Files
Jungfraujoch/tests/StreamWriterTest.cpp
T
leonarski_fandFilip Leonarski 19b135186d
Build Packages / Create release (push) Successful in 7s
Build Packages / build:portable:macos-arm64:nocuda (push) Successful in 4m56s
Build Packages / build:rugnux:linux-aarch64:cuda (push) Successful in 9m10s
Build Packages / build:portable:linux-x86_64:nocuda (push) Successful in 13m51s
Build Packages / build:portable:linux-x86_64:cuda (push) Successful in 16m56s
Build Packages / build:jfjoch:rocky8:nocuda (push) Successful in 20m59s
Build Packages / build:portable:windows-x86_64:nocuda (push) Successful in 21m25s
Build Packages / build:jfjoch:rocky9:nocuda (push) Successful in 21m17s
Build Packages / build:jfjoch:ubuntu2204:nocuda (push) Successful in 14m14s
Build Packages / HDF5 consumer tests (DIALS, XDS) (push) Successful in 25m59s
Build Packages / build:portable:windows-x86_64:cuda (push) Successful in 25m48s
Build Packages / build:jfjoch:ubuntu2404:nocuda (push) Successful in 15m52s
Build Packages / build:jfjoch:rocky8:cuda-sls9 (push) Successful in 19m33s
Build Packages / Generate python client (push) Successful in 29s
Build Packages / Build documentation (push) Successful in 1m19s
Build Packages / build:jfjoch:ubuntu2204:cuda (push) Successful in 20m49s
Build Packages / build:jfjoch:rocky9:cuda-sls9 (push) Successful in 25m31s
Build Packages / build:jfjoch:rocky8:cuda (push) Successful in 25m24s
Build Packages / build:jfjoch:rocky9:cuda (push) Successful in 26m5s
Build Packages / build:jfjoch:ubuntu2404:cuda (push) Successful in 22m19s
Build Packages / Unit tests (push) Failing after 2h28m24s
Build Packages / Validation Battery (push) Successful in 40m33s
v1.0.0-rc.175 (#85)
* Rugnux: Processes a collection live while jfjoch_broker writes it (`rugnux <master.h5> <broker URL>`).
* Rugnux: Processing starts on the first images and confirms its choices once all are in, so analysis can begin during collection.
* Rugnux: Faster processing, especially on servers with several GPUs and CPU sockets and on computers without a GPU (`--no-gpu`, macOS).
* Rugnux: Uses much less GPU memory, so large data sets stay on the GPU of smaller cards, and a data set too large for the card is processed on the CPU instead.
* Rugnux: On macOS, Bragg integration runs on the Apple GPU.
* Rugnux: Better handling of twinning and pseudo-symmetry, with the twin hypothesis written as `<prefix>_twin_<SG>.mtz`, and a wrong `--model` no longer confirms the data or decides the hand.
* Rugnux: `--model` accepts predicted (AlphaFold) models and falls back to molecular replacement when the model is not placed as a rigid body (experimental `--mr` forces it), settling space group, hand and twinning and writing a different answer as `<prefix>_mr.mtz`.
* Rugnux: Molecular replacement is faster and settles the space group in more cases with several copies in the asymmetric unit.
* Rugnux: Deposition-ready mmCIF, a richer report, and numbered steps on the terminal with the full log in `<prefix>_log.txt`; `_P1.mtz`, `--estimate-beam-center` and `--beam-center-search` are removed.
* Rugnux: The beam-stop shadow is found correctly on tilted detectors.
* Rugnux: Stills files that record a stationary goniometer axis get the default stills geometry refinement.
* jfjoch_viewer: "Analyze dataset" uses the same defaults as a bare `rugnux <file>`, and shows the results of processed CBF, marCCD and SMV sweeps.
* jfjoch_viewer: The Files panel refreshes on F5 and every few seconds, and the plot type and other selections stay put during live collection.
* jfjoch_writer: Memory stays bounded and old data files are closed efficiently on long runs, any failure ends the run in an error state the broker is told about, and the master file appears only after its data files; the ZeroMQ pusher logs when a stalled writer makes it drop images.
* Fixes: The TCP image pusher no longer blocks the broker's receiver, GPU out-of-memory failures stop with a clear message, and the XDS plugin masks module and chip edges.

---------

Co-authored-by: Filip Leonarski <jungfrau@filips-mac-mini.home>
Reviewed-on: #85
Co-authored-by: Filip Leonarski <filip.leonarski@psi.ch>
2026-10-16 15:34:43 +02:00

299 lines
13 KiB
C++

// SPDX-FileCopyrightText: 2024 Filip Leonarski, Paul Scherrer Institute <filip.leonarski@psi.ch>
// SPDX-License-Identifier: GPL-3.0-only
#include <catch2/catch_all.hpp>
#include <filesystem>
#include <future>
#include <thread>
#include <sys/stat.h>
#include "../writer/StreamWriter.h"
#include "../image_pusher/ZMQStream2Pusher.h"
#include "../receiver/JFJochReceiverService.h"
#include "../image_pusher/ZMQWriterNotificationPuller.h"
#include "../image_puller/TestImagePuller.h"
TEST_CASE("StreamWriterTest_ZMQ", "[StreamWriter]") {
RegisterHDF5Filter();
Logger logger("StreamWriterTest_ZMQ");
DiffractionExperiment x(DetJF(2));
x.FilePrefix("subdir/StreamWriterTest").NumTriggers(1).ImagesPerTrigger(5)
.UseInternalPacketGenerator(true).Raw().PedestalG0Frames(0).OverwriteExistingFiles(true);
PixelMask pixel_mask(x);
JFModuleGainCalibration gain;
AcquisitionDeviceGroup aq_devices;
for (int i = 0; i < x.GetDataStreamsNum(); i++)
aq_devices.AddHLSDevice(64);
ZMQStream2Pusher pusher({"ipc://*"});
JFJochReceiverService fpga_receiver_service(aq_devices, logger, pusher);
std::unique_ptr<StreamWriter> writer;
REQUIRE(x.GetImageNum() == 5);
auto pusher_addr = pusher.GetAddress();
REQUIRE(pusher_addr.size() == 1);
ZMQImagePuller puller(pusher_addr[0]);
REQUIRE_NOTHROW(writer = std::make_unique<StreamWriter>(logger, puller));
CHECK(writer->GetStatistics().state == StreamWriterState::Idle);
REQUIRE_NOTHROW(fpga_receiver_service.Start(x, pixel_mask, nullptr));
REQUIRE_NOTHROW(writer->Run());
REQUIRE_NOTHROW(fpga_receiver_service.Stop());
REQUIRE(fpga_receiver_service.GetStatus()->images_collected == 5);
REQUIRE(fpga_receiver_service.GetStatus()->images_sent == 5);
CHECK(writer->GetStatistics().state == StreamWriterState::Finalized);
CHECK(writer->GetStatistics().processed_images == 5);
CHECK(writer->GetStatistics().file_prefix == x.GetFilePrefix());
// HDF5 file can be opened
std::unique_ptr<HDF5ReadOnlyFile> file;
REQUIRE_NOTHROW(file = std::make_unique<HDF5ReadOnlyFile>("subdir/StreamWriterTest_data_000001.h5"));
std::unique_ptr<HDF5DataSet> dataset;
REQUIRE_NOTHROW(dataset = std::make_unique<HDF5DataSet>(*file, "/entry/data/data"));
std::unique_ptr<HDF5DataSpace> dataspace;
REQUIRE_NOTHROW(dataspace = std::make_unique<HDF5DataSpace>(*dataset));
REQUIRE(dataspace->GetNumOfDimensions() == 3);
REQUIRE(dataspace->GetDimensions()[0] == 5);
REQUIRE(dataspace->GetDimensions()[1] == RAW_MODULE_COLS);
REQUIRE(dataspace->GetDimensions()[2] == 2*RAW_MODULE_LINES);
REQUIRE(std::filesystem::remove("subdir/StreamWriterTest_master.h5"));
REQUIRE(std::filesystem::remove("subdir/StreamWriterTest_data_000001.h5"));
REQUIRE(std::filesystem::remove("subdir"));
}
TEST_CASE("StreamWriterTest_ZMQ_Update", "[StreamWriter]") {
RegisterHDF5Filter();
Logger logger("StreamWriterTest_ZMQ_Update");
DatasetSettings d;
d.FilePrefix("subdir/StreamWriterTest2").NumTriggers(1).ImagesPerTrigger(5).RunName("run1").RunNumber(256);
DiffractionExperiment x(DetJF(2));
x.UseInternalPacketGenerator(true).Raw().PedestalG0Frames(0)
.ImportDatasetSettings(d).OverwriteExistingFiles(true);
PixelMask pixel_mask(x);
JFModuleGainCalibration gain;
AcquisitionDeviceGroup aq_devices;
for (int i = 0; i < x.GetDataStreamsNum(); i++)
aq_devices.AddHLSDevice(64);
ZMQStream2Pusher pusher({"ipc://*"});
pusher.WriterNotificationSocket("ipc://*");
JFJochReceiverService fpga_receiver_service(aq_devices, logger, pusher);
std::unique_ptr<StreamWriter> writer;
REQUIRE(x.GetImageNum() == 5);
auto pusher_addr = pusher.GetAddress();
REQUIRE(pusher_addr.size() == 1);
ZMQImagePuller puller(pusher_addr[0]);
REQUIRE_NOTHROW(writer = std::make_unique<StreamWriter>(logger, puller));
CHECK(writer->GetStatistics().state == StreamWriterState::Idle);
REQUIRE_NOTHROW(fpga_receiver_service.Start(x, pixel_mask, nullptr));
REQUIRE_NOTHROW(writer->Run());
REQUIRE_NOTHROW(fpga_receiver_service.Stop());
REQUIRE(fpga_receiver_service.GetStatus()->images_collected == 5);
REQUIRE(fpga_receiver_service.GetStatus()->images_sent == 5);
CHECK(writer->GetStatistics().state == StreamWriterState::Finalized);
CHECK(writer->GetStatistics().processed_images == 5);
CHECK(writer->GetStatistics().file_prefix == x.GetFilePrefix());
// HDF5 file can be opened
std::unique_ptr<HDF5ReadOnlyFile> file;
REQUIRE_NOTHROW(file = std::make_unique<HDF5ReadOnlyFile>("subdir/StreamWriterTest2_data_000001.h5"));
std::unique_ptr<HDF5DataSet> dataset;
REQUIRE_NOTHROW(dataset = std::make_unique<HDF5DataSet>(*file, "/entry/data/data"));
std::unique_ptr<HDF5DataSpace> dataspace;
REQUIRE_NOTHROW(dataspace = std::make_unique<HDF5DataSpace>(*dataset));
REQUIRE(dataspace->GetNumOfDimensions() == 3);
REQUIRE(dataspace->GetDimensions()[0] == 5);
REQUIRE(dataspace->GetDimensions()[1] == RAW_MODULE_COLS);
REQUIRE(dataspace->GetDimensions()[2] == 2*RAW_MODULE_LINES);
REQUIRE(std::filesystem::remove("subdir/StreamWriterTest2_master.h5"));
REQUIRE(std::filesystem::remove("subdir/StreamWriterTest2_data_000001.h5"));
REQUIRE(std::filesystem::remove("subdir"));
}
TEST_CASE("StreamWriterTest_ZMQ_Update_NoNotification", "[StreamWriter]") {
// This tests simulates what happens if writer notification about writing end is missing
// Expected end result: receiver ends with an exception
RegisterHDF5Filter();
Logger logger("StreamWriterTest_ZMQ_Update_NoNotification");
DatasetSettings d;
d.FilePrefix("subdir/StreamWriterTest3").NumTriggers(1).ImagesPerTrigger(5).RunName("run1").RunNumber(256);
DiffractionExperiment x(DetJF(2));
x.UseInternalPacketGenerator(true).Raw().PedestalG0Frames(0)
.ImportDatasetSettings(d).OverwriteExistingFiles(true);
PixelMask pixel_mask(x);
JFModuleGainCalibration gain;
AcquisitionDeviceGroup aq_devices;
for (int i = 0; i < x.GetDataStreamsNum(); i++)
aq_devices.AddHLSDevice(64);
ZMQStream2Pusher pusher({"ipc://*"});
pusher.WriterNotificationSocket("ipc://*");
JFJochReceiverService fpga_receiver_service(aq_devices, logger, pusher);
std::unique_ptr<StreamWriter> writer;
REQUIRE(x.GetImageNum() == 5);
auto pusher_addr = pusher.GetAddress();
REQUIRE(pusher_addr.size() == 1);
ZMQImagePuller puller(pusher_addr[0]);
REQUIRE_NOTHROW(writer = std::make_unique<StreamWriter>(logger, puller));
writer->DebugSkipWriteNotification(true);
CHECK(writer->GetStatistics().state == StreamWriterState::Idle);
REQUIRE_NOTHROW(fpga_receiver_service.Start(x, pixel_mask, nullptr));
REQUIRE_NOTHROW(writer->Run());
JFJochReceiverOutput r;
REQUIRE_NOTHROW(r = fpga_receiver_service.Stop());
REQUIRE(!r.writer_err.empty());
REQUIRE(fpga_receiver_service.GetStatus()->images_collected == 5);
REQUIRE(fpga_receiver_service.GetStatus()->images_sent == 5);
CHECK(writer->GetStatistics().state == StreamWriterState::Finalized);
CHECK(writer->GetStatistics().processed_images == 5);
CHECK(writer->GetStatistics().file_prefix == x.GetFilePrefix());
// HDF5 file can be opened
std::unique_ptr<HDF5ReadOnlyFile> file;
REQUIRE_NOTHROW(file = std::make_unique<HDF5ReadOnlyFile>("subdir/StreamWriterTest3_data_000001.h5"));
std::unique_ptr<HDF5DataSet> dataset;
REQUIRE_NOTHROW(dataset = std::make_unique<HDF5DataSet>(*file, "/entry/data/data"));
std::unique_ptr<HDF5DataSpace> dataspace;
REQUIRE_NOTHROW(dataspace = std::make_unique<HDF5DataSpace>(*dataset));
REQUIRE(dataspace->GetNumOfDimensions() == 3);
REQUIRE(dataspace->GetDimensions()[0] == 5);
REQUIRE(dataspace->GetDimensions()[1] == RAW_MODULE_COLS);
REQUIRE(dataspace->GetDimensions()[2] == 2*RAW_MODULE_LINES);
REQUIRE(std::filesystem::remove("subdir/StreamWriterTest3_master.h5"));
REQUIRE(std::filesystem::remove("subdir/StreamWriterTest3_data_000001.h5"));
REQUIRE(std::filesystem::remove("subdir"));
}
// Two writers, the second one writing every other data file: the master, written by the first, must
// appear only after the second has closed its data files - here the second one starts only after the
// whole collection has been sent, so it is still writing when END goes out.
TEST_CASE("StreamWriterTest_ZMQ_MasterAfterAllDataFiles", "[StreamWriter]") {
RegisterHDF5Filter();
Logger logger("StreamWriterTest_ZMQ_MasterAfterAllDataFiles");
DatasetSettings d;
d.FilePrefix("subdir4/StreamWriterTest4").NumTriggers(1).ImagesPerTrigger(4).ImagesPerFile(2)
.RunName("run1").RunNumber(257);
DiffractionExperiment x(DetJF(2));
x.UseInternalPacketGenerator(true).Raw().PedestalG0Frames(0)
.ImportDatasetSettings(d).OverwriteExistingFiles(true);
PixelMask pixel_mask(x);
AcquisitionDeviceGroup aq_devices;
for (int i = 0; i < x.GetDataStreamsNum(); i++)
aq_devices.AddHLSDevice(64);
ZMQStream2Pusher pusher({"ipc://*", "ipc://*"});
pusher.WriterNotificationSocket("ipc://*");
JFJochReceiverService fpga_receiver_service(aq_devices, logger, pusher);
auto pusher_addr = pusher.GetAddress();
REQUIRE(pusher_addr.size() == 2);
ZMQImagePuller puller0(pusher_addr[0]), puller1(pusher_addr[1]);
StreamWriter writer0(logger, puller0), writer1(logger, puller1);
REQUIRE_NOTHROW(fpga_receiver_service.Start(x, pixel_mask, nullptr));
auto w0 = std::async(std::launch::async, [&] { return writer0.Run(); });
for (int i = 0; i < 1200; i++) {
const auto status = fpga_receiver_service.GetStatus();
if (status && status->images_sent == 4)
break;
std::this_thread::sleep_for(std::chrono::milliseconds(100));
}
auto stop = std::async(std::launch::async, [&] { return fpga_receiver_service.Stop(); });
std::this_thread::sleep_for(std::chrono::seconds(2));
auto w1 = std::async(std::launch::async, [&] { return writer1.Run(); });
JFJochReceiverOutput r;
REQUIRE_NOTHROW(r = stop.get());
w0.get();
w1.get();
CHECK(r.writer_err.empty());
auto changed = [](const std::string &path) {
struct stat st{};
REQUIRE(stat(path.c_str(), &st) == 0);
return std::make_pair(st.st_ctim.tv_sec, st.st_ctim.tv_nsec);
};
const auto master = changed("subdir4/StreamWriterTest4_master.h5");
for (int f = 1; f <= 2; f++)
CHECK(changed("subdir4/StreamWriterTest4_data_00000" + std::to_string(f) + ".h5") <= master);
std::filesystem::remove_all("subdir4");
}
// A failure that is not a JFJochException must end the run the same way as one that is: error state,
// END processed, state Finalized. Before this test the writer caught only its own exception type at the
// per-image and END sites, and anything else - here the JSON library refusing a sample name that is not
// UTF-8 while publishing a file-done notification - left Run() with the state stuck in Receiving, no END
// acknowledgement and no notification, and on TCP a broker waiting for that acknowledgement for ever.
TEST_CASE("StreamWriterTest_NonJFJochException_EndsRunInErrorState", "[StreamWriter]") {
RegisterHDF5Filter();
Logger logger("StreamWriterTest_NonJFJochException");
DiffractionExperiment x(DetJF(2));
x.FilePrefix("subdir5/StreamWriterTest5").NumTriggers(1).ImagesPerTrigger(2).ImagesPerFile(1)
.Compression(CompressionAlgorithm::NO_COMPRESSION).OverwriteExistingFiles(true)
.SampleName("\xff\xfe not utf-8");
StartMessage start_message;
x.FillMessage(start_message);
start_message.write_master_file = true;
TestImagePuller puller;
// The file-done PUB socket is what makes the writer serialise the sample name to JSON.
StreamWriter writer(logger, puller, "ipc://*");
puller.Put({.cbor = std::make_shared<CBORStream2DeserializerOutput>(start_message)});
std::vector<int16_t> image(x.GetPixelsNum());
for (int i = 0; i < 2; i++) {
DataMessage message{};
message.image = CompressedImage(image, x.GetXPixelsNum(), x.GetYPixelsNum());
message.number = i;
puller.Put({.cbor = std::make_shared<CBORStream2DeserializerOutput>(message)});
}
EndMessage end_message{};
end_message.max_image_number = 2;
puller.Put({.cbor = std::make_shared<CBORStream2DeserializerOutput>(end_message)});
StreamWriterOutput out;
REQUIRE_NOTHROW(out = writer.Run());
// Image 0 filled its one-image data file; closing it published the notification, which threw.
CHECK(out.stats.state == StreamWriterState::Finalized);
CHECK(out.stats.processed_images == 0);
CHECK(out.data_file_stats.empty());
// The master is never written for a run that ended in error; the data file was renamed into
// place before the notification was attempted.
CHECK(!std::filesystem::exists("subdir5/StreamWriterTest5_master.h5"));
CHECK(std::filesystem::exists("subdir5/StreamWriterTest5_data_000001.h5"));
std::filesystem::remove_all("subdir5");
}