Files
Jungfraujoch/receiver/JFJochReceiverService.cpp
T
leonarski_fandClaude Fable 5 77e53ac714 Fix exception handling in acquisition failure paths
Follow-ups to the Idle-vs-Error acquisition failure split:

- JFJochServices::Stop: raise the critical detector fault BEFORE the ordinary
  receiver/writer exception. Both can be set at once (detector not idle, then
  receiver->Stop() also throws); throwing the ordinary one first masked the
  detector fault and returned the broker to Idle instead of Error, skipping the
  required re-initialisation.
- SLSDetectorWrapper::Stop/Deactivate: rethrow a JFJochException as-is (as Start
  already does) so InternalStop's JFJochCriticalException is not downgraded to an
  ordinary JFJochException by the generic catch.
- JFJochServices::Start: wrap the whole best-effort receiver cleanup (Cancel +
  Stop) in the catch so a throwing Cancel cannot replace the original detector
  exception before it is re-raised.

Also fix the "cannot create indexing pool" failure seen when retrying
initialize after a failed acquisition start: a failed start leaves the receiver
(and its GPU resources) alive until the next Start, but the retried initialize
rebuilds the indexer pool first, so the fresh GPU indexer had to coexist with
the stale receiver and its init failed. JFJochReceiverService::Indexing now
releases the previous receiver before (re)building the pool (safe: only runs
when the receiver is idle), which also removes a dangling pool pointer held by
that receiver.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-08 14:36:25 +02:00

341 lines
14 KiB
C++

// SPDX-FileCopyrightText: 2024 Filip Leonarski, Paul Scherrer Institute <filip.leonarski@psi.ch>
// SPDX-License-Identifier: GPL-3.0-only
#include "JFJochReceiverService.h"
#include "JFJochReceiverFPGA.h"
#include "JFJochReceiverLite.h"
#include "../preview/JFJochJPEG.h"
#include "../preview/JFJochTIFF.h"
JFJochReceiverService::JFJochReceiverService(AcquisitionDeviceGroup &in_aq_devices,
Logger &in_logger, ImagePusher &pusher,
size_t send_buffer_size_MiB)
: aq_devices(in_aq_devices),
logger(in_logger),
image_buffer(send_buffer_size_MiB * 1024 * 1024),
image_pusher(pusher),
spot_finding_settings(DiffractionExperiment::DefaultDataProcessingSettings()) {
}
JFJochReceiverService &JFJochReceiverService::NumThreads(int64_t input) {
if (input <= 0)
throw JFJochException(JFJochExceptionCategory::InputParameterInvalid, "Thread number must be above zero");
nthreads = input;
return *this;
}
void JFJochReceiverService::FinalizeMeasurementChangeState() {
std::unique_lock ul(state_mutex);
state = ReceiverState::Idle;
measurement_done.notify_all();
}
void JFJochReceiverService::FinalizeMeasurement() {
try {
receiver->StopReceiver();
} catch (...) {
FinalizeMeasurementChangeState();
throw;
}
FinalizeMeasurementChangeState();
}
std::optional<JFJochReceiverStatus> JFJochReceiverService::GetStatus() {
return receiver_status.GetStatus();
}
void JFJochReceiverService::Start(const DiffractionExperiment &experiment,
const PixelMask &pixel_mask,
const JFCalibration *calibration,
std::shared_ptr<ImagePuller> puller) {
std::unique_lock ul_state(state_mutex); // unique lock, as it will destroy and create receiver object
if (state != ReceiverState::Idle)
throw JFJochException(JFJochExceptionCategory::WrongDAQState, "Receiver not idle, cannot start");
try {
auto nthreads_local = nthreads;
if (experiment.IsCPUSummation())
nthreads_local = 4;
// First clean-up old measurement
receiver.reset();
preview_image.Configure(experiment, pixel_mask);
switch (experiment.GetDetectorType()) {
case DetectorType::EIGER:
case DetectorType::JUNGFRAU:
receiver = std::make_unique<JFJochReceiverFPGA>(experiment, pixel_mask,
calibration,
aq_devices, image_pusher,
logger,
nthreads_local,
spot_finding_settings,
receiver_status,
plots,
image_buffer,
zmq_preview_socket.get(),
zmq_metadata_socket.get(),
indexer_thread_pool.get());
break;
case DetectorType::DECTRIS:
if (puller)
image_puller = puller;
else {
image_puller = std::make_shared<ZMQImagePuller>(
experiment.GetDetectorSetup().GetDECTRISStream2Addr());
}
receiver = std::make_unique<JFJochReceiverLite>(experiment,
pixel_mask,
*image_puller,
image_pusher,
logger,
nthreads_local,
spot_finding_settings,
receiver_status,
plots,
image_buffer,
zmq_preview_socket.get(),
zmq_metadata_socket.get(),
indexer_thread_pool.get());
break;
}
measurement = std::async(std::launch::async, &JFJochReceiverService::FinalizeMeasurement, this);
state = ReceiverState::Running;
} catch (const JFJochException &e) {
logger.ErrorException(e);
throw;
}
}
void JFJochReceiverService::Cancel(bool silent) {
std::unique_lock ul(state_mutex);
if (state == ReceiverState::Running)
receiver->Cancel(silent);
}
JFJochReceiverOutput JFJochReceiverService::Stop() {
std::unique_lock ul(state_mutex);
measurement_done.wait(ul, [this] { return (state != ReceiverState::Running); });
if (state != ReceiverState::Idle)
throw JFJochException(JFJochExceptionCategory::WrongReceiverState, "Receiver in weird state");
try {
if (measurement.valid())
measurement.get();
} catch (JFJochException &e) {
logger.ErrorException(e);
throw;
}
if (!receiver) {
logger.Warning("Request to stop while receiver not running");
throw JFJochException(JFJochExceptionCategory::WrongReceiverState, "Receiver idle, cannot stop");
}
return receiver->GetFinalStatistics();
}
void JFJochReceiverService::SetSpotFindingSettings(const SpotFindingSettings &settings) {
try {
std::unique_lock ul(state_mutex);
DiffractionExperiment::CheckDataProcessingSettings(settings);
spot_finding_settings = settings;
if (state != ReceiverState::Idle)
receiver->SetSpotFindingSettings(settings);
} catch (std::exception &e) {
logger.ErrorException(e);
throw;
}
}
MultiLinePlot JFJochReceiverService::GetDataProcessingPlot(const PlotRequest &request) {
return plots.GetPlots(request);
}
void JFJochReceiverService::GetPlotRaw(std::vector<float> &v, PlotType type, const std::string &roi) {
plots.GetPlotRaw(v, type, roi);
}
std::vector<AcquisitionDeviceNetConfig> JFJochReceiverService::GetNetworkConfig() {
return aq_devices.GetNetworkConfig();
}
void JFJochReceiverService::LoadInternalGeneratorImage(const DiffractionExperiment &experiment,
const std::vector<uint16_t> &image,
uint64_t image_number) {
std::vector<uint16_t> raw_geom, eiger_geom;
const uint16_t *frame;
if (image.size() == RAW_MODULE_SIZE * experiment.GetModulesNum()) {
frame = image.data();
} else if (image.size() == experiment.GetPixelsNum()) {
raw_geom.resize(RAW_MODULE_SIZE * experiment.GetModulesNum());
ConvertedToRawGeometry(experiment, raw_geom.data(), image.data());
frame = raw_geom.data();
} else
throw JFJochException(JFJochExceptionCategory::InputParameterInvalid,
"Size of input array with raw expected image is wrong");
for (int i = 0; i < experiment.GetDataStreamsNum(); i++) {
uint32_t module0 = experiment.GetFirstModuleOfDataStream(i);
switch (experiment.GetDetectorSetup().GetDetectorType()) {
case DetectorType::EIGER:
eiger_geom.resize(RAW_MODULE_SIZE);
for (int m = 0; m < experiment.GetModulesNum(i); m++) {
RawToEigerInput(eiger_geom.data(), frame + (module0 + m) * RAW_MODULE_SIZE);
aq_devices[i].SetInternalGeneratorFrame(eiger_geom.data(),
m + experiment.GetModulesNum(i) * image_number);
}
break;
case DetectorType::JUNGFRAU:
for (int m = 0; m < experiment.GetModulesNum(i); m++)
aq_devices[i].SetInternalGeneratorFrame(frame + (module0 + m) * RAW_MODULE_SIZE,
m + experiment.GetModulesNum(i) * image_number);
break;
default:
throw JFJochException(JFJochExceptionCategory::InputParameterInvalid, "Detector not supported");
}
}
}
void JFJochReceiverService::GetXFELEventCode(std::vector<uint64_t> &v) const {
plots.GetXFELEventCode(v);
}
void JFJochReceiverService::GetXFELPulseID(std::vector<uint64_t> &v) const {
plots.GetXFELPulseID(v);
}
std::vector<DeviceStatus> JFJochReceiverService::GetDeviceStatus() const {
return aq_devices.GetDeviceStatus();
}
std::optional<float> JFJochReceiverService::GetProgress() const {
return receiver_status.GetProgress();
}
JFJochReceiverService &JFJochReceiverService::PreviewSocket(const std::string &addr, const std::optional<int32_t> &watermark) {
if (!addr.empty()) {
logger.Info("ZeroMQ preview socket available at {}", addr);
zmq_preview_socket = std::make_unique<ZMQPreviewSocket>(addr, watermark);
}
return *this;
}
JFJochReceiverService &JFJochReceiverService::MetadataSocket(const std::string &addr) {
if (!addr.empty()) {
logger.Info("ZeroMQ metadata socket available at {}", addr);
zmq_metadata_socket = std::make_unique<ZMQMetadataSocket>(addr);
}
return *this;
}
std::string JFJochReceiverService::GetPreviewSocketAddress() const {
if (zmq_preview_socket)
return zmq_preview_socket->GetAddress();
return "";
}
std::string JFJochReceiverService::GetMetadataSocketAddress() const {
if (zmq_metadata_socket)
return zmq_metadata_socket->GetAddress();
return "";
}
JFJochReceiverService &JFJochReceiverService::PreviewSocketSettings(const ZMQPreviewSettings &input) {
if (zmq_preview_socket)
zmq_preview_socket->ImportSettings(input);
return *this;
}
JFJochReceiverService &JFJochReceiverService::MetadataSocketSettings(const ZMQMetadataSettings &input) {
if (zmq_metadata_socket)
zmq_metadata_socket->ImportSettings(input);
return *this;
}
ZMQPreviewSettings JFJochReceiverService::GetPreviewSocketSettings() const {
if (zmq_preview_socket)
return zmq_preview_socket->GetSettings();
return {};
}
ZMQMetadataSettings JFJochReceiverService::GetMetadataSocketSettings() const {
if (zmq_metadata_socket)
return zmq_metadata_socket->GetSettings();
return {};
}
void JFJochReceiverService::GetStartMessageFromBuffer(std::vector<uint8_t> &v) {
image_buffer.GetStartMessage(v);
}
bool JFJochReceiverService::GetImageFromBuffer(std::vector<uint8_t> &v, int64_t image_number) {
return image_buffer.GetImage(v, image_number);
}
std::string JFJochReceiverService::GetJPEGFromBuffer(const PreviewImageSettings &settings, int64_t image_number) {
std::vector<uint8_t> cbor_image;
if (!image_buffer.GetImage(cbor_image, image_number))
return {};
return preview_image.GenerateImage(settings, cbor_image);
}
std::string JFJochReceiverService::GetTIFFFromBuffer(int64_t image_number) {
std::vector<uint8_t> cbor_image;
if (!image_buffer.GetImage(cbor_image, image_number))
return {};
return PreviewImage::GenerateTIFF(cbor_image);
}
ImageBufferStatus JFJochReceiverService::GetImageBufferStatus() const {
return image_buffer.GetStatus();
}
void JFJochReceiverService::ClearImageBuffer() {
std::unique_lock ul(state_mutex);
// Clearing image buffer during data collection could be catastrophic, so better protect here, even if redundant
// with JFJochStateMachine
if (state == ReceiverState::Idle)
image_buffer.Finalize(std::chrono::milliseconds(2500));
else
throw JFJochException(JFJochExceptionCategory::WrongDAQState,
"Cannot clear image buffer during data collection");
}
JFJochReceiverService &JFJochReceiverService::Indexing(const IndexingSettings &input) {
std::unique_lock ul(state_mutex);
// Clearing image buffer during data collection could be catastrophic, so better protect here, even if redundant
// with JFJochStateMachine
if (state == ReceiverState::Idle) {
// Release the previous run's receiver first. It holds a raw pointer to the indexer pool and
// its own GPU resources; keeping it alive while we rebuild the pool means a fresh GPU indexer
// has to coexist with a stale receiver (e.g. after a failed acquisition start, where the
// receiver is stopped but not destroyed until the next Start), which can make the GPU indexer
// initialisation fail. Destroying the receiver before the pool also avoids the dangling
// pointer. Safe here: state is Idle, so no measurement is using it.
receiver.reset();
logger.Info("Resetting indexing thread pool");
indexer_thread_pool.reset();
if (input.GetAlgorithm() != IndexingAlgorithmEnum::None) {
logger.Info("Creating indexing thread pool...");
indexer_thread_pool = std::make_unique<IndexerThreadPool>(input);
logger.Info(" ... done");
}
return *this;
} else
throw JFJochException(JFJochExceptionCategory::WrongDAQState,
"Cannot change indexing settings during data collection");
}
ImagePusherStatus JFJochReceiverService::GetImagePusherStatus() const {
return image_pusher.GetStatus();
}