blob: 45284e2a253ac6cb1f7a4a3b94588997ca801aa5 [file] [log] [blame]
// Copyright (c) 2013 The Chromium Authors. All rights reserved.
// Use of this source code is governed by a BSD-style license that can be
// found in the LICENSE file.
#include "net/disk_cache/simple/simple_entry_impl.h"
#include <algorithm>
#include <cstring>
#include <limits>
#include <utility>
#include <vector>
#include "base/bind.h"
#include "base/bind_helpers.h"
#include "base/callback.h"
#include "base/location.h"
#include "base/logging.h"
#include "base/single_thread_task_runner.h"
#include "base/task_runner.h"
#include "base/task_runner_util.h"
#include "base/threading/thread_task_runner_handle.h"
#include "base/time/time.h"
#include "base/trace_event/memory_usage_estimator.h"
#include "net/base/io_buffer.h"
#include "net/base/net_errors.h"
#include "net/base/prioritized_task_runner.h"
#include "net/disk_cache/backend_cleanup_tracker.h"
#include "net/disk_cache/net_log_parameters.h"
#include "net/disk_cache/simple/simple_backend_impl.h"
#include "net/disk_cache/simple/simple_histogram_enums.h"
#include "net/disk_cache/simple/simple_histogram_macros.h"
#include "net/disk_cache/simple/simple_index.h"
#include "net/disk_cache/simple/simple_net_log_parameters.h"
#include "net/disk_cache/simple/simple_synchronous_entry.h"
#include "net/disk_cache/simple/simple_util.h"
#include "net/log/net_log.h"
#include "net/log/net_log_source_type.h"
#include "starboard/memory.h"
#include "third_party/zlib/zlib.h"
namespace disk_cache {
namespace {
// An entry can store sparse data taking up to 1 / kMaxSparseDataSizeDivisor of
// the cache.
const int64_t kMaxSparseDataSizeDivisor = 10;
// Used in histograms, please only add entries at the end.
enum SimpleEntryWriteResult {
SIMPLE_ENTRY_WRITE_RESULT_SUCCESS = 0,
SIMPLE_ENTRY_WRITE_RESULT_INVALID_ARGUMENT = 1,
SIMPLE_ENTRY_WRITE_RESULT_OVER_MAX_SIZE = 2,
SIMPLE_ENTRY_WRITE_RESULT_BAD_STATE = 3,
SIMPLE_ENTRY_WRITE_RESULT_SYNC_WRITE_FAILURE = 4,
SIMPLE_ENTRY_WRITE_RESULT_FAST_EMPTY_RETURN = 5,
SIMPLE_ENTRY_WRITE_RESULT_MAX = 6,
};
void RecordReadResult(net::CacheType cache_type, SimpleReadResult result) {
SIMPLE_CACHE_UMA(ENUMERATION,
"ReadResult", cache_type, result, READ_RESULT_MAX);
}
void RecordWriteResult(net::CacheType cache_type,
SimpleEntryWriteResult result) {
SIMPLE_CACHE_UMA(ENUMERATION, "WriteResult2", cache_type, result,
SIMPLE_ENTRY_WRITE_RESULT_MAX);
}
void RecordHeaderSize(net::CacheType cache_type, int size) {
SIMPLE_CACHE_UMA(COUNTS_10000, "HeaderSize", cache_type, size);
}
int g_open_entry_count = 0;
void AdjustOpenEntryCountBy(net::CacheType cache_type, int offset) {
g_open_entry_count += offset;
SIMPLE_CACHE_UMA(COUNTS_10000,
"GlobalOpenEntryCount", cache_type, g_open_entry_count);
}
void InvokeCallbackIfBackendIsAlive(
const base::WeakPtr<SimpleBackendImpl>& backend,
net::CompletionOnceCallback completion_callback,
int result) {
DCHECK(!completion_callback.is_null());
if (!backend.get())
return;
std::move(completion_callback).Run(result);
}
// If |sync_possible| is false, and callback is available, posts rv to it and
// return net::ERR_IO_PENDING; otherwise just passes through rv.
int PostToCallbackIfNeeded(bool sync_possible,
net::CompletionOnceCallback callback,
int rv) {
if (!sync_possible && !callback.is_null()) {
base::ThreadTaskRunnerHandle::Get()->PostTask(
FROM_HERE, base::BindOnce(std::move(callback), rv));
return net::ERR_IO_PENDING;
} else {
return rv;
}
}
} // namespace
using base::Closure;
using base::OnceClosure;
using base::FilePath;
using base::Time;
using base::TaskRunner;
// Static function called by base::trace_event::EstimateMemoryUsage() to
// estimate the memory of SimpleEntryOperation.
// This needs to be in disk_cache namespace.
size_t EstimateMemoryUsage(const SimpleEntryOperation& op) {
return 0;
}
// A helper class to insure that RunNextOperationIfNeeded() is called when
// exiting the current stack frame.
class SimpleEntryImpl::ScopedOperationRunner {
public:
explicit ScopedOperationRunner(SimpleEntryImpl* entry) : entry_(entry) {
}
~ScopedOperationRunner() {
entry_->RunNextOperationIfNeeded();
}
private:
SimpleEntryImpl* const entry_;
};
SimpleEntryImpl::ActiveEntryProxy::~ActiveEntryProxy() = default;
SimpleEntryImpl::SimpleEntryImpl(
net::CacheType cache_type,
const FilePath& path,
scoped_refptr<BackendCleanupTracker> cleanup_tracker,
const uint64_t entry_hash,
OperationsMode operations_mode,
SimpleBackendImpl* backend,
SimpleFileTracker* file_tracker,
net::NetLog* net_log,
uint32_t entry_priority)
: cleanup_tracker_(std::move(cleanup_tracker)),
backend_(backend->AsWeakPtr()),
file_tracker_(file_tracker),
cache_type_(cache_type),
path_(path),
entry_hash_(entry_hash),
use_optimistic_operations_(operations_mode == OPTIMISTIC_OPERATIONS),
is_initial_stream1_read_(true),
last_used_(Time::Now()),
last_modified_(last_used_),
sparse_data_size_(0),
open_count_(0),
doom_state_(DOOM_NONE),
optimistic_create_pending_doom_state_(CREATE_NORMAL),
state_(STATE_UNINITIALIZED),
synchronous_entry_(NULL),
prioritized_task_runner_(backend_->prioritized_task_runner()),
net_log_(
net::NetLogWithSource::Make(net_log,
net::NetLogSourceType::DISK_CACHE_ENTRY)),
stream_0_data_(base::MakeRefCounted<net::GrowableIOBuffer>()),
entry_priority_(entry_priority) {
static_assert(arraysize(data_size_) == arraysize(crc32s_end_offset_),
"arrays should be the same size");
static_assert(arraysize(data_size_) == arraysize(crc32s_),
"arrays should be the same size");
static_assert(arraysize(data_size_) == arraysize(have_written_),
"arrays should be the same size");
static_assert(arraysize(data_size_) == arraysize(crc_check_state_),
"arrays should be the same size");
ResetEntry();
net_log_.BeginEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY,
CreateNetLogSimpleEntryConstructionCallback(this));
}
void SimpleEntryImpl::SetActiveEntryProxy(
std::unique_ptr<ActiveEntryProxy> active_entry_proxy) {
DCHECK(!active_entry_proxy_);
active_entry_proxy_ = std::move(active_entry_proxy);
}
net::Error SimpleEntryImpl::OpenEntry(Entry** out_entry,
net::CompletionOnceCallback callback) {
DCHECK(backend_.get());
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_OPEN_CALL);
bool have_index = backend_->index()->initialized();
// This enumeration is used in histograms, add entries only at end.
enum OpenEntryIndexEnum {
INDEX_NOEXIST = 0,
INDEX_MISS = 1,
INDEX_HIT = 2,
INDEX_MAX = 3,
};
OpenEntryIndexEnum open_entry_index_enum = INDEX_NOEXIST;
if (have_index) {
if (backend_->index()->Has(entry_hash_))
open_entry_index_enum = INDEX_HIT;
else
open_entry_index_enum = INDEX_MISS;
}
SIMPLE_CACHE_UMA(ENUMERATION,
"OpenEntryIndexState", cache_type_,
open_entry_index_enum, INDEX_MAX);
// If entry is not known to the index, initiate fast failover to the network.
if (open_entry_index_enum == INDEX_MISS) {
net_log_.AddEventWithNetErrorCode(
net::NetLogEventType::SIMPLE_CACHE_ENTRY_OPEN_END, net::ERR_FAILED);
return net::ERR_FAILED;
}
pending_operations_.push(SimpleEntryOperation::OpenOperation(
this, have_index, std::move(callback), out_entry));
RunNextOperationIfNeeded();
return net::ERR_IO_PENDING;
}
net::Error SimpleEntryImpl::CreateEntry(Entry** out_entry,
net::CompletionOnceCallback callback) {
DCHECK(backend_.get());
DCHECK_EQ(entry_hash_, simple_util::GetEntryHashKey(key_));
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_CREATE_CALL);
bool have_index = backend_->index()->initialized();
net::Error ret_value = net::ERR_FAILED;
if (use_optimistic_operations_ &&
state_ == STATE_UNINITIALIZED && pending_operations_.size() == 0) {
net_log_.AddEvent(
net::NetLogEventType::SIMPLE_CACHE_ENTRY_CREATE_OPTIMISTIC);
ReturnEntryToCaller(out_entry);
pending_operations_.push(SimpleEntryOperation::CreateOperation(
this, have_index, CompletionOnceCallback(),
static_cast<Entry**>(NULL)));
ret_value = net::OK;
// If we are optimistically returning before a preceeding doom, we need to
// wait for that IO, about which we will be notified externally.
if (optimistic_create_pending_doom_state_ != CREATE_NORMAL) {
DCHECK_EQ(CREATE_OPTIMISTIC_PENDING_DOOM,
optimistic_create_pending_doom_state_);
state_ = STATE_IO_PENDING;
}
} else {
pending_operations_.push(SimpleEntryOperation::CreateOperation(
this, have_index, std::move(callback), out_entry));
ret_value = net::ERR_IO_PENDING;
}
// We insert the entry in the index before creating the entry files in the
// SimpleSynchronousEntry, because this way the worst scenario is when we
// have the entry in the index but we don't have the created files yet, this
// way we never leak files. CreationOperationComplete will remove the entry
// from the index if the creation fails.
backend_->index()->Insert(entry_hash_);
RunNextOperationIfNeeded();
return ret_value;
}
net::Error SimpleEntryImpl::DoomEntry(net::CompletionOnceCallback callback) {
if (doom_state_ != DOOM_NONE)
return net::OK;
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_DOOM_CALL);
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_DOOM_BEGIN);
MarkAsDoomed(DOOM_QUEUED);
if (backend_.get()) {
if (optimistic_create_pending_doom_state_ == CREATE_NORMAL) {
backend_->OnDoomStart(entry_hash_);
} else {
DCHECK_EQ(STATE_IO_PENDING, state_);
DCHECK_EQ(CREATE_OPTIMISTIC_PENDING_DOOM,
optimistic_create_pending_doom_state_);
// If we are in this state, we went ahead with making the entry even
// though the backend was already keeping track of a doom, so it can't
// keep track of ours. So we delay notifying it until
// NotifyDoomBeforeCreateComplete is called. Since this path is invoked
// only when the queue of post-doom callbacks was previously empty, while
// the CompletionOnceCallback for the op is posted,
// NotifyDoomBeforeCreateComplete() will be the first thing running after
// the previous doom completes, so at that point we can immediately grab
// a spot in entries_pending_doom_.
optimistic_create_pending_doom_state_ =
CREATE_OPTIMISTIC_PENDING_DOOM_FOLLOWED_BY_DOOM;
}
}
pending_operations_.push(
SimpleEntryOperation::DoomOperation(this, std::move(callback)));
RunNextOperationIfNeeded();
return net::ERR_IO_PENDING;
}
void SimpleEntryImpl::SetCreatePendingDoom() {
DCHECK_EQ(CREATE_NORMAL, optimistic_create_pending_doom_state_);
optimistic_create_pending_doom_state_ = CREATE_OPTIMISTIC_PENDING_DOOM;
}
void SimpleEntryImpl::NotifyDoomBeforeCreateComplete() {
DCHECK_EQ(STATE_IO_PENDING, state_);
DCHECK_NE(CREATE_NORMAL, optimistic_create_pending_doom_state_);
if (backend_.get() && optimistic_create_pending_doom_state_ ==
CREATE_OPTIMISTIC_PENDING_DOOM_FOLLOWED_BY_DOOM)
backend_->OnDoomStart(entry_hash_);
state_ = STATE_UNINITIALIZED;
optimistic_create_pending_doom_state_ = CREATE_NORMAL;
RunNextOperationIfNeeded();
}
void SimpleEntryImpl::SetKey(const std::string& key) {
key_ = key;
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_SET_KEY,
net::NetLog::StringCallback("key", &key));
}
void SimpleEntryImpl::Doom() {
DoomEntry(CompletionOnceCallback());
}
void SimpleEntryImpl::Close() {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
DCHECK_LT(0, open_count_);
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_CLOSE_CALL);
if (--open_count_ > 0) {
DCHECK(!HasOneRef());
Release(); // Balanced in ReturnEntryToCaller().
return;
}
pending_operations_.push(SimpleEntryOperation::CloseOperation(this));
DCHECK(!HasOneRef());
Release(); // Balanced in ReturnEntryToCaller().
RunNextOperationIfNeeded();
}
std::string SimpleEntryImpl::GetKey() const {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
return key_;
}
Time SimpleEntryImpl::GetLastUsed() const {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
return last_used_;
}
Time SimpleEntryImpl::GetLastModified() const {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
return last_modified_;
}
int32_t SimpleEntryImpl::GetDataSize(int stream_index) const {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
DCHECK_LE(0, data_size_[stream_index]);
return data_size_[stream_index];
}
int SimpleEntryImpl::ReadData(int stream_index,
int offset,
net::IOBuffer* buf,
int buf_len,
CompletionOnceCallback callback) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_READ_CALL,
CreateNetLogReadWriteDataCallback(stream_index, offset,
buf_len, false));
}
if (stream_index < 0 || stream_index >= kSimpleEntryStreamCount ||
buf_len < 0) {
if (net_log_.IsCapturing()) {
net_log_.AddEvent(
net::NetLogEventType::SIMPLE_CACHE_ENTRY_READ_END,
CreateNetLogReadWriteCompleteCallback(net::ERR_INVALID_ARGUMENT));
}
RecordReadResult(cache_type_, READ_RESULT_INVALID_ARGUMENT);
return net::ERR_INVALID_ARGUMENT;
}
// If this is the only operation, bypass the queue, and also see if there is
// in-memory data to handle it synchronously. In principle, multiple reads can
// be parallelized, but past studies have shown that parallelizable ones
// happen <1% of the time, so it's probably not worth the effort.
bool alone_in_queue =
pending_operations_.size() == 0 && state_ == STATE_READY;
if (alone_in_queue) {
return ReadDataInternal(/*sync_possible = */ true, stream_index, offset,
buf, buf_len, std::move(callback));
}
pending_operations_.push(SimpleEntryOperation::ReadOperation(
this, stream_index, offset, buf_len, buf, std::move(callback)));
RunNextOperationIfNeeded();
return net::ERR_IO_PENDING;
}
int SimpleEntryImpl::WriteData(int stream_index,
int offset,
net::IOBuffer* buf,
int buf_len,
CompletionOnceCallback callback,
bool truncate) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_WRITE_CALL,
CreateNetLogReadWriteDataCallback(stream_index, offset,
buf_len, truncate));
}
if (stream_index < 0 || stream_index >= kSimpleEntryStreamCount ||
offset < 0 || buf_len < 0) {
if (net_log_.IsCapturing()) {
net_log_.AddEvent(
net::NetLogEventType::SIMPLE_CACHE_ENTRY_WRITE_END,
CreateNetLogReadWriteCompleteCallback(net::ERR_INVALID_ARGUMENT));
}
RecordWriteResult(cache_type_, SIMPLE_ENTRY_WRITE_RESULT_INVALID_ARGUMENT);
return net::ERR_INVALID_ARGUMENT;
}
if (backend_.get() && offset + buf_len > backend_->GetMaxFileSize()) {
if (net_log_.IsCapturing()) {
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_WRITE_END,
CreateNetLogReadWriteCompleteCallback(net::ERR_FAILED));
}
RecordWriteResult(cache_type_, SIMPLE_ENTRY_WRITE_RESULT_OVER_MAX_SIZE);
return net::ERR_FAILED;
}
ScopedOperationRunner operation_runner(this);
// Stream 0 data is kept in memory, so can be written immediatly if there are
// no IO operations pending.
if (stream_index == 0 && state_ == STATE_READY &&
pending_operations_.size() == 0)
return SetStream0Data(buf, offset, buf_len, truncate);
// We can only do optimistic Write if there is no pending operations, so
// that we are sure that the next call to RunNextOperationIfNeeded will
// actually run the write operation that sets the stream size. It also
// prevents from previous possibly-conflicting writes that could be stacked
// in the |pending_operations_|. We could optimize this for when we have
// only read operations enqueued, but past studies have shown that that such
// parallelizable cases are very rare.
const bool optimistic =
(use_optimistic_operations_ && state_ == STATE_READY &&
pending_operations_.size() == 0);
CompletionOnceCallback op_callback;
scoped_refptr<net::IOBuffer> op_buf;
int ret_value = net::ERR_FAILED;
if (!optimistic) {
op_buf = buf;
op_callback = std::move(callback);
ret_value = net::ERR_IO_PENDING;
} else {
// TODO(morlovich,pasko): For performance, don't use a copy of an IOBuffer
// here to avoid paying the price of the RefCountedThreadSafe atomic
// operations.
if (buf) {
op_buf = base::MakeRefCounted<IOBuffer>(buf_len);
SbMemoryCopy(op_buf->data(), buf->data(), buf_len);
}
op_callback = CompletionOnceCallback();
ret_value = buf_len;
if (net_log_.IsCapturing()) {
net_log_.AddEvent(
net::NetLogEventType::SIMPLE_CACHE_ENTRY_WRITE_OPTIMISTIC,
CreateNetLogReadWriteCompleteCallback(buf_len));
}
}
pending_operations_.push(SimpleEntryOperation::WriteOperation(
this, stream_index, offset, buf_len, op_buf.get(), truncate, optimistic,
std::move(op_callback)));
return ret_value;
}
int SimpleEntryImpl::ReadSparseData(int64_t offset,
net::IOBuffer* buf,
int buf_len,
CompletionOnceCallback callback) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_READ_SPARSE_CALL,
CreateNetLogSparseOperationCallback(offset, buf_len));
}
ScopedOperationRunner operation_runner(this);
pending_operations_.push(SimpleEntryOperation::ReadSparseOperation(
this, offset, buf_len, buf, std::move(callback)));
return net::ERR_IO_PENDING;
}
int SimpleEntryImpl::WriteSparseData(int64_t offset,
net::IOBuffer* buf,
int buf_len,
CompletionOnceCallback callback) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(
net::NetLogEventType::SIMPLE_CACHE_ENTRY_WRITE_SPARSE_CALL,
CreateNetLogSparseOperationCallback(offset, buf_len));
}
ScopedOperationRunner operation_runner(this);
pending_operations_.push(SimpleEntryOperation::WriteSparseOperation(
this, offset, buf_len, buf, std::move(callback)));
return net::ERR_IO_PENDING;
}
int SimpleEntryImpl::GetAvailableRange(int64_t offset,
int len,
int64_t* start,
CompletionOnceCallback callback) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
ScopedOperationRunner operation_runner(this);
pending_operations_.push(SimpleEntryOperation::GetAvailableRangeOperation(
this, offset, len, start, std::move(callback)));
return net::ERR_IO_PENDING;
}
bool SimpleEntryImpl::CouldBeSparse() const {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
// TODO(morlovich): Actually check.
return true;
}
void SimpleEntryImpl::CancelSparseIO() {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
// The Simple Cache does not return distinct objects for the same non-doomed
// entry, so there's no need to coordinate which object is performing sparse
// I/O. Therefore, CancelSparseIO and ReadyForSparseIO succeed instantly.
}
net::Error SimpleEntryImpl::ReadyForSparseIO(CompletionOnceCallback callback) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
// The simple Cache does not return distinct objects for the same non-doomed
// entry, so there's no need to coordinate which object is performing sparse
// I/O. Therefore, CancelSparseIO and ReadyForSparseIO succeed instantly.
return net::OK;
}
void SimpleEntryImpl::SetLastUsedTimeForTest(base::Time time) {
// Note that we do not update |last_used_| here as it gets overwritten
// by UpdateDataFromEntryStat with the current time. It would be more involved
// to make that value stick and is not needed by the current tests.
backend_->index()->SetLastUsedTimeForTest(entry_hash_, time);
}
size_t SimpleEntryImpl::EstimateMemoryUsage() const {
// TODO(xunjieli): crbug.com/669108. It'd be nice to have the rest of |entry|
// measured, but the ownership of SimpleSynchronousEntry isn't straightforward
return sizeof(SimpleSynchronousEntry) +
base::trace_event::EstimateMemoryUsage(pending_operations_) +
(stream_0_data_ ? stream_0_data_->capacity() : 0) +
(stream_1_prefetch_data_ ? stream_1_prefetch_data_->capacity() : 0);
}
void SimpleEntryImpl::SetPriority(uint32_t entry_priority) {
entry_priority_ = entry_priority;
}
SimpleEntryImpl::~SimpleEntryImpl() {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
DCHECK_EQ(0U, pending_operations_.size());
// STATE_IO_PENDING is possible here in one corner case: the entry had
// dispatched the final Close() operation to SimpleSynchronousEntry, and the
// only thing keeping |this| alive were the callbacks for that
// PostTaskAndReply. If at that point the message loop is shut down, all
// outstanding tasks get destroyed, dropping the last reference without
// CloseOperationComplete ever getting to run to exit from IO_PENDING.
DCHECK(state_ == STATE_UNINITIALIZED || state_ == STATE_FAILURE ||
state_ == STATE_IO_PENDING);
DCHECK(!synchronous_entry_);
net_log_.EndEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY);
}
void SimpleEntryImpl::PostClientCallback(net::CompletionOnceCallback callback,
int result) {
if (callback.is_null())
return;
// Note that the callback is posted rather than directly invoked to avoid
// reentrancy issues.
base::ThreadTaskRunnerHandle::Get()->PostTask(
FROM_HERE, base::BindOnce(&InvokeCallbackIfBackendIsAlive, backend_,
std::move(callback), result));
}
void SimpleEntryImpl::ResetEntry() {
// If we're doomed, we can't really do anything else with the entry, since
// we no longer own the name and are disconnected from the active entry table.
// We preserve doom_state_ accross this entry for this same reason.
state_ = doom_state_ == DOOM_COMPLETED ? STATE_FAILURE : STATE_UNINITIALIZED;
std::memset(crc32s_end_offset_, 0, sizeof(crc32s_end_offset_));
std::memset(crc32s_, 0, sizeof(crc32s_));
std::memset(have_written_, 0, sizeof(have_written_));
std::memset(data_size_, 0, sizeof(data_size_));
for (size_t i = 0; i < arraysize(crc_check_state_); ++i) {
crc_check_state_[i] = CRC_CHECK_NEVER_READ_AT_ALL;
}
}
void SimpleEntryImpl::ReturnEntryToCaller(Entry** out_entry) {
DCHECK(out_entry);
++open_count_;
AddRef(); // Balanced in Close()
if (!backend_.get()) {
// This method can be called when an asynchronous operation completed.
// If the backend no longer exists, the callback won't be invoked, and so we
// must close ourselves to avoid leaking. As well, there's no guarantee the
// client-provided pointer (|out_entry|) hasn't been freed, and no point
// dereferencing it, either.
Close();
return;
}
*out_entry = this;
}
void SimpleEntryImpl::MarkAsDoomed(DoomState new_state) {
DCHECK_NE(DOOM_NONE, new_state);
doom_state_ = new_state;
if (!backend_.get())
return;
backend_->index()->Remove(entry_hash_);
active_entry_proxy_.reset();
}
void SimpleEntryImpl::RunNextOperationIfNeeded() {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
SIMPLE_CACHE_UMA(CUSTOM_COUNTS,
"EntryOperationsPending", cache_type_,
pending_operations_.size(), 0, 100, 20);
if (!pending_operations_.empty() && state_ != STATE_IO_PENDING) {
SimpleEntryOperation operation = std::move(pending_operations_.front());
pending_operations_.pop();
switch (operation.type()) {
case SimpleEntryOperation::TYPE_OPEN:
OpenEntryInternal(operation.have_index(), operation.ReleaseCallback(),
operation.out_entry());
break;
case SimpleEntryOperation::TYPE_CREATE:
CreateEntryInternal(operation.have_index(), operation.ReleaseCallback(),
operation.out_entry());
break;
case SimpleEntryOperation::TYPE_CLOSE:
CloseInternal();
break;
case SimpleEntryOperation::TYPE_READ:
ReadDataInternal(/* sync_possible= */ false, operation.index(),
operation.offset(), operation.buf(),
operation.length(), operation.ReleaseCallback());
break;
case SimpleEntryOperation::TYPE_WRITE:
WriteDataInternal(operation.index(), operation.offset(),
operation.buf(), operation.length(),
operation.ReleaseCallback(), operation.truncate());
break;
case SimpleEntryOperation::TYPE_READ_SPARSE:
ReadSparseDataInternal(operation.sparse_offset(), operation.buf(),
operation.length(), operation.ReleaseCallback());
break;
case SimpleEntryOperation::TYPE_WRITE_SPARSE:
WriteSparseDataInternal(operation.sparse_offset(), operation.buf(),
operation.length(),
operation.ReleaseCallback());
break;
case SimpleEntryOperation::TYPE_GET_AVAILABLE_RANGE:
GetAvailableRangeInternal(operation.sparse_offset(), operation.length(),
operation.out_start(),
operation.ReleaseCallback());
break;
case SimpleEntryOperation::TYPE_DOOM:
DoomEntryInternal(operation.ReleaseCallback());
break;
default:
NOTREACHED();
}
// |this| may have been deleted.
}
}
void SimpleEntryImpl::OpenEntryInternal(bool have_index,
net::CompletionOnceCallback callback,
Entry** out_entry) {
ScopedOperationRunner operation_runner(this);
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_OPEN_BEGIN);
if (state_ == STATE_READY) {
ReturnEntryToCaller(out_entry);
PostClientCallback(std::move(callback), net::OK);
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_OPEN_END,
CreateNetLogSimpleEntryCreationCallback(this, net::OK));
return;
}
if (state_ == STATE_FAILURE) {
PostClientCallback(std::move(callback), net::ERR_FAILED);
net_log_.AddEvent(
net::NetLogEventType::SIMPLE_CACHE_ENTRY_OPEN_END,
CreateNetLogSimpleEntryCreationCallback(this, net::ERR_FAILED));
return;
}
DCHECK_EQ(STATE_UNINITIALIZED, state_);
DCHECK(!synchronous_entry_);
state_ = STATE_IO_PENDING;
const base::TimeTicks start_time = base::TimeTicks::Now();
std::unique_ptr<SimpleEntryCreationResults> results(
new SimpleEntryCreationResults(SimpleEntryStat(
last_used_, last_modified_, data_size_, sparse_data_size_)));
base::OnceClosure task = base::BindOnce(
&SimpleSynchronousEntry::OpenEntry, cache_type_, path_, key_, entry_hash_,
have_index, start_time, file_tracker_, results.get());
base::OnceClosure reply = base::BindOnce(
&SimpleEntryImpl::CreationOperationComplete, this, std::move(callback),
start_time, base::Passed(&results), out_entry,
net::NetLogEventType::SIMPLE_CACHE_ENTRY_OPEN_END);
prioritized_task_runner_->PostTaskAndReply(FROM_HERE, std::move(task),
std::move(reply), entry_priority_);
}
void SimpleEntryImpl::CreateEntryInternal(bool have_index,
net::CompletionOnceCallback callback,
Entry** out_entry) {
ScopedOperationRunner operation_runner(this);
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_CREATE_BEGIN);
if (state_ != STATE_UNINITIALIZED) {
// There is already an active normal entry.
net_log_.AddEvent(
net::NetLogEventType::SIMPLE_CACHE_ENTRY_CREATE_END,
CreateNetLogSimpleEntryCreationCallback(this, net::ERR_FAILED));
PostClientCallback(std::move(callback), net::ERR_FAILED);
return;
}
DCHECK_EQ(STATE_UNINITIALIZED, state_);
DCHECK(!synchronous_entry_);
state_ = STATE_IO_PENDING;
// Since we don't know the correct values for |last_used_| and
// |last_modified_| yet, we make this approximation.
last_used_ = last_modified_ = base::Time::Now();
// If creation succeeds, we should mark all streams to be saved on close.
for (int i = 0; i < kSimpleEntryStreamCount; ++i)
have_written_[i] = true;
const base::TimeTicks start_time = base::TimeTicks::Now();
std::unique_ptr<SimpleEntryCreationResults> results(
new SimpleEntryCreationResults(SimpleEntryStat(
last_used_, last_modified_, data_size_, sparse_data_size_)));
OnceClosure task = base::BindOnce(
&SimpleSynchronousEntry::CreateEntry, cache_type_, path_, key_,
entry_hash_, have_index, start_time, file_tracker_, results.get());
OnceClosure reply = base::BindOnce(
&SimpleEntryImpl::CreationOperationComplete, this, std::move(callback),
start_time, base::Passed(&results), out_entry,
net::NetLogEventType::SIMPLE_CACHE_ENTRY_CREATE_END);
prioritized_task_runner_->PostTaskAndReply(FROM_HERE, std::move(task),
std::move(reply), entry_priority_);
}
void SimpleEntryImpl::CloseInternal() {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
typedef SimpleSynchronousEntry::CRCRecord CRCRecord;
std::unique_ptr<std::vector<CRCRecord>> crc32s_to_write(
new std::vector<CRCRecord>());
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_CLOSE_BEGIN);
if (state_ == STATE_READY) {
DCHECK(synchronous_entry_);
state_ = STATE_IO_PENDING;
for (int i = 0; i < kSimpleEntryStreamCount; ++i) {
if (have_written_[i]) {
if (GetDataSize(i) == crc32s_end_offset_[i]) {
int32_t crc = GetDataSize(i) == 0 ? crc32(0, Z_NULL, 0) : crc32s_[i];
crc32s_to_write->push_back(CRCRecord(i, true, crc));
} else {
crc32s_to_write->push_back(CRCRecord(i, false, 0));
}
}
}
} else {
DCHECK(STATE_UNINITIALIZED == state_ || STATE_FAILURE == state_);
}
if (synchronous_entry_) {
Closure task = base::Bind(
&SimpleSynchronousEntry::Close, base::Unretained(synchronous_entry_),
SimpleEntryStat(last_used_, last_modified_, data_size_,
sparse_data_size_),
base::Passed(&crc32s_to_write), base::RetainedRef(stream_0_data_));
Closure reply = base::Bind(&SimpleEntryImpl::CloseOperationComplete, this);
synchronous_entry_ = NULL;
prioritized_task_runner_->PostTaskAndReply(
FROM_HERE, std::move(task), std::move(reply), entry_priority_);
for (int i = 0; i < kSimpleEntryStreamCount; ++i) {
if (!have_written_[i]) {
SIMPLE_CACHE_UMA(ENUMERATION,
"CheckCRCResult", cache_type_,
crc_check_state_[i], CRC_CHECK_MAX);
}
}
} else {
CloseOperationComplete();
}
}
int SimpleEntryImpl::ReadDataInternal(bool sync_possible,
int stream_index,
int offset,
net::IOBuffer* buf,
int buf_len,
net::CompletionOnceCallback callback) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
ScopedOperationRunner operation_runner(this);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_READ_BEGIN,
CreateNetLogReadWriteDataCallback(stream_index, offset,
buf_len, false));
}
if (state_ == STATE_FAILURE || state_ == STATE_UNINITIALIZED) {
RecordReadResult(cache_type_, READ_RESULT_BAD_STATE);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_READ_END,
CreateNetLogReadWriteCompleteCallback(net::ERR_FAILED));
}
// Note that the API states that client-provided callbacks for entry-level
// (i.e. non-backend) operations (e.g. read, write) are invoked even if
// the backend was already destroyed.
return PostToCallbackIfNeeded(sync_possible, std::move(callback),
net::ERR_FAILED);
}
DCHECK_EQ(STATE_READY, state_);
if (offset >= GetDataSize(stream_index) || offset < 0 || !buf_len) {
RecordReadResult(cache_type_, sync_possible
? READ_RESULT_NONBLOCK_EMPTY_RETURN
: READ_RESULT_FAST_EMPTY_RETURN);
// If there is nothing to read, we bail out before setting state_ to
// STATE_IO_PENDING (so ScopedOperationRunner might start us on next op
// here).
return PostToCallbackIfNeeded(sync_possible, std::move(callback), 0);
}
buf_len = std::min(buf_len, GetDataSize(stream_index) - offset);
// Since stream 0 data is kept in memory, it is read immediately.
if (stream_index == 0) {
int rv = ReadFromBuffer(stream_0_data_.get(), offset, buf_len, buf);
return PostToCallbackIfNeeded(sync_possible, std::move(callback), rv);
}
// Sometimes we can read in-ram prefetched stream 1 data immediately, too.
if (stream_index == 1) {
if (is_initial_stream1_read_) {
SIMPLE_CACHE_UMA(BOOLEAN, "ReadStream1FromPrefetched", cache_type_,
stream_1_prefetch_data_ != nullptr);
}
is_initial_stream1_read_ = false;
if (stream_1_prefetch_data_) {
int rv =
ReadFromBuffer(stream_1_prefetch_data_.get(), offset, buf_len, buf);
return PostToCallbackIfNeeded(sync_possible, std::move(callback), rv);
}
}
state_ = STATE_IO_PENDING;
if (doom_state_ == DOOM_NONE && backend_.get())
backend_->index()->UseIfExists(entry_hash_);
SimpleSynchronousEntry::ReadRequest read_req(stream_index, offset, buf_len);
// Figure out if we should be computing the checksum for this read,
// and whether we should be verifying it, too.
if (crc32s_end_offset_[stream_index] == offset) {
read_req.request_update_crc = true;
read_req.previous_crc32 =
offset == 0 ? crc32(0, Z_NULL, 0) : crc32s_[stream_index];
// We can't verify the checksum if we already overwrote part of the file.
// (It may still make sense to compute it if the overwritten area and the
// about-to-read-in area are adjoint).
read_req.request_verify_crc = !have_written_[stream_index];
}
std::unique_ptr<SimpleSynchronousEntry::ReadResult> result =
std::make_unique<SimpleSynchronousEntry::ReadResult>();
std::unique_ptr<SimpleEntryStat> entry_stat(new SimpleEntryStat(
last_used_, last_modified_, data_size_, sparse_data_size_));
OnceClosure task = base::BindOnce(
&SimpleSynchronousEntry::ReadData, base::Unretained(synchronous_entry_),
read_req, entry_stat.get(), base::RetainedRef(buf), result.get());
OnceClosure reply = base::BindOnce(
&SimpleEntryImpl::ReadOperationComplete, this, stream_index, offset,
std::move(callback), base::Passed(&entry_stat), base::Passed(&result));
prioritized_task_runner_->PostTaskAndReply(FROM_HERE, std::move(task),
std::move(reply), entry_priority_);
return net::ERR_IO_PENDING;
}
void SimpleEntryImpl::WriteDataInternal(int stream_index,
int offset,
net::IOBuffer* buf,
int buf_len,
net::CompletionOnceCallback callback,
bool truncate) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
ScopedOperationRunner operation_runner(this);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_WRITE_BEGIN,
CreateNetLogReadWriteDataCallback(stream_index, offset,
buf_len, truncate));
}
if (state_ == STATE_FAILURE || state_ == STATE_UNINITIALIZED) {
RecordWriteResult(cache_type_, SIMPLE_ENTRY_WRITE_RESULT_BAD_STATE);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_WRITE_END,
CreateNetLogReadWriteCompleteCallback(net::ERR_FAILED));
}
if (!callback.is_null()) {
base::ThreadTaskRunnerHandle::Get()->PostTask(
FROM_HERE, base::BindOnce(std::move(callback), net::ERR_FAILED));
}
// |this| may be destroyed after return here.
return;
}
DCHECK_EQ(STATE_READY, state_);
// Since stream 0 data is kept in memory, it will be written immediatly.
if (stream_index == 0) {
int ret_value = SetStream0Data(buf, offset, buf_len, truncate);
if (!callback.is_null()) {
base::ThreadTaskRunnerHandle::Get()->PostTask(
FROM_HERE, base::BindOnce(std::move(callback), ret_value));
}
return;
}
// Ignore zero-length writes that do not change the file size.
if (buf_len == 0) {
int32_t data_size = data_size_[stream_index];
if (truncate ? (offset == data_size) : (offset <= data_size)) {
RecordWriteResult(cache_type_,
SIMPLE_ENTRY_WRITE_RESULT_FAST_EMPTY_RETURN);
if (!callback.is_null()) {
base::ThreadTaskRunnerHandle::Get()->PostTask(
FROM_HERE, base::BindOnce(std::move(callback), 0));
}
return;
}
}
state_ = STATE_IO_PENDING;
if (doom_state_ == DOOM_NONE && backend_.get())
backend_->index()->UseIfExists(entry_hash_);
// Any stream 1 write invalidates the prefetched data.
if (stream_index == 1)
stream_1_prefetch_data_ = nullptr;
bool request_update_crc = false;
uint32_t initial_crc = 0;
if (offset < crc32s_end_offset_[stream_index]) {
// If a range for which the crc32 was already computed is rewritten, the
// computation of the crc32 need to start from 0 again.
crc32s_end_offset_[stream_index] = 0;
}
if (crc32s_end_offset_[stream_index] == offset) {
request_update_crc = true;
initial_crc = (offset != 0) ? crc32s_[stream_index] : crc32(0, Z_NULL, 0);
}
// |entry_stat| needs to be initialized before modifying |data_size_|.
std::unique_ptr<SimpleEntryStat> entry_stat(new SimpleEntryStat(
last_used_, last_modified_, data_size_, sparse_data_size_));
if (truncate) {
data_size_[stream_index] = offset + buf_len;
} else {
data_size_[stream_index] = std::max(offset + buf_len,
GetDataSize(stream_index));
}
std::unique_ptr<SimpleSynchronousEntry::WriteResult> write_result =
std::make_unique<SimpleSynchronousEntry::WriteResult>();
// Since we don't know the correct values for |last_used_| and
// |last_modified_| yet, we make this approximation.
last_used_ = last_modified_ = base::Time::Now();
have_written_[stream_index] = true;
// Writing on stream 1 affects the placement of stream 0 in the file, the EOF
// record will have to be rewritten.
if (stream_index == 1)
have_written_[0] = true;
// Retain a reference to |buf| in |reply| instead of |task|, so that we can
// reduce cross thread malloc/free pairs. The cross thread malloc/free pair
// increases the apparent memory usage due to the thread cached free list.
// TODO(morlovich): Remove the doom_state_ argument to WriteData, since with
// renaming rather than delete, creating a new stream 2 of doomed entry will
// just work.
OnceClosure task = base::BindOnce(
&SimpleSynchronousEntry::WriteData, base::Unretained(synchronous_entry_),
SimpleSynchronousEntry::WriteRequest(
stream_index, offset, buf_len, initial_crc, truncate,
doom_state_ != DOOM_NONE, request_update_crc),
base::Unretained(buf), entry_stat.get(), write_result.get());
OnceClosure reply = base::BindOnce(
&SimpleEntryImpl::WriteOperationComplete, this, stream_index,
std::move(callback), base::Passed(&entry_stat),
base::Passed(&write_result), base::RetainedRef(buf));
prioritized_task_runner_->PostTaskAndReply(FROM_HERE, std::move(task),
std::move(reply), entry_priority_);
}
void SimpleEntryImpl::ReadSparseDataInternal(
int64_t sparse_offset,
net::IOBuffer* buf,
int buf_len,
net::CompletionOnceCallback callback) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
ScopedOperationRunner operation_runner(this);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(
net::NetLogEventType::SIMPLE_CACHE_ENTRY_READ_SPARSE_BEGIN,
CreateNetLogSparseOperationCallback(sparse_offset, buf_len));
}
if (state_ == STATE_FAILURE || state_ == STATE_UNINITIALIZED) {
if (net_log_.IsCapturing()) {
net_log_.AddEvent(
net::NetLogEventType::SIMPLE_CACHE_ENTRY_READ_SPARSE_END,
CreateNetLogReadWriteCompleteCallback(net::ERR_FAILED));
}
if (!callback.is_null()) {
base::ThreadTaskRunnerHandle::Get()->PostTask(
FROM_HERE, base::BindOnce(std::move(callback), net::ERR_FAILED));
}
// |this| may be destroyed after return here.
return;
}
DCHECK_EQ(STATE_READY, state_);
state_ = STATE_IO_PENDING;
std::unique_ptr<int> result(new int());
std::unique_ptr<base::Time> last_used(new base::Time());
OnceClosure task = base::BindOnce(
&SimpleSynchronousEntry::ReadSparseData,
base::Unretained(synchronous_entry_),
SimpleSynchronousEntry::SparseRequest(sparse_offset, buf_len),
base::RetainedRef(buf), last_used.get(), result.get());
OnceClosure reply = base::BindOnce(
&SimpleEntryImpl::ReadSparseOperationComplete, this, std::move(callback),
base::Passed(&last_used), base::Passed(&result));
prioritized_task_runner_->PostTaskAndReply(FROM_HERE, std::move(task),
std::move(reply), entry_priority_);
}
void SimpleEntryImpl::WriteSparseDataInternal(
int64_t sparse_offset,
net::IOBuffer* buf,
int buf_len,
net::CompletionOnceCallback callback) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
ScopedOperationRunner operation_runner(this);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(
net::NetLogEventType::SIMPLE_CACHE_ENTRY_WRITE_SPARSE_BEGIN,
CreateNetLogSparseOperationCallback(sparse_offset, buf_len));
}
if (state_ == STATE_FAILURE || state_ == STATE_UNINITIALIZED) {
if (net_log_.IsCapturing()) {
net_log_.AddEvent(
net::NetLogEventType::SIMPLE_CACHE_ENTRY_WRITE_SPARSE_END,
CreateNetLogReadWriteCompleteCallback(net::ERR_FAILED));
}
if (!callback.is_null()) {
base::ThreadTaskRunnerHandle::Get()->PostTask(
FROM_HERE, base::BindOnce(std::move(callback), net::ERR_FAILED));
}
// |this| may be destroyed after return here.
return;
}
DCHECK_EQ(STATE_READY, state_);
state_ = STATE_IO_PENDING;
uint64_t max_sparse_data_size = std::numeric_limits<int64_t>::max();
if (backend_.get()) {
uint64_t max_cache_size = backend_->index()->max_size();
max_sparse_data_size = max_cache_size / kMaxSparseDataSizeDivisor;
}
std::unique_ptr<SimpleEntryStat> entry_stat(new SimpleEntryStat(
last_used_, last_modified_, data_size_, sparse_data_size_));
last_used_ = last_modified_ = base::Time::Now();
std::unique_ptr<int> result(new int());
OnceClosure task = base::BindOnce(
&SimpleSynchronousEntry::WriteSparseData,
base::Unretained(synchronous_entry_),
SimpleSynchronousEntry::SparseRequest(sparse_offset, buf_len),
base::RetainedRef(buf), max_sparse_data_size, entry_stat.get(),
result.get());
OnceClosure reply = base::BindOnce(
&SimpleEntryImpl::WriteSparseOperationComplete, this, std::move(callback),
base::Passed(&entry_stat), base::Passed(&result));
prioritized_task_runner_->PostTaskAndReply(FROM_HERE, std::move(task),
std::move(reply), entry_priority_);
}
void SimpleEntryImpl::GetAvailableRangeInternal(
int64_t sparse_offset,
int len,
int64_t* out_start,
net::CompletionOnceCallback callback) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
ScopedOperationRunner operation_runner(this);
if (state_ == STATE_FAILURE || state_ == STATE_UNINITIALIZED) {
if (!callback.is_null()) {
base::ThreadTaskRunnerHandle::Get()->PostTask(
FROM_HERE, base::BindOnce(std::move(callback), net::ERR_FAILED));
}
// |this| may be destroyed after return here.
return;
}
DCHECK_EQ(STATE_READY, state_);
state_ = STATE_IO_PENDING;
std::unique_ptr<int> result(new int());
OnceClosure task =
base::BindOnce(&SimpleSynchronousEntry::GetAvailableRange,
base::Unretained(synchronous_entry_),
SimpleSynchronousEntry::SparseRequest(sparse_offset, len),
out_start, result.get());
OnceClosure reply =
base::BindOnce(&SimpleEntryImpl::GetAvailableRangeOperationComplete, this,
std::move(callback), base::Passed(&result));
prioritized_task_runner_->PostTaskAndReply(FROM_HERE, std::move(task),
std::move(reply), entry_priority_);
}
void SimpleEntryImpl::DoomEntryInternal(net::CompletionOnceCallback callback) {
if (doom_state_ == DOOM_COMPLETED) {
// During the time we were sitting on a queue, some operation failed
// and cleaned our files up, so we don't have to do anything.
DoomOperationComplete(std::move(callback), state_, net::OK);
return;
}
if (!backend_) {
// If there's no backend, we want to truncate the files rather than delete
// or rename them. Either op will update the entry directory's mtime, which
// will likely force a full index rebuild on the next startup; this is
// clearly an undesirable cost. Instead, the lesser evil is to set the entry
// files to length zero, leaving the invalid entry in the index. On the next
// attempt to open the entry, it will fail asynchronously (since the magic
// numbers will not be found), and the files will actually be removed.
// Since there is no backend, new entries to conflict with us also can't be
// created.
prioritized_task_runner_->PostTaskAndReplyWithResult(
FROM_HERE,
base::BindOnce(&SimpleSynchronousEntry::TruncateEntryFiles, path_,
entry_hash_),
base::BindOnce(&SimpleEntryImpl::DoomOperationComplete, this,
std::move(callback),
// Return to STATE_FAILURE after dooming, since no
// operation can succeed on the truncated entry files.
STATE_FAILURE),
entry_priority_);
state_ = STATE_IO_PENDING;
return;
}
if (synchronous_entry_) {
// If there is a backing object, we have to go through its instance methods,
// so that it can rename itself and keep track of the altenative name.
prioritized_task_runner_->PostTaskAndReplyWithResult(
FROM_HERE,
base::BindOnce(&SimpleSynchronousEntry::Doom,
base::Unretained(synchronous_entry_)),
base::BindOnce(&SimpleEntryImpl::DoomOperationComplete, this,
std::move(callback), state_),
entry_priority_);
} else {
DCHECK_EQ(STATE_UNINITIALIZED, state_);
// If nothing is open, we can just delete the files. We know they have the
// base names, since if we ever renamed them our doom_state_ would be
// DOOM_COMPLETED, and we would exit at function entry.
prioritized_task_runner_->PostTaskAndReplyWithResult(
FROM_HERE,
base::BindOnce(&SimpleSynchronousEntry::DeleteEntryFiles, path_,
cache_type_, entry_hash_),
base::BindOnce(&SimpleEntryImpl::DoomOperationComplete, this,
std::move(callback), state_),
entry_priority_);
}
state_ = STATE_IO_PENDING;
}
void SimpleEntryImpl::CreationOperationComplete(
net::CompletionOnceCallback completion_callback,
const base::TimeTicks& start_time,
std::unique_ptr<SimpleEntryCreationResults> in_results,
Entry** out_entry,
net::NetLogEventType end_event_type) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
DCHECK_EQ(state_, STATE_IO_PENDING);
DCHECK(in_results);
ScopedOperationRunner operation_runner(this);
SIMPLE_CACHE_UMA(BOOLEAN,
"EntryCreationResult", cache_type_,
in_results->result == net::OK);
if (in_results->result != net::OK) {
if (in_results->result != net::ERR_FILE_EXISTS) {
// Here we keep index up-to-date, but don't remove ourselves from active
// entries since we may have queued operations, and it would be
// problematic to run further Creates, Opens, or Dooms if we are not
// the active entry. We can only do this because OpenEntryInternal
// and CreateEntryInternal have to start from STATE_UNINITIALIZED, so
// nothing else is going on which may be confused.
if (backend_)
backend_->index()->Remove(entry_hash_);
}
net_log_.AddEventWithNetErrorCode(end_event_type, net::ERR_FAILED);
PostClientCallback(std::move(completion_callback), net::ERR_FAILED);
ResetEntry();
return;
}
// Make sure to keep the index up-to-date. We likely already did this when
// CreateEntry was called, but it's possible we were sitting on a queue
// after an op that removed us.
if (backend_ && doom_state_ == DOOM_NONE)
backend_->index()->Insert(entry_hash_);
// If out_entry is NULL, it means we already called ReturnEntryToCaller from
// the optimistic Create case.
if (out_entry)
ReturnEntryToCaller(out_entry);
state_ = STATE_READY;
synchronous_entry_ = in_results->sync_entry;
// Copy over any pre-fetched data and its CRCs.
for (int stream = 0; stream < 2; ++stream) {
const SimpleStreamPrefetchData& prefetched =
in_results->stream_prefetch_data[stream];
if (prefetched.data.get()) {
if (stream == 0)
stream_0_data_ = prefetched.data;
else
stream_1_prefetch_data_ = prefetched.data;
// The crc was read in SimpleSynchronousEntry.
crc_check_state_[stream] = CRC_CHECK_DONE;
crc32s_[stream] = prefetched.stream_crc32;
crc32s_end_offset_[stream] = in_results->entry_stat.data_size(stream);
}
}
// If this entry was opened by hash, key_ could still be empty. If so, update
// it with the key read from the synchronous entry.
if (key_.empty()) {
SetKey(synchronous_entry_->key());
} else {
// This should only be triggered when creating an entry. In the open case
// the key is either copied from the arguments to open, or checked
// in the synchronous entry.
DCHECK_EQ(key_, synchronous_entry_->key());
}
UpdateDataFromEntryStat(in_results->entry_stat);
SIMPLE_CACHE_UMA(TIMES,
"EntryCreationTime", cache_type_,
(base::TimeTicks::Now() - start_time));
AdjustOpenEntryCountBy(cache_type_, 1);
net_log_.AddEvent(end_event_type);
PostClientCallback(std::move(completion_callback), net::OK);
}
void SimpleEntryImpl::EntryOperationComplete(
net::CompletionOnceCallback completion_callback,
const SimpleEntryStat& entry_stat,
int result) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
DCHECK(synchronous_entry_);
DCHECK_EQ(STATE_IO_PENDING, state_);
if (result < 0) {
state_ = STATE_FAILURE;
MarkAsDoomed(DOOM_COMPLETED);
} else {
state_ = STATE_READY;
UpdateDataFromEntryStat(entry_stat);
}
if (!completion_callback.is_null()) {
base::ThreadTaskRunnerHandle::Get()->PostTask(
FROM_HERE, base::BindOnce(std::move(completion_callback), result));
}
RunNextOperationIfNeeded();
}
void SimpleEntryImpl::ReadOperationComplete(
int stream_index,
int offset,
net::CompletionOnceCallback completion_callback,
std::unique_ptr<SimpleEntryStat> entry_stat,
std::unique_ptr<SimpleSynchronousEntry::ReadResult> read_result) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
DCHECK(synchronous_entry_);
DCHECK_EQ(STATE_IO_PENDING, state_);
DCHECK(read_result);
int result = read_result->result;
if (result > 0 &&
crc_check_state_[stream_index] == CRC_CHECK_NEVER_READ_AT_ALL) {
crc_check_state_[stream_index] = CRC_CHECK_NEVER_READ_TO_END;
}
if (read_result->crc_updated) {
if (result > 0) {
DCHECK_EQ(crc32s_end_offset_[stream_index], offset);
crc32s_end_offset_[stream_index] += result;
crc32s_[stream_index] = read_result->updated_crc32;
}
if (read_result->crc_performed_verify)
crc_check_state_[stream_index] = CRC_CHECK_DONE;
}
if (result < 0) {
crc32s_end_offset_[stream_index] = 0;
} else {
if (crc_check_state_[stream_index] == CRC_CHECK_NEVER_READ_TO_END &&
offset + result == GetDataSize(stream_index)) {
crc_check_state_[stream_index] = CRC_CHECK_NOT_DONE;
}
}
RecordReadResultConsideringChecksum(read_result);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_READ_END,
CreateNetLogReadWriteCompleteCallback(result));
}
EntryOperationComplete(std::move(completion_callback), *entry_stat, result);
}
void SimpleEntryImpl::WriteOperationComplete(
int stream_index,
net::CompletionOnceCallback completion_callback,
std::unique_ptr<SimpleEntryStat> entry_stat,
std::unique_ptr<SimpleSynchronousEntry::WriteResult> write_result,
net::IOBuffer* buf) {
int result = write_result->result;
if (result >= 0)
RecordWriteResult(cache_type_, SIMPLE_ENTRY_WRITE_RESULT_SUCCESS);
else
RecordWriteResult(cache_type_,
SIMPLE_ENTRY_WRITE_RESULT_SYNC_WRITE_FAILURE);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_WRITE_END,
CreateNetLogReadWriteCompleteCallback(result));
}
if (result < 0)
crc32s_end_offset_[stream_index] = 0;
if (result > 0 && write_result->crc_updated) {
crc32s_end_offset_[stream_index] += result;
crc32s_[stream_index] = write_result->updated_crc32;
}
EntryOperationComplete(std::move(completion_callback), *entry_stat, result);
}
void SimpleEntryImpl::ReadSparseOperationComplete(
net::CompletionOnceCallback completion_callback,
std::unique_ptr<base::Time> last_used,
std::unique_ptr<int> result) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
DCHECK(synchronous_entry_);
DCHECK(result);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_READ_SPARSE_END,
CreateNetLogReadWriteCompleteCallback(*result));
}
SimpleEntryStat entry_stat(*last_used, last_modified_, data_size_,
sparse_data_size_);
EntryOperationComplete(std::move(completion_callback), entry_stat, *result);
}
void SimpleEntryImpl::WriteSparseOperationComplete(
net::CompletionOnceCallback completion_callback,
std::unique_ptr<SimpleEntryStat> entry_stat,
std::unique_ptr<int> result) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
DCHECK(synchronous_entry_);
DCHECK(result);
if (net_log_.IsCapturing()) {
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_WRITE_SPARSE_END,
CreateNetLogReadWriteCompleteCallback(*result));
}
EntryOperationComplete(std::move(completion_callback), *entry_stat, *result);
}
void SimpleEntryImpl::GetAvailableRangeOperationComplete(
net::CompletionOnceCallback completion_callback,
std::unique_ptr<int> result) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
DCHECK(synchronous_entry_);
DCHECK(result);
SimpleEntryStat entry_stat(last_used_, last_modified_, data_size_,
sparse_data_size_);
EntryOperationComplete(std::move(completion_callback), entry_stat, *result);
}
void SimpleEntryImpl::DoomOperationComplete(
net::CompletionOnceCallback callback,
State state_to_restore,
int result) {
state_ = state_to_restore;
doom_state_ = DOOM_COMPLETED;
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_DOOM_END);
PostClientCallback(std::move(callback), result);
RunNextOperationIfNeeded();
if (backend_)
backend_->OnDoomComplete(entry_hash_);
}
void SimpleEntryImpl::RecordReadResultConsideringChecksum(
const std::unique_ptr<SimpleSynchronousEntry::ReadResult>& read_result)
const {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
DCHECK(synchronous_entry_);
DCHECK_EQ(STATE_IO_PENDING, state_);
if (read_result->result >= 0) {
RecordReadResult(cache_type_, READ_RESULT_SUCCESS);
} else {
if (read_result->crc_updated && read_result->crc_performed_verify &&
!read_result->crc_verify_ok)
RecordReadResult(cache_type_, READ_RESULT_SYNC_CHECKSUM_FAILURE);
else
RecordReadResult(cache_type_, READ_RESULT_SYNC_READ_FAILURE);
}
}
void SimpleEntryImpl::CloseOperationComplete() {
DCHECK(!synchronous_entry_);
DCHECK_EQ(0, open_count_);
DCHECK(STATE_IO_PENDING == state_ || STATE_FAILURE == state_ ||
STATE_UNINITIALIZED == state_);
net_log_.AddEvent(net::NetLogEventType::SIMPLE_CACHE_ENTRY_CLOSE_END);
AdjustOpenEntryCountBy(cache_type_, -1);
ResetEntry();
RunNextOperationIfNeeded();
}
void SimpleEntryImpl::UpdateDataFromEntryStat(
const SimpleEntryStat& entry_stat) {
DCHECK_CALLED_ON_VALID_THREAD(io_thread_checker_);
DCHECK(synchronous_entry_);
DCHECK_EQ(STATE_READY, state_);
last_used_ = entry_stat.last_used();
last_modified_ = entry_stat.last_modified();
for (int i = 0; i < kSimpleEntryStreamCount; ++i) {
data_size_[i] = entry_stat.data_size(i);
}
sparse_data_size_ = entry_stat.sparse_data_size();
if (doom_state_ == DOOM_NONE && backend_.get()) {
backend_->index()->UpdateEntrySize(
entry_hash_, base::checked_cast<uint32_t>(GetDiskUsage()));
}
}
int64_t SimpleEntryImpl::GetDiskUsage() const {
int64_t file_size = 0;
for (int i = 0; i < kSimpleEntryStreamCount; ++i) {
file_size +=
simple_util::GetFileSizeFromDataSize(key_.size(), data_size_[i]);
}
file_size += sparse_data_size_;
return file_size;
}
int SimpleEntryImpl::ReadFromBuffer(net::GrowableIOBuffer* in_buf,
int offset,
int buf_len,
net::IOBuffer* out_buf) {
DCHECK_GE(buf_len, 0);
SbMemoryCopy(out_buf->data(), in_buf->data() + offset, buf_len);
UpdateDataFromEntryStat(SimpleEntryStat(base::Time::Now(), last_modified_,
data_size_, sparse_data_size_));
RecordReadResult(cache_type_, READ_RESULT_SUCCESS);
return buf_len;
}
int SimpleEntryImpl::SetStream0Data(net::IOBuffer* buf,
int offset,
int buf_len,
bool truncate) {
// Currently, stream 0 is only used for HTTP headers, and always writes them
// with a single, truncating write. Detect these writes and record the size
// changes of the headers. Also, support writes to stream 0 that have
// different access patterns, as required by the API contract.
// All other clients of the Simple Cache are encouraged to use stream 1.
have_written_[0] = true;
int data_size = GetDataSize(0);
if (offset == 0 && truncate) {
stream_0_data_->SetCapacity(buf_len);
SbMemoryCopy(stream_0_data_->data(), buf->data(), buf_len);
data_size_[0] = buf_len;
} else {
const int buffer_size =
truncate ? offset + buf_len : std::max(offset + buf_len, data_size);
stream_0_data_->SetCapacity(buffer_size);
// If |stream_0_data_| was extended, the extension until offset needs to be
// zero-filled.
const int fill_size = offset <= data_size ? 0 : offset - data_size;
if (fill_size > 0)
SbMemorySet(stream_0_data_->data() + data_size, 0, fill_size);
if (buf)
SbMemoryCopy(stream_0_data_->data() + offset, buf->data(), buf_len);
data_size_[0] = buffer_size;
}
RecordHeaderSize(cache_type_, data_size_[0]);
base::Time modification_time = base::Time::Now();
// Reset checksum; SimpleSynchronousEntry::Close will compute it for us,
// and do it off the I/O thread.
crc32s_end_offset_[0] = 0;
UpdateDataFromEntryStat(
SimpleEntryStat(modification_time, modification_time, data_size_,
sparse_data_size_));
RecordWriteResult(cache_type_, SIMPLE_ENTRY_WRITE_RESULT_SUCCESS);
return buf_len;
}
} // namespace disk_cache