Files
Jungfraujoch/image_analysis/spot_finding/StrongPixelSet.cpp
T
leonarski_fandClaude Opus 5 4bdb229fb8
Build Packages / build:viewer-tgz:cpu (push) Successful in 7m46s
Build Packages / build:viewer-tgz:cuda (push) Successful in 9m14s
Build Packages / build:rpm (ubuntu2404_nocuda) (push) Successful in 13m51s
Build Packages / build:rpm (rocky8_nocuda) (push) Successful in 14m17s
Build Packages / build:rpm (ubuntu2204_nocuda) (push) Successful in 14m14s
Build Packages / build:rpm (rocky9_nocuda) (push) Successful in 14m43s
Build Packages / build:rpm (rocky8_sls9) (push) Successful in 14m45s
Build Packages / build:rpm (rocky8) (push) Successful in 11m44s
Build Packages / build:rpm (rocky9_sls9) (push) Successful in 13m24s
Build Packages / XDS test (durin plugin) (push) Successful in 8m33s
Build Packages / Generate python client (push) Successful in 28s
Build Packages / Build documentation (push) Successful in 1m4s
Build Packages / Create release (push) Skipped
Build Packages / build:rpm (rocky9) (push) Successful in 12m45s
Build Packages / build:rpm (ubuntu2204) (push) Successful in 12m25s
Build Packages / build:rpm (ubuntu2404) (push) Successful in 13m1s
Build Packages / DIALS test (push) Successful in 14m29s
Build Packages / XDS test (neggia plugin) (push) Successful in 8m17s
Build Packages / XDS test (JFJoch plugin) (push) Successful in 9m5s
Build Packages / Unit tests (push) Successful in 1h16m19s
Build Packages / build:windows:nocuda (push) Failing after 2s
Build Packages / build:windows:cuda (push) Failing after 3s
spot_finding: find connected components on the GPU
The spot finder flagged strong pixels on the device and then labelled them on the
host, so every frame sent the packed bitmask back - 2.26 MB on a large detector -
and the host walked all of it to recover a few hundred pixels. Do the labelling on
the device instead: compact the bitmask into a flat-index-sorted list, find each
pixel's backward neighbours by binary search, union them lock-free with path
halving, then label, accumulate and filter in one kernel. Only the spot list comes
back, and only one stream synchronisation per frame.

The gain in the ordinary case is modest - about a quarter off per-image spot
finding - because the host algorithm is genuinely fast on a normal frame. What
justifies it is the frame that is not ordinary. The host labels a sorted sparse
list through a window spanning two detector lines, so its cost is quadratic in how
many strong pixels share a line. A lit band of detector rows - a hot module, a
panel edge - costs 33 ms at two rows and 377 ms at fifteen, all of it under the
pixel cap that was supposed to bound this, and none of it maskable when the cause
is a diffraction ring rather than a defect: a ring runs tangent to a row at its
top and bottom, which is exactly the shape that hurts. The device version is flat
at 0.05 to 0.64 ms across every geometry tried, so an online run no longer stalls
a quarter of a second on an ice ring. Rejecting an over-cap frame is now free too,
since the count is known before any pixel is written.

Also label once and filter three times. The per-image minimum-pixel search runs the
extraction at three settings, but that setting only decides which components are
kept - it does not change the components - so the search itself need not be
repeated. This helps the host path as much as the device one.

The resolution mask moves to the device as a bit mask, uploaded when the limits
change rather than per frame, since the compaction needs it there.

Parity is asserted permanently rather than argued: five cases covering realistic
frames, occupancy from a hundred pixels to past the cap, the pathological
geometries including rings, the resolution mask, and a hundred-repeat determinism
check - requiring the same partition, the same spot order, and identical counts.
The centroid is a float sum and therefore order-dependent, so the device walks each
component from its root in ascending order and fuses its multiply-add the way the
host's does; note that whether the host fuses at all depends on the architecture
flags, so exact centroid equality is asserted where the compiler fuses and a
two-ulp bound otherwise. Making those accumulators integer would remove that
dependence entirely and is worth doing separately.

Regression set: all 37 crystals identical to the last printed digit. Unit suite
passes with the new cases.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-02 13:34:43 +02:00

152 lines
5.1 KiB
C++

// SPDX-FileCopyrightText: 2024 Filip Leonarski, Paul Scherrer Institute <filip.leonarski@psi.ch>
// SPDX-License-Identifier: GPL-3.0-only
// SparseCCL code taken from https://github.com/acts-project/traccc/blob/main/core/include/traccc/clusterization/detail/sparse_ccl.hpp
// (c) 2021-2022 CERN for the benefit of the ACTS project
// Mozilla Public License Version 2.0
#include <bitset>
#include "StrongPixelSet.h"
StrongPixelSet::StrongPixelSet() : strong_pixel_count(0) {
pixels.reserve(max_strong_pixel_per_module);
}
void StrongPixelSet::AddStrongPixel(uint16_t col, uint16_t line, int32_t photons) {
pixels.push_back(strong_pixel{.col = col, .line = line, .counts = photons});
++strong_pixel_count;
}
bool is_far_enough(strong_pixel pixel0, strong_pixel pixel1) {
return (pixel1.line - pixel0.line) > 1;
}
bool is_adjacent(strong_pixel pixel0, strong_pixel pixel1) {
auto line_diff = pixel0.line - pixel1.line;
auto col_diff = pixel0.col - pixel1.col;
return line_diff <= 1 && line_diff >= -1 && col_diff <= 1 && col_diff >= -1;
}
uint32_t StrongPixelSet::find_root(uint32_t e) {
uint32_t r = e;
while (L[r] != r)
r = L[r];
return r;
}
uint32_t StrongPixelSet::make_union(uint32_t e1, uint32_t e2) {
uint32_t e;
if (e1 < e2) {
e = e1;
L[e2] = e;
} else {
e = e2;
L[e1] = e;
}
return e;
}
std::vector<DiffractionSpot> StrongPixelSet::sparseccl() {
L.resize(pixels.size());
unsigned int labels = 0;
// first scan: pixel association
uint32_t start_j = 0;
for (uint32_t i = 0; i < pixels.size(); ++i) {
L[i] = i;
uint32_t ai = i;
for (uint32_t j = start_j; j < i; ++j) {
if (is_adjacent(pixels[i], pixels[j])) {
ai = make_union(ai, find_root(j));
} else if (is_far_enough(pixels[j], pixels[i])) {
++start_j;
}
}
}
// second scan: transitive closure
for (uint32_t i = 0; i < L.size(); ++i) {
if (L[i] == i) {
L[i] = labels++;
} else {
L[i] = L[L[i]];
}
}
std::vector<DiffractionSpot> spots(labels);
for (uint32_t i = 0; i < L.size(); i++)
spots[L[i]].AddPixel(pixels[i].col, pixels[i].line, pixels[i].counts);
return spots;
}
void StrongPixelSet::FindComponentsImage(const SpotFindingSettings &settings, std::vector<DiffractionSpot> &spots) {
// Avoid spot finding, when more than 65536 strong pixel count (will be super slow)
if (!pixels.empty() && (strong_pixel_count < UINT16_MAX)) {
for (const auto &spot: sparseccl()) {
if (spot.PixelCount() <= settings.max_pix_per_spot)
spots.push_back(spot);
}
}
}
void StrongPixelSet::FindSpots(const DiffractionExperiment &experiment, const SpotFindingSettings &settings,
std::vector<DiffractionSpot> &spots, uint16_t module_number) {
// Avoid spot finding, when more than 65536 strong pixel count (will be super slow)
if (!pixels.empty() && (strong_pixel_count < UINT16_MAX)) {
for (const auto &spot: sparseccl()) {
if ((spot.PixelCount() <= settings.max_pix_per_spot)
&& (spot.PixelCount() >= settings.min_pix_per_spot.value_or(2))) {
auto s = spot;
s.ConvertToImageCoordinates(experiment, module_number);
spots.push_back(s);
}
}
}
}
void StrongPixelSet::ReadFPGAOutput(const DiffractionExperiment & experiment,
const DeviceOutput &output) {
// Too many strong pixels will kill performance in data processing, so protection is needed
// Also if there are no strong pixels, there is no point in looking for them
if ((output.spot_finding_result.strong_pixel_count == 0) ||
(output.spot_finding_result.strong_pixel_count > max_strong_pixel_per_module)) {
// If max strong pixel per module condition kicks-in, still report correct strong pixel count
strong_pixel_count = output.spot_finding_result.strong_pixel_count;
return;
}
auto pixel_depth = experiment.GetByteDepthImage();
auto out_ptr = (uint32_t *) output.spot_finding_result.strong_pixel;
for (int i = 0; i < RAW_MODULE_SIZE / (8 * sizeof(out_ptr[0])); i++) {
size_t npixel = i * 8 * sizeof(out_ptr[0]);
size_t line = npixel / RAW_MODULE_COLS;
if (out_ptr[i] != 0) {
std::bitset<32> bitset(out_ptr[i]);
for (int j = 0; j < 32; j++) {
if (bitset.test(j)) {
size_t col = (npixel | j) % RAW_MODULE_COLS;
if (pixel_depth == 2)
AddStrongPixel(col, line, output.pixels[npixel | j]);
else if (pixel_depth == 1)
AddStrongPixel(col, line, ((int8_t *)output.pixels)[npixel | j]);
else if (pixel_depth == 4)
AddStrongPixel(col, line, ((int32_t *)output.pixels)[npixel | j]);
}
}
}
}
}
uint32_t StrongPixelSet::GetStrongPixelCount() const {
return strong_pixel_count;
}