NextGraph
/
rocksdb
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
3963 Commits
4 Branches
0 Tags
169 MiB
Tag: Branch: Tree: 9d6503f88d
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '9d6503f88d'
${ noResults }
rocksdb/db/plain_table_db_test.cc

1107 lines
37 KiB
Raw Normal View History Unescape

A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
// 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.
// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same 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.
Block plain_table_db_test in ROCKSDB_LITE Summary: Block plain_table_db_test in ROCKSDB_LITE since plain table is not supported in ROCKSDB_LITE Test Plan: plain_table_db_test Reviewers: igor, sdong, yhchiang Reviewed By: yhchiang Subscribers: dhruba Differential Revision: https://reviews.facebook.net/D42159
9 years ago
#ifndef ROCKSDB_LITE
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
#include <algorithm>
#include <set>
#include "db/db_impl.h"
#include "db/filename.h"
#include "db/version_set.h"
#include "db/write_batch_internal.h"
#include "rocksdb/cache.h"
#include "rocksdb/compaction_filter.h"
Clean up arena API Summary: Easy thing goes first. This patch moves arena to internal dir; based on which, the coming patch will deal with memtable_rep. Test Plan: make check Reviewers: haobo, sdong, dhruba CC: leveldb Differential Revision: https://reviews.facebook.net/D15615
11 years ago
#include "rocksdb/db.h"
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
#include "rocksdb/env.h"
Clean up arena API Summary: Easy thing goes first. This patch moves arena to internal dir; based on which, the coming patch will deal with memtable_rep. Test Plan: make check Reviewers: haobo, sdong, dhruba CC: leveldb Differential Revision: https://reviews.facebook.net/D15615
11 years ago
#include "rocksdb/filter_policy.h"
#include "rocksdb/slice_transform.h"
#include "rocksdb/table.h"
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
#include "table/meta_blocks.h"
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
#include "table/bloom_block.h"
First phase API clean up Summary: Addressed all the issues in https://reviews.facebook.net/D15447. Now most table-related modules are hidden from user land. Test Plan: make check Reviewers: sdong, haobo, dhruba CC: leveldb Differential Revision: https://reviews.facebook.net/D15525
11 years ago
#include "table/plain_table_factory.h"
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
#include "table/plain_table_reader.h"
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
#include "util/hash.h"
#include "util/logging.h"
#include "util/mutexlock.h"
rocksdb: Remove #include "util/string_util.h" from util/testharness.h Summary: 1. Manually deleted #include "util/string_util.h" from util/testharness.h 2. ``` % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -E 'say $F[0] if /: error:/' build.log | sort -u | xargs sed -i '/#include "util\/testharness.h"/i #include "util\/string_util.h"' ``` Test Plan: Make sure make all completes with no errors. ``` % make all -j55 ``` Reviewers: meyering, igor, sdong Reviewed By: sdong Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35493
10 years ago
#include "util/string_util.h"
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
#include "util/testharness.h"
#include "util/testutil.h"
#include "utilities/merge_operators.h"
using std::unique_ptr;
namespace rocksdb {
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
class PlainTableDBTest : public testing::Test {
Fix the valgrind issues
11 years ago
protected:
private:
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
std::string dbname_;
Env* env_;
DB* db_;
Options last_options_;
Fix the valgrind issues
11 years ago
public:
PlainTableDBTest() : env_(Env::Default()) {
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
dbname_ = test::TmpDir() + "/plain_table_db_test";
rocksdb: Replace ASSERT* with EXPECT* in functions that does not return void value Summary: gtest does not use exceptions to fail a unit test by design, and `ASSERT*`s are implemented using `return`. As a consequence we cannot use `ASSERT*` in a function that does not return `void` value ([[ https://code.google.com/p/googletest/wiki/AdvancedGuide#Assertion_Placement | 1]]), and have to fix our existing code. This diff does this in a generic way, with no manual changes. In order to detect all existing `ASSERT*` that are used in functions that doesn't return void value, I change the code to generate compile errors for such cases. In `util/testharness.h` I defined `EXPECT*` assertions, the same way as `ASSERT*`, and redefined `ASSERT*` to return `void`. Then executed: ```lang=bash % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -e 'print "-- -number=".$F[1]." ".$F[0]."\n" if /: error:/' \ build.log | xargs -L 1 perl -spi -e 's/ASSERT/EXPECT/g if $. == $number' % make format ``` After that I reverted back change to `ASSERT*` in `util/testharness.h`. But preserved introduced `EXPECT*`, which is the same as `ASSERT*`. This will be deleted once switched to gtest. This diff is independent and contains manual changes only in `util/testharness.h`. Test Plan: Make sure all tests are passing. ```lang=bash % USE_CLANG=1 make check ``` Reviewers: igor, lgalanis, sdong, yufei.zhu, rven, meyering Reviewed By: meyering Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D33333
10 years ago
EXPECT_OK(DestroyDB(dbname_, Options()));
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
db_ = nullptr;
Reopen();
}
~PlainTableDBTest() {
delete db_;
rocksdb: Replace ASSERT* with EXPECT* in functions that does not return void value Summary: gtest does not use exceptions to fail a unit test by design, and `ASSERT*`s are implemented using `return`. As a consequence we cannot use `ASSERT*` in a function that does not return `void` value ([[ https://code.google.com/p/googletest/wiki/AdvancedGuide#Assertion_Placement | 1]]), and have to fix our existing code. This diff does this in a generic way, with no manual changes. In order to detect all existing `ASSERT*` that are used in functions that doesn't return void value, I change the code to generate compile errors for such cases. In `util/testharness.h` I defined `EXPECT*` assertions, the same way as `ASSERT*`, and redefined `ASSERT*` to return `void`. Then executed: ```lang=bash % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -e 'print "-- -number=".$F[1]." ".$F[0]."\n" if /: error:/' \ build.log | xargs -L 1 perl -spi -e 's/ASSERT/EXPECT/g if $. == $number' % make format ``` After that I reverted back change to `ASSERT*` in `util/testharness.h`. But preserved introduced `EXPECT*`, which is the same as `ASSERT*`. This will be deleted once switched to gtest. This diff is independent and contains manual changes only in `util/testharness.h`. Test Plan: Make sure all tests are passing. ```lang=bash % USE_CLANG=1 make check ``` Reviewers: igor, lgalanis, sdong, yufei.zhu, rven, meyering Reviewed By: meyering Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D33333
10 years ago
EXPECT_OK(DestroyDB(dbname_, Options()));
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
}
// Return the current option configuration.
Options CurrentOptions() {
Options options;
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
PlainTableOptions plain_table_options;
plain_table_options.user_key_len = 0;
plain_table_options.bloom_bits_per_key = 2;
plain_table_options.hash_table_ratio = 0.8;
plain_table_options.index_sparseness = 3;
plain_table_options.huge_page_tlb_size = 0;
plain_table_options.encoding_type = kPrefix;
plain_table_options.full_scan_mode = false;
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
plain_table_options.store_index_in_file = false;
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
options.table_factory.reset(NewPlainTableFactory(plain_table_options));
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
options.memtable_factory.reset(NewHashLinkListRepFactory(4, 0, 3, true));
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
Consolidate SliceTransform object ownership Summary: (1) Fix SanitizeOptions() to also check HashLinkList. The current dynamic case just happens to work because the 2 classes have the same layout. (2) Do not delete SliceTransform object in HashSkipListFactory and HashLinkListFactory destructor. Reason: SanitizeOptions() enforces prefix_extractor and SliceTransform to be the same object when Hash**Factory is used. This makes the behavior strange: when Hash**Factory is used, prefix_extractor will be released by RocksDB. If other memtable factory is used, prefix_extractor should be released by user. Test Plan: db_bench && make asan_check Reviewers: haobo, igor, sdong Reviewed By: igor CC: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D16587
11 years ago
options.prefix_extractor.reset(NewFixedPrefixTransform(8));
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
options.allow_mmap_reads = true;
return options;
}
DBImpl* dbfull() {
return reinterpret_cast<DBImpl*>(db_);
}
void Reopen(Options* options = nullptr) {
ASSERT_OK(TryReopen(options));
}
void Close() {
delete db_;
db_ = nullptr;
}
void DestroyAndReopen(Options* options = nullptr) {
//Destroy using last options
Destroy(&last_options_);
ASSERT_OK(TryReopen(options));
}
void Destroy(Options* options) {
delete db_;
db_ = nullptr;
ASSERT_OK(DestroyDB(dbname_, *options));
}
Status PureReopen(Options* options, DB** db) {
return DB::Open(*options, dbname_, db);
}
Status TryReopen(Options* options = nullptr) {
delete db_;
db_ = nullptr;
Options opts;
if (options != nullptr) {
opts = *options;
} else {
opts = CurrentOptions();
opts.create_if_missing = true;
}
last_options_ = opts;
return DB::Open(opts, dbname_, &db_);
}
Status Put(const Slice& k, const Slice& v) {
return db_->Put(WriteOptions(), k, v);
}
Status Delete(const std::string& k) {
return db_->Delete(WriteOptions(), k);
}
std::string Get(const std::string& k, const Snapshot* snapshot = nullptr) {
kill ReadOptions.prefix and .prefix_seek Summary: also add an override option total_order_iteration if you want to use full iterator with prefix_extractor Test Plan: make all check Reviewers: igor, haobo, sdong, yhchiang Reviewed By: haobo CC: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D17805
11 years ago
ReadOptions options;
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
options.snapshot = snapshot;
std::string result;
Status s = db_->Get(options, k, &result);
if (s.IsNotFound()) {
result = "NOT_FOUND";
} else if (!s.ok()) {
result = s.ToString();
}
return result;
}
int NumTableFilesAtLevel(int level) {
std::string property;
rocksdb: Replace ASSERT* with EXPECT* in functions that does not return void value Summary: gtest does not use exceptions to fail a unit test by design, and `ASSERT*`s are implemented using `return`. As a consequence we cannot use `ASSERT*` in a function that does not return `void` value ([[ https://code.google.com/p/googletest/wiki/AdvancedGuide#Assertion_Placement | 1]]), and have to fix our existing code. This diff does this in a generic way, with no manual changes. In order to detect all existing `ASSERT*` that are used in functions that doesn't return void value, I change the code to generate compile errors for such cases. In `util/testharness.h` I defined `EXPECT*` assertions, the same way as `ASSERT*`, and redefined `ASSERT*` to return `void`. Then executed: ```lang=bash % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -e 'print "-- -number=".$F[1]." ".$F[0]."\n" if /: error:/' \ build.log | xargs -L 1 perl -spi -e 's/ASSERT/EXPECT/g if $. == $number' % make format ``` After that I reverted back change to `ASSERT*` in `util/testharness.h`. But preserved introduced `EXPECT*`, which is the same as `ASSERT*`. This will be deleted once switched to gtest. This diff is independent and contains manual changes only in `util/testharness.h`. Test Plan: Make sure all tests are passing. ```lang=bash % USE_CLANG=1 make check ``` Reviewers: igor, lgalanis, sdong, yufei.zhu, rven, meyering Reviewed By: meyering Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D33333
10 years ago
EXPECT_TRUE(db_->GetProperty(
"rocksdb.num-files-at-level" + NumberToString(level), &property));
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
return atoi(property.c_str());
}
// Return spread of files per level
std::string FilesPerLevel() {
std::string result;
Turn on -Wshorten-64-to-32 and fix all the errors Summary: We need to turn on -Wshorten-64-to-32 for mobile. See D1671432 (internal phabricator) for details. This diff turns on the warning flag and fixes all the errors. There were also some interesting errors that I might call bugs, especially in plain table. Going forward, I think it makes sense to have this flag turned on and be very very careful when converting 64-bit to 32-bit variables. Test Plan: compiles Reviewers: ljin, rven, yhchiang, sdong Reviewed By: yhchiang Subscribers: bobbaldwin, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D28689
10 years ago
size_t last_non_zero_offset = 0;
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
for (int level = 0; level < db_->NumberLevels(); level++) {
int f = NumTableFilesAtLevel(level);
char buf[100];
snprintf(buf, sizeof(buf), "%s%d", (level ? "," : ""), f);
result += buf;
if (f > 0) {
last_non_zero_offset = result.size();
}
}
result.resize(last_non_zero_offset);
return result;
}
std::string IterStatus(Iterator* iter) {
std::string result;
if (iter->Valid()) {
result = iter->key().ToString() + "->" + iter->value().ToString();
} else {
result = "(invalid)";
}
return result;
}
};
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
TEST_F(PlainTableDBTest, Empty) {
Fix the valgrind issues
11 years ago
ASSERT_TRUE(dbfull() != nullptr);
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
ASSERT_EQ("NOT_FOUND", Get("0000000000000foo"));
}
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
extern const uint64_t kPlainTableMagicNumber;
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
class TestPlainTableReader : public PlainTableReader {
public:
introduce ImmutableOptions Summary: As a preparation to support updating some options dynamically, I'd like to first introduce ImmutableOptions, which is a subset of Options that cannot be changed during the course of a DB lifetime without restart. ColumnFamily will keep both Options and ImmutableOptions. Any component below ColumnFamily should only take ImmutableOptions in their constructor. Other options should be taken from APIs, which will be allowed to adjust dynamically. I am yet to make changes to memtable and other related classes to take ImmutableOptions in their ctor. That can be done in a seprate diff as this one is already pretty big. Test Plan: make all check Reviewers: yhchiang, igor, sdong Reviewed By: sdong Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D22545
10 years ago
TestPlainTableReader(const EnvOptions& env_options,
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
const InternalKeyComparator& icomparator,
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
EncodingType encoding_type, uint64_t file_size,
int bloom_bits_per_key, double hash_table_ratio,
size_t index_sparseness,
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
const TableProperties* table_properties,
Move rate_limiter, write buffering, most perf context instrumentation and most random kill out of Env Summary: We want to keep Env a think layer for better portability. Less platform dependent codes should be moved out of Env. In this patch, I create a wrapper of file readers and writers, and put rate limiting, write buffering, as well as most perf context instrumentation and random kill out of Env. It will make it easier to maintain multiple Env in the future. Test Plan: Run all existing unit tests. Reviewers: anthony, kradhakrishnan, IslamAbdelRahman, yhchiang, igor Reviewed By: igor Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D42321
9 years ago
unique_ptr<RandomAccessFileReader>&& file,
introduce ImmutableOptions Summary: As a preparation to support updating some options dynamically, I'd like to first introduce ImmutableOptions, which is a subset of Options that cannot be changed during the course of a DB lifetime without restart. ColumnFamily will keep both Options and ImmutableOptions. Any component below ColumnFamily should only take ImmutableOptions in their constructor. Other options should be taken from APIs, which will be allowed to adjust dynamically. I am yet to make changes to memtable and other related classes to take ImmutableOptions in their ctor. That can be done in a seprate diff as this one is already pretty big. Test Plan: make all check Reviewers: yhchiang, igor, sdong Reviewed By: sdong Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D22545
10 years ago
const ImmutableCFOptions& ioptions,
Move rate_limiter, write buffering, most perf context instrumentation and most random kill out of Env Summary: We want to keep Env a think layer for better portability. Less platform dependent codes should be moved out of Env. In this patch, I create a wrapper of file readers and writers, and put rate limiting, write buffering, as well as most perf context instrumentation and random kill out of Env. It will make it easier to maintain multiple Env in the future. Test Plan: Run all existing unit tests. Reviewers: anthony, kradhakrishnan, IslamAbdelRahman, yhchiang, igor Reviewed By: igor Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D42321
9 years ago
bool* expect_bloom_not_match, bool store_index_in_file)
introduce ImmutableOptions Summary: As a preparation to support updating some options dynamically, I'd like to first introduce ImmutableOptions, which is a subset of Options that cannot be changed during the course of a DB lifetime without restart. ColumnFamily will keep both Options and ImmutableOptions. Any component below ColumnFamily should only take ImmutableOptions in their constructor. Other options should be taken from APIs, which will be allowed to adjust dynamically. I am yet to make changes to memtable and other related classes to take ImmutableOptions in their ctor. That can be done in a seprate diff as this one is already pretty big. Test Plan: make all check Reviewers: yhchiang, igor, sdong Reviewed By: sdong Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D22545
10 years ago
: PlainTableReader(ioptions, std::move(file), env_options, icomparator,
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
encoding_type, file_size, table_properties),
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
expect_bloom_not_match_(expect_bloom_not_match) {
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
Status s = MmapDataFile();
rocksdb: Replace ASSERT* with EXPECT* in functions that does not return void value Summary: gtest does not use exceptions to fail a unit test by design, and `ASSERT*`s are implemented using `return`. As a consequence we cannot use `ASSERT*` in a function that does not return `void` value ([[ https://code.google.com/p/googletest/wiki/AdvancedGuide#Assertion_Placement | 1]]), and have to fix our existing code. This diff does this in a generic way, with no manual changes. In order to detect all existing `ASSERT*` that are used in functions that doesn't return void value, I change the code to generate compile errors for such cases. In `util/testharness.h` I defined `EXPECT*` assertions, the same way as `ASSERT*`, and redefined `ASSERT*` to return `void`. Then executed: ```lang=bash % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -e 'print "-- -number=".$F[1]." ".$F[0]."\n" if /: error:/' \ build.log | xargs -L 1 perl -spi -e 's/ASSERT/EXPECT/g if $. == $number' % make format ``` After that I reverted back change to `ASSERT*` in `util/testharness.h`. But preserved introduced `EXPECT*`, which is the same as `ASSERT*`. This will be deleted once switched to gtest. This diff is independent and contains manual changes only in `util/testharness.h`. Test Plan: Make sure all tests are passing. ```lang=bash % USE_CLANG=1 make check ``` Reviewers: igor, lgalanis, sdong, yufei.zhu, rven, meyering Reviewed By: meyering Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D33333
10 years ago
EXPECT_TRUE(s.ok());
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
s = PopulateIndex(const_cast<TableProperties*>(table_properties),
bloom_bits_per_key, hash_table_ratio, index_sparseness,
2 * 1024 * 1024);
rocksdb: Replace ASSERT* with EXPECT* in functions that does not return void value Summary: gtest does not use exceptions to fail a unit test by design, and `ASSERT*`s are implemented using `return`. As a consequence we cannot use `ASSERT*` in a function that does not return `void` value ([[ https://code.google.com/p/googletest/wiki/AdvancedGuide#Assertion_Placement | 1]]), and have to fix our existing code. This diff does this in a generic way, with no manual changes. In order to detect all existing `ASSERT*` that are used in functions that doesn't return void value, I change the code to generate compile errors for such cases. In `util/testharness.h` I defined `EXPECT*` assertions, the same way as `ASSERT*`, and redefined `ASSERT*` to return `void`. Then executed: ```lang=bash % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -e 'print "-- -number=".$F[1]." ".$F[0]."\n" if /: error:/' \ build.log | xargs -L 1 perl -spi -e 's/ASSERT/EXPECT/g if $. == $number' % make format ``` After that I reverted back change to `ASSERT*` in `util/testharness.h`. But preserved introduced `EXPECT*`, which is the same as `ASSERT*`. This will be deleted once switched to gtest. This diff is independent and contains manual changes only in `util/testharness.h`. Test Plan: Make sure all tests are passing. ```lang=bash % USE_CLANG=1 make check ``` Reviewers: igor, lgalanis, sdong, yufei.zhu, rven, meyering Reviewed By: meyering Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D33333
10 years ago
EXPECT_TRUE(s.ok());
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
TableProperties* props = const_cast<TableProperties*>(table_properties);
if (store_index_in_file) {
auto bloom_version_ptr = props->user_collected_properties.find(
PlainTablePropertyNames::kBloomVersion);
rocksdb: Replace ASSERT* with EXPECT* in functions that does not return void value Summary: gtest does not use exceptions to fail a unit test by design, and `ASSERT*`s are implemented using `return`. As a consequence we cannot use `ASSERT*` in a function that does not return `void` value ([[ https://code.google.com/p/googletest/wiki/AdvancedGuide#Assertion_Placement | 1]]), and have to fix our existing code. This diff does this in a generic way, with no manual changes. In order to detect all existing `ASSERT*` that are used in functions that doesn't return void value, I change the code to generate compile errors for such cases. In `util/testharness.h` I defined `EXPECT*` assertions, the same way as `ASSERT*`, and redefined `ASSERT*` to return `void`. Then executed: ```lang=bash % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -e 'print "-- -number=".$F[1]." ".$F[0]."\n" if /: error:/' \ build.log | xargs -L 1 perl -spi -e 's/ASSERT/EXPECT/g if $. == $number' % make format ``` After that I reverted back change to `ASSERT*` in `util/testharness.h`. But preserved introduced `EXPECT*`, which is the same as `ASSERT*`. This will be deleted once switched to gtest. This diff is independent and contains manual changes only in `util/testharness.h`. Test Plan: Make sure all tests are passing. ```lang=bash % USE_CLANG=1 make check ``` Reviewers: igor, lgalanis, sdong, yufei.zhu, rven, meyering Reviewed By: meyering Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D33333
10 years ago
EXPECT_TRUE(bloom_version_ptr != props->user_collected_properties.end());
EXPECT_EQ(bloom_version_ptr->second, std::string("1"));
introduce ImmutableOptions Summary: As a preparation to support updating some options dynamically, I'd like to first introduce ImmutableOptions, which is a subset of Options that cannot be changed during the course of a DB lifetime without restart. ColumnFamily will keep both Options and ImmutableOptions. Any component below ColumnFamily should only take ImmutableOptions in their constructor. Other options should be taken from APIs, which will be allowed to adjust dynamically. I am yet to make changes to memtable and other related classes to take ImmutableOptions in their ctor. That can be done in a seprate diff as this one is already pretty big. Test Plan: make all check Reviewers: yhchiang, igor, sdong Reviewed By: sdong Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D22545
10 years ago
if (ioptions.bloom_locality > 0) {
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
auto num_blocks_ptr = props->user_collected_properties.find(
PlainTablePropertyNames::kNumBloomBlocks);
rocksdb: Replace ASSERT* with EXPECT* in functions that does not return void value Summary: gtest does not use exceptions to fail a unit test by design, and `ASSERT*`s are implemented using `return`. As a consequence we cannot use `ASSERT*` in a function that does not return `void` value ([[ https://code.google.com/p/googletest/wiki/AdvancedGuide#Assertion_Placement | 1]]), and have to fix our existing code. This diff does this in a generic way, with no manual changes. In order to detect all existing `ASSERT*` that are used in functions that doesn't return void value, I change the code to generate compile errors for such cases. In `util/testharness.h` I defined `EXPECT*` assertions, the same way as `ASSERT*`, and redefined `ASSERT*` to return `void`. Then executed: ```lang=bash % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -e 'print "-- -number=".$F[1]." ".$F[0]."\n" if /: error:/' \ build.log | xargs -L 1 perl -spi -e 's/ASSERT/EXPECT/g if $. == $number' % make format ``` After that I reverted back change to `ASSERT*` in `util/testharness.h`. But preserved introduced `EXPECT*`, which is the same as `ASSERT*`. This will be deleted once switched to gtest. This diff is independent and contains manual changes only in `util/testharness.h`. Test Plan: Make sure all tests are passing. ```lang=bash % USE_CLANG=1 make check ``` Reviewers: igor, lgalanis, sdong, yufei.zhu, rven, meyering Reviewed By: meyering Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D33333
10 years ago
EXPECT_TRUE(num_blocks_ptr != props->user_collected_properties.end());
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
}
}
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
}
virtual ~TestPlainTableReader() {}
private:
virtual bool MatchBloom(uint32_t hash) const override {
bool ret = PlainTableReader::MatchBloom(hash);
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
if (*expect_bloom_not_match_) {
rocksdb: Replace ASSERT* with EXPECT* in functions that does not return void value Summary: gtest does not use exceptions to fail a unit test by design, and `ASSERT*`s are implemented using `return`. As a consequence we cannot use `ASSERT*` in a function that does not return `void` value ([[ https://code.google.com/p/googletest/wiki/AdvancedGuide#Assertion_Placement | 1]]), and have to fix our existing code. This diff does this in a generic way, with no manual changes. In order to detect all existing `ASSERT*` that are used in functions that doesn't return void value, I change the code to generate compile errors for such cases. In `util/testharness.h` I defined `EXPECT*` assertions, the same way as `ASSERT*`, and redefined `ASSERT*` to return `void`. Then executed: ```lang=bash % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -e 'print "-- -number=".$F[1]." ".$F[0]."\n" if /: error:/' \ build.log | xargs -L 1 perl -spi -e 's/ASSERT/EXPECT/g if $. == $number' % make format ``` After that I reverted back change to `ASSERT*` in `util/testharness.h`. But preserved introduced `EXPECT*`, which is the same as `ASSERT*`. This will be deleted once switched to gtest. This diff is independent and contains manual changes only in `util/testharness.h`. Test Plan: Make sure all tests are passing. ```lang=bash % USE_CLANG=1 make check ``` Reviewers: igor, lgalanis, sdong, yufei.zhu, rven, meyering Reviewed By: meyering Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D33333
10 years ago
EXPECT_TRUE(!ret);
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
} else {
rocksdb: Replace ASSERT* with EXPECT* in functions that does not return void value Summary: gtest does not use exceptions to fail a unit test by design, and `ASSERT*`s are implemented using `return`. As a consequence we cannot use `ASSERT*` in a function that does not return `void` value ([[ https://code.google.com/p/googletest/wiki/AdvancedGuide#Assertion_Placement | 1]]), and have to fix our existing code. This diff does this in a generic way, with no manual changes. In order to detect all existing `ASSERT*` that are used in functions that doesn't return void value, I change the code to generate compile errors for such cases. In `util/testharness.h` I defined `EXPECT*` assertions, the same way as `ASSERT*`, and redefined `ASSERT*` to return `void`. Then executed: ```lang=bash % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -e 'print "-- -number=".$F[1]." ".$F[0]."\n" if /: error:/' \ build.log | xargs -L 1 perl -spi -e 's/ASSERT/EXPECT/g if $. == $number' % make format ``` After that I reverted back change to `ASSERT*` in `util/testharness.h`. But preserved introduced `EXPECT*`, which is the same as `ASSERT*`. This will be deleted once switched to gtest. This diff is independent and contains manual changes only in `util/testharness.h`. Test Plan: Make sure all tests are passing. ```lang=bash % USE_CLANG=1 make check ``` Reviewers: igor, lgalanis, sdong, yufei.zhu, rven, meyering Reviewed By: meyering Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D33333
10 years ago
EXPECT_TRUE(ret);
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
}
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
return ret;
}
bool* expect_bloom_not_match_;
};
extern const uint64_t kPlainTableMagicNumber;
class TestPlainTableFactory : public PlainTableFactory {
public:
explicit TestPlainTableFactory(bool* expect_bloom_not_match,
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
const PlainTableOptions& options)
: PlainTableFactory(options),
bloom_bits_per_key_(options.bloom_bits_per_key),
hash_table_ratio_(options.hash_table_ratio),
index_sparseness_(options.index_sparseness),
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
store_index_in_file_(options.store_index_in_file),
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
expect_bloom_not_match_(expect_bloom_not_match) {}
introduce ImmutableOptions Summary: As a preparation to support updating some options dynamically, I'd like to first introduce ImmutableOptions, which is a subset of Options that cannot be changed during the course of a DB lifetime without restart. ColumnFamily will keep both Options and ImmutableOptions. Any component below ColumnFamily should only take ImmutableOptions in their constructor. Other options should be taken from APIs, which will be allowed to adjust dynamically. I am yet to make changes to memtable and other related classes to take ImmutableOptions in their ctor. That can be done in a seprate diff as this one is already pretty big. Test Plan: make all check Reviewers: yhchiang, igor, sdong Reviewed By: sdong Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D22545
10 years ago
Status NewTableReader(const ImmutableCFOptions& ioptions,
const EnvOptions& env_options,
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
const InternalKeyComparator& internal_comparator,
Move rate_limiter, write buffering, most perf context instrumentation and most random kill out of Env Summary: We want to keep Env a think layer for better portability. Less platform dependent codes should be moved out of Env. In this patch, I create a wrapper of file readers and writers, and put rate limiting, write buffering, as well as most perf context instrumentation and random kill out of Env. It will make it easier to maintain multiple Env in the future. Test Plan: Run all existing unit tests. Reviewers: anthony, kradhakrishnan, IslamAbdelRahman, yhchiang, igor Reviewed By: igor Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D42321
9 years ago
unique_ptr<RandomAccessFileReader>&& file,
uint64_t file_size,
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
unique_ptr<TableReader>* table) const override {
TableProperties* props = nullptr;
auto s = ReadTableProperties(file.get(), file_size, kPlainTableMagicNumber,
introduce ImmutableOptions Summary: As a preparation to support updating some options dynamically, I'd like to first introduce ImmutableOptions, which is a subset of Options that cannot be changed during the course of a DB lifetime without restart. ColumnFamily will keep both Options and ImmutableOptions. Any component below ColumnFamily should only take ImmutableOptions in their constructor. Other options should be taken from APIs, which will be allowed to adjust dynamically. I am yet to make changes to memtable and other related classes to take ImmutableOptions in their ctor. That can be done in a seprate diff as this one is already pretty big. Test Plan: make all check Reviewers: yhchiang, igor, sdong Reviewed By: sdong Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D22545
10 years ago
ioptions.env, ioptions.info_log, &props);
rocksdb: Replace ASSERT* with EXPECT* in functions that does not return void value Summary: gtest does not use exceptions to fail a unit test by design, and `ASSERT*`s are implemented using `return`. As a consequence we cannot use `ASSERT*` in a function that does not return `void` value ([[ https://code.google.com/p/googletest/wiki/AdvancedGuide#Assertion_Placement | 1]]), and have to fix our existing code. This diff does this in a generic way, with no manual changes. In order to detect all existing `ASSERT*` that are used in functions that doesn't return void value, I change the code to generate compile errors for such cases. In `util/testharness.h` I defined `EXPECT*` assertions, the same way as `ASSERT*`, and redefined `ASSERT*` to return `void`. Then executed: ```lang=bash % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -e 'print "-- -number=".$F[1]." ".$F[0]."\n" if /: error:/' \ build.log | xargs -L 1 perl -spi -e 's/ASSERT/EXPECT/g if $. == $number' % make format ``` After that I reverted back change to `ASSERT*` in `util/testharness.h`. But preserved introduced `EXPECT*`, which is the same as `ASSERT*`. This will be deleted once switched to gtest. This diff is independent and contains manual changes only in `util/testharness.h`. Test Plan: Make sure all tests are passing. ```lang=bash % USE_CLANG=1 make check ``` Reviewers: igor, lgalanis, sdong, yufei.zhu, rven, meyering Reviewed By: meyering Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D33333
10 years ago
EXPECT_TRUE(s.ok());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
if (store_index_in_file_) {
BlockHandle bloom_block_handle;
s = FindMetaBlock(file.get(), file_size, kPlainTableMagicNumber,
introduce ImmutableOptions Summary: As a preparation to support updating some options dynamically, I'd like to first introduce ImmutableOptions, which is a subset of Options that cannot be changed during the course of a DB lifetime without restart. ColumnFamily will keep both Options and ImmutableOptions. Any component below ColumnFamily should only take ImmutableOptions in their constructor. Other options should be taken from APIs, which will be allowed to adjust dynamically. I am yet to make changes to memtable and other related classes to take ImmutableOptions in their ctor. That can be done in a seprate diff as this one is already pretty big. Test Plan: make all check Reviewers: yhchiang, igor, sdong Reviewed By: sdong Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D22545
10 years ago
ioptions.env, BloomBlockBuilder::kBloomBlock,
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
&bloom_block_handle);
rocksdb: Replace ASSERT* with EXPECT* in functions that does not return void value Summary: gtest does not use exceptions to fail a unit test by design, and `ASSERT*`s are implemented using `return`. As a consequence we cannot use `ASSERT*` in a function that does not return `void` value ([[ https://code.google.com/p/googletest/wiki/AdvancedGuide#Assertion_Placement | 1]]), and have to fix our existing code. This diff does this in a generic way, with no manual changes. In order to detect all existing `ASSERT*` that are used in functions that doesn't return void value, I change the code to generate compile errors for such cases. In `util/testharness.h` I defined `EXPECT*` assertions, the same way as `ASSERT*`, and redefined `ASSERT*` to return `void`. Then executed: ```lang=bash % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -e 'print "-- -number=".$F[1]." ".$F[0]."\n" if /: error:/' \ build.log | xargs -L 1 perl -spi -e 's/ASSERT/EXPECT/g if $. == $number' % make format ``` After that I reverted back change to `ASSERT*` in `util/testharness.h`. But preserved introduced `EXPECT*`, which is the same as `ASSERT*`. This will be deleted once switched to gtest. This diff is independent and contains manual changes only in `util/testharness.h`. Test Plan: Make sure all tests are passing. ```lang=bash % USE_CLANG=1 make check ``` Reviewers: igor, lgalanis, sdong, yufei.zhu, rven, meyering Reviewed By: meyering Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D33333
10 years ago
EXPECT_TRUE(s.ok());
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
BlockHandle index_block_handle;
s = FindMetaBlock(
introduce ImmutableOptions Summary: As a preparation to support updating some options dynamically, I'd like to first introduce ImmutableOptions, which is a subset of Options that cannot be changed during the course of a DB lifetime without restart. ColumnFamily will keep both Options and ImmutableOptions. Any component below ColumnFamily should only take ImmutableOptions in their constructor. Other options should be taken from APIs, which will be allowed to adjust dynamically. I am yet to make changes to memtable and other related classes to take ImmutableOptions in their ctor. That can be done in a seprate diff as this one is already pretty big. Test Plan: make all check Reviewers: yhchiang, igor, sdong Reviewed By: sdong Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D22545
10 years ago
file.get(), file_size, kPlainTableMagicNumber, ioptions.env,
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
PlainTableIndexBuilder::kPlainTableIndexBlock, &index_block_handle);
rocksdb: Replace ASSERT* with EXPECT* in functions that does not return void value Summary: gtest does not use exceptions to fail a unit test by design, and `ASSERT*`s are implemented using `return`. As a consequence we cannot use `ASSERT*` in a function that does not return `void` value ([[ https://code.google.com/p/googletest/wiki/AdvancedGuide#Assertion_Placement | 1]]), and have to fix our existing code. This diff does this in a generic way, with no manual changes. In order to detect all existing `ASSERT*` that are used in functions that doesn't return void value, I change the code to generate compile errors for such cases. In `util/testharness.h` I defined `EXPECT*` assertions, the same way as `ASSERT*`, and redefined `ASSERT*` to return `void`. Then executed: ```lang=bash % USE_CLANG=1 make all -j55 -k 2> build.log % perl -naF: -e 'print "-- -number=".$F[1]." ".$F[0]."\n" if /: error:/' \ build.log | xargs -L 1 perl -spi -e 's/ASSERT/EXPECT/g if $. == $number' % make format ``` After that I reverted back change to `ASSERT*` in `util/testharness.h`. But preserved introduced `EXPECT*`, which is the same as `ASSERT*`. This will be deleted once switched to gtest. This diff is independent and contains manual changes only in `util/testharness.h`. Test Plan: Make sure all tests are passing. ```lang=bash % USE_CLANG=1 make check ``` Reviewers: igor, lgalanis, sdong, yufei.zhu, rven, meyering Reviewed By: meyering Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D33333
10 years ago
EXPECT_TRUE(s.ok());
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
}
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
auto& user_props = props->user_collected_properties;
auto encoding_type_prop =
user_props.find(PlainTablePropertyNames::kEncodingType);
assert(encoding_type_prop != user_props.end());
EncodingType encoding_type = static_cast<EncodingType>(
DecodeFixed32(encoding_type_prop->second.c_str()));
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
std::unique_ptr<PlainTableReader> new_reader(new TestPlainTableReader(
introduce ImmutableOptions Summary: As a preparation to support updating some options dynamically, I'd like to first introduce ImmutableOptions, which is a subset of Options that cannot be changed during the course of a DB lifetime without restart. ColumnFamily will keep both Options and ImmutableOptions. Any component below ColumnFamily should only take ImmutableOptions in their constructor. Other options should be taken from APIs, which will be allowed to adjust dynamically. I am yet to make changes to memtable and other related classes to take ImmutableOptions in their ctor. That can be done in a seprate diff as this one is already pretty big. Test Plan: make all check Reviewers: yhchiang, igor, sdong Reviewed By: sdong Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D22545
10 years ago
env_options, internal_comparator, encoding_type, file_size,
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
bloom_bits_per_key_, hash_table_ratio_, index_sparseness_, props,
introduce ImmutableOptions Summary: As a preparation to support updating some options dynamically, I'd like to first introduce ImmutableOptions, which is a subset of Options that cannot be changed during the course of a DB lifetime without restart. ColumnFamily will keep both Options and ImmutableOptions. Any component below ColumnFamily should only take ImmutableOptions in their constructor. Other options should be taken from APIs, which will be allowed to adjust dynamically. I am yet to make changes to memtable and other related classes to take ImmutableOptions in their ctor. That can be done in a seprate diff as this one is already pretty big. Test Plan: make all check Reviewers: yhchiang, igor, sdong Reviewed By: sdong Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D22545
10 years ago
std::move(file), ioptions, expect_bloom_not_match_,
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
store_index_in_file_));
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
*table = std::move(new_reader);
return s;
}
private:
int bloom_bits_per_key_;
double hash_table_ratio_;
size_t index_sparseness_;
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
bool store_index_in_file_;
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
bool* expect_bloom_not_match_;
};
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
TEST_F(PlainTableDBTest, Flush) {
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
for (size_t huge_page_tlb_size = 0; huge_page_tlb_size <= 2 * 1024 * 1024;
huge_page_tlb_size += 2 * 1024 * 1024) {
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
for (EncodingType encoding_type : {kPlain, kPrefix}) {
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
for (int bloom_bits = 0; bloom_bits <= 117; bloom_bits += 117) {
for (int total_order = 0; total_order <= 1; total_order++) {
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
for (int store_index_in_file = 0; store_index_in_file <= 1;
++store_index_in_file) {
if (!bloom_bits && store_index_in_file) {
continue;
}
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
Options options = CurrentOptions();
options.create_if_missing = true;
// Set only one bucket to force bucket conflict.
// Test index interval for the same prefix to be 1, 2 and 4
if (total_order) {
options.prefix_extractor.reset();
PlainTableOptions plain_table_options;
plain_table_options.user_key_len = 0;
plain_table_options.bloom_bits_per_key = bloom_bits;
plain_table_options.hash_table_ratio = 0;
plain_table_options.index_sparseness = 2;
plain_table_options.huge_page_tlb_size = huge_page_tlb_size;
plain_table_options.encoding_type = encoding_type;
plain_table_options.full_scan_mode = false;
plain_table_options.store_index_in_file = store_index_in_file;
options.table_factory.reset(
NewPlainTableFactory(plain_table_options));
} else {
PlainTableOptions plain_table_options;
plain_table_options.user_key_len = 0;
plain_table_options.bloom_bits_per_key = bloom_bits;
plain_table_options.hash_table_ratio = 0.75;
plain_table_options.index_sparseness = 16;
plain_table_options.huge_page_tlb_size = huge_page_tlb_size;
plain_table_options.encoding_type = encoding_type;
plain_table_options.full_scan_mode = false;
plain_table_options.store_index_in_file = store_index_in_file;
options.table_factory.reset(
NewPlainTableFactory(plain_table_options));
}
DestroyAndReopen(&options);
Add DB property "rocksdb.estimate-table-readers-mem" Summary: Add a DB Property "rocksdb.estimate-table-readers-mem" to return estimated memory usage by all loaded table readers, other than allocated from block cache. Refactor the property codes to allow getting property from a version, with DB mutex not acquired. Test Plan: Add several checks of this new property in existing codes for various cases. Reviewers: yhchiang, ljin Reviewed By: ljin Subscribers: xjin, igor, leveldb Differential Revision: https://reviews.facebook.net/D20733
10 years ago
uint64_t int_num;
ASSERT_TRUE(dbfull()->GetIntProperty(
"rocksdb.estimate-table-readers-mem", &int_num));
ASSERT_EQ(int_num, 0U);
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
ASSERT_OK(Put("1000000000000foo", "v1"));
ASSERT_OK(Put("0000000000000bar", "v2"));
ASSERT_OK(Put("1000000000000foo", "v3"));
dbfull()->TEST_FlushMemTable();
Add DB property "rocksdb.estimate-table-readers-mem" Summary: Add a DB Property "rocksdb.estimate-table-readers-mem" to return estimated memory usage by all loaded table readers, other than allocated from block cache. Refactor the property codes to allow getting property from a version, with DB mutex not acquired. Test Plan: Add several checks of this new property in existing codes for various cases. Reviewers: yhchiang, ljin Reviewed By: ljin Subscribers: xjin, igor, leveldb Differential Revision: https://reviews.facebook.net/D20733
10 years ago
ASSERT_TRUE(dbfull()->GetIntProperty(
"rocksdb.estimate-table-readers-mem", &int_num));
ASSERT_GT(int_num, 0U);
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
TablePropertiesCollection ptc;
reinterpret_cast<DB*>(dbfull())->GetPropertiesOfAllTables(&ptc);
ASSERT_EQ(1U, ptc.size());
auto row = ptc.begin();
auto tp = row->second;
if (!store_index_in_file) {
ASSERT_EQ(total_order ? "4" : "12",
(tp->user_collected_properties)
.at("plain_table_hash_table_size"));
ASSERT_EQ("0", (tp->user_collected_properties)
.at("plain_table_sub_index_size"));
} else {
ASSERT_EQ("0", (tp->user_collected_properties)
.at("plain_table_hash_table_size"));
ASSERT_EQ("0", (tp->user_collected_properties)
.at("plain_table_sub_index_size"));
}
ASSERT_EQ("v3", Get("1000000000000foo"));
ASSERT_EQ("v2", Get("0000000000000bar"));
}
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
}
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
}
}
}
}
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
TEST_F(PlainTableDBTest, Flush2) {
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
for (size_t huge_page_tlb_size = 0; huge_page_tlb_size <= 2 * 1024 * 1024;
huge_page_tlb_size += 2 * 1024 * 1024) {
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
for (EncodingType encoding_type : {kPlain, kPrefix}) {
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
for (int bloom_bits = 0; bloom_bits <= 117; bloom_bits += 117) {
for (int total_order = 0; total_order <= 1; total_order++) {
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
for (int store_index_in_file = 0; store_index_in_file <= 1;
++store_index_in_file) {
if (encoding_type == kPrefix && total_order) {
continue;
}
if (!bloom_bits && store_index_in_file) {
continue;
}
if (total_order && store_index_in_file) {
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
continue;
}
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
bool expect_bloom_not_match = false;
Options options = CurrentOptions();
options.create_if_missing = true;
// Set only one bucket to force bucket conflict.
// Test index interval for the same prefix to be 1, 2 and 4
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
PlainTableOptions plain_table_options;
Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB" This reverts commit 7dafa3a1d7e63d4cc4a7d717ab958efc616a2892.
11 years ago
if (total_order) {
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
options.prefix_extractor = nullptr;
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
plain_table_options.hash_table_ratio = 0;
plain_table_options.index_sparseness = 2;
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
} else {
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
plain_table_options.hash_table_ratio = 0.75;
plain_table_options.index_sparseness = 16;
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
}
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
plain_table_options.user_key_len = kPlainTableVariableLength;
plain_table_options.bloom_bits_per_key = bloom_bits;
plain_table_options.huge_page_tlb_size = huge_page_tlb_size;
plain_table_options.encoding_type = encoding_type;
plain_table_options.store_index_in_file = store_index_in_file;
options.table_factory.reset(new TestPlainTableFactory(
&expect_bloom_not_match, plain_table_options));
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
DestroyAndReopen(&options);
ASSERT_OK(Put("0000000000000bar", "b"));
ASSERT_OK(Put("1000000000000foo", "v1"));
dbfull()->TEST_FlushMemTable();
ASSERT_OK(Put("1000000000000foo", "v2"));
dbfull()->TEST_FlushMemTable();
ASSERT_EQ("v2", Get("1000000000000foo"));
ASSERT_OK(Put("0000000000000eee", "v3"));
dbfull()->TEST_FlushMemTable();
ASSERT_EQ("v3", Get("0000000000000eee"));
ASSERT_OK(Delete("0000000000000bar"));
dbfull()->TEST_FlushMemTable();
ASSERT_EQ("NOT_FOUND", Get("0000000000000bar"));
ASSERT_OK(Put("0000000000000eee", "v5"));
ASSERT_OK(Put("9000000000000eee", "v5"));
dbfull()->TEST_FlushMemTable();
ASSERT_EQ("v5", Get("0000000000000eee"));
// Test Bloom Filter
if (bloom_bits > 0) {
// Neither key nor value should exist.
expect_bloom_not_match = true;
ASSERT_EQ("NOT_FOUND", Get("5_not00000000bar"));
// Key doesn't exist any more but prefix exists.
if (total_order) {
ASSERT_EQ("NOT_FOUND", Get("1000000000000not"));
ASSERT_EQ("NOT_FOUND", Get("0000000000000not"));
}
expect_bloom_not_match = false;
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
}
}
Adding option to save PlainTable index and bloom filter in SST file. Summary: Adding option to save PlainTable index and bloom filter in SST file. If there is no bloom block and/or index block, PlainTableReader builds new ones. Otherwise PlainTableReader just use these blocks. Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19527
10 years ago
}
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
}
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
}
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
}
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
}
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
TEST_F(PlainTableDBTest, Iterator) {
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
for (size_t huge_page_tlb_size = 0; huge_page_tlb_size <= 2 * 1024 * 1024;
huge_page_tlb_size += 2 * 1024 * 1024) {
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
for (EncodingType encoding_type : {kPlain, kPrefix}) {
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
for (int bloom_bits = 0; bloom_bits <= 117; bloom_bits += 117) {
for (int total_order = 0; total_order <= 1; total_order++) {
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
if (encoding_type == kPrefix && total_order == 1) {
continue;
}
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
bool expect_bloom_not_match = false;
Options options = CurrentOptions();
options.create_if_missing = true;
// Set only one bucket to force bucket conflict.
// Test index interval for the same prefix to be 1, 2 and 4
if (total_order) {
options.prefix_extractor = nullptr;
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
PlainTableOptions plain_table_options;
plain_table_options.user_key_len = 16;
plain_table_options.bloom_bits_per_key = bloom_bits;
plain_table_options.hash_table_ratio = 0;
plain_table_options.index_sparseness = 2;
plain_table_options.huge_page_tlb_size = huge_page_tlb_size;
plain_table_options.encoding_type = encoding_type;
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
options.table_factory.reset(new TestPlainTableFactory(
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
&expect_bloom_not_match, plain_table_options));
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
} else {
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
PlainTableOptions plain_table_options;
plain_table_options.user_key_len = 16;
plain_table_options.bloom_bits_per_key = bloom_bits;
plain_table_options.hash_table_ratio = 0.75;
plain_table_options.index_sparseness = 16;
plain_table_options.huge_page_tlb_size = huge_page_tlb_size;
plain_table_options.encoding_type = encoding_type;
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
options.table_factory.reset(new TestPlainTableFactory(
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
&expect_bloom_not_match, plain_table_options));
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
}
DestroyAndReopen(&options);
ASSERT_OK(Put("1000000000foo002", "v_2"));
ASSERT_OK(Put("0000000000000bar", "random"));
ASSERT_OK(Put("1000000000foo001", "v1"));
ASSERT_OK(Put("3000000000000bar", "bar_v"));
ASSERT_OK(Put("1000000000foo003", "v__3"));
ASSERT_OK(Put("1000000000foo004", "v__4"));
ASSERT_OK(Put("1000000000foo005", "v__5"));
ASSERT_OK(Put("1000000000foo007", "v__7"));
ASSERT_OK(Put("1000000000foo008", "v__8"));
dbfull()->TEST_FlushMemTable();
ASSERT_EQ("v1", Get("1000000000foo001"));
ASSERT_EQ("v__3", Get("1000000000foo003"));
Iterator* iter = dbfull()->NewIterator(ReadOptions());
iter->Seek("1000000000foo000");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("1000000000foo001", iter->key().ToString());
ASSERT_EQ("v1", iter->value().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Next();
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("1000000000foo002", iter->key().ToString());
ASSERT_EQ("v_2", iter->value().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Next();
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("1000000000foo003", iter->key().ToString());
ASSERT_EQ("v__3", iter->value().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Next();
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("1000000000foo004", iter->key().ToString());
ASSERT_EQ("v__4", iter->value().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("3000000000000bar");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("3000000000000bar", iter->key().ToString());
ASSERT_EQ("bar_v", iter->value().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("1000000000foo000");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("1000000000foo001", iter->key().ToString());
ASSERT_EQ("v1", iter->value().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("1000000000foo005");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("1000000000foo005", iter->key().ToString());
ASSERT_EQ("v__5", iter->value().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("1000000000foo006");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("1000000000foo007", iter->key().ToString());
ASSERT_EQ("v__7", iter->value().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("1000000000foo008");
Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB" This reverts commit 7dafa3a1d7e63d4cc4a7d717ab958efc616a2892.
11 years ago
ASSERT_TRUE(iter->Valid());
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
ASSERT_EQ("1000000000foo008", iter->key().ToString());
ASSERT_EQ("v__8", iter->value().ToString());
Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB Summary: Add an option to allocate a piece of memory from huge page TLB. Add options to trigger it in dynamic bloom, plain table indexes andhash linked list hash table. Test Plan: make all check Reviewers: haobo, ljin Reviewed By: haobo CC: nkg-, dhruba, leveldb, igor, yhchiang Differential Revision: https://reviews.facebook.net/D18357
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
if (total_order == 0) {
iter->Seek("1000000000foo009");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("3000000000000bar", iter->key().ToString());
Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB" This reverts commit 7dafa3a1d7e63d4cc4a7d717ab958efc616a2892.
11 years ago
}
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
// Test Bloom Filter
if (bloom_bits > 0) {
if (!total_order) {
// Neither key nor value should exist.
expect_bloom_not_match = true;
iter->Seek("2not000000000bar");
ASSERT_TRUE(!iter->Valid());
ASSERT_EQ("NOT_FOUND", Get("2not000000000bar"));
expect_bloom_not_match = false;
} else {
expect_bloom_not_match = true;
ASSERT_EQ("NOT_FOUND", Get("2not000000000bar"));
expect_bloom_not_match = false;
}
}
delete iter;
}
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
}
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
}
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
}
}
Turn on -Wmissing-prototypes Summary: Compiling for iOS has by default turned on -Wmissing-prototypes, which causes rocksdb to fail compiling. This diff turns on -Wmissing-prototypes in our compile options and cleans up all functions with missing prototypes. Test Plan: compiles Reviewers: dhruba, haobo, ljin, sdong Reviewed By: ljin CC: leveldb Differential Revision: https://reviews.facebook.net/D17649
11 years ago
namespace {
PlainTableIterator not to store copied key in std::string Summary: Move PlainTableIterator's copied key from std::string local buffer to avoid paying the extra costs in std::string related to sharing. Reuse the same buffer class in DbIter. Move the class to dbformat.h. This patch improves iterator performance significantly. Running this benchmark: ./table_reader_bench --num_keys2=17 --iterator --plain_table --time_unit=nanosecond The average latency is improved to about 750 nanoseconds from 1100 nanoseconds. Test Plan: Add a unit test. make all check Reviewers: haobo, ljin Reviewed By: haobo CC: igor, yhchiang, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D17547
11 years ago
std::string MakeLongKey(size_t length, char c) {
return std::string(length, c);
}
Turn on -Wmissing-prototypes Summary: Compiling for iOS has by default turned on -Wmissing-prototypes, which causes rocksdb to fail compiling. This diff turns on -Wmissing-prototypes in our compile options and cleans up all functions with missing prototypes. Test Plan: compiles Reviewers: dhruba, haobo, ljin, sdong Reviewed By: ljin CC: leveldb Differential Revision: https://reviews.facebook.net/D17649
11 years ago
} // namespace
PlainTableIterator not to store copied key in std::string Summary: Move PlainTableIterator's copied key from std::string local buffer to avoid paying the extra costs in std::string related to sharing. Reuse the same buffer class in DbIter. Move the class to dbformat.h. This patch improves iterator performance significantly. Running this benchmark: ./table_reader_bench --num_keys2=17 --iterator --plain_table --time_unit=nanosecond The average latency is improved to about 750 nanoseconds from 1100 nanoseconds. Test Plan: Add a unit test. make all check Reviewers: haobo, ljin Reviewed By: haobo CC: igor, yhchiang, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D17547
11 years ago
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
TEST_F(PlainTableDBTest, IteratorLargeKeys) {
PlainTableIterator not to store copied key in std::string Summary: Move PlainTableIterator's copied key from std::string local buffer to avoid paying the extra costs in std::string related to sharing. Reuse the same buffer class in DbIter. Move the class to dbformat.h. This patch improves iterator performance significantly. Running this benchmark: ./table_reader_bench --num_keys2=17 --iterator --plain_table --time_unit=nanosecond The average latency is improved to about 750 nanoseconds from 1100 nanoseconds. Test Plan: Add a unit test. make all check Reviewers: haobo, ljin Reviewed By: haobo CC: igor, yhchiang, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D17547
11 years ago
Options options = CurrentOptions();
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
PlainTableOptions plain_table_options;
plain_table_options.user_key_len = 0;
plain_table_options.bloom_bits_per_key = 0;
plain_table_options.hash_table_ratio = 0;
options.table_factory.reset(NewPlainTableFactory(plain_table_options));
PlainTableIterator not to store copied key in std::string Summary: Move PlainTableIterator's copied key from std::string local buffer to avoid paying the extra costs in std::string related to sharing. Reuse the same buffer class in DbIter. Move the class to dbformat.h. This patch improves iterator performance significantly. Running this benchmark: ./table_reader_bench --num_keys2=17 --iterator --plain_table --time_unit=nanosecond The average latency is improved to about 750 nanoseconds from 1100 nanoseconds. Test Plan: Add a unit test. make all check Reviewers: haobo, ljin Reviewed By: haobo CC: igor, yhchiang, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D17547
11 years ago
options.create_if_missing = true;
options.prefix_extractor.reset();
DestroyAndReopen(&options);
std::string key_list[] = {
MakeLongKey(30, '0'),
MakeLongKey(16, '1'),
MakeLongKey(32, '2'),
MakeLongKey(60, '3'),
MakeLongKey(90, '4'),
MakeLongKey(50, '5'),
MakeLongKey(26, '6')
};
for (size_t i = 0; i < 7; i++) {
Add rocksdb::ToString() to address cases where std::to_string is not available. Summary: In some environment such as android, the c++ library does not have std::to_string. This path adds rocksdb::ToString(), which wraps std::to_string when std::to_string is not available, and implements std::to_string in the other case. Test Plan: make dbg -j32 ./db_test make clean make dbg OPT=-DOS_ANDROID -j32 ./db_test Reviewers: ljin, sdong, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D29181
10 years ago
ASSERT_OK(Put(key_list[i], ToString(i)));
PlainTableIterator not to store copied key in std::string Summary: Move PlainTableIterator's copied key from std::string local buffer to avoid paying the extra costs in std::string related to sharing. Reuse the same buffer class in DbIter. Move the class to dbformat.h. This patch improves iterator performance significantly. Running this benchmark: ./table_reader_bench --num_keys2=17 --iterator --plain_table --time_unit=nanosecond The average latency is improved to about 750 nanoseconds from 1100 nanoseconds. Test Plan: Add a unit test. make all check Reviewers: haobo, ljin Reviewed By: haobo CC: igor, yhchiang, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D17547
11 years ago
}
dbfull()->TEST_FlushMemTable();
kill ReadOptions.prefix and .prefix_seek Summary: also add an override option total_order_iteration if you want to use full iterator with prefix_extractor Test Plan: make all check Reviewers: igor, haobo, sdong, yhchiang Reviewed By: haobo CC: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D17805
11 years ago
Iterator* iter = dbfull()->NewIterator(ReadOptions());
PlainTableIterator not to store copied key in std::string Summary: Move PlainTableIterator's copied key from std::string local buffer to avoid paying the extra costs in std::string related to sharing. Reuse the same buffer class in DbIter. Move the class to dbformat.h. This patch improves iterator performance significantly. Running this benchmark: ./table_reader_bench --num_keys2=17 --iterator --plain_table --time_unit=nanosecond The average latency is improved to about 750 nanoseconds from 1100 nanoseconds. Test Plan: Add a unit test. make all check Reviewers: haobo, ljin Reviewed By: haobo CC: igor, yhchiang, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D17547
11 years ago
iter->Seek(key_list[0]);
for (size_t i = 0; i < 7; i++) {
ASSERT_TRUE(iter->Valid());
ASSERT_EQ(key_list[i], iter->key().ToString());
Add rocksdb::ToString() to address cases where std::to_string is not available. Summary: In some environment such as android, the c++ library does not have std::to_string. This path adds rocksdb::ToString(), which wraps std::to_string when std::to_string is not available, and implements std::to_string in the other case. Test Plan: make dbg -j32 ./db_test make clean make dbg OPT=-DOS_ANDROID -j32 ./db_test Reviewers: ljin, sdong, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D29181
10 years ago
ASSERT_EQ(ToString(i), iter->value().ToString());
PlainTableIterator not to store copied key in std::string Summary: Move PlainTableIterator's copied key from std::string local buffer to avoid paying the extra costs in std::string related to sharing. Reuse the same buffer class in DbIter. Move the class to dbformat.h. This patch improves iterator performance significantly. Running this benchmark: ./table_reader_bench --num_keys2=17 --iterator --plain_table --time_unit=nanosecond The average latency is improved to about 750 nanoseconds from 1100 nanoseconds. Test Plan: Add a unit test. make all check Reviewers: haobo, ljin Reviewed By: haobo CC: igor, yhchiang, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D17547
11 years ago
iter->Next();
}
ASSERT_TRUE(!iter->Valid());
delete iter;
}
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
namespace {
std::string MakeLongKeyWithPrefix(size_t length, char c) {
return "00000000" + std::string(length - 8, c);
}
} // namespace
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
TEST_F(PlainTableDBTest, IteratorLargeKeysWithPrefix) {
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
Options options = CurrentOptions();
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
PlainTableOptions plain_table_options;
plain_table_options.user_key_len = 16;
plain_table_options.bloom_bits_per_key = 0;
plain_table_options.hash_table_ratio = 0.8;
plain_table_options.index_sparseness = 3;
plain_table_options.huge_page_tlb_size = 0;
plain_table_options.encoding_type = kPrefix;
options.table_factory.reset(NewPlainTableFactory(plain_table_options));
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
options.create_if_missing = true;
DestroyAndReopen(&options);
std::string key_list[] = {
MakeLongKeyWithPrefix(30, '0'), MakeLongKeyWithPrefix(16, '1'),
MakeLongKeyWithPrefix(32, '2'), MakeLongKeyWithPrefix(60, '3'),
MakeLongKeyWithPrefix(90, '4'), MakeLongKeyWithPrefix(50, '5'),
MakeLongKeyWithPrefix(26, '6')};
for (size_t i = 0; i < 7; i++) {
Add rocksdb::ToString() to address cases where std::to_string is not available. Summary: In some environment such as android, the c++ library does not have std::to_string. This path adds rocksdb::ToString(), which wraps std::to_string when std::to_string is not available, and implements std::to_string in the other case. Test Plan: make dbg -j32 ./db_test make clean make dbg OPT=-DOS_ANDROID -j32 ./db_test Reviewers: ljin, sdong, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D29181
10 years ago
ASSERT_OK(Put(key_list[i], ToString(i)));
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
}
dbfull()->TEST_FlushMemTable();
Iterator* iter = dbfull()->NewIterator(ReadOptions());
iter->Seek(key_list[0]);
for (size_t i = 0; i < 7; i++) {
ASSERT_TRUE(iter->Valid());
ASSERT_EQ(key_list[i], iter->key().ToString());
Add rocksdb::ToString() to address cases where std::to_string is not available. Summary: In some environment such as android, the c++ library does not have std::to_string. This path adds rocksdb::ToString(), which wraps std::to_string when std::to_string is not available, and implements std::to_string in the other case. Test Plan: make dbg -j32 ./db_test make clean make dbg OPT=-DOS_ANDROID -j32 ./db_test Reviewers: ljin, sdong, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D29181
10 years ago
ASSERT_EQ(ToString(i), iter->value().ToString());
PlainTable to encode to avoid to rewrite prefix when it is the same as the previous key Summary: Add a encoding feature of PlainTable to encode PlainTable's keys to save some bytes for the same prefixes. The data format is documented in table/plain_table_factory.h Test Plan: Add unit test coverage in plain_table_db_test Reviewers: yhchiang, igor, dhruba, ljin, haobo Reviewed By: haobo Subscribers: nkg-, leveldb Differential Revision: https://reviews.facebook.net/D18735
11 years ago
iter->Next();
}
ASSERT_TRUE(!iter->Valid());
delete iter;
}
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
TEST_F(PlainTableDBTest, IteratorReverseSuffixComparator) {
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Options options = CurrentOptions();
options.create_if_missing = true;
// Set only one bucket to force bucket conflict.
// Test index interval for the same prefix to be 1, 2 and 4
Add ComparatorDBTest to test non-default comparators Summary: Add some helper functions to make sure DB works well for non-default comparators. Add a test for SimpleSuffixReverseComparator. Test Plan: Run the new test Reviewers: ljin, rven, yhchiang, igor Reviewed By: igor Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D27831
10 years ago
test::SimpleSuffixReverseComparator comp;
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
options.comparator = &comp;
DestroyAndReopen(&options);
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
ASSERT_OK(Put("1000000000foo002", "v_2"));
ASSERT_OK(Put("0000000000000bar", "random"));
ASSERT_OK(Put("1000000000foo001", "v1"));
ASSERT_OK(Put("3000000000000bar", "bar_v"));
ASSERT_OK(Put("1000000000foo003", "v__3"));
ASSERT_OK(Put("1000000000foo004", "v__4"));
ASSERT_OK(Put("1000000000foo005", "v__5"));
ASSERT_OK(Put("1000000000foo007", "v__7"));
ASSERT_OK(Put("1000000000foo008", "v__8"));
dbfull()->TEST_FlushMemTable();
ASSERT_EQ("v1", Get("1000000000foo001"));
ASSERT_EQ("v__3", Get("1000000000foo003"));
kill ReadOptions.prefix and .prefix_seek Summary: also add an override option total_order_iteration if you want to use full iterator with prefix_extractor Test Plan: make all check Reviewers: igor, haobo, sdong, yhchiang Reviewed By: haobo CC: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D17805
11 years ago
Iterator* iter = dbfull()->NewIterator(ReadOptions());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
iter->Seek("1000000000foo009");
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
ASSERT_TRUE(iter->Valid());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
ASSERT_EQ("1000000000foo008", iter->key().ToString());
ASSERT_EQ("v__8", iter->value().ToString());
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
iter->Next();
ASSERT_TRUE(iter->Valid());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
ASSERT_EQ("1000000000foo007", iter->key().ToString());
ASSERT_EQ("v__7", iter->value().ToString());
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
iter->Next();
ASSERT_TRUE(iter->Valid());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
ASSERT_EQ("1000000000foo005", iter->key().ToString());
ASSERT_EQ("v__5", iter->value().ToString());
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
iter->Next();
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("1000000000foo004", iter->key().ToString());
ASSERT_EQ("v__4", iter->value().ToString());
iter->Seek("3000000000000bar");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("3000000000000bar", iter->key().ToString());
ASSERT_EQ("bar_v", iter->value().ToString());
iter->Seek("1000000000foo005");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("1000000000foo005", iter->key().ToString());
ASSERT_EQ("v__5", iter->value().ToString());
iter->Seek("1000000000foo006");
ASSERT_TRUE(iter->Valid());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
ASSERT_EQ("1000000000foo005", iter->key().ToString());
ASSERT_EQ("v__5", iter->value().ToString());
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
iter->Seek("1000000000foo008");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("1000000000foo008", iter->key().ToString());
ASSERT_EQ("v__8", iter->value().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
iter->Seek("1000000000foo000");
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("3000000000000bar", iter->key().ToString());
delete iter;
}
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
TEST_F(PlainTableDBTest, HashBucketConflict) {
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
for (size_t huge_page_tlb_size = 0; huge_page_tlb_size <= 2 * 1024 * 1024;
huge_page_tlb_size += 2 * 1024 * 1024) {
for (unsigned char i = 1; i <= 3; i++) {
Options options = CurrentOptions();
options.create_if_missing = true;
// Set only one bucket to force bucket conflict.
// Test index interval for the same prefix to be 1, 2 and 4
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
PlainTableOptions plain_table_options;
plain_table_options.user_key_len = 16;
plain_table_options.bloom_bits_per_key = 0;
plain_table_options.hash_table_ratio = 0;
plain_table_options.index_sparseness = 2 ^ i;
plain_table_options.huge_page_tlb_size = huge_page_tlb_size;
options.table_factory.reset(NewPlainTableFactory(plain_table_options));
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
DestroyAndReopen(&options);
ASSERT_OK(Put("5000000000000fo0", "v1"));
ASSERT_OK(Put("5000000000000fo1", "v2"));
ASSERT_OK(Put("5000000000000fo2", "v"));
ASSERT_OK(Put("2000000000000fo0", "v3"));
ASSERT_OK(Put("2000000000000fo1", "v4"));
ASSERT_OK(Put("2000000000000fo2", "v"));
ASSERT_OK(Put("2000000000000fo3", "v"));
Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB Summary: Add an option to allocate a piece of memory from huge page TLB. Add options to trigger it in dynamic bloom, plain table indexes andhash linked list hash table. Test Plan: make all check Reviewers: haobo, ljin Reviewed By: haobo CC: nkg-, dhruba, leveldb, igor, yhchiang Differential Revision: https://reviews.facebook.net/D18357
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
dbfull()->TEST_FlushMemTable();
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
ASSERT_EQ("v1", Get("5000000000000fo0"));
ASSERT_EQ("v2", Get("5000000000000fo1"));
ASSERT_EQ("v3", Get("2000000000000fo0"));
ASSERT_EQ("v4", Get("2000000000000fo1"));
Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB Summary: Add an option to allocate a piece of memory from huge page TLB. Add options to trigger it in dynamic bloom, plain table indexes andhash linked list hash table. Test Plan: make all check Reviewers: haobo, ljin Reviewed By: haobo CC: nkg-, dhruba, leveldb, igor, yhchiang Differential Revision: https://reviews.facebook.net/D18357
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
ASSERT_EQ("NOT_FOUND", Get("5000000000000bar"));
ASSERT_EQ("NOT_FOUND", Get("2000000000000bar"));
ASSERT_EQ("NOT_FOUND", Get("5000000000000fo8"));
ASSERT_EQ("NOT_FOUND", Get("2000000000000fo8"));
Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB Summary: Add an option to allocate a piece of memory from huge page TLB. Add options to trigger it in dynamic bloom, plain table indexes andhash linked list hash table. Test Plan: make all check Reviewers: haobo, ljin Reviewed By: haobo CC: nkg-, dhruba, leveldb, igor, yhchiang Differential Revision: https://reviews.facebook.net/D18357
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
ReadOptions ro;
Iterator* iter = dbfull()->NewIterator(ro);
Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB Summary: Add an option to allocate a piece of memory from huge page TLB. Add options to trigger it in dynamic bloom, plain table indexes andhash linked list hash table. Test Plan: make all check Reviewers: haobo, ljin Reviewed By: haobo CC: nkg-, dhruba, leveldb, igor, yhchiang Differential Revision: https://reviews.facebook.net/D18357
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("5000000000000fo0");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("5000000000000fo0", iter->key().ToString());
iter->Next();
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("5000000000000fo1", iter->key().ToString());
iter->Seek("5000000000000fo1");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("5000000000000fo1", iter->key().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("2000000000000fo0");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("2000000000000fo0", iter->key().ToString());
iter->Next();
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("2000000000000fo1", iter->key().ToString());
iter->Seek("2000000000000fo1");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("2000000000000fo1", iter->key().ToString());
iter->Seek("2000000000000bar");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("2000000000000fo0", iter->key().ToString());
iter->Seek("5000000000000bar");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("5000000000000fo0", iter->key().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("2000000000000fo8");
ASSERT_TRUE(!iter->Valid() ||
options.comparator->Compare(iter->key(), "20000001") > 0);
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("5000000000000fo8");
ASSERT_TRUE(!iter->Valid());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("1000000000000fo2");
ASSERT_TRUE(!iter->Valid());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("3000000000000fo2");
ASSERT_TRUE(!iter->Valid());
Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB Summary: Add an option to allocate a piece of memory from huge page TLB. Add options to trigger it in dynamic bloom, plain table indexes andhash linked list hash table. Test Plan: make all check Reviewers: haobo, ljin Reviewed By: haobo CC: nkg-, dhruba, leveldb, igor, yhchiang Differential Revision: https://reviews.facebook.net/D18357
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("8000000000000fo2");
ASSERT_TRUE(!iter->Valid());
delete iter;
}
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
}
}
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
TEST_F(PlainTableDBTest, HashBucketConflictReverseSuffixComparator) {
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
for (size_t huge_page_tlb_size = 0; huge_page_tlb_size <= 2 * 1024 * 1024;
huge_page_tlb_size += 2 * 1024 * 1024) {
for (unsigned char i = 1; i <= 3; i++) {
Options options = CurrentOptions();
options.create_if_missing = true;
Add ComparatorDBTest to test non-default comparators Summary: Add some helper functions to make sure DB works well for non-default comparators. Add a test for SimpleSuffixReverseComparator. Test Plan: Run the new test Reviewers: ljin, rven, yhchiang, igor Reviewed By: igor Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D27831
10 years ago
test::SimpleSuffixReverseComparator comp;
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
options.comparator = &comp;
// Set only one bucket to force bucket conflict.
// Test index interval for the same prefix to be 1, 2 and 4
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
PlainTableOptions plain_table_options;
plain_table_options.user_key_len = 16;
plain_table_options.bloom_bits_per_key = 0;
plain_table_options.hash_table_ratio = 0;
plain_table_options.index_sparseness = 2 ^ i;
plain_table_options.huge_page_tlb_size = huge_page_tlb_size;
options.table_factory.reset(NewPlainTableFactory(plain_table_options));
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
DestroyAndReopen(&options);
ASSERT_OK(Put("5000000000000fo0", "v1"));
ASSERT_OK(Put("5000000000000fo1", "v2"));
ASSERT_OK(Put("5000000000000fo2", "v"));
ASSERT_OK(Put("2000000000000fo0", "v3"));
ASSERT_OK(Put("2000000000000fo1", "v4"));
ASSERT_OK(Put("2000000000000fo2", "v"));
ASSERT_OK(Put("2000000000000fo3", "v"));
Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB Summary: Add an option to allocate a piece of memory from huge page TLB. Add options to trigger it in dynamic bloom, plain table indexes andhash linked list hash table. Test Plan: make all check Reviewers: haobo, ljin Reviewed By: haobo CC: nkg-, dhruba, leveldb, igor, yhchiang Differential Revision: https://reviews.facebook.net/D18357
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
dbfull()->TEST_FlushMemTable();
Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB Summary: Add an option to allocate a piece of memory from huge page TLB. Add options to trigger it in dynamic bloom, plain table indexes andhash linked list hash table. Test Plan: make all check Reviewers: haobo, ljin Reviewed By: haobo CC: nkg-, dhruba, leveldb, igor, yhchiang Differential Revision: https://reviews.facebook.net/D18357
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
ASSERT_EQ("v1", Get("5000000000000fo0"));
ASSERT_EQ("v2", Get("5000000000000fo1"));
ASSERT_EQ("v3", Get("2000000000000fo0"));
ASSERT_EQ("v4", Get("2000000000000fo1"));
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
ASSERT_EQ("NOT_FOUND", Get("5000000000000bar"));
ASSERT_EQ("NOT_FOUND", Get("2000000000000bar"));
ASSERT_EQ("NOT_FOUND", Get("5000000000000fo8"));
ASSERT_EQ("NOT_FOUND", Get("2000000000000fo8"));
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
ReadOptions ro;
Iterator* iter = dbfull()->NewIterator(ro);
Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB Summary: Add an option to allocate a piece of memory from huge page TLB. Add options to trigger it in dynamic bloom, plain table indexes andhash linked list hash table. Test Plan: make all check Reviewers: haobo, ljin Reviewed By: haobo CC: nkg-, dhruba, leveldb, igor, yhchiang Differential Revision: https://reviews.facebook.net/D18357
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("5000000000000fo1");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("5000000000000fo1", iter->key().ToString());
iter->Next();
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("5000000000000fo0", iter->key().ToString());
iter->Seek("5000000000000fo1");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("5000000000000fo1", iter->key().ToString());
iter->Seek("2000000000000fo1");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("2000000000000fo1", iter->key().ToString());
iter->Next();
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("2000000000000fo0", iter->key().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("2000000000000fo1");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("2000000000000fo1", iter->key().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("2000000000000var");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("2000000000000fo3", iter->key().ToString());
iter->Seek("5000000000000var");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("5000000000000fo2", iter->key().ToString());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
std::string seek_key = "2000000000000bar";
iter->Seek(seek_key);
ASSERT_TRUE(!iter->Valid() ||
options.prefix_extractor->Transform(iter->key()) !=
options.prefix_extractor->Transform(seek_key));
Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB Summary: Add an option to allocate a piece of memory from huge page TLB. Add options to trigger it in dynamic bloom, plain table indexes andhash linked list hash table. Test Plan: make all check Reviewers: haobo, ljin Reviewed By: haobo CC: nkg-, dhruba, leveldb, igor, yhchiang Differential Revision: https://reviews.facebook.net/D18357
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("1000000000000fo2");
ASSERT_TRUE(!iter->Valid());
Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB Summary: Add an option to allocate a piece of memory from huge page TLB. Add options to trigger it in dynamic bloom, plain table indexes andhash linked list hash table. Test Plan: make all check Reviewers: haobo, ljin Reviewed By: haobo CC: nkg-, dhruba, leveldb, igor, yhchiang Differential Revision: https://reviews.facebook.net/D18357
11 years ago
Revert "Revert "Allow allocating dynamic bloom, plain table indexes and hash linked list from huge page TLB"" And make the default 0 for hash linked list memtable This reverts commit d69dc64be78a8da3ce661454655966d11ff61bb6.
11 years ago
iter->Seek("3000000000000fo2");
ASSERT_TRUE(!iter->Valid());
iter->Seek("8000000000000fo2");
ASSERT_TRUE(!iter->Valid());
delete iter;
}
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
}
}
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
TEST_F(PlainTableDBTest, NonExistingKeyToNonEmptyBucket) {
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
Options options = CurrentOptions();
options.create_if_missing = true;
// Set only one bucket to force bucket conflict.
// Test index interval for the same prefix to be 1, 2 and 4
Add PlainTableOptions Summary: Since we have a lot of options for PlainTable, add a struct PlainTableOptions to manage them Test Plan: make all check Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D20175
10 years ago
PlainTableOptions plain_table_options;
plain_table_options.user_key_len = 16;
plain_table_options.bloom_bits_per_key = 0;
plain_table_options.hash_table_ratio = 0;
plain_table_options.index_sparseness = 5;
options.table_factory.reset(NewPlainTableFactory(plain_table_options));
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
DestroyAndReopen(&options);
ASSERT_OK(Put("5000000000000fo0", "v1"));
ASSERT_OK(Put("5000000000000fo1", "v2"));
ASSERT_OK(Put("5000000000000fo2", "v3"));
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
dbfull()->TEST_FlushMemTable();
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
ASSERT_EQ("v1", Get("5000000000000fo0"));
ASSERT_EQ("v2", Get("5000000000000fo1"));
ASSERT_EQ("v3", Get("5000000000000fo2"));
ASSERT_EQ("NOT_FOUND", Get("8000000000000bar"));
ASSERT_EQ("NOT_FOUND", Get("1000000000000bar"));
kill ReadOptions.prefix and .prefix_seek Summary: also add an override option total_order_iteration if you want to use full iterator with prefix_extractor Test Plan: make all check Reviewers: igor, haobo, sdong, yhchiang Reviewed By: haobo CC: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D17805
11 years ago
Iterator* iter = dbfull()->NewIterator(ReadOptions());
Add more black-box tests for PlainTable and explicitly support total order mode Summary: 1. Add some more implementation-aware tests for PlainTable 2. move from a hard-coded one index per 16 rows in one prefix to a configurable number. Also, make hash table ratio = 0 means binary search only. Also fixes some divide 0 risks. 3. Explicitly support total order (only use binary search) 4. some code cleaning up. Test Plan: make all check Reviewers: haobo, kailiu Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D16023
11 years ago
iter->Seek("5000000000000bar");
ASSERT_TRUE(iter->Valid());
ASSERT_EQ("5000000000000fo0", iter->key().ToString());
iter->Seek("5000000000000fo8");
ASSERT_TRUE(!iter->Valid());
iter->Seek("1000000000000fo2");
ASSERT_TRUE(!iter->Valid());
iter->Seek("8000000000000fo2");
ASSERT_TRUE(!iter->Valid());
delete iter;
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
}
static std::string Key(int i) {
char buf[100];
snprintf(buf, sizeof(buf), "key_______%06d", i);
return std::string(buf);
}
static std::string RandomString(Random* rnd, int len) {
std::string r;
test::RandomString(rnd, len, &r);
return r;
}
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
TEST_F(PlainTableDBTest, CompactionTrigger) {
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
Options options = CurrentOptions();
Arena usage to be calculated using malloc_usable_size() Summary: malloc_usable_size() gets a better estimation of memory usage. It is already used to calculate block cache memory usage. Use it in arena too. Test Plan: Run all unit tests Reviewers: anthony, kradhakrishnan, rven, IslamAbdelRahman, yhchiang Reviewed By: yhchiang Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D43317
9 years ago
options.write_buffer_size = 120 << 10; // 100KB
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
options.num_levels = 3;
options.level0_file_num_compaction_trigger = 3;
Reopen(&options);
Random rnd(301);
for (int num = 0; num < options.level0_file_num_compaction_trigger - 1;
num++) {
std::vector<std::string> values;
Arena usage to be calculated using malloc_usable_size() Summary: malloc_usable_size() gets a better estimation of memory usage. It is already used to calculate block cache memory usage. Use it in arena too. Test Plan: Run all unit tests Reviewers: anthony, kradhakrishnan, rven, IslamAbdelRahman, yhchiang Reviewed By: yhchiang Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D43317
9 years ago
// Write 120KB (10 values, each 12K)
for (int i = 0; i < 10; i++) {
values.push_back(RandomString(&rnd, 12000));
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
ASSERT_OK(Put(Key(i), values[i]));
}
Arena usage to be calculated using malloc_usable_size() Summary: malloc_usable_size() gets a better estimation of memory usage. It is already used to calculate block cache memory usage. Use it in arena too. Test Plan: Run all unit tests Reviewers: anthony, kradhakrishnan, rven, IslamAbdelRahman, yhchiang Reviewed By: yhchiang Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D43317
9 years ago
ASSERT_OK(Put(Key(999), ""));
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
dbfull()->TEST_WaitForFlushMemTable();
ASSERT_EQ(NumTableFilesAtLevel(0), num + 1);
}
//generate one more file in level-0, and should trigger level-0 compaction
std::vector<std::string> values;
for (int i = 0; i < 12; i++) {
values.push_back(RandomString(&rnd, 10000));
ASSERT_OK(Put(Key(i), values[i]));
}
Arena usage to be calculated using malloc_usable_size() Summary: malloc_usable_size() gets a better estimation of memory usage. It is already used to calculate block cache memory usage. Use it in arena too. Test Plan: Run all unit tests Reviewers: anthony, kradhakrishnan, rven, IslamAbdelRahman, yhchiang Reviewed By: yhchiang Subscribers: leveldb, dhruba Differential Revision: https://reviews.facebook.net/D43317
9 years ago
ASSERT_OK(Put(Key(999), ""));
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
dbfull()->TEST_WaitForCompact();
ASSERT_EQ(NumTableFilesAtLevel(0), 0);
ASSERT_EQ(NumTableFilesAtLevel(1), 1);
}
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
TEST_F(PlainTableDBTest, AdaptiveTable) {
Add a table factory that can read DB with both of PlainTable and BlockBasedTable in it Summary: The new table factory is used if users want to convert a DB from one table format to the other. A user can use this table to open a DB written using one table format and write new files to another table format. Test Plan: add a unit test Reviewers: haobo, igor Reviewed By: igor Subscribers: dhruba, ljin, yhchiang, leveldb Differential Revision: https://reviews.facebook.net/D19017
11 years ago
Options options = CurrentOptions();
options.create_if_missing = true;
options.table_factory.reset(NewPlainTableFactory());
DestroyAndReopen(&options);
ASSERT_OK(Put("1000000000000foo", "v1"));
ASSERT_OK(Put("0000000000000bar", "v2"));
ASSERT_OK(Put("1000000000000foo", "v3"));
dbfull()->TEST_FlushMemTable();
options.create_if_missing = false;
std::shared_ptr<TableFactory> dummy_factory;
Change order of parameters in adaptive table factory Summary: This is minor, but if we put the writing talbe factory as the third parameter, when we add a new table format, we'll have a situation: 1) block based factory 2) plain table factory 3) output factory 4) new format factory I think it makes more sense to have output as the first parameter. Also, fixed a NewAdaptiveTableFactory() call in unit test Test Plan: unit test Reviewers: sdong Reviewed By: sdong Subscribers: leveldb Differential Revision: https://reviews.facebook.net/D19119
11 years ago
std::shared_ptr<TableFactory> block_based_factory(
NewBlockBasedTableFactory());
options.table_factory.reset(NewAdaptiveTableFactory(
block_based_factory, dummy_factory, dummy_factory));
Add a table factory that can read DB with both of PlainTable and BlockBasedTable in it Summary: The new table factory is used if users want to convert a DB from one table format to the other. A user can use this table to open a DB written using one table format and write new files to another table format. Test Plan: add a unit test Reviewers: haobo, igor Reviewed By: igor Subscribers: dhruba, ljin, yhchiang, leveldb Differential Revision: https://reviews.facebook.net/D19017
11 years ago
Reopen(&options);
ASSERT_EQ("v3", Get("1000000000000foo"));
ASSERT_EQ("v2", Get("0000000000000bar"));
ASSERT_OK(Put("2000000000000foo", "v4"));
ASSERT_OK(Put("3000000000000bar", "v5"));
dbfull()->TEST_FlushMemTable();
ASSERT_EQ("v4", Get("2000000000000foo"));
ASSERT_EQ("v5", Get("3000000000000bar"));
Reopen(&options);
ASSERT_EQ("v3", Get("1000000000000foo"));
ASSERT_EQ("v2", Get("0000000000000bar"));
ASSERT_EQ("v4", Get("2000000000000foo"));
ASSERT_EQ("v5", Get("3000000000000bar"));
options.table_factory.reset(NewBlockBasedTableFactory());
Reopen(&options);
ASSERT_NE("v3", Get("1000000000000foo"));
options.table_factory.reset(NewPlainTableFactory());
Reopen(&options);
ASSERT_NE("v5", Get("3000000000000bar"));
}
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
} // namespace rocksdb
int main(int argc, char** argv) {
rocksdb: switch to gtest Summary: Our existing test notation is very similar to what is used in gtest. It makes it easy to adopt what is different. In this diff I modify existing [[ https://code.google.com/p/googletest/wiki/Primer#Test_Fixtures:_Using_the_Same_Data_Configuration_for_Multiple_Te | test fixture ]] classes to inherit from `testing::Test`. Also for unit tests that use fixture class, `TEST` is replaced with `TEST_F` as required in gtest. There are several custom `main` functions in our existing tests. To make this transition easier, I modify all `main` functions to fallow gtest notation. But eventually we can remove them and use implementation of `main` that gtest provides. ```lang=bash % cat ~/transform #!/bin/sh files=$(git ls-files '*test\.cc') for file in $files do if grep -q "rocksdb::test::RunAllTests()" $file then if grep -Eq '^class \w+Test {' $file then perl -pi -e 's/^(class \w+Test) {/${1}: public testing::Test {/g' $file perl -pi -e 's/^(TEST)/${1}_F/g' $file fi perl -pi -e 's/(int main.*\{)/${1}::testing::InitGoogleTest(&argc, argv);/g' $file perl -pi -e 's/rocksdb::test::RunAllTests/RUN_ALL_TESTS/g' $file fi done % sh ~/transform % make format ``` Second iteration of this diff contains only scripted changes. Third iteration contains manual changes to fix last errors and make it compilable. Test Plan: Build and notice no errors. ```lang=bash % USE_CLANG=1 make check -j55 ``` Tests are still testing. Reviewers: meyering, sdong, rven, igor Reviewed By: igor Subscribers: dhruba, leveldb Differential Revision: https://reviews.facebook.net/D35157
10 years ago
::testing::InitGoogleTest(&argc, argv);
return RUN_ALL_TESTS();
A Simple Plain Table Summary: A Simple plain table format. No block structure. When creating the table reader, scanning the full table to create indexes. Test Plan:Add unit test Reviewers:haobo,dhruba,kailiu CC: Task ID: # Blame Rev:
11 years ago
}
Block plain_table_db_test in ROCKSDB_LITE Summary: Block plain_table_db_test in ROCKSDB_LITE since plain table is not supported in ROCKSDB_LITE Test Plan: plain_table_db_test Reviewers: igor, sdong, yhchiang Reviewed By: yhchiang Subscribers: dhruba Differential Revision: https://reviews.facebook.net/D42159
9 years ago
#else
#include <stdio.h>
int main(int argc, char** argv) {
fprintf(stderr, "SKIPPED as plain table is not supported in ROCKSDB_LITE\n");
return 0;
}
#endif // !ROCKSDB_LITE