Files
versioned-map/VersionedMap.cpp

1694 lines
53 KiB
C++

#include "VersionedMap.h"
#include "Internal.h"
#include "KeyCompare.h"
#include "PrintMutation.h"
#include "RootSet.h"
#include <assert.h>
#include <atomic>
#include <cstdint>
#include <inttypes.h>
#include <optional>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/mman.h>
#include <sys/types.h>
#include <unistd.h>
#include <memcheck.h>
static_assert(std::is_standard_layout_v<weaselab::VersionedMap::MutationType>);
static_assert(std::is_standard_layout_v<weaselab::VersionedMap::Key>);
static_assert(std::is_standard_layout_v<weaselab::VersionedMap::Mutation>);
static_assert(std::is_standard_layout_v<weaselab::VersionedMap::Iterator>);
static_assert(std::bidirectional_iterator<weaselab::VersionedMap::Iterator>);
static_assert(std::is_standard_layout_v<
weaselab::VersionedMap::Iterator::VersionedMutation>);
void *mmapSafe(void *addr, size_t len, int prot, int flags, int fd,
off_t offset) {
void *result = mmap(addr, len, prot, flags, fd, offset);
if (result == MAP_FAILED) {
int err = errno; // GCOVR_EXCL_LINE
fprintf( // GCOVR_EXCL_LINE
stderr, // GCOVR_EXCL_LINE
"Error calling mmap(%p, %zu, %d, %d, %d, %jd): %d %s\n", // GCOVR_EXCL_LINE
addr, len, prot, flags, fd, (intmax_t)offset, err, // GCOVR_EXCL_LINE
strerror(err)); // GCOVR_EXCL_LINE
fflush(stderr); // GCOVR_EXCL_LINE
abort(); // GCOVR_EXCL_LINE
}
return result;
}
void mprotectSafe(void *p, size_t s, int prot) {
if (mprotect(p, s, prot) != 0) {
int err = errno; // GCOVR_EXCL_LINE
fprintf(stderr, // GCOVR_EXCL_LINE
"Error calling mprotect(%p, %zu, %d): %s\n", // GCOVR_EXCL_LINE
p, // GCOVR_EXCL_LINE
s, // GCOVR_EXCL_LINE
prot, // GCOVR_EXCL_LINE
strerror(err)); // GCOVR_EXCL_LINE
fflush(stderr); // GCOVR_EXCL_LINE
abort(); // GCOVR_EXCL_LINE
}
}
void munmapSafe(void *ptr, size_t size) {
if (munmap(ptr, size) != 0) {
int err = errno; // GCOVR_EXCL_LINE
fprintf(stderr, "Error calling munmap(%p, %zu): %s\n", // GCOVR_EXCL_LINE
ptr, // GCOVR_EXCL_LINE
size, // GCOVR_EXCL_LINE
strerror(err)); // GCOVR_EXCL_LINE
fflush(stderr); // GCOVR_EXCL_LINE
abort(); // GCOVR_EXCL_LINE
}
}
namespace weaselab {
// 96 is enough for an entire search path in a tree with a size that
// overflows int. See
// https://en.wikipedia.org/wiki/Random_binary_tree#The_longest_path
constexpr int kPathLengthUpperBound = 96;
struct Entry {
// If there is a point mutation at key, then pointVersion is >= 0 and key has
// not been modified since pointVersion. Otherwise it's negative.
int64_t pointVersion;
// If there is a range mutation ending at key, then rangeVersion is >= 0 and
// the range has not been modified since rangeVersion. Otherwise it's
// negative.
int64_t rangeVersion;
int keyLen;
// Negative if this key is cleared. Only meaningful if this is a point
// mutation.
int valLen;
mutable int refCount;
uint32_t priority;
// True if the entry is a point mutation. If false, this entry's key should be
// read through to the underlying data structure.
bool pointMutation() const { return pointVersion >= 0; }
bool pointSet() const { return pointVersion >= 0 && valLen >= 0; }
bool pointClear() const { return pointVersion >= 0 && valLen < 0; }
// True if mutations in (pred, this) are cleared. If false, (pred, this)
// should be read through to the underlying data structure.
bool clearTo() const { return rangeVersion >= 0; }
// There's an extra zero byte past the end of getKey, used for
// reconstructing logical mutations without copies.
const uint8_t *getKey() const { return (const uint8_t *)(this + 1); }
const uint8_t *getVal() const {
return (const uint8_t *)(this + 1) + 1 + keyLen;
}
Entry *addref() const {
++refCount;
#if DEBUG_VERBOSE
if (debugVerboseEnabled) {
printf("addref %p to %d\n", this, refCount);
}
#endif
return (Entry *)this;
}
void delref() const {
#if DEBUG_VERBOSE
if (debugVerboseEnabled) {
printf("delref %p to %d\n", this, refCount - 1);
}
#endif
if (--refCount == 0) {
safe_free((void *)this, sizeof(Entry) + keyLen + 1 + std::max(valLen, 0));
}
}
static Entry *make(int64_t pointVersion, int64_t rangeVersion,
const uint8_t *key, int keyLen, const uint8_t *val,
int valLen, uint32_t priority) {
auto e =
(Entry *)safe_malloc(sizeof(Entry) + keyLen + 1 + std::max(valLen, 0));
e->pointVersion = pointVersion;
e->rangeVersion = rangeVersion;
e->keyLen = keyLen;
e->valLen = valLen;
e->refCount = 1;
e->priority = priority;
if (keyLen > 0) {
memcpy((uint8_t *)e->getKey(), key, keyLen);
}
((uint8_t *)e->getKey())[keyLen] = 0;
if (valLen > 0) {
memcpy((uint8_t *)e->getVal(), val, valLen);
}
return e;
}
};
struct UpdateInfo {
UpdateInfo() : version(kVersionIfNotUpdated) {}
int64_t version;
constexpr static int64_t kVersionIfNotUpdated = 0x7fffffffffffffff;
bool updated() const { return version != UpdateInfo::kVersionIfNotUpdated; }
bool updated(int64_t at) const { return version <= at; }
};
static_assert(std::atomic<UpdateInfo>::is_always_lock_free);
struct Node {
union {
std::atomic<UpdateInfo> updateInfo;
uint32_t nextFree;
};
Entry *entry;
// [left/right, older/newer]. Logically this is only 1 aux pointer since we
// only store one updateInfo, but this encoding let's us write a branch-free
// `child` function, which really helps with the effective ILP of the bulk
// firstGeq function.
uint32_t pointer[2][2];
};
// Limit mmap to 32 GiB so valgrind doesn't complain.
// https://bugs.kde.org/show_bug.cgi?id=229500
constexpr size_t kMapSize = size_t(32) * (1 << 30);
const size_t kPageSize = sysconf(_SC_PAGESIZE);
const uint32_t kNodesPerPage = kPageSize / sizeof(Node);
const uint32_t kMinAddressable = kNodesPerPage;
constexpr uint32_t kUpsizeBytes = 1 << 20;
constexpr uint32_t kUpsizeNodes = kUpsizeBytes / sizeof(Node);
static_assert(kUpsizeNodes * sizeof(Node) == kUpsizeBytes);
struct BitSet {
explicit BitSet(uint32_t size)
: words((uint64_t *)safe_calloc(size / 64 + 1, 8)), size(size) {}
bool test(uint32_t i) const {
return words[i >> 6] & (uint64_t(1) << (i & 63));
}
// Returns former value
bool set(uint32_t i) {
const auto prev = words[i >> 6];
const auto mask = uint64_t(1) << (i & 63);
words[i >> 6] |= mask;
max_ = std::max(i, max_);
return prev & mask;
}
// Returns 0 if set is empty
uint32_t max() const { return max_; }
template <class F>
void iterateAbsentApproxBackwards(F f, uint32_t begin, uint32_t end) const {
// TODO can this be improved? We can do something with a word at a time
// instead of a bit at a time. The first attempt at doing so benchmarked as
// slower.
assert(begin != 0);
for (uint32_t i = end - 1; i >= begin; --i) {
if (!test(i)) {
f(i);
}
}
}
~BitSet() { safe_free(words, (size / 64 + 1) * 8); }
private:
uint32_t max_ = 0;
uint64_t *const words;
const uint32_t size;
};
int64_t mmapBytes = 0;
int64_t peakMmapBytes = 0;
struct MemManager {
MemManager()
: base((Node *)mmapSafe(nullptr, kMapSize, PROT_NONE,
MAP_PRIVATE | MAP_ANONYMOUS, -1, 0)) {
if (kPageSize % sizeof(Node) != 0) {
fprintf(stderr, // GCOVR_EXCL_LINE
"kPageSize not a multiple of Node size\n"); // GCOVR_EXCL_LINE
abort(); // GCOVR_EXCL_LINE
}
if (kUpsizeBytes % kPageSize != 0) {
fprintf(stderr, // GCOVR_EXCL_LINE
"kUpsizeBytes not a multiple of kPageSize\n"); // GCOVR_EXCL_LINE
abort(); // GCOVR_EXCL_LINE
}
}
~MemManager() {
gc(nullptr, 0, 0);
munmapSafe(base, kMapSize);
}
Node *const base;
uint32_t allocate() {
if (freeList != 0) {
uint32_t result = freeList;
freeList = base[result].nextFree;
assert(base[result].entry == nullptr);
return result;
}
if (next == firstUnaddressable) {
mprotectSafe(base + firstUnaddressable, kUpsizeBytes,
PROT_READ | PROT_WRITE);
VALGRIND_MAKE_MEM_NOACCESS(base + firstUnaddressable, kUpsizeBytes);
firstUnaddressable += kUpsizeNodes;
#if SHOW_MEMORY
mmapBytes = getBytes();
peakMmapBytes = std::max(peakMmapBytes, mmapBytes);
#endif
if (firstUnaddressable > kMapSize / sizeof(Node)) {
fprintf( // GCOVR_EXCL_LINE
stderr, // GCOVR_EXCL_LINE
"Out of memory: firstUnaddressable > kMapSize / " // GCOVR_EXCL_LINE
"sizeof(Node)\n"); // GCOVR_EXCL_LINE
abort(); // GCOVR_EXCL_LINE
}
}
VALGRIND_MAKE_MEM_UNDEFINED(base + next, sizeof(Node));
return next++;
}
void gc(const uint32_t *roots, int numRoots, int64_t oldestVersion) {
// Calculate reachable set
BitSet reachable{next};
// Each node has at most 3 children and nodes along the search path aren't
// in the stack, so we need 2 * kPathLengthUpperBound
uint32_t stack[2 * kPathLengthUpperBound];
int stackIndex = 0;
auto tryPush = [&]([[maybe_unused]] uint32_t parent, uint32_t child) {
if (!reachable.set(child)) {
#if DEBUG_VERBOSE
if (debugVerboseEnabled) {
printf(" GC: reach: %u (parent %u)\n", child, parent);
}
#endif
assert(stackIndex < int(sizeof(stack) / sizeof(stack[0])));
stack[stackIndex++] = child;
}
};
for (int i = 0; i < numRoots; ++i) {
if (roots[i] == 0) {
continue;
}
tryPush(0, roots[i]);
while (stackIndex > 0) {
uint32_t p = stack[--stackIndex];
auto &node = base[p];
auto updateInfo = node.updateInfo.load(std::memory_order_relaxed);
if (updateInfo.updated()) {
if (node.pointer[0][1] != 0) {
tryPush(p, node.pointer[0][1]);
}
if (node.pointer[1][1] != 0) {
tryPush(p, node.pointer[1][1]);
}
}
if (!updateInfo.updated(oldestVersion)) {
if (node.pointer[0][0] != 0) {
tryPush(p, node.pointer[0][0]);
}
if (node.pointer[1][0] != 0) {
tryPush(p, node.pointer[1][0]);
}
}
}
}
uint32_t max = reachable.max();
if (max == 0) {
max = kMinAddressable - 1;
}
assert(max < next);
// Rebuild free list to prefer leftward nodes
freeList = 0;
reachable.iterateAbsentApproxBackwards(
[&](uint32_t i) {
if (base[i].entry != nullptr) {
#if DEBUG_VERBOSE
if (debugVerboseEnabled) {
printf("Collecting %u while building free list\n", i);
}
#endif
base[i].entry->delref();
base[i].entry = nullptr;
}
base[i].nextFree = freeList;
freeList = i;
},
kMinAddressable, max + 1);
// Entries to the right of max don't need to be in the freelist. They're
// allocated by pointer bumping.
for (uint32_t i = max + 1; i < next; ++i) {
if (base[i].entry != nullptr) {
#if DEBUG_VERBOSE
if (debugVerboseEnabled) {
printf("Collecting %u while shrinking right\n", i);
}
#endif
base[i].entry->delref();
base[i].entry = nullptr;
}
}
uint32_t newFirstUnaddressable = (max / kNodesPerPage + 1) * kNodesPerPage;
if (newFirstUnaddressable < firstUnaddressable) {
mprotectSafe(base + newFirstUnaddressable,
(firstUnaddressable - newFirstUnaddressable) * sizeof(Node),
PROT_NONE);
firstUnaddressable = newFirstUnaddressable;
#if SHOW_MEMORY
mmapBytes = getBytes();
#endif
}
next = max + 1;
}
int64_t getBytes() const {
return (firstUnaddressable - kMinAddressable) * sizeof(Node);
}
private:
uint32_t next = kMinAddressable;
uint32_t firstUnaddressable = kMinAddressable;
uint32_t freeList = 0;
};
auto operator<=>(const VersionedMap::Key &lhs, const Node &rhs) {
int cl = std::min(lhs.len, rhs.entry->keyLen);
if (cl > 0) {
int c = memcmp(lhs.p, rhs.entry->getKey(), cl);
if (c != 0) {
return c <=> 0;
}
}
return lhs.len <=> rhs.entry->keyLen;
}
constexpr int orderToInt(std::strong_ordering o) {
return o == std::strong_ordering::less ? -1
: o == std::strong_ordering::equal ? 0
: 1;
}
struct Finger {
void push(uint32_t node, bool dir) {
searchPath[searchPathSize_] = node;
direction[searchPathSize_] = dir;
++searchPathSize_;
}
void pop() {
assert(searchPathSize_ > 0);
--searchPathSize_;
}
uint32_t backNode() const {
assert(searchPathSize_ > 0);
return searchPath[searchPathSize_ - 1];
}
uint32_t &backNodeRef() {
assert(searchPathSize_ > 0);
return searchPath[searchPathSize_ - 1];
}
bool backDirection() const {
assert(searchPathSize_ > 0);
return direction[searchPathSize_ - 1];
}
uint32_t searchPathSize() const { return searchPathSize_; }
void setSearchPathSizeUnsafe(int size) { searchPathSize_ = size; }
Finger() { clear(); }
void clear() {
#ifndef NDEBUG
memset(searchPath, 0, sizeof(searchPath));
memset(direction, 0, sizeof(direction));
#endif
searchPathSize_ = 0;
}
void copyTo(Finger &result) {
#ifndef NDEBUG
memset(result.searchPath, 0, sizeof(searchPath));
memset(result.direction, 0, sizeof(direction));
#endif
memcpy(result.searchPath, searchPath,
searchPathSize_ * sizeof(searchPath[0]));
memcpy(result.direction, direction, searchPathSize_ * sizeof(direction[0]));
result.searchPathSize_ = searchPathSize_;
}
Finger(const Finger &) = delete;
Finger &operator=(const Finger &) = delete;
Finger(Finger &&) = delete;
Finger &operator=(Finger &&) = delete;
bool operator==(const Finger &other) const {
bool result = searchPathSize_ == other.searchPathSize_ &&
(searchPathSize_ == 0 || backNode() == other.backNode());
#ifndef NDEBUG
auto expected = searchPathSize_ == other.searchPathSize_ &&
memcmp(searchPath, other.searchPath,
searchPathSize_ * sizeof(searchPath[0])) == 0 &&
(searchPathSize_ == 0 ||
memcmp(direction + 1, other.direction + 1,
(searchPathSize_ - 1) * sizeof(direction[0])) == 0);
assert(result == expected);
#endif
return result;
}
private:
uint32_t searchPath[kPathLengthUpperBound];
bool direction[kPathLengthUpperBound];
int searchPathSize_;
};
VersionedMap::Key keyAfter(VersionedMap::Key k, Arena &arena) {
uint8_t *result = new (arena) uint8_t[k.len + 1];
memcpy(result, k.p, k.len);
result[k.len] = 0;
return {result, k.len + 1};
}
struct __attribute__((__visibility__("hidden"))) VersionedMap::Impl {
// The last node is allowed to be 0, in which case this is the search path of
// where an entry would exist
template <std::memory_order kOrder, bool kDirection>
void move(Finger &finger, int64_t at) const {
uint32_t c;
if (finger.backNode() != 0 &&
(c = child<kOrder>(finger.backNode(), kDirection, at)) != 0) {
finger.push(c, kDirection);
while ((c = child<kOrder>(finger.backNode(), !kDirection, at)) != 0) {
finger.push(c, !kDirection);
}
} else {
while (finger.searchPathSize() > 1 &&
finger.backDirection() == kDirection) {
finger.pop();
}
finger.pop();
}
}
template <std::memory_order kOrder>
uint32_t child(uint32_t node, bool which, int64_t at) const {
assert(node != 0);
static_assert(kOrder == std::memory_order_acquire ||
kOrder == std::memory_order_relaxed);
auto &n = mm.base[node];
uint32_t result;
assert(at < UpdateInfo::kVersionIfNotUpdated);
auto updateInfo = n.updateInfo.load(kOrder);
result = n.pointer[which][updateInfo.updated(at)];
assert(result == 0 || result >= kMinAddressable);
#ifndef NDEBUG
if (result != 0) {
assert(mm.base[result].entry != nullptr);
}
#endif
return result;
}
// Returns the node that results from setting `which` to `child` on `node`
uint32_t update(uint32_t node, bool which, uint32_t child, int64_t version) {
assert(node == 0 || node >= kMinAddressable);
assert(child == 0 || child >= kMinAddressable);
if (this->child<std::memory_order_relaxed>(node, which, version) == child) {
return node;
}
auto &n = mm.base[node];
auto updateInfo = n.updateInfo.load(std::memory_order_relaxed);
const bool updated = updateInfo.updated();
auto doCopy = [&]() {
uint32_t copy = mm.allocate();
#if DEBUG_VERBOSE
if (debugVerboseEnabled) {
printf("Copy %u to %u\n", node, copy);
}
#endif
auto &c = mm.base[copy];
c.entry = n.entry->addref();
c.pointer[which][0] = child;
c.pointer[!which][0] =
this->child<std::memory_order_relaxed>(node, !which, latestVersion);
c.updateInfo.store(UpdateInfo{}, std::memory_order_relaxed);
assert(copy == 0 || copy >= kMinAddressable);
return copy;
};
if (n.entry->pointVersion == version || n.entry->rangeVersion == version) {
// This node is not yet published to concurrent readers
n.pointer[which][0] = child;
assert(node == 0 || node >= kMinAddressable);
return node;
}
if (updateInfo.version == version) {
// Not a data race since concurrent readers are reading at a version <
// `updateInfo.version`
n.pointer[which][1] = child;
assert(node == 0 || node >= kMinAddressable);
return node;
}
if (updated) {
// We already used this node's in-place update
return doCopy();
} else {
n.pointer[which][1] = child;
n.pointer[!which][1] = n.pointer[!which][0];
updateInfo.version = version;
n.updateInfo.store(updateInfo, std::memory_order_release); // Must be last
assert(node == 0 || node >= kMinAddressable);
return node;
}
}
void rotate(uint32_t &n, int64_t at, bool right) {
auto l = child<std::memory_order_relaxed>(n, !right, at);
n = update(
l, right,
update(n, !right, child<std::memory_order_relaxed>(l, right, at), at),
at);
}
struct Val {
const uint8_t *p;
int len;
};
// Initialize finger to the insertion path of `key`. If `key` is not present,
// then the finger ends on a null entry.
template <std::memory_order kOrder, class T>
void search(Key key, T root, int64_t version, Finger &finger) const {
// Prevent integer promotion etc
static_assert(std::is_same_v<T, uint32_t>);
finger.clear();
bool ignored = false;
finger.push(root, ignored);
for (;;) {
auto n = finger.backNode();
if (n == 0) {
break;
}
auto c = key <=> mm.base[n];
if (c == 0) {
// No duplicates
break;
}
finger.push(child<kOrder>(n, c > 0, version), c > 0);
}
}
// If `val` is true, then this is a point set at `latestVersion`.
// If `endRange` is true, then this is a range end marker at `latestVersion`.
// Otherwise it's the beginning of a range at `latestVersion`.
// `finger` is a valid finger to the insertion path of `key` in the latest
// version (which can be obtained with `search`)
void insert(Key key, std::optional<Val> val, bool endRange, Finger &finger) {
const bool inserted = finger.backNode() == 0;
int64_t pointVersion, rangeVersion;
if (val.has_value()) {
// Point set
pointVersion = latestVersion;
if (inserted) {
Finger copy;
finger.copyTo(copy);
move<std::memory_order_relaxed, true>(copy, latestVersion);
if (copy.searchPathSize() == 0) {
rangeVersion = -1; // Sentinel for "no mutation ending here"
} else {
rangeVersion = mm.base[copy.backNode()].entry->rangeVersion;
}
} else {
auto *entry = mm.base[finger.backNode()].entry;
rangeVersion = entry->rangeVersion;
}
} else if (endRange) {
rangeVersion = latestVersion;
if (inserted) {
val = {nullptr, -1};
pointVersion = -1; // Sentinel for "no point mutation here"
#ifndef NDEBUG
// If we inserted this, there would be adjacent clears and so the
// range would not be canonical
Finger copy;
finger.copyTo(copy);
move<std::memory_order_relaxed, true>(copy, latestVersion);
assert(copy.searchPathSize() == 0 ||
mm.base[copy.backNode()].entry->rangeVersion < 0);
#endif
} else {
auto *entry = mm.base[finger.backNode()].entry;
val = {entry->getVal(), entry->valLen};
pointVersion = entry->pointVersion;
}
} else {
// Beginning of a clear range
pointVersion = -1; // Sentinel for "no point mutation here"
if (inserted) {
// If there were a clear range here, it wouldn't be canonical
rangeVersion = -1; // Sentinel for "no mutation ending here"
} else {
auto *entry = mm.base[finger.backNode()].entry;
rangeVersion = entry->rangeVersion;
}
val = {nullptr, -1};
}
// TODO check for noop?
// Prepare new node
const uint32_t node = newNode(
pointVersion, rangeVersion, key.p, key.len, val->p, val->len,
inserted ? gRandom.next() : mm.base[finger.backNode()].entry->priority);
if (!inserted) {
auto &n = mm.base[node];
n.pointer[0][0] = child<std::memory_order_relaxed>(finger.backNode(),
false, latestVersion);
n.pointer[1][0] = child<std::memory_order_relaxed>(finger.backNode(),
true, latestVersion);
}
finger.backNodeRef() = node;
uint32_t oldSize = finger.searchPathSize();
if (inserted) {
// Rotate
for (;;) {
const uint32_t node = finger.backNode();
oldSize = finger.searchPathSize();
if (finger.searchPathSize() == 1) {
// Made it to the root
latestRoot = node;
break;
}
const bool direction = finger.backDirection();
finger.pop();
auto &parent = finger.backNodeRef();
parent = update(parent, direction, node, latestVersion);
if (mm.base[node].entry->priority > mm.base[parent].entry->priority) {
rotate(parent, latestVersion, !direction);
} else {
break;
}
}
}
// Propagate
for (;;) {
const uint32_t node = finger.backNode();
if (finger.searchPathSize() == 1) {
// Made it to the root
latestRoot = node;
break;
}
const bool direction = finger.backDirection();
finger.pop();
const auto old = finger.backNode();
auto &parent = finger.backNodeRef();
parent = update(parent, direction, node, latestVersion);
if (parent == old) {
break;
}
}
finger.setSearchPathSizeUnsafe(oldSize);
#ifndef NDEBUG
{
Finger expected;
search<std::memory_order_relaxed>(key, latestRoot, latestVersion,
expected);
assert(finger == expected);
}
#endif
}
// Removes `finger` from the tree, and leaves `finger` pointing to insertion
// point of its former entry.
void remove(Finger &finger) {
// True if finger is pointing to an entry > than the entry we're removing
// after we rotate it down
// Rotate down until we can remove the entry
for (;;) {
auto &node = finger.backNodeRef();
const auto l =
child<std::memory_order_relaxed>(node, false, latestVersion);
const auto r =
child<std::memory_order_relaxed>(node, true, latestVersion);
if (l == 0 && r == 0) {
// TODO we can avoid some rotations if we stop when l or r == 0
node = 0;
break;
} else {
const bool direction = (l == 0 ? 0 : mm.base[l].entry->priority) >
(r == 0 ? 0 : mm.base[r].entry->priority);
rotate(node, latestVersion, direction);
assert(node != 0);
finger.push(
child<std::memory_order_relaxed>(node, direction, latestVersion),
direction);
}
}
// propagate up the search path, all the way to the root since we may have
// more rotations to do even if an update doesn't change a node pointer
auto node = finger.backNode();
assert(node == 0);
const auto oldSize = finger.searchPathSize();
for (;;) {
if (finger.searchPathSize() == 1) {
// Made it to the root
latestRoot = node;
break;
}
const bool direction = finger.backDirection();
finger.pop();
auto &parent = finger.backNodeRef();
[[maybe_unused]] auto old = parent;
parent = update(parent, direction, node, latestVersion);
node = parent;
}
finger.setSearchPathSizeUnsafe(oldSize);
}
uint32_t newNode(int64_t version, int64_t rangeVersion, const uint8_t *key,
int keyLen, const uint8_t *val, int valLen,
uint32_t priority) {
auto result = mm.allocate();
auto &node = mm.base[result];
node.pointer[0][0] = 0;
node.pointer[1][0] = 0;
node.updateInfo.store(UpdateInfo{}, std::memory_order_relaxed);
node.entry =
Entry::make(version, rangeVersion, key, keyLen, val, valLen, priority);
return result;
}
void setOldestVersion(int64_t oldestVersion) {
mallocBytesDelta = 0;
this->oldestVersion = oldestVersion;
roots.setOldestVersion(oldestVersion);
mm.gc(roots.roots(), roots.rootCount(), oldestVersion);
totalMallocBytes += mallocBytesDelta;
}
int64_t getBytes() const { return totalMallocBytes + mm.getBytes(); }
void printInOrder(int64_t version) const;
void printInOrderHelper(int64_t version, uint32_t node, int depth) const;
int accumulatedFuel = 0;
void scanAndRemoveOldEntries(int fuel) {
accumulatedFuel += fuel;
#ifdef NDEBUG
// This is here for performance reasons, since we want to amortize the cost
// of searching for continueKey. In tests, we want to exercise the rest of
// the code often.
if (accumulatedFuel < 500) {
return;
}
#endif
// Get a finger to the first entry > continueKey, or the last entry if
// continueKey is the empty string
if (latestRoot == 0) {
// Tree is empty
return;
}
Finger finger;
if (continueKey.len == 0) {
// Set finger to last entry in tree
bool ignored = false;
finger.push(latestRoot, ignored);
uint32_t c;
while ((c = child<std::memory_order_relaxed>(finger.backNode(), true,
latestVersion)) != 0) {
finger.push(c, true);
}
} else {
search<std::memory_order_relaxed>(continueKey, latestRoot, latestVersion,
finger);
move<std::memory_order_relaxed, true>(finger, latestVersion);
if (finger.searchPathSize() == 0) {
continueKey = {nullptr, 0};
return;
}
}
assert(finger.backNode() != 0);
int64_t rangeVersion = mm.base[finger.backNode()].entry->rangeVersion;
move<std::memory_order_relaxed, false>(finger, latestVersion);
if (finger.searchPathSize() == 0) {
continueKey = {nullptr, 0};
return;
}
// Phew. Ok now we have a finger to the next entry to consider removing, and
// the range version terminated at this entry.
for (; accumulatedFuel > 0; --accumulatedFuel) {
const auto &n = mm.base[finger.backNode()];
if (rangeVersion < 0 && std::max(n.entry->pointVersion,
n.entry->rangeVersion) < oldestVersion) {
remove(finger);
if (latestRoot == 0) {
return;
}
} else {
rangeVersion = n.entry->rangeVersion;
}
move<std::memory_order_relaxed, false>(finger, latestVersion);
if (finger.searchPathSize() == 0) {
continueKey = {nullptr, 0};
return;
}
}
continueArena = Arena();
const auto &n = mm.base[finger.backNode()];
uint8_t *data = new (continueArena) uint8_t[n.entry->keyLen];
memcpy(data, n.entry->getKey(), n.entry->keyLen);
continueKey = {data, n.entry->keyLen};
}
void addMutations(const Mutation *mutations, int numMutations,
int64_t version) {
mallocBytesDelta = 0;
assert(latestVersion < version);
latestVersion = version;
latestRoot = roots.roots()[roots.rootCount() - 1];
// TODO tune?
scanAndRemoveOldEntries(2 * numMutations + 10);
Arena arena;
// TODO Improve ILP?
for (int i = 0; i < numMutations; ++i) {
const auto &m = mutations[i];
switch (m.type) {
case Set: {
Finger iter;
search<std::memory_order_relaxed>({m.param1, m.param1Len}, latestRoot,
latestVersion, iter);
insert({m.param1, m.param1Len}, {{m.param2, m.param2Len}},
/*endRange*/ false, iter);
} break;
case Clear: {
// TODO we can avoid some insertions here. Complexity is getting out of
// hand though.
if (m.param2Len == 0) {
Finger iter;
search<std::memory_order_relaxed>({m.param1, m.param1Len}, latestRoot,
latestVersion, iter);
const bool found = iter.searchPathSize() > 0 && iter.backNode() != 0;
bool engulfLeft = found && mm.base[iter.backNode()].entry->clearTo();
bool engulfRight = false;
const Entry *next;
Finger copy;
if (iter.searchPathSize() > 0) {
iter.copyTo(copy);
move<std::memory_order_relaxed, true>(copy, latestVersion);
next = copy.searchPathSize() > 0 ? mm.base[copy.backNode()].entry
: nullptr;
if (next && next->clearTo()) {
engulfRight = true;
if (!found) {
engulfLeft = true;
}
}
}
if (engulfLeft && engulfRight) {
insert({next->getKey(), next->keyLen}, {}, /*endRange*/ true, copy);
if (found) {
move<std::memory_order_relaxed, false>(copy, latestVersion);
// Point to the same entry as iter pointed to, but it's not
// invalidated
remove(copy);
}
} else if (engulfLeft) {
assert(found);
remove(iter);
insert(keyAfter({m.param1, m.param1Len}, arena), {},
/*endRange*/ true, iter);
} else if (engulfRight) {
insert({m.param1, m.param1Len}, {}, /*endRange*/ false, iter);
move<std::memory_order_relaxed, true>(iter, latestVersion);
assert(iter.searchPathSize() > 0 &&
mm.base[iter.backNode()].entry == next);
insert({next->getKey(), next->keyLen}, {}, /*endRange*/ true, iter);
} else {
insert({m.param1, m.param1Len}, {{nullptr, -1}}, /*endRange*/ false,
iter);
}
} else {
// TODO ILP these
Finger begin;
search<std::memory_order_relaxed>({m.param1, m.param1Len}, latestRoot,
latestVersion, begin);
const bool foundBegin =
begin.searchPathSize() > 0 && begin.backNode() != 0;
Finger end;
search<std::memory_order_relaxed>({m.param2, m.param2Len}, latestRoot,
latestVersion, end);
const bool foundEnd = end.searchPathSize() > 0 && end.backNode() != 0;
// Check if we can engulf on the left
bool engulfLeft;
Finger copy;
begin.copyTo(copy);
move<std::memory_order_relaxed, true>(copy, latestVersion);
if (foundBegin) {
engulfLeft = begin.searchPathSize() > 0 &&
mm.base[begin.backNode()].entry->clearTo();
} else {
engulfLeft = copy.searchPathSize() > 0 &&
mm.base[copy.backNode()].entry->clearTo();
}
// Check if we can engulf on the right
bool engulfRight = false;
if (!foundEnd) {
end.copyTo(copy);
move<std::memory_order_relaxed, true>(copy, latestVersion);
const auto *next = copy.searchPathSize() > 0
? mm.base[copy.backNode()].entry
: nullptr;
engulfRight = next && next->clearTo();
}
if (engulfLeft && foundBegin) {
remove(begin);
} else if (!engulfLeft) {
insert({m.param1, m.param1Len}, {}, /*rangeEntry*/ false, begin);
}
move<std::memory_order_relaxed, true>(begin, latestVersion);
while (begin.searchPathSize() > 0 &&
mm.base[begin.backNode()] < Key{m.param2, m.param2Len}) {
remove(begin);
move<std::memory_order_relaxed, true>(begin, latestVersion);
}
#ifndef NDEBUG
if (foundEnd) {
[[maybe_unused]] bool beginEqEnd =
mm.base[begin.backNode()] <=> Key{m.param2, m.param2Len} == 0;
assert(beginEqEnd);
}
#endif
if (engulfRight) {
if (foundEnd) {
remove(begin);
move<std::memory_order_relaxed, true>(begin, latestVersion);
}
assert(begin.searchPathSize() > 0 && begin.backNode() != 0);
insert({mm.base[begin.backNode()].entry->getKey(),
mm.base[begin.backNode()].entry->keyLen},
{}, /*rangeEntry*/ true, begin);
} else {
if (!foundEnd) {
// TODO remove this search
search<std::memory_order_relaxed>(
{m.param2, m.param2Len}, latestRoot, latestVersion, begin);
}
insert({m.param2, m.param2Len}, {}, /*rangeEntry*/ true, begin);
}
}
} break;
default: // GCOVR_EXCL_LINE
assert(false); // GCOVR_EXCL_LINE
}
}
roots.add(latestRoot, latestVersion);
totalMallocBytes += mallocBytesDelta;
// Check the "latestRoot is only meaningful in the callstack of
// addMutations" property
VALGRIND_MAKE_MEM_UNDEFINED(&latestRoot, sizeof(latestRoot));
}
struct StepwiseFirstGeq {
const VersionedMap::Impl *map;
const weaselab::VersionedMap::Key *key;
int64_t version;
weaselab::VersionedMap::Iterator *iterator;
void begin(uint32_t root);
bool step();
void end();
};
void firstGeq(const Key *key, const int64_t *version, Iterator *iterator,
int count) const;
void firstGeq(const Key *key, Iterator *iterator, int count) const;
// State used to resume scanning and removing old entries in `addMutations`
Key continueKey;
Arena continueArena;
MemManager mm;
RootSet roots;
// Only meaningful within the callstack of `addMutations`
uint32_t latestRoot;
int64_t oldestVersion = 0;
int64_t latestVersion = 0;
int64_t totalMallocBytes = sizeof(Impl);
};
VersionedMap::Impl *internal_makeImpl(int64_t version) {
mallocBytesDelta = 0;
auto *result =
new (safe_malloc(sizeof(VersionedMap::Impl))) VersionedMap::Impl();
result->totalMallocBytes = mallocBytesDelta;
result->latestVersion = version;
return result;
}
VersionedMap::VersionedMap(int64_t version)
: impl(internal_makeImpl(version)) {}
VersionedMap::~VersionedMap() {
if (impl != nullptr) {
impl->~Impl();
safe_free(impl, sizeof(*impl));
}
}
VersionedMap::VersionedMap(VersionedMap &&other) noexcept {
impl = std::exchange(other.impl, nullptr);
}
VersionedMap &VersionedMap::operator=(VersionedMap &&other) noexcept {
impl = std::exchange(other.impl, nullptr);
return *this;
}
void VersionedMap::addMutations(const Mutation *mutations, int numMutations,
int64_t version) {
impl->addMutations(mutations, numMutations, version);
}
struct VersionedMap::Iterator::Impl {
Finger finger;
int64_t version;
const VersionedMap::Impl *map;
// State for materializing mutations associated with the entry at `finger`.
// Cases:
// - If finger is a set and the end of a clear, then mutation[0] is the clear
// and mutation[1] is the set.
// - If finger is a set and not the end of a clear, then mutation[0] is the
// set
// - If finger is a clear and not a set, then mutation[0] is the clear
int mutationCount;
int mutationIndex;
VersionedMutation mutations[2];
void copyTo(Impl &result) {
result.map = map;
result.version = version;
result.mutationCount = mutationCount;
result.mutationIndex = mutationIndex;
result.mutations[0] = mutations[0];
result.mutations[1] = mutations[1];
finger.copyTo(result.finger);
}
bool equals(const Impl &other) const {
assert(map == other.map);
assert(version == other.version);
return finger == other.finger && mutationIndex == other.mutationIndex;
}
};
VersionedMap::Iterator::~Iterator() {
if (impl != nullptr) {
impl->~Impl();
safe_free(impl, sizeof(*impl));
}
}
VersionedMap::Iterator::Iterator(const Iterator &other)
: impl(new(safe_malloc(sizeof(Impl))) Impl()) {
other.impl->copyTo(*impl);
}
VersionedMap::Iterator &
VersionedMap::Iterator::operator=(const Iterator &other) {
if (impl != nullptr) {
impl->~Impl();
safe_free(impl, sizeof(*impl));
}
impl = new (safe_malloc(sizeof(Impl))) Impl();
other.impl->copyTo(*impl);
return *this;
}
VersionedMap::Iterator::Iterator(Iterator &&other) noexcept
: impl(std::exchange(other.impl, nullptr)) {}
VersionedMap::Iterator &
VersionedMap::Iterator::operator=(Iterator &&other) noexcept {
if (impl != nullptr) {
impl->~Impl();
safe_free(impl, sizeof(*impl));
}
impl = std::exchange(other.impl, nullptr);
return *this;
}
VersionedMap::Iterator::VersionedMutation
VersionedMap::Iterator::operator*() const {
#if DEBUG_VERBOSE
if (debugVerboseEnabled) {
printf("Dereference %u\n", impl->finger.backNode());
}
#endif
assert(impl->finger.searchPathSize() != 0);
assert(impl->mutationIndex < impl->mutationCount);
assert(impl->mutationIndex >= 0);
return impl->mutations[impl->mutationIndex];
}
void materializeMutations(VersionedMap::Iterator::Impl *impl, const Entry *prev,
const Entry *next) {
if (prev == nullptr) {
Finger copy;
impl->finger.copyTo(copy);
impl->map->move<std::memory_order_acquire, false>(copy, impl->version);
if (copy.searchPathSize() > 0) {
prev = impl->map->mm.base[copy.backNode()].entry;
} else {
assert(!impl->map->mm.base[impl->finger.backNode()].entry->clearTo());
}
}
if (next == nullptr) {
Finger copy;
impl->finger.copyTo(copy);
impl->map->move<std::memory_order_acquire, true>(copy, impl->version);
if (copy.searchPathSize() > 0) {
next = impl->map->mm.base[copy.backNode()].entry;
}
}
const auto &entry = *impl->map->mm.base[impl->finger.backNode()].entry;
impl->mutationCount = 0;
if (entry.clearTo()) {
impl->mutations[impl->mutationCount++] = {
prev->getKey(),
entry.getKey(),
prev->pointSet() ? prev->keyLen + 1 : prev->keyLen,
entry.keyLen,
VersionedMap::Clear,
entry.rangeVersion};
}
if (entry.pointMutation()) {
if (entry.valLen < 0 /* pointClear */) {
assert(next == nullptr || !next->clearTo());
impl->mutations[impl->mutationCount++] = {
entry.getKey(), nullptr, entry.keyLen, 0,
VersionedMap::Clear, entry.pointVersion};
} else {
impl->mutations[impl->mutationCount++] = {
entry.getKey(), entry.getVal(), entry.keyLen,
entry.valLen, VersionedMap::Set, entry.pointVersion};
}
}
}
VersionedMap::Iterator &VersionedMap::Iterator::operator++() {
if (impl->mutationIndex < impl->mutationCount - 1) {
++impl->mutationIndex;
return *this;
}
do {
const auto &entry = *impl->map->mm.base[impl->finger.backNode()].entry;
impl->map->move<std::memory_order_acquire, true>(impl->finger,
impl->version);
if (impl->finger.searchPathSize() == 0) {
break;
}
materializeMutations(impl, &entry, nullptr);
} while (impl->mutationCount == 0);
impl->mutationIndex = 0;
return *this;
}
VersionedMap::Iterator VersionedMap::Iterator::operator++(int) {
auto result = *this;
// TODO Interposable call
++*this;
return result;
}
VersionedMap::Iterator &VersionedMap::Iterator::operator--() {
if (impl->mutationIndex > 0) {
--impl->mutationIndex;
return *this;
}
// Handle decrementing end
if (impl->finger.searchPathSize() == 0) {
bool ignored = false;
impl->finger.push(
impl->map->roots.getThreadSafeHandle().rootForVersion(impl->version),
ignored);
assert(impl->finger.backNode() != 0);
uint32_t c;
while ((c = impl->map->child<std::memory_order_acquire>(
impl->finger.backNode(), true, impl->version)) != 0) {
impl->finger.push(c, true);
}
const Entry *next = nullptr;
for (;;) {
materializeMutations(impl, nullptr, next);
if (impl->mutationCount > 0) {
break;
}
next = impl->map->mm.base[impl->finger.backNode()].entry;
impl->map->move<std::memory_order_acquire, false>(impl->finger,
impl->version);
}
impl->mutationIndex = impl->mutationCount - 1;
return *this;
}
do {
const Entry *entry = impl->map->mm.base[impl->finger.backNode()].entry;
impl->map->move<std::memory_order_acquire, false>(impl->finger,
impl->version);
if (impl->finger.searchPathSize() == 0) {
break;
}
materializeMutations(impl, nullptr, entry);
} while (impl->mutationCount == 0);
impl->mutationIndex = impl->mutationCount - 1;
return *this;
}
VersionedMap::Iterator VersionedMap::Iterator::operator--(int) {
auto result = *this;
// TODO Interposable call
--*this;
return result;
}
bool VersionedMap::Iterator::operator==(const Iterator &other) const {
if (impl == nullptr || other.impl == nullptr) {
return impl == other.impl;
}
return impl->equals(*other.impl);
}
bool geq(const VersionedMap::Iterator::VersionedMutation &m,
const VersionedMap::Key &k) {
if (m.type == VersionedMap::Set || m.param2Len == 0) {
return VersionedMap::Key{m.param1, m.param1Len} >= k;
} else {
return VersionedMap::Key{m.param2, m.param2Len} > k;
}
}
void VersionedMap::Impl::StepwiseFirstGeq::begin(uint32_t root) {
if (iterator->impl != nullptr) {
iterator->impl->~Impl();
new (iterator->impl) Iterator::Impl();
} else {
iterator->impl = new (safe_malloc(sizeof(Iterator::Impl))) Iterator::Impl();
}
Finger &finger = iterator->impl->finger;
finger.clear();
bool ignored = false;
finger.push(root, ignored);
}
bool VersionedMap::Impl::StepwiseFirstGeq::step() {
Finger &finger = iterator->impl->finger;
auto n = finger.backNode();
if (n == 0) {
return true;
}
auto c = *key <=> map->mm.base[n];
if (c == 0) {
// No duplicates
return true;
}
finger.push(map->child<std::memory_order_acquire>(n, c > 0, version), c > 0);
return false;
}
void VersionedMap::Impl::StepwiseFirstGeq::end() {
Finger &finger = iterator->impl->finger;
if (finger.searchPathSize() > 0 && finger.backNode() == 0) {
map->move<std::memory_order_acquire, true>(finger, version);
if (finger.searchPathSize() > 0) {
assert(finger.backNode() != 0);
}
}
iterator->impl->version = version;
iterator->impl->map = map;
const Entry *prev = nullptr;
for (;;) {
if (finger.searchPathSize() == 0) {
break;
} else {
materializeMutations(iterator->impl, prev, nullptr);
for (int j = 0; j < iterator->impl->mutationCount; ++j) {
if (geq(iterator->impl->mutations[j], *key)) {
iterator->impl->mutationIndex = j;
goto loopEnd;
}
}
}
prev = iterator->impl->map->mm.base[finger.backNode()].entry;
iterator->impl->map->move<std::memory_order_acquire, true>(
finger, iterator->impl->version);
}
loopEnd:;
}
constexpr int kStackAllocThreshold = 2;
void VersionedMap::Impl::firstGeq(const weaselab::VersionedMap::Key *key,
const int64_t *version, Iterator *iterator,
int count) const {
if (count == 0) {
return;
}
// Use stack allocation for small count
Arena arena;
StepwiseFirstGeq stepwiseStackAlloc[kStackAllocThreshold];
int nextJobStackAllocation[kStackAllocThreshold];
StepwiseFirstGeq *stepwise;
int *nextJob;
if (count <= kStackAllocThreshold) {
stepwise = stepwiseStackAlloc;
nextJob = nextJobStackAllocation;
} else {
stepwise = new (arena) StepwiseFirstGeq[count];
nextJob = new (arena) int[count];
}
auto handle = roots.getThreadSafeHandle();
for (int i = 0; i < count; ++i) {
stepwise[i].map = this;
stepwise[i].key = &key[i];
stepwise[i].version = version[i];
stepwise[i].iterator = &iterator[i];
stepwise[i].begin(handle.rootForVersion(version[i]));
nextJob[i] = i + 1;
}
nextJob[count - 1] = 0;
int prevJob = count - 1;
int job = 0;
for (;;) {
if (stepwise[job].step()) {
stepwise[job].end();
if (job == prevJob) {
break;
}
nextJob[prevJob] = nextJob[job];
job = prevJob;
}
prevJob = job;
job = nextJob[job];
}
}
void VersionedMap::Impl::firstGeq(const weaselab::VersionedMap::Key *key,
Iterator *iterator, int count) const {
if (count == 0) {
return;
}
// Use stack allocation for small count
Arena arena;
StepwiseFirstGeq stepwiseStackAlloc[kStackAllocThreshold];
int nextJobStackAllocation[kStackAllocThreshold];
StepwiseFirstGeq *stepwise;
int *nextJob;
if (count <= kStackAllocThreshold) {
stepwise = stepwiseStackAlloc;
nextJob = nextJobStackAllocation;
} else {
stepwise = new (arena) StepwiseFirstGeq[count];
nextJob = new (arena) int[count];
}
const uint32_t root = roots.roots()[roots.rootCount() - 1];
assert(root == roots.getThreadSafeHandle().rootForVersion(latestVersion));
for (int i = 0; i < count; ++i) {
stepwise[i].map = this;
stepwise[i].key = &key[i];
stepwise[i].version = latestVersion;
stepwise[i].iterator = &iterator[i];
stepwise[i].begin(root);
nextJob[i] = i + 1;
}
nextJob[count - 1] = 0;
int prevJob = count - 1;
int job = 0;
for (;;) {
if (stepwise[job].step()) {
stepwise[job].end();
if (job == prevJob) {
break;
}
nextJob[prevJob] = nextJob[job];
job = prevJob;
}
prevJob = job;
job = nextJob[job];
}
}
bool VersionedMap::Iterator::operator!=(const Iterator &other) const {
if (impl == nullptr || other.impl == nullptr) {
return impl != other.impl;
}
return !impl->equals(*other.impl);
}
void VersionedMap::firstGeq(const Key *key, const int64_t *version,
Iterator *iterator, int count) const {
impl->firstGeq(key, version, iterator, count);
}
void VersionedMap::firstGeq(const Key *key, Iterator *iterator,
int count) const {
impl->firstGeq(key, iterator, count);
}
VersionedMap::Iterator VersionedMap::begin(int64_t version) const {
VersionedMap::Iterator result;
result.impl = new (safe_malloc(sizeof(Iterator::Impl))) Iterator::Impl();
result.impl->version = version;
bool ignored = false;
result.impl->finger.push(
impl->roots.getThreadSafeHandle().rootForVersion(version), ignored);
if (result.impl->finger.backNode() == 0) {
result.impl->finger.pop();
} else {
uint32_t c;
while ((c = impl->child<std::memory_order_acquire>(
result.impl->finger.backNode(), false, version)) != 0) {
result.impl->finger.push(c, false);
}
}
result.impl->map = impl;
const Entry *prev = nullptr;
for (;;) {
if (result.impl->finger.searchPathSize() > 0) {
materializeMutations(result.impl, prev, nullptr);
if (result.impl->mutationCount > 0) {
break;
}
} else {
break;
}
prev = result.impl->map->mm.base[result.impl->finger.backNode()].entry;
result.impl->map->move<std::memory_order_acquire, true>(
result.impl->finger, result.impl->version);
}
result.impl->mutationIndex = 0;
return result;
}
VersionedMap::Iterator VersionedMap::end(int64_t version) const {
VersionedMap::Iterator result;
result.impl = new (safe_malloc(sizeof(Iterator::Impl))) Iterator::Impl();
result.impl->map = impl;
result.impl->mutationIndex = 0;
result.impl->version = version;
return result;
}
int64_t VersionedMap::getVersion() const { return impl->latestVersion; }
int64_t VersionedMap::getOldestVersion() const { return impl->oldestVersion; }
void VersionedMap::setOldestVersion(int64_t oldestVersion) {
impl->setOldestVersion(oldestVersion);
}
int64_t VersionedMap::getBytes() const { return impl->getBytes(); }
// ==================== END IMPLEMENTATION ====================
// GCOVR_EXCL_START
#ifdef NDEBUG
inline
#endif
void
VersionedMap::Impl::printInOrder(int64_t version) const {
printInOrderHelper(version,
roots.getThreadSafeHandle().rootForVersion(version), 0);
}
#ifdef NDEBUG
inline
#endif
void
VersionedMap::Impl::printInOrderHelper(int64_t version, uint32_t node,
int depth) const {
if (node == 0) {
return;
}
printInOrderHelper(version,
child<std::memory_order_relaxed>(node, true, version),
depth + 1);
for (int i = 0; i < depth; ++i) {
printf(" ");
}
printf("node %u: ", node);
printBinary({mm.base[node].entry->getKey(), mm.base[node].entry->keyLen});
if (mm.base[node].entry->pointSet()) {
printf(" -> '");
printBinary({mm.base[node].entry->getVal(), mm.base[node].entry->valLen});
printf("' @ %" PRId64, mm.base[node].entry->pointVersion);
}
if (mm.base[node].entry->pointClear()) {
printf(" <cleared @ %" PRId64 ">", mm.base[node].entry->pointVersion);
}
if (mm.base[node].entry->clearTo()) {
printf(" <clearTo @ %" PRId64 ">", mm.base[node].entry->rangeVersion);
}
if (mm.base[node].entry->pointVersion < 0 &&
mm.base[node].entry->rangeVersion < 0) {
printf(" <noop>");
}
printf("\n");
VersionedMap::Impl::printInOrderHelper(
version, child<std::memory_order_relaxed>(node, false, version),
depth + 1);
}
VersionedMap::Impl *cast(const VersionedMap &m) {
VersionedMap::Impl *result;
memcpy(&result, &m, sizeof(void *));
return result;
}
#if SHOW_MEMORY
struct __attribute__((visibility("default"))) PeakPrinter {
~PeakPrinter() {
printf("--- versioned_map ---\n");
printf("mmap bytes: %g\n", double(mmapBytes));
printf("Peak mmap bytes: %g\n", double(peakMmapBytes));
}
} peakPrinter2;
#endif
} // namespace weaselab
#ifdef ENABLE_MAIN
#include <nanobench.h>
void breakpoint_me() {}
int main() {
{
weaselab::VersionedMap versionedMap{0};
{
weaselab::VersionedMap::Mutation m[] = {
{(const uint8_t *)"a", 1, (const uint8_t *)"b", 1,
weaselab::VersionedMap::Clear},
};
versionedMap.addMutations(m, sizeof(m) / sizeof(m[0]), 1);
}
{
weaselab::VersionedMap::Mutation m[] = {
{(const uint8_t *)"b", 1, nullptr, 0, weaselab::VersionedMap::Clear},
};
versionedMap.addMutations(m, sizeof(m) / sizeof(m[0]), 2);
}
const int64_t v = versionedMap.getVersion();
cast(versionedMap)->printInOrder(v);
weaselab::VersionedMap::Key k = {(const uint8_t *)"a", 2};
weaselab::VersionedMap::Iterator iter;
versionedMap.firstGeq(&k, &v, &iter, 1);
versionedMap.setOldestVersion(2);
breakpoint_me();
for (auto end = versionedMap.end(v); iter != end; ++iter) {
printMutation(*iter);
}
}
return 0;
}
#endif
// GCOVR_EXCL_STOP