mx/Jungfraujoch
3
2
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 82 Wiki Activity
Files
51706641fae540d90f09c26dcff667dc05ffe060
Jungfraujoch/tests
T
History
leonarski_fandClaude Opus 5.5 51706641fa RESULT-CHANGING: molecular replacement's transform FFT on the GPU wherever it fits the search's budget 
The molecular transform's FFT went to the GPU only where its buffers took at most a quarter of the card
(MapToFPhiGPUFits), the last quarter rule left in molecular replacement. On a large cell that is never:
4nwv's 768^3 box needs 5.45 GB, so on a 16 GB card it ran on the CPU in FFTW, single-threaded - 12.5 s
of the search's 31 s.

Now it is judged, as every other part of the search, against the search's budget (80% of the card less
what is resident; the search holds nothing else yet when it runs), and the device writes only the
cropped transform the search keeps (CropHalfL, moved to ModelFFT), so it needs 4.46 GB instead of 5.45.
cuFFT and FFTW agree to float rounding, not bit for bit, so wherever the transform moves to the GPU the
search's functions can differ in their last bits: on 4nwv on 16 GB and 8.55 GB cards it moves (it now
does on any card above about 5.6 GB, before above 21.8 GB); the small boxes were on the GPU before and
give the same values (the same kernel, cropped).

Measured on the RTX 5080, timing benchmark: 4nwv's transform 14.1 -> 2.2 s, MR 31.6 -> 20.1 s, run
93.0 -> 89.2 s; MR's peak device memory unchanged at 8.7 GB (the FFT's own 5.2 GB comes before anything
else is held). On this card the reflection files, maps, placed models, reports and the MR candidate lines
of myob, cytc and 4nwv still came out byte-identical to rc175-metal-merge. [MolecularReplacement] passes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-15 13:41:48 +02:00
..
crystfel
v1.0.0.rc-161 (#71)
2026-08-13 17:03:10 +02:00
data
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
test_data
v1.0.0-rc.160 (#70)
2026-07-19 09:39:28 +02:00
xds
XDS plugin 1.1.0: mask module/chip edges, pass saturation through; recommend Durin
2026-10-08 10:56:34 +02:00
xds_durin
XDS plugin 1.1.0: mask module/chip edges, pass saturation through; recommend Durin
2026-10-08 10:56:34 +02:00
xds_neggia
XDS plugin 1.1.0: mask module/chip edges, pass saturation through; recommend Durin
2026-10-08 10:56:34 +02:00
AcquisitionCountersTest.cpp
v1.0.0-rc.36
2025-05-05 19:32:22 +02:00
AdaptiveSpotFinderCPUTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
AdaptiveSpotFinderGPUTest.cpp
GPU azimuthal profile: sum the corrected values in fixed point
2026-10-08 12:33:34 +02:00
AdaptiveThresholdTest.cpp
v1.0.0-rc.166 (#76)
2026-09-02 21:17:31 +02:00
AnisotropyAnalysisTest.cpp
Anisotropy verdict: the directional limits are a second hypothesis beside the tensor
2026-10-11 02:21:32 +02:00
AutoIncrVectorTest.cpp
v1.0.0-rc.139 (#49)
2026-04-29 09:50:50 +02:00
AzimuthalIntegrationTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
BackgroundGainTest.cpp
rugnux: background clip scaled by the CCD gain measured in the pre-scan
2026-10-09 21:18:12 +02:00
BandwidthEstimateTest.cpp
Move SpotWidth to bragg_integration/, SpindleCuspLoss and LeftoverLattices to indexing/
2026-10-07 14:05:37 +02:00
BeamCenterFFTGPUTest.cpp
v1.0.0-rc.169 (#79)
2026-09-15 17:09:31 +02:00
BeamCenterFFTTest.cpp
v1.0.0-rc.169 (#79)
2026-09-15 17:09:31 +02:00
BeamCenterFromBackgroundTest.cpp
1.0.0-rc.174 (#84)
2026-10-06 14:03:18 +02:00
BearerTokensTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
BraggIntegrationEngineCompressedImageTest.cpp
1.0.0-rc.174 (#84)
2026-10-06 14:03:18 +02:00
BraggIntegrationEngineCPUTest.cpp
rugnux two-pass: the integration radius decided on the pre-pass frames, before the canonical pass
2026-10-11 13:01:23 +02:00
BraggIntegrationEngineGPUTest.cpp
1.0.0-rc.174 (#84)
2026-10-06 14:03:18 +02:00
BraggIntegrationEngineMetalTest.cpp
Metal Bragg integration engine on macOS, output unchanged (bit-identical to the CPU engine)
2026-10-14 23:18:42 +02:00
BraggStencilTest.cpp
1.0.0-rc.174 (#84)
2026-10-06 14:03:18 +02:00
BrokerHttpAuthTest.cpp
Broker: a collection can be followed from /start, its start message comes in the event stream
2026-10-08 17:03:09 +02:00
BSLZ4DecoderGPUFuzzTest.cpp
v1.0.0.rc-161 (#71)
2026-08-13 17:03:10 +02:00
BSLZ4DecoderGPUTest.cpp
v1.0.0.rc-161 (#71)
2026-08-13 17:03:10 +02:00
CalcBraggPredictionTest.cpp
CalcBraggPredictionTest: skip the GPU cases on a machine without a card
2026-10-13 07:54:56 +02:00
CalibrationTest.cpp
v1.0.0-rc.166 (#76)
2026-09-02 21:17:31 +02:00
CBORTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
CCTest.cpp
v1.0.0-rc.146 (#56)
2026-05-28 18:48:35 +02:00
CheckImageOutput.h
v1.0.0-rc.36
2025-05-05 19:32:22 +02:00
CMakeLists.txt
Merge rc175-metal-integration into rc175: Metal Bragg integration engine on macOS; CPU profile fit in single precision with a shared exp (result-changing at ulp level on the CPU path only); CPU partial scaling loop reads its working copy in order
2026-10-15 09:25:26 +02:00
CoordTest.cpp
v1.0.0-rc.41
2025-06-10 18:14:04 +02:00
CorrectionSurfaceGPUTest.cpp
1.0.0-rc.174 (#84)
2026-10-06 14:03:18 +02:00
CrystalLatticeTest.cpp
v1.0.0-rc.166 (#76)
2026-09-02 21:17:31 +02:00
CrystalSettingTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
CudaHandledErrorTest.cpp
Indexer failures are thrown, not read as "no lattice"; GPU memory errors say what to do
2026-10-10 18:56:00 +02:00
CUDAMemHelpersTest.cpp
Merge cufft-cache into rc175: cuFFT plans reused within a rugnux run and destroyed when it ends, output unchanged
2026-10-14 22:12:53 +02:00
DectrisSimplonClientTest.cpp
DectrisSimplonClient: separate connections for commands and status reads
2026-10-08 22:53:01 +02:00
DetectorGeometryTest.cpp
v1.0.0-rc.103 (#8)
2025-11-19 09:40:50 +01:00
DetectorSettingsTest.cpp
v1.0.0-rc.139 (#49)
2026-04-29 09:50:50 +02:00
DetectorSetupTest.cpp
v1.0.0.rc-162 (#72)
2026-08-25 08:21:39 +02:00
DetGeomCalibTest.cpp
v1.0.0-rc.166 (#76)
2026-09-02 21:17:31 +02:00
DiagnosticOutputTest.cpp
rugnux: _plot.txt gets a graded per-frame disposition column
2026-10-13 15:44:04 +02:00
DiffractionExperimentTest.cpp
v1.0.0-rc.166 (#76)
2026-09-02 21:17:31 +02:00
DiffractionGeometryTest.cpp
v1.0.0-rc.166 (#76)
2026-09-02 21:17:31 +02:00
DiffractionSpotTest.cpp
v1.0.0.rc-161 (#71)
2026-08-13 17:03:10 +02:00
DirectHostReadGPUTest.cpp
GPU: kernels read host buffers in place on unified-memory systems
2026-10-14 21:49:37 +02:00
enospc_shim.c
SHIM library improvements from the HDF Group
2026-05-08 11:39:51 +02:00
FPGAFrameGeneratorTest.cpp
version 1.0.0-rc.25
2024-11-22 21:25:20 +01:00
FPGAHLSModulesTest.cpp
v1.0.0-rc.91
2025-10-20 20:43:44 +02:00
FPGAHostWriterTest.cpp
v1.0.0-rc.34
2025-04-14 11:52:06 +02:00
FPGAIntegrationTest.cpp
v1.0.0-rc.97
2025-11-09 12:42:27 +01:00
FPGANetworkTest.cpp
v1.0.0-rc.72
2025-09-08 20:28:59 +02:00
FPGAPTPTest.cpp
v1.0.0-rc.72
2025-09-08 20:28:59 +02:00
FPGASpotFindingUnitTest.cpp
v.1.0.0-rc.32
2025-03-24 12:16:33 +01:00
FrameTransformationTest.cpp
v1.0.0-rc.172 (#82)
2026-09-22 06:48:37 +02:00
FrenchWilsonTest.cpp
1.0.0-rc.174 (#84)
2026-10-06 14:03:18 +02:00
GoniometerAxisTest.cpp
v1.0.0-rc49
2025-06-18 15:19:18 +02:00
GridScanSettingsTest.cpp
v1.0.0-rc.166 (#76)
2026-09-02 21:17:31 +02:00
HDF5WritingTest.cpp
v1.0.0-rc.167 (#77)
2026-09-09 07:25:13 +02:00
HistogramTest.cpp
v1.0.0-rc.121 (#28)
2025-12-12 21:24:20 +01:00
HKLKeyTest.cpp
v1.0.0-rc.124 (#31)
2026-02-01 13:29:33 +01:00
HotPixelFinderTest.cpp
rugnux pre-scan: projection and defective pixels per card, the sample read once, beam stop memoised
2026-10-12 08:11:05 +02:00
ImageBufferTest.cpp
v1.0.0-rc.128 (#35)
2026-03-03 22:24:44 +01:00
ImageMetadataTest.cpp
v1.0.0-rc.103 (#8)
2025-11-19 09:40:50 +01:00
ImagePreprocessorGPUFusedTest.cpp
v1.0.0.rc-162 (#72)
2026-08-25 08:21:39 +02:00
ImageSpotFinderCPUTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
ImageSpotFinderGPUTest.cpp
v1.0.0.rc-161 (#71)
2026-08-13 17:03:10 +02:00
IndexingUnitTest.cpp
rugnux: doubled-cell hypothesis - the supercell probe reads its class to the limit, and a class that stands clear is merged in the doubled cell as well
2026-10-10 10:47:55 +02:00
JFCalibrationTest.cpp
v1.0.0-rc.36
2025-05-05 19:32:22 +02:00
JFConversionTest.cpp
version 1.0.0-rc.25
2024-11-22 21:25:20 +01:00
jfjoch_hdf5_enospc_test.cpp
v1.0.0.rc-162 (#72)
2026-08-25 08:21:39 +02:00
JFJochReaderTest.cpp
HDF5 reader: read variable-length string attributes and datasets
2026-10-09 21:15:01 +02:00
JFJochReceiverIntegrationTest.cpp
v1.0.0-rc.150 (#60)
2026-06-15 20:24:15 +02:00
JFJochReceiverLiteTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
JFJochReceiverPlotsTest.cpp
v1.0.0-rc.153 (#63)
2026-06-23 20:29:49 +02:00
JFJochReceiverProcessingTest.cpp
v1.0.0-rc.160 (#70)
2026-07-19 09:39:28 +02:00
JFJochStateMachineTest.cpp
v1.0.0-rc.169 (#79)
2026-09-15 17:09:31 +02:00
JFPedestalTest.cpp
version 1.0.0-rc.25
2024-11-22 21:25:20 +01:00
JPEGTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
LatticeReductionTest.cpp
v1.0.0-rc.153 (#63)
2026-06-23 20:29:49 +02:00
LatticeSearchTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
LePageLatticeTest.cpp
v1.0.0-rc.167 (#77)
2026-09-09 07:25:13 +02:00
LiveCollectionTest.cpp
Broker: a collection can be followed from /start, its start message comes in the event stream
2026-10-08 17:03:09 +02:00
LoadReferenceMtzFreeFlagsTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
LoadReferenceSfCifTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
LossyFilterTest.cpp
version 1.0.0-rc.25
2024-11-22 21:25:20 +01:00
MergeScaleTest.cpp
RESULT-CHANGING (owner approved): stills crystals indexed in a lattice containing the run's are put on the run's cell
2026-10-14 13:15:13 +02:00
ModelMaskGPUTest.cpp
ModelMaskGPU: gemmi's shrink step, so fine rigid-body zones stay on the GPU
2026-10-09 12:38:37 +02:00
ModelScaleGPUTest.cpp
Model validation: scale fits on the GPU, change-of-basis probe gridded once per frame on all threads
2026-10-12 21:40:54 +02:00
ModelScalingTest.cpp
--model: the bulk-solvent b_sol ceiling follows the model's atomic B
2026-10-10 20:47:04 +02:00
ModelValidationTest.cpp
Molecular replacement: one GPU plan per merge on model validation's budget, the transform cropped, output unchanged
2026-10-15 13:33:39 +02:00
ModuleSummationTest.cpp
version 1.0.0-rc.27 minor fixes
2024-12-04 20:58:51 +01:00
MolecularReplacementTest.cpp
RESULT-CHANGING: molecular replacement's transform FFT on the GPU wherever it fits the search's budget
2026-10-15 13:41:48 +02:00
MovingAverageTest.cpp
v1.0.0-rc.31
2025-03-02 13:15:28 +01:00
MultiLatticeSearchTest.cpp
v1.0.0-rc.148 (#58)
2026-06-08 08:30:35 +02:00
nxmx_geometry_dials_test.py
v1.0.0.rc-162 (#72)
2026-08-25 08:21:39 +02:00
PairSortGPUTest.cpp
RotationScaleMerge: sort the outlier band's log-spread pairs on the device
2026-10-08 07:27:25 +02:00
ParallelForTest.cpp
rugnux: on a machine of several sockets, keep to the sockets the GPUs are on
2026-10-11 09:27:32 +02:00
PedestalCalcTest.cpp
v1.0.0-rc.97
2025-11-09 12:42:27 +01:00
PixelMaskTest.cpp
v1.0.0.rc-161 (#71)
2026-08-13 17:03:10 +02:00
PortableTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
PostRefineTest.cpp
rugnux hypothesis arms read their residual on the pre-pass frames
2026-10-11 07:12:05 +02:00
PreflightTest.cpp
v1.0.0-rc.165 (#75)
2026-08-27 22:16:54 +02:00
PreviewCounterTest.cpp
version 1.0.0-rc.25
2024-11-22 21:25:20 +01:00
RawToConvertedGeometryTest.cpp
v1.0.0-rc.97
2025-11-09 12:42:27 +01:00
ReceiveRingBufferTest.cpp
Writer: received images live in a byte-bounded ring buffer until written and republished
2026-10-13 12:51:34 +02:00
RegressionTest.cpp
1.0.0-rc.89
2025-10-01 22:54:40 +02:00
ReindexAmbiguityTest.cpp
Model validation expands strict NCS; the written cell follows the model's indexing
2026-10-10 13:35:49 +02:00
ResolutionShellsTest.cpp
v1.0.0-rc.166 (#76)
2026-09-02 21:17:31 +02:00
ResultReportTest.cpp
rugnux --model: molecular replacement where the rigid body does not place the model
2026-10-12 22:20:36 +02:00
RfreeFlagsTest.cpp
v1.0.0-rc.172 (#82)
2026-09-22 06:48:37 +02:00
RingsFromProfileTest.cpp
v1.0.0-rc.166 (#76)
2026-09-02 21:17:31 +02:00
ROIIntegrationCPUTest.cpp
v1.0.0-rc.153 (#63)
2026-06-23 20:29:49 +02:00
ROIIntegrationGPUTest.cpp
v1.0.0-rc.153 (#63)
2026-06-23 20:29:49 +02:00
ROIMapTest.cpp
v1.0.0-rc.153 (#63)
2026-06-23 20:29:49 +02:00
RotationIndexerTest.cpp
Indexer failures are thrown, not read as "no lattice"; GPU memory errors say what to do
2026-10-10 18:56:00 +02:00
RotationIngestGPUTest.cpp
Rotation scaling on the GPU: two-pass ingest, decided before the card is touched, output unchanged
2026-10-15 11:13:41 +02:00
RotationScaleWalkTest.cpp
rugnux: the rotation-scale walk stops where the next step would move the sweep end by under half a frame
2026-10-13 19:42:25 +02:00
RugnuxLargeTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
RugnuxTest.cpp
rugnux -X FFT is the algorithm: cuFFT on a GPU, FFTW without one; -X cuFFT names the GPU implementation
2026-10-14 10:13:23 +02:00
SearchSpaceGroupTest.cpp
Space group: a screw one axial reflection could not confirm is reported open
2026-10-13 14:43:44 +02:00
SearchSpaceGroupTwinTest.cpp
v1.0.0-rc.169 (#79)
2026-09-15 17:09:31 +02:00
ShadowFinderTest.cpp
rugnux pre-scan: projection and defective pixels per card, the sample read once, beam stop memoised
2026-10-12 08:11:05 +02:00
SphericalHarmonicSurfaceTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
SpindleBlindFractionTest.cpp
v1.0.0-rc.167 (#77)
2026-09-09 07:25:13 +02:00
SpindleCuspLossTest.cpp
Move SpotWidth to bragg_integration/, SpindleCuspLoss and LeftoverLattices to indexing/
2026-10-07 14:05:37 +02:00
SpotExtractorGPUParityTest.cpp
v1.0.0.rc-161 (#71)
2026-08-13 17:03:10 +02:00
SpotFootprintTest.cpp
1.0.0-rc.174 (#84)
2026-10-06 14:03:18 +02:00
SpotTilesTest.cpp
SpotTiles_MatchWholeFrame_GPU: wait for the upload before the squares are cut
2026-10-14 22:35:48 +02:00
SpotUtilsTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
StatusVectorTest.cpp
v1.0.0-rc.127 (#34)
2026-03-02 15:57:12 +01:00
StoredImageDecodeGPUTest.cpp
marCCD upload: stage the stored pixels instead of page-locking the frame in place
2026-10-13 05:21:33 +02:00
StreamWriterTest.cpp
Writer: a failure that is not a JFJochException ends the run in error state too
2026-10-13 12:27:08 +02:00
StrongPixelSetTest.cpp
v1.0.0-rc.166 (#76)
2026-09-02 21:17:31 +02:00
SweepLayoutTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00
SyntheticMergedReflections.h
v1.0.0.rc-161 (#71)
2026-08-13 17:03:10 +02:00
TCPImagePusherTest.cpp
Writer: received images live in a byte-bounded ring buffer until written and republished
2026-10-13 12:51:34 +02:00
TestData.h
v1.0.0-rc.153 (#63)
2026-06-23 20:29:49 +02:00
ThreadSafeFIFOSetTest.cpp
v1.0.0-rc.135 (#44)
2026-04-16 11:59:59 +02:00
TIFFTest.cpp
v1.0.0-rc.153 (#63)
2026-06-23 20:29:49 +02:00
TimeTest.cpp
v1.0.0-rc.96 (#1)
2025-11-02 13:45:57 +01:00
TopPixelsTest.cpp
v1.0.0-rc.121 (#28)
2025-12-12 21:24:20 +01:00
TranslationalNCSTest.cpp
v1.0.0-rc.172 (#82)
2026-09-22 06:48:37 +02:00
TwinningAnalysisTest.cpp
rugnux: write the twin hypothesis beside the answer as <prefix>_twin_<SG>.mtz/.cif
2026-10-09 19:40:22 +02:00
UnitCellTest.cpp
v1.0.0-rc.146 (#56)
2026-05-28 18:48:35 +02:00
WilsonOutliersTest.cpp
RotationScaleMerge: the Wilson outlier test on the device in the GPU build
2026-10-08 07:07:32 +02:00
WriteReflectionsTest.cpp
RESULT-CHANGING (owner approved): stills unmerged MTZ written only for the crystals on the run's cell
2026-10-14 13:15:13 +02:00
XDSPluginTest.cpp
XDSPluginTest: expect a signed image's INTx_MAX to pass through as saturated
2026-10-08 12:36:40 +02:00
XtalOptimizerTest.cpp
v1.0.0-rc.166 (#76)
2026-09-02 21:17:31 +02:00
XtalRefineCeres.h
1.0.0-rc.174 (#84)
2026-10-06 14:03:18 +02:00
XtalRefineTest.cpp
1.0.0-rc.174 (#84)
2026-10-06 14:03:18 +02:00
ZMQImagePusherTest.cpp
Writer: received images live in a byte-bounded ring buffer until written and republished
2026-10-13 12:51:34 +02:00
ZMQMetadataSocketTest.cpp
v1.0.0-rc.133
2026-03-26 20:50:33 +01:00
ZMQPreviewSocketTest.cpp
version 1.0.0-rc.25
2024-11-22 21:25:20 +01:00
ZSTDCompressorTest.cpp
v1.0.0-rc.173 (#83)
2026-09-29 15:57:32 +02:00