blob: 6edbc9092773c7afa2cfbef2400bf214c46e5d98 [file] [log] [blame]
/*
*
* Copyright 2015, Google Inc.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are
* met:
*
* * Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following disclaimer
* in the documentation and/or other materials provided with the
* distribution.
* * Neither the name of Google Inc. nor the names of its
* contributors may be used to endorse or promote products derived from
* this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
*/
#include <grpc++/server_context.h>
#include <mutex>
#include <grpc++/completion_queue.h>
#include <grpc++/impl/call.h>
#include <grpc++/support/time.h>
#include <grpc/compression.h>
#include <grpc/grpc.h>
#include <grpc/support/alloc.h>
#include <grpc/support/log.h>
#include "src/core/lib/surface/call.h"
namespace grpc {
// CompletionOp
class ServerContext::CompletionOp final : public CallOpSetInterface {
public:
// initial refs: one in the server context, one in the cq
CompletionOp()
: has_tag_(false),
tag_(nullptr),
refs_(2),
finalized_(false),
cancelled_(0) {}
void FillOps(grpc_op* ops, size_t* nops) override;
bool FinalizeResult(void** tag, bool* status) override;
bool CheckCancelled(CompletionQueue* cq) {
cq->TryPluck(this);
return CheckCancelledNoPluck();
}
bool CheckCancelledAsync() { return CheckCancelledNoPluck(); }
void set_tag(void* tag) {
has_tag_ = true;
tag_ = tag;
}
void Unref();
private:
bool CheckCancelledNoPluck() {
std::lock_guard<std::mutex> g(mu_);
return finalized_ ? (cancelled_ != 0) : false;
}
bool has_tag_;
void* tag_;
std::mutex mu_;
int refs_;
bool finalized_;
int cancelled_;
};
void ServerContext::CompletionOp::Unref() {
std::unique_lock<std::mutex> lock(mu_);
if (--refs_ == 0) {
lock.unlock();
delete this;
}
}
void ServerContext::CompletionOp::FillOps(grpc_op* ops, size_t* nops) {
ops->op = GRPC_OP_RECV_CLOSE_ON_SERVER;
ops->data.recv_close_on_server.cancelled = &cancelled_;
ops->flags = 0;
ops->reserved = NULL;
*nops = 1;
}
bool ServerContext::CompletionOp::FinalizeResult(void** tag, bool* status) {
std::unique_lock<std::mutex> lock(mu_);
finalized_ = true;
bool ret = false;
if (has_tag_) {
*tag = tag_;
ret = true;
}
if (!*status) cancelled_ = 1;
if (--refs_ == 0) {
lock.unlock();
delete this;
}
return ret;
}
// ServerContext body
ServerContext::ServerContext()
: completion_op_(nullptr),
has_notify_when_done_tag_(false),
async_notify_when_done_tag_(nullptr),
deadline_(gpr_inf_future(GPR_CLOCK_REALTIME)),
call_(nullptr),
cq_(nullptr),
sent_initial_metadata_(false),
compression_level_set_(false) {}
ServerContext::ServerContext(gpr_timespec deadline, grpc_metadata* metadata,
size_t metadata_count)
: completion_op_(nullptr),
has_notify_when_done_tag_(false),
async_notify_when_done_tag_(nullptr),
deadline_(deadline),
call_(nullptr),
cq_(nullptr),
sent_initial_metadata_(false),
compression_level_set_(false) {
for (size_t i = 0; i < metadata_count; i++) {
client_metadata_.map()->insert(
std::pair<grpc::string_ref, grpc::string_ref>(
StringRefFromSlice(&metadata[i].key),
StringRefFromSlice(&metadata[i].value)));
}
}
ServerContext::~ServerContext() {
if (call_) {
grpc_call_destroy(call_);
}
if (completion_op_) {
completion_op_->Unref();
}
}
void ServerContext::BeginCompletionOp(Call* call) {
GPR_ASSERT(!completion_op_);
completion_op_ = new CompletionOp();
if (has_notify_when_done_tag_) {
completion_op_->set_tag(async_notify_when_done_tag_);
}
call->PerformOps(completion_op_);
}
void ServerContext::AddInitialMetadata(const grpc::string& key,
const grpc::string& value) {
initial_metadata_.insert(std::make_pair(key, value));
}
void ServerContext::AddTrailingMetadata(const grpc::string& key,
const grpc::string& value) {
trailing_metadata_.insert(std::make_pair(key, value));
}
void ServerContext::TryCancel() const {
grpc_call_error err = grpc_call_cancel_with_status(
call_, GRPC_STATUS_CANCELLED, "Cancelled on the server side", NULL);
if (err != GRPC_CALL_OK) {
gpr_log(GPR_ERROR, "TryCancel failed with: %d", err);
}
}
bool ServerContext::IsCancelled() const {
if (has_notify_when_done_tag_) {
// when using async API, but the result is only valid
// if the tag has already been delivered at the completion queue
return completion_op_ && completion_op_->CheckCancelledAsync();
} else {
// when using sync API
return completion_op_ && completion_op_->CheckCancelled(cq_);
}
}
void ServerContext::set_compression_algorithm(
grpc_compression_algorithm algorithm) {
char* algorithm_name = NULL;
if (!grpc_compression_algorithm_name(algorithm, &algorithm_name)) {
gpr_log(GPR_ERROR, "Name for compression algorithm '%d' unknown.",
algorithm);
abort();
}
GPR_ASSERT(algorithm_name != NULL);
AddInitialMetadata(GRPC_COMPRESSION_REQUEST_ALGORITHM_MD_KEY, algorithm_name);
}
grpc::string ServerContext::peer() const {
grpc::string peer;
if (call_) {
char* c_peer = grpc_call_get_peer(call_);
peer = c_peer;
gpr_free(c_peer);
}
return peer;
}
const struct census_context* ServerContext::census_context() const {
return grpc_census_call_get_context(call_);
}
void ServerContext::SetLoadReportingCosts(
const std::vector<grpc::string>& cost_data) {
if (call_ == nullptr) return;
grpc_load_reporting_cost_context* cost_ctx =
static_cast<grpc_load_reporting_cost_context*>(
gpr_malloc(sizeof(*cost_ctx)));
cost_ctx->values_count = cost_data.size();
cost_ctx->values = static_cast<grpc_slice*>(
gpr_malloc(sizeof(*cost_ctx->values) * cost_ctx->values_count));
for (size_t i = 0; i < cost_ctx->values_count; ++i) {
cost_ctx->values[i] =
grpc_slice_from_copied_buffer(cost_data[i].data(), cost_data[i].size());
}
grpc_call_set_load_reporting_cost_context(call_, cost_ctx);
}
} // namespace grpc