New upstream version 6.0.20upstream

5 files changed, 483 insertions, 0 deletions

diff --git a/src/mongo/db/query/util/deferred.h b/src/mongo/db/query/util/deferred.h
new file mode 100644
index 00000000000..a2609bb6b49
--- /dev/null
+++ b/src/mongo/db/query/util/deferred.h
@@ -0,0 +1,118 @@
+/**
+ *    Copyright (C) 2023-present MongoDB, Inc.
+ *
+ *    This program is free software: you can redistribute it and/or modify
+ *    it under the terms of the Server Side Public License, version 1,
+ *    as published by MongoDB, Inc.
+ *
+ *    This program is distributed in the hope that it will be useful,
+ *    but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *    Server Side Public License for more details.
+ *
+ *    You should have received a copy of the Server Side Public License
+ *    along with this program. If not, see
+ *    <http://www.mongodb.com/licensing/server-side-public-license>.
+ *
+ *    As a special exception, the copyright holders give permission to link the
+ *    code of portions of this program with the OpenSSL library under certain
+ *    conditions as described in each individual source file and distribute
+ *    linked combinations including the program with the OpenSSL library. You
+ *    must comply with the Server Side Public License in all respects for
+ *    all of the code used other than as permitted herein. If you modify file(s)
+ *    with this exception, you may extend this exception to your version of the
+ *    file(s), but you are not obligated to do so. If you do not wish to do so,
+ *    delete this exception statement from your version. If you delete this
+ *    exception statement from all source files in the program, then also delete
+ *    it in the license file.
+ */
+
+#pragma once
+
+#include <functional>
+
+namespace mongo {
+
+/**
+ * A template class that provides a way to defer the initialization of an object until its value is
+ * actually required. This is also commonly referred to as lazy initialization.
+ *
+ * Dangers:
+ * - This implementation is currently not thread safe, and it shouldn't be used in multi-threaded
+ *   fashion.
+ * - Be careful about using this for lazy initialization of data members and capturing the 'this'
+ *   variable. Code like this will result in buggy/unsafe move constructors, which would have a
+ *   dangling reference to the moved-from type:
+ *
+ *   class MyType {
+ *       int x;
+ *       // !!! Dangling 'this' when moved !!!
+ *       Deferred<int> xSquared{[this]() { return this->x * this-> x; };
+ *   };
+ *   Instead, it is better to do something like this:
+ *   class MyType {
+ *       int xSquared() const {
+ *          return *_xSquared.get(_x);
+ *       }
+ *
+ *       int _x;
+ *       Deferred<int, int> _xSquared{[](int x) { return x * x; };
+ *   };
+ * - As a similar danger, the value is only computed once. if you initialize it with arguments like
+ *   the above 'xSquared()' implementation, then be cogniscent that the value will never change. If
+ *   '_x' changes, '_xSquared' will not.
+ *
+ * A Deferred class can be constructed with either an initial value (eager initialization) or a
+ * function that will generate the value when needed.
+ */
+template <typename T, typename... Args>
+class Deferred {
+public:
+    /**
+     * Instantiates a Deffered<T> with the given data - no callbacks or lazy initialization.
+     */
+    Deferred(T data) : _data(data) {}
+
+    /**
+     * Stores a function to compute a T later. Please note the warnings described in this class
+     * comment.
+     */
+    Deferred(std::function<T(Args&&...)> initializer) : _initializer(std::move(initializer)) {}
+
+    /**
+     * Returns a pointer to the managed object. Initializes the object if it hasn't done so already.
+     */
+    T& get(Args&&... args) const {
+        if (_initializer) {
+            _data = _initializer(std::forward<Args>(args)...);
+            _initializer = nullptr;
+        }
+        return _data;
+    }
+
+    /**
+     * Dereferences the pointer to the managed object. Note this is only a valid shortcut if there
+     * are no arguments to '_initializer'.
+     */
+    T* operator->() const {
+        return &get();
+    }
+
+    /**
+     * Returns a referenced to the managed object. Initializes the object if it hasn't done so
+     * already. Note this is only a valid shortcut if there are no arguments to '_initializer'.
+     */
+    const T& operator*() const {
+        return get();
+    }
+
+    bool isInitialized() const {
+        return _initializer ? false : true;
+    }
+
+private:
+    mutable T _data;
+    mutable std::function<T(Args&&...)> _initializer;
+};
+
+}  // namespace mongo
diff --git a/src/mongo/db/query/util/deferred_test.cpp b/src/mongo/db/query/util/deferred_test.cpp
new file mode 100644
index 00000000000..de256394787
--- /dev/null
+++ b/src/mongo/db/query/util/deferred_test.cpp
@@ -0,0 +1,98 @@
+/**
+ *    Copyright (C) 2023-present MongoDB, Inc.
+ *
+ *    This program is free software: you can redistribute it and/or modify
+ *    it under the terms of the Server Side Public License, version 1,
+ *    as published by MongoDB, Inc.
+ *
+ *    This program is distributed in the hope that it will be useful,
+ *    but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *    Server Side Public License for more details.
+ *
+ *    You should have received a copy of the Server Side Public License
+ *    along with this program. If not, see
+ *    <http://www.mongodb.com/licensing/server-side-public-license>.
+ *
+ *    As a special exception, the copyright holders give permission to link the
+ *    code of portions of this program with the OpenSSL library under certain
+ *    conditions as described in each individual source file and distribute
+ *    linked combinations including the program with the OpenSSL library. You
+ *    must comply with the Server Side Public License in all respects for
+ *    all of the code used other than as permitted herein. If you modify file(s)
+ *    with this exception, you may extend this exception to your version of the
+ *    file(s), but you are not obligated to do so. If you do not wish to do so,
+ *    delete this exception statement from your version. If you delete this
+ *    exception statement from all source files in the program, then also delete
+ *    it in the license file.
+ */
+
+#include "mongo/db/query/util/deferred.h"
+
+#include "mongo/unittest/unittest.h"
+
+namespace mongo {
+using std::string;
+using namespace std::string_literals;
+
+
+TEST(DeferredTest, EagerInitialization) {
+    Deferred<string> eager{"someString"};
+    ASSERT_TRUE(eager.isInitialized());
+    ASSERT_EQ(eager.get(), "someString"s);
+    ASSERT_EQ(*eager, "someString"s);
+}
+
+TEST(DeferredTest, DeferredInitialization) {
+    size_t initializationCount = 0;
+    Deferred<string> deferred{[&]() {
+        initializationCount++;
+        return "someString"s;
+    }};
+    ASSERT_FALSE(deferred.isInitialized());
+
+    // Ensure the deferred object wasn't initialized on creation.
+    ASSERT_EQ(initializationCount, 0);
+
+    // Ensure that the deferred object is initialized on pointer dereferences.
+    ASSERT_FALSE(deferred->empty());
+    ASSERT_TRUE(deferred.isInitialized());
+
+    ASSERT_EQ(initializationCount, 1);
+
+    // Ensure that the content of the deferred object is equal to its raw counterpart, while also
+    // verifing that it is initialized at most once.
+    ASSERT_EQ(deferred.get(), "someString"s);
+    ASSERT_EQ(initializationCount, 1);
+}
+
+TEST(DeferredTest, DeferredInitializationWithOneArgument) {
+    size_t initializationCount = 0;
+    Deferred<string, const string&> deferred{[&](const string& input) {
+        initializationCount++;
+        return "{" + input + "}";
+    }};
+
+    // Ensure the deferred object wasn't initialized on creation.
+    ASSERT_EQ(initializationCount, 0);
+
+    // Ensure that the content of the deferred object is equal to its raw counterpart, while also
+    // verifing that it is initialized at most once.
+    ASSERT_EQ(deferred.get("more curlies"), "{more curlies}"s);
+    ASSERT_EQ(initializationCount, 1);
+
+    // Note that the value is cached, so it's not really valid to call it with a different argument.
+    ASSERT_EQ(deferred.get("merganser"), "{more curlies}"s);
+    ASSERT_EQ(initializationCount, 1);
+}
+
+TEST(DeferredTest, DeferredInitializationWithTwoArgs) {
+    Deferred<string, const string&, const string&> deferred{
+        [&](const auto& input, const auto& prefix) { return prefix + input; }};
+
+    ASSERT_EQ(deferred.get("cowbell", "more "), "more cowbell"s);
+    ASSERT_EQ(deferred.get("cowbell", "more "), "more cowbell"s);
+    ASSERT_EQ(deferred.get("cowbell", "less?"), "more cowbell"s);
+    ASSERT_EQ(deferred.get("tests", "better"), "more cowbell"s);
+}
+}  // namespace mongo
diff --git a/src/mongo/db/query/util/memory_util.cpp b/src/mongo/db/query/util/memory_util.cpp
new file mode 100644
index 00000000000..8a206deb2b7
--- /dev/null
+++ b/src/mongo/db/query/util/memory_util.cpp
@@ -0,0 +1,128 @@
+/**
+ *    Copyright (C) 2021-present MongoDB, Inc.
+ *
+ *    This program is free software: you can redistribute it and/or modify
+ *    it under the terms of the Server Side Public License, version 1,
+ *    as published by MongoDB, Inc.
+ *
+ *    This program is distributed in the hope that it will be useful,
+ *    but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *    Server Side Public License for more details.
+ *
+ *    You should have received a copy of the Server Side Public License
+ *    along with this program. If not, see
+ *    <http://www.mongodb.com/licensing/server-side-public-license>.
+ *
+ *    As a special exception, the copyright holders give permission to link the
+ *    code of portions of this program with the OpenSSL library under certain
+ *    conditions as described in each individual source file and distribute
+ *    linked combinations including the program with the OpenSSL library. You
+ *    must comply with the Server Side Public License in all respects for
+ *    all of the code used other than as permitted herein. If you modify file(s)
+ *    with this exception, you may extend this exception to your version of the
+ *    file(s), but you are not obligated to do so. If you do not wish to do so,
+ *    delete this exception statement from your version. If you delete this
+ *    exception statement from all source files in the program, then also delete
+ *    it in the license file.
+ */
+
+#define MONGO_LOGV2_DEFAULT_COMPONENT ::mongo::logv2::LogComponent::kQuery
+
+#include "mongo/db/query/util/memory_util.h"
+
+#include <cstddef>
+#include <pcrecpp.h>
+
+#include "mongo/logv2/log.h"
+#include "mongo/util/processinfo.h"
+
+
+namespace mongo::memory_util {
+
+StatusWith<MemoryUnits> parseUnitString(const std::string& strUnit) {
+    if (strUnit.empty()) {
+        return Status(ErrorCodes::Error{6007010}, "Unit value cannot be empty");
+    }
+
+    if (strUnit[0] == '%') {
+        return MemoryUnits::kPercent;
+    } else if (strUnit[0] == 'M' || strUnit[0] == 'm') {
+        return MemoryUnits::kMB;
+    } else if (strUnit[0] == 'G' || strUnit[0] == 'g') {
+        return MemoryUnits::kGB;
+    }
+
+    return Status(ErrorCodes::Error{6007011}, "Incorrect unit value");
+}
+
+StatusWith<MemorySize> MemorySize::parse(const std::string& str) {
+    pcrecpp::RE_Options opt;
+    opt.set_caseless(true);
+    // Looks for a floating point number with followed by a unit suffix (MB, GB, %).
+    pcrecpp::RE re("\\s*(\\d+\\.?\\d*)\\s*(MB|GB|%)\\s*", opt);
+
+    double size{};
+    std::string strUnit{};
+    if (!re.FullMatch(str, &size, &strUnit)) {
+        return {ErrorCodes::Error{6007012}, "Unable to parse memory size string"};
+    }
+
+    auto statusWithUnit = parseUnitString(strUnit);
+    if (!statusWithUnit.isOK()) {
+        return statusWithUnit.getStatus();
+    }
+    return MemorySize{size, statusWithUnit.getValue()};
+}
+
+size_t convertToSizeInBytes(const MemorySize& memSize) {
+    constexpr size_t kBytesInMB = 1024 * 1024;
+    constexpr size_t kMBytesInGB = 1024;
+
+    double sizeInMB = memSize.size;
+
+    switch (memSize.units) {
+        case MemoryUnits::kPercent:
+            sizeInMB *= ProcessInfo::getMemSizeMB() / 100.0;
+            break;
+        case MemoryUnits::kMB:
+            break;
+        case MemoryUnits::kGB:
+            sizeInMB *= kMBytesInGB;
+            break;
+    }
+
+    return static_cast<size_t>(sizeInMB * kBytesInMB);
+}
+
+size_t getRequestedMemSizeInBytes(const MemorySize& memSize) {
+    size_t planCacheSize = convertToSizeInBytes(memSize);
+    uassert(5968001,
+            "Cache size must be at least 1KB * number of cores",
+            planCacheSize >= 1024 * ProcessInfo::getNumCores());
+    return planCacheSize;
+}
+
+/**
+ * Sets upper limit on a storage structure's size. Either that structure's maximumSize or to
+ * percentage of the total system's memory (both known at call site), whichever is smaller.
+ */
+size_t capMemorySize(size_t requestedSizeBytes,
+                     size_t maximumSizeGB,
+                     double percentTotalSystemMemory) {
+    constexpr size_t kBytesInGB = 1024 * 1024 * 1024;
+    // Express maximum size in bytes.
+    const size_t maximumSizeBytes = maximumSizeGB * kBytesInGB;
+    const memory_util::MemorySize limitToProcessSize{percentTotalSystemMemory,
+                                                     memory_util::MemoryUnits::kPercent};
+    const size_t limitToProcessSizeInBytes = convertToSizeInBytes(limitToProcessSize);
+
+    // The size will be capped by the minimum of the two values defined above.
+    const size_t upperLimit = std::min(maximumSizeBytes, limitToProcessSizeInBytes);
+
+    if (requestedSizeBytes > upperLimit) {
+        requestedSizeBytes = upperLimit;
+    }
+    return requestedSizeBytes;
+}
+}  // namespace mongo::memory_util
diff --git a/src/mongo/db/query/util/memory_util.h b/src/mongo/db/query/util/memory_util.h
new file mode 100644
index 00000000000..345780b4c84
--- /dev/null
+++ b/src/mongo/db/query/util/memory_util.h
@@ -0,0 +1,66 @@
+/**
+ *    Copyright (C) 2021-present MongoDB, Inc.
+ *
+ *    This program is free software: you can redistribute it and/or modify
+ *    it under the terms of the Server Side Public License, version 1,
+ *    as published by MongoDB, Inc.
+ *
+ *    This program is distributed in the hope that it will be useful,
+ *    but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *    Server Side Public License for more details.
+ *
+ *    You should have received a copy of the Server Side Public License
+ *    along with this program. If not, see
+ *    <http://www.mongodb.com/licensing/server-side-public-license>.
+ *
+ *    As a special exception, the copyright holders give permission to link the
+ *    code of portions of this program with the OpenSSL library under certain
+ *    conditions as described in each individual source file and distribute
+ *    linked combinations including the program with the OpenSSL library. You
+ *    must comply with the Server Side Public License in all respects for
+ *    all of the code used other than as permitted herein. If you modify file(s)
+ *    with this exception, you may extend this exception to your version of the
+ *    file(s), but you are not obligated to do so. If you do not wish to do so,
+ *    delete this exception statement from your version. If you delete this
+ *    exception statement from all source files in the program, then also delete
+ *    it in the license file.
+ */
+
+#pragma once
+
+#include <string>
+
+#include "mongo/base/error_codes.h"
+#include "mongo/base/status_with.h"
+
+namespace mongo::memory_util {
+
+/**
+ * Defines units of memory.
+ */
+enum class MemoryUnits {
+    kPercent,
+    kMB,
+    kGB,
+};
+
+/**
+ * Represents parsed memory size parameter.
+ */
+struct MemorySize {
+    static StatusWith<MemorySize> parse(const std::string& str);
+
+    const double size;
+    const MemoryUnits units;
+};
+
+StatusWith<MemoryUnits> parseUnitString(const std::string& strUnit);
+size_t convertToSizeInBytes(const MemorySize& memSize);
+size_t capMemorySize(size_t requestedSizeBytes,
+                     size_t maximumSizeGB,
+                     double percentTotalSystemMemory);
+size_t getRequestedMemSizeInBytes(const MemorySize& memSize);
+
+
+}  // namespace mongo::memory_util
diff --git a/src/mongo/db/query/util/memory_util_test.cpp b/src/mongo/db/query/util/memory_util_test.cpp
new file mode 100644
index 00000000000..78f7b3098d6
--- /dev/null
+++ b/src/mongo/db/query/util/memory_util_test.cpp
@@ -0,0 +1,73 @@
+/**
+ *    Copyright (C) 2021-present MongoDB, Inc.
+ *
+ *    This program is free software: you can redistribute it and/or modify
+ *    it under the terms of the Server Side Public License, version 1,
+ *    as published by MongoDB, Inc.
+ *
+ *    This program is distributed in the hope that it will be useful,
+ *    but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *    Server Side Public License for more details.
+ *
+ *    You should have received a copy of the Server Side Public License
+ *    along with this program. If not, see
+ *    <http://www.mongodb.com/licensing/server-side-public-license>.
+ *
+ *    As a special exception, the copyright holders give permission to link the
+ *    code of portions of this program with the OpenSSL library under certain
+ *    conditions as described in each individual source file and distribute
+ *    linked combinations including the program with the OpenSSL library. You
+ *    must comply with the Server Side Public License in all respects for
+ *    all of the code used other than as permitted herein. If you modify file(s)
+ *    with this exception, you may extend this exception to your version of the
+ *    file(s), but you are not obligated to do so. If you do not wish to do so,
+ *    delete this exception statement from your version. If you delete this
+ *    exception statement from all source files in the program, then also delete
+ *    it in the license file.
+ */
+
+#include "mongo/db/query/util/memory_util.h"
+
+#include "mongo/unittest/unittest.h"
+
+namespace mongo::memory_util {
+
+bool operator==(const MemorySize& lhs, const MemorySize& rhs) {
+    constexpr double kEpsilon = 1e-10;
+    return std::abs(lhs.size - rhs.size) < kEpsilon && lhs.units == rhs.units;
+}
+
+TEST(MemorySizeTest, ParseUnitStringPercent) {
+    ASSERT_TRUE(MemoryUnits::kPercent == parseUnitString("%"));
+}
+
+TEST(MemorySizeTest, ParseUnitStringMB) {
+    ASSERT_TRUE(MemoryUnits::kMB == parseUnitString("MB"));
+    ASSERT_TRUE(MemoryUnits::kMB == parseUnitString("mb"));
+    ASSERT_TRUE(MemoryUnits::kMB == parseUnitString("mB"));
+    ASSERT_TRUE(MemoryUnits::kMB == parseUnitString("Mb"));
+}
+
+TEST(MemorySizeTest, ParseUnitStringGB) {
+    ASSERT_TRUE(MemoryUnits::kGB == parseUnitString("GB"));
+    ASSERT_TRUE(MemoryUnits::kGB == parseUnitString("gb"));
+    ASSERT_TRUE(MemoryUnits::kGB == parseUnitString("gB"));
+    ASSERT_TRUE(MemoryUnits::kGB == parseUnitString("Gb"));
+}
+
+TEST(MemorySizeTest, ParseUnitStringIncorrectValue) {
+    ASSERT_NOT_OK(parseUnitString("").getStatus());
+    ASSERT_NOT_OK(parseUnitString(" ").getStatus());
+    ASSERT_NOT_OK(parseUnitString("KB").getStatus());
+}
+
+TEST(MemorySizeTest, ParseMemorySize) {
+    ASSERT_TRUE((MemorySize{10.0, MemoryUnits::kPercent}) == MemorySize::parse("10%"));
+    ASSERT_TRUE((MemorySize{300.0, MemoryUnits::kMB}) == MemorySize::parse("300MB"));
+    ASSERT_TRUE((MemorySize{4.0, MemoryUnits::kGB}) == MemorySize::parse("4GB"));
+    ASSERT_TRUE((MemorySize{5.1, MemoryUnits::kPercent}) == MemorySize::parse(" 5.1%"));
+    ASSERT_TRUE((MemorySize{11.1, MemoryUnits::kMB}) == MemorySize::parse("11.1 mb"));
+    ASSERT_TRUE((MemorySize{12.1, MemoryUnits::kGB}) == MemorySize::parse("  12.1    Gb  "));
+}
+}  // namespace mongo::memory_util