HttpCache::Transaction layer allowing parallel validation

net/http/http_cache_transaction.cc

 // Copyright (c) 2012 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/http/http_cache_transaction.h"
 
 #include "build/build_config.h"  // For OS_POSIX
 
 #if defined(OS_POSIX)
 #include <unistd.h>
@@ skipping 151 matching lines @@
 
 HttpCache::Transaction::Transaction(RequestPriority priority, HttpCache* cache)
     : next_state_(STATE_NONE),
       request_(NULL),
       priority_(priority),
       cache_(cache->GetWeakPtr()),
       entry_(NULL),
       new_entry_(NULL),
       new_response_(NULL),
       mode_(NONE),
+      original_mode_(NONE),
       reading_(false),
       invalid_range_(false),
       truncated_(false),
       is_sparse_(false),
       range_requested_(false),
       handling_206_(false),
       cache_pending_(false),
       done_reading_(false),
       vary_mismatch_(false),
       couldnt_conditionalize_request_(false),
       bypass_lock_for_test_(false),
       fail_conditionalization_for_test_(false),
+      validating_cannot_proceed_(false),
       io_buf_len_(0),
       read_offset_(0),
       effective_load_flags_(0),
       write_len_(0),
       cache_entry_status_(CacheEntryStatus::ENTRY_UNDEFINED),
       validation_cause_(VALIDATION_CAUSE_UNDEFINED),
       total_received_bytes_(0),
       total_sent_bytes_(0),
       websocket_handshake_stream_base_create_helper_(NULL),
       weak_factory_(this) {
@@ skipping 376 matching lines @@
     ConnectionAttempts* out) const {
   ConnectionAttempts new_connection_attempts;
   if (network_trans_)
     network_trans_->GetConnectionAttempts(&new_connection_attempts);
 
   out->swap(new_connection_attempts);
   out->insert(out->begin(), old_connection_attempts_.begin(),
               old_connection_attempts_.end());
 }
 
+void HttpCache::Transaction::SetValidatingCannotProceed() {
+  validating_cannot_proceed_ = true;
+  entry_ = nullptr;
+}
+
 size_t HttpCache::Transaction::EstimateMemoryUsage() const {
   // TODO(xunjieli): Consider improving the coverage. crbug.com/669108.
   return 0;
 }
 
 //-----------------------------------------------------------------------------
 
 // A few common patterns: (Foo* means Foo -> FooComplete)
 //
 // 1. Not-cached entry:
 //   Start():
 //   GetBackend* -> InitEntry -> OpenEntry* -> CreateEntry* -> AddToEntry* ->
 //   SendRequest* -> SuccessfulSendRequest -> OverwriteCachedResponse ->
 //   CacheWriteResponse* -> TruncateCachedData* -> TruncateCachedMetadata* ->
 //   PartialHeadersReceived
 //
 //   Read():
 //   NetworkRead* -> CacheWriteData*
 //
 // 2. Cached entry, no validation:
 //   Start():
 //   GetBackend* -> InitEntry -> OpenEntry* -> AddToEntry* -> CacheReadResponse*
 //   -> CacheDispatchValidation -> BeginPartialCacheValidation() ->
-//   BeginCacheValidation() -> SetupEntryForRead()
+//   BeginCacheValidation() -> SetupEntryForRead() -> WaitBeforeRead*
 //
 //   Read():
 //   CacheReadData*
 //
 // 3. Cached entry, validation (304):
 //   Start():
 //   GetBackend* -> InitEntry -> OpenEntry* -> AddToEntry* -> CacheReadResponse*
 //   -> CacheDispatchValidation -> BeginPartialCacheValidation() ->
 //   BeginCacheValidation() -> SendRequest* -> SuccessfulSendRequest ->
 //   UpdateCachedResponse -> CacheWriteUpdatedResponse* ->
 //   UpdateCachedResponseComplete -> OverwriteCachedResponse ->
-//   PartialHeadersReceived
+//   PartialHeadersReceived -> WaitBeforeRead*
 //
 //   Read():
 //   CacheReadData*
 //
 // 4. Cached entry, validation and replace (200):
 //   Start():
 //   GetBackend* -> InitEntry -> OpenEntry* -> AddToEntry* -> CacheReadResponse*
 //   -> CacheDispatchValidation -> BeginPartialCacheValidation() ->
 //   BeginCacheValidation() -> SendRequest* -> SuccessfulSendRequest ->
 //   OverwriteCachedResponse -> CacheWriteResponse* -> DoTruncateCachedData* ->
@@ skipping 79 matching lines @@
 //   BeginPartialCacheValidation() -> BeginCacheValidation() ->
 //   SetupEntryForRead()
 //
 //   Read():
 //   CacheReadData*
 //
 // 14. Cached entry more than 5 minutes old, unused_since_prefetch is true:
 //   Like examples 2-4, only CacheToggleUnusedSincePrefetch* is inserted between
 //   CacheReadResponse* and CacheDispatchValidation.
 int HttpCache::Transaction::DoLoop(int result) {
+  // If its the start state machine and it cannot proceed due to failure by
+  // another transaction in writing the response, restart this transaction.
+  if (!reading_ && validating_cannot_proceed_) {
+    validating_cannot_proceed_ = false;
+    RestartAfterValidationStarted();
+  }
+
+  int rv = DoLoopImpl(result);
+
+  // If its the end of the Start() state machine, the transaction might have to
+  // wait before it can read from the cache if there is another transaction
+  // writing to the cache currently. This is being done to enable parallelizing
+  // the validation phase of a transaction while another transaction is still
+  // writing the response to the cache.
+  // Checking reading_ will make sure that the control is still in the start
+  // state machine and also that we do not enter the wait state for
+  // partial requests that have already started reading and re-entered the
+  // start() state machine.
+  if (!reading_ && rv == OK) {
+    next_state_ = STATE_WAIT_BEFORE_READ;
+    rv = DoLoopImpl(rv);
+  }
+
+  if (rv != ERR_IO_PENDING && !callback_.is_null()) {
+    read_buf_ = nullptr;  // Release the buffer before invoking the callback.
+    base::ResetAndReturn(&callback_).Run(rv);
+  }
+
+  return rv;
+}
+
+int HttpCache::Transaction::DoLoopImpl(int result) {
   DCHECK(next_state_ != STATE_NONE);
 
   int rv = result;
   do {
     State state = next_state_;
     next_state_ = STATE_NONE;
     switch (state) {
       case STATE_GET_BACKEND:
         DCHECK_EQ(OK, rv);
         rv = DoGetBackend();
@@ skipping 119 matching lines @@
         DCHECK_EQ(OK, rv);
         rv = DoPartialHeadersReceived();
         break;
       case STATE_CACHE_READ_METADATA:
         DCHECK_EQ(OK, rv);
         rv = DoCacheReadMetadata();
         break;
       case STATE_CACHE_READ_METADATA_COMPLETE:
         rv = DoCacheReadMetadataComplete(rv);
         break;
+      case STATE_WAIT_BEFORE_READ:
+        DCHECK_EQ(OK, rv);
+        rv = DoWaitBeforeRead();
+        break;
+      case STATE_WAIT_BEFORE_READ_COMPLETE:
+        rv = DoWaitBeforeReadComplete(rv);
+        break;
       case STATE_NETWORK_READ:
         DCHECK_EQ(OK, rv);
         rv = DoNetworkRead();
         break;
       case STATE_NETWORK_READ_COMPLETE:
         rv = DoNetworkReadComplete(rv);
         break;
       case STATE_CACHE_READ_DATA:
         DCHECK_EQ(OK, rv);
         rv = DoCacheReadData();
@@ skipping 14 matching lines @@
       case STATE_CACHE_WRITE_TRUNCATED_RESPONSE_COMPLETE:
         rv = DoCacheWriteTruncatedResponseComplete(rv);
         break;
       default:
         NOTREACHED() << "bad state";
         rv = ERR_FAILED;
         break;
     }
   } while (rv != ERR_IO_PENDING && next_state_ != STATE_NONE);
 
-  if (rv != ERR_IO_PENDING && !callback_.is_null()) {
-    read_buf_ = NULL;  // Release the buffer before invoking the callback.
-    base::ResetAndReturn(&callback_).Run(rv);
-  }
-
   return rv;
 }
 
 int HttpCache::Transaction::DoGetBackend() {
   cache_pending_ = true;
   next_state_ = STATE_GET_BACKEND_COMPLETE;
   net_log_.BeginEvent(NetLogEventType::HTTP_CACHE_GET_BACKEND);
   return cache_->GetBackendForTransaction(this);
 }
 
@@ skipping 55 matching lines @@
       partial_->RestoreHeaders(&custom_request_->extra_headers);
       partial_.reset();
     }
     next_state_ = STATE_SEND_REQUEST;
   } else {
     next_state_ = STATE_INIT_ENTRY;
   }
 
   // This is only set if we have something to do with the response.
   range_requested_ = (partial_.get() != NULL);
-
+  original_mode_ = mode_;
   return OK;
 }
 
 int HttpCache::Transaction::DoInitEntry() {
   TRACE_EVENT0("io", "HttpCacheTransaction::DoInitEntry");
   DCHECK(!new_entry_);
 
   if (!cache_.get())
     return ERR_UNEXPECTED;
 
@@ skipping 511 matching lines @@
     UpdateCacheEntryStatus(CacheEntryStatus::ENTRY_NOT_IN_CACHE);
   }
 
   // Invalidate any cached GET with a successful PUT or DELETE.
   if (mode_ == WRITE &&
       (request_->method == "PUT" || request_->method == "DELETE")) {
     if (NonErrorResponse(new_response->headers->response_code())) {
       int ret = cache_->DoomEntry(cache_key_, NULL);
       DCHECK_EQ(OK, ret);
     }
-    cache_->DoneWritingToEntry(entry_, true);
+    cache_->DoneWritingToEntry(entry_, true, this);
     entry_ = NULL;
     mode_ = NONE;
   }
 
   // Invalidate any cached GET with a successful POST.
   if (!(effective_load_flags_ & LOAD_DISABLE_CACHE) &&
       request_->method == "POST" &&
       NonErrorResponse(new_response->headers->response_code())) {
     cache_->DoomMainEntryForUrl(request_->url);
   }
@@ skipping 12 matching lines @@
     if (new_response->headers->response_code() == 304 || handling_206_) {
       UpdateCacheEntryStatus(CacheEntryStatus::ENTRY_VALIDATED);
       next_state_ = STATE_UPDATE_CACHED_RESPONSE;
       return OK;
     }
     UpdateCacheEntryStatus(CacheEntryStatus::ENTRY_UPDATED);
     mode_ = WRITE;
   }
 
   next_state_ = STATE_OVERWRITE_CACHED_RESPONSE;
+
+  if (!entry_)
+    return OK;
+
+  // Invalidate any current entry with a successful response if this transaction
+  // cannot write to this entry.
+  if (new_response->headers->response_code() != 304 &&
+      (entry_->writer || !entry_->readers.empty())) {
+    DCHECK_EQ(entry_->headers_transaction, this);
+    cache_->DoneResponseHeaders(entry_, this, false);
+    entry_ = nullptr;
+    mode_ = NONE;
+    return OK;
+  }
+
   return OK;
 }
 
 // We received 304 or 206 and we want to update the cached response headers.
 int HttpCache::Transaction::DoUpdateCachedResponse() {
   TRACE_EVENT0("io", "HttpCacheTransaction::DoUpdateCachedResponse");
   next_state_ = STATE_UPDATE_CACHED_RESPONSE_COMPLETE;
   int rv = OK;
   // Update the cached response based on the headers and properties of
   // new_response_.
@@ skipping 49 matching lines @@
     DCHECK(!handling_206_);
     // We got a "not modified" response and already updated the corresponding
     // cache entry above.
     //
     // By closing the cached entry now, we make sure that the 304 rather than
     // the cached 200 response, is what will be returned to the user.
     DoneWritingToEntry(true);
   } else if (entry_ && !handling_206_) {
     DCHECK_EQ(READ_WRITE, mode_);
     if (!partial_ || partial_->IsLastRange()) {
-      cache_->ConvertWriterToReader(entry_);
       mode_ = READ;
     }
     // We no longer need the network transaction, so destroy it.
     ResetNetworkTransaction();
   } else if (entry_ && handling_206_ && truncated_ &&
              partial_->initial_validation()) {
     // We just finished the validation of a truncated entry, and the server
     // is willing to resume the operation. Now we go back and start serving
     // the first part to the user.
     ResetNetworkTransaction();
@@ skipping 116 matching lines @@
       next_state_ = STATE_CACHE_READ_DATA;
     }
   } else if (mode_ != NONE) {
     // We are about to return the headers for a byte-range request to the user,
     // so let's fix them.
     partial_->FixResponseHeaders(response_.headers.get(), true);
   }
   return OK;
 }
 
+int HttpCache::Transaction::DoWaitBeforeRead() {
+  if (!entry_)
+    return OK;
+
+  next_state_ = STATE_WAIT_BEFORE_READ_COMPLETE;
+
+  // If |this| is the writer or there is no response body to be written or read,
+  // it does not need to wait.
+  if (entry_->writer == this || request_->method == "HEAD")

[Randy Smith (Not in Mondays), 2017/03/17 18:13:06]

I'm a bit concerned about races here.  The transition to writer from
validating_transaction completes after a PostTask.  Can we be in a situation
where that happens after this test, and if so, what's the result?

[shivanisha, 2017/03/20 18:14:49]

Actually, after the state simplification since a transaction only becomes writer
after this function, this condition can be removed.
A related change:
(We just have to make sure that after completing DoWaitBeforeReadComplete is
executed, this function does not get executed again due to the DoLoop code. For
that, I am adding a boolean is_wait_before_read_complete_ and checking it in
DoLoop in addition to !reading_ before setting the state to
STATE_WAIT_BEFORE_READ)

+    return OK;
+
+  // If |this| is in readers, then no need to wait.

[Randy Smith (Not in Mondays), 2017/03/17 18:13:06]

This comment doesn't seem to fit the following lines?

[shivanisha, 2017/03/20 18:14:49]

Expanded the comment to be more clear. The if condition basically wants to
exclude transactions that are already in readers, which can happen to
transactions with read-only permissions.

+  if (entry_->headers_transaction == this)

[Randy Smith (Not in Mondays), 2017/03/17 18:13:05]

When does this happen?  Naively, I think of WAIT_BEFORE_READ as being the state
reflection in HttpCache::Transaction between the completion of being a
headers_transaction and the start of being a reader/writer, so we should always
be in the done_headers_queue at this point.  What am I missing?

[shivanisha, 2017/03/20 18:14:49]

At this point a transaction should always be a headers_transaction and will move
to done_headers_queue in DoneResponseHeaders(). The only exception is the
transactions that are read-only, they would be in readers at this point, thus
this condition.

+    return cache_->DoneResponseHeaders(entry_, this, true);
+
+  return OK;
+}
+
+int HttpCache::Transaction::DoWaitBeforeReadComplete(int rv) {
+  if (rv == ERR_CACHE_RACE) {
+    RestartAfterValidationStarted();
+    return OK;
+  }
+
+  if (rv != OK)
+    return rv;
+
+  // If successful then this must be either be the writer or a reader as the
+  // response must have completely been written to the cache.
+  if (entry_->writer == this)
+    return OK;
+
+  mode_ = READ;
+  if (network_trans_)
+    ResetNetworkTransaction();
+  return OK;
+}
+
+void HttpCache::Transaction::RestartAfterValidationStarted() {

[shivanisha, 2017/03/16 15:43:26]

DCHECK(!reading_);

[shivanisha, 2017/03/20 18:14:49]

done

+  next_state_ = STATE_INIT_ENTRY;
+  cache_entry_status_ = CacheEntryStatus::ENTRY_UNDEFINED;
+  entry_ = nullptr;
+  mode_ = original_mode_;
+  if (network_trans_)
+    network_trans_.reset();
+}
+
 int HttpCache::Transaction::DoCacheReadMetadata() {
   TRACE_EVENT0("io", "HttpCacheTransaction::DoCacheReadMetadata");
   DCHECK(entry_);
   DCHECK(!response_.metadata.get());
   next_state_ = STATE_CACHE_READ_METADATA_COMPLETE;
 
   response_.metadata =
       new IOBufferWithSize(entry_->disk_entry->GetDataSize(kMetadataIndex));
 
   net_log_.BeginEvent(NetLogEventType::HTTP_CACHE_READ_INFO);
@@ skipping 809 matching lines @@
   if (partial_) {
     if (truncated_ || is_sparse_ || !invalid_range_) {
       // We are going to return the saved response headers to the caller, so
       // we may need to adjust them first.
       next_state_ = STATE_PARTIAL_HEADERS_RECEIVED;
       return OK;
     } else {
       partial_.reset();
     }
   }
-  cache_->ConvertWriterToReader(entry_);
+
+  // This might or might not be able to add it as an active reader based on
+  // whether this is the active writer or not. If not, it will be added to a
+  // wait queue in DoWaitBeforeRead.
   mode_ = READ;
 
   if (request_->method == "HEAD")
     FixHeadersForHead();
 
   if (entry_->disk_entry->GetDataSize(kMetadataIndex))
     next_state_ = STATE_CACHE_READ_METADATA;
   return OK;
 }
 
@@ skipping 65 matching lines @@
   }
   return OK;
 }
 
 void HttpCache::Transaction::DoneWritingToEntry(bool success) {
   if (!entry_)
     return;
 
   RecordHistograms();
 
-  cache_->DoneWritingToEntry(entry_, success);
+  cache_->DoneWritingToEntry(entry_, success, this);
   entry_ = NULL;
   mode_ = NONE;  // switch to 'pass through' mode
 }
 
 int HttpCache::Transaction::OnCacheReadError(int result, bool restart) {
   DLOG(ERROR) << "ReadData failed: " << result;
   const int result_for_histogram = std::max(0, -result);
   if (restart) {
     UMA_HISTOGRAM_SPARSE_SLOWLY("HttpCache.ReadErrorRestartable",
                                 result_for_histogram);
@@ skipping 335 matching lines @@
     default:
       NOTREACHED();
   }
 }
 
 void HttpCache::Transaction::OnIOComplete(int result) {
   DoLoop(result);
 }
 
 }  // namespace net