOptionsUniversal();
|
|
options->hard_rate_limit = 0;
|
|
options->soft_rate_limit = 0;
|
|
options->purge_redundant_kvs_while_flush = false;
|
|
options->max_mem_compaction_level = 0;
|
|
options->compaction_filter = nullptr;
|
|
|
|
char* new_options_ptr = new char[sizeof(ColumnFamilyOptions)];
|
|
ColumnFamilyOptions* new_options =
|
|
new (new_options_ptr) ColumnFamilyOptions();
|
|
FillWithSpecialChar(new_options_ptr, sizeof(ColumnFamilyOptions),
|
|
kColumnFamilyOptionsExcluded);
|
|
|
|
// Need to update the option string if a new option is added.
|
|
ASSERT_OK(GetColumnFamilyOptionsFromString(
|
|
*options,
|
|
"compaction_filter_factory=mpudlojcujCompactionFilterFactory;"
|
|
"table_factory=PlainTable;"
|
|
"prefix_extractor=rocksdb.CappedPrefix.13;"
|
|
"comparator=leveldb.BytewiseComparator;"
|
|
"compression_per_level=kBZip2Compression:kBZip2Compression:"
|
|
"kBZip2Compression:kNoCompression:kZlibCompression:kBZip2Compression:"
|
|
"kSnappyCompression;"
|
|
"max_bytes_for_level_base=986;"
|
|
"bloom_locality=8016;"
|
|
"target_file_size_base=4294976376;"
|
|
"memtable_huge_page_size=2557;"
|
|
"max_successive_merges=5497;"
|
|
"max_sequential_skip_in_iterations=4294971408;"
|
|
"arena_block_size=1893;"
|
|
"target_file_size_multiplier=35;"
|
|
"min_write_buffer_number_to_merge=9;"
|
|
"max_write_buffer_number=84;"
|
|
"write_buffer_size=1653;"
|
|
"max_compaction_bytes=64;"
|
|
"max_bytes_for_level_multiplier=60;"
|
|
"memtable_factory=SkipListFactory;"
|
|
"compression=kNoCompression;"
|
|
"compression_opts=5:6:7:8:9:true;"
|
|
"bottommost_compression_opts=4:5:6:7:8:true;"
|
|
"bottommost_compression=kDisableCompressionOption;"
|
|
"level0_stop_writes_trigger=33;"
|
|
"num_levels=99;"
|
|
"level0_slowdown_writes_trigger=22;"
|
|
"level0_file_num_compaction_trigger=14;"
|
|
"compaction_filter=urxcqstuwnCompactionFilter;"
|
|
"soft_rate_limit=530.615385;"
|
|
"soft_pending_compaction_bytes_limit=0;"
|
|
"max_write_buffer_number_to_maintain=84;"
|
|
"max_write_buffer_size_to_maintain=2147483648;"
|
|
"merge_operator=aabcxehazrMergeOperator;"
|
|
"memtable_prefix_bloom_size_ratio=0.4642;"
|
|
"memtable_whole_key_filtering=true;"
|
|
"memtable_insert_with_hint_prefix_extractor=rocksdb.CappedPrefix.13;"
|
|
"paranoid_file_checks=true;"
|
|
"force_consistency_checks=true;"
|
|
"inplace_update_num_locks=7429;"
|
|
"optimize_filters_for_hits=false;"
|
|
"level_compaction_dynamic_level_bytes=false;"
|
|
"inplace_update_support=false;"
|
|
"compaction_style=kCompactionStyleFIFO;"
|
|
"compaction_pri=kMinOverlappingRatio;"
|
|
"hard_pending_compaction_bytes_limit=0;"
|
|
"disable_auto_compactions=false;"
|
|
"report_bg_io_stats=true;"
|
|
"ttl=60;"
|
|
"periodic_compaction_seconds=3600;"
|
|
"sample_for_compression=0;"
|
|
"compaction_options_fifo={max_table_files_size=3;allow_"
|
|
"compaction=false;};",
|
|
new_options));
|
|
|
|
ASSERT_EQ(unset_bytes_base,
|
|
NumUnsetBytes(new_options_ptr, sizeof(ColumnFamilyOptions),
|
|
kColumnFamilyOptionsExcluded));
|
|
|
|
ColumnFamilyOptions rnd_filled_options = *new_options;
|
|
|
|
options->~ColumnFamilyOptions();
|
|
new_options->~ColumnFamilyOptions();
|
|
|
|
delete[] options_ptr;
|
|
delete[] new_options_ptr;
|
|
|
|
// Test copying to mutabable and immutable options and copy back the mutable
|
|
// part.
|
|
const OffsetGap kMutableCFOptionsExcluded = {
|
|
{offset_of(&MutableCFOptions::prefix_extractor),
|
|
sizeof(std::shared_ptr<const SliceTransform>)},
|
|
{offset_of(&MutableCFOptions::max_bytes_for_level_multiplier_additional),
|
|
sizeof(std::vector<int>)},
|
|
{offset_of(&MutableCFOptions::max_file_size),
|
|
sizeof(std::vector<uint64_t>)},
|
|
};
|
|
|
|
// For all memory used for options, pre-fill every char. Otherwise, the
|
|
// padding bytes might be different so that byte-wise comparison doesn't
|
|
// general equal results even if objects are equal.
|
|
const char kMySpecialChar = 'x';
|
|
char* mcfo1_ptr = new char[sizeof(MutableCFOptions)];
|
|
FillWithSpecialChar(mcfo1_ptr, sizeof(MutableCFOptions),
|
|
kMutableCFOptionsExcluded, kMySpecialChar);
|
|
char* mcfo2_ptr = new char[sizeof(MutableCFOptions)];
|
|
FillWithSpecialChar(mcfo2_ptr, sizeof(MutableCFOptions),
|
|
kMutableCFOptionsExcluded, kMySpecialChar);
|
|
|
|
// A clean column family options is constructed after filling the same special
|
|
// char as the initial one. So that the padding bytes are the same.
|
|
char* cfo_clean_ptr = new char[sizeof(ColumnFamilyOptions)];
|
|
FillWithSpecialChar(cfo_clean_ptr, sizeof(ColumnFamilyOptions),
|
|
kColumnFamilyOptionsExcluded);
|
|
rnd_filled_options.num_levels = 66;
|
|
ColumnFamilyOptions* cfo_clean = new (cfo_clean_ptr) ColumnFamilyOptions();
|
|
|
|
MutableCFOptions* mcfo1 =
|
|
new (mcfo1_ptr) MutableCFOptions(rnd_filled_options);
|
|
ColumnFamilyOptions cfo_back = BuildColumnFamilyOptions(*cfo_clean, *mcfo1);
|
|
MutableCFOptions* mcfo2 = new (mcfo2_ptr) MutableCFOptions(cfo_back);
|
|
|
|
ASSERT_TRUE(CompareBytes(mcfo1_ptr, mcfo2_ptr, sizeof(MutableCFOptions),
|
|
kMutableCFOptionsExcluded));
|
|
|
|
cfo_clean->~ColumnFamilyOptions();
|
|
mcfo1->~MutableCFOptions();
|
|
mcfo2->~MutableCFOptions();
|
|
delete[] mcfo1_ptr;
|
|
delete[] mcfo2_ptr;
|
|
delete[] cfo_clean_ptr;
|
|
}
|
|
#endif // !__clang__
|
|
#endif // OS_LINUX || OS_WIN
|
|
#endif // !ROCKSDB_LITE
|
|
|
|
} // namespace ROCKSDB_NAMESPACE
|
|
|
|
int main(int argc, char** argv) {
|
|
::testing::InitGoogleTest(&argc, argv);
|
|
#ifdef GFLAGS
|
|
ParseCommandLineFlags(&argc, &argv, true);
|
|
#endif // GFLAGS
|
|
return RUN_ALL_TESTS();
|
|
}
|