Build Packages / build:portable:linux-x86_64:nocuda (push) Successful in 16m43s
Build Packages / build:jfjoch:ubuntu2404:nocuda (push) Successful in 28m46s
Build Packages / build:jfjoch:rocky8:cuda-sls9 (push) Successful in 32m5s
Build Packages / build:jfjoch:rocky9:cuda (push) Successful in 39m44s
Build Packages / build:portable:macos-arm64:nocuda (push) Successful in 5m7s
Build Packages / Create release (push) Successful in 8s
Build Packages / Build documentation (push) Successful in 49s
Build Packages / build:portable:linux-x86_64:cuda (push) Successful in 22m30s
Build Packages / build:jfjoch:rocky8:nocuda (push) Successful in 31m54s
Build Packages / build:rugnux:linux-aarch64:cuda (push) Successful in 11m29s
Build Packages / build:portable:windows-x86_64:nocuda (push) Successful in 23m51s
Build Packages / build:jfjoch:ubuntu2204:nocuda (push) Successful in 26m7s
Build Packages / build:jfjoch:ubuntu2204:cuda (push) Successful in 29m38s
Build Packages / build:portable:windows-x86_64:cuda (push) Successful in 28m6s
Build Packages / HDF5 consumer tests (DIALS, XDS) (push) Successful in 29m44s
Build Packages / build:jfjoch:rocky9:nocuda (push) Successful in 33m22s
Build Packages / Generate python client (push) Successful in 47s
Build Packages / build:jfjoch:rocky9:cuda-sls9 (push) Successful in 34m6s
Build Packages / Unit tests (push) Successful in 1h2m47s
Build Packages / build:jfjoch:rocky8:cuda (push) Successful in 35m59s
Build Packages / build:jfjoch:ubuntu2404:cuda (push) Successful in 34m8s
Build Packages / Validation Battery (push) Successful in 38m31s
* Rugnux: Refinement several times faster, one protocol on the CPU and the GPU, parts of it on the Apple GPU; its random draws are the same on every platform. * Rugnux: Refinement follows common practice: the free set decides only the restraint weights and the displacement model (isotropic, TLS or anisotropic, by Hamilton's test), everything else is decided on the working set; lower R-free for placed and predicted models. * Rugnux: Restraint dictionaries via `--restraints` (or among the inputs) and more built-in ligands; a model whose space group is a subgroup of the data's is refined in its own group; reflections plus a model select `--mode refine` on their own. * Rugnux: `-C` fixes the crystal system on the axes as typed; `-S` alone fixes the space group only, the cell is still found and reduced; a space group or cell stored in the input file is ignored, with a line in the log. * Rugnux: More reliable multi-sweep merging; `--mode scale` merges several stills runs as one; `_plot.txt` and the report name each input; a second crystal form is reported for stills given a cell. * jfjoch_broker (experimental): Live Rugnux on FPGA detectors (JUNGFRAU) takes the same path as a DECTRIS stream, rotation and serial alike, model step included, with analysis threads per GPU; frames waiting for the analysis can be viewed; no host stage drops a frame - a source faster than the analysis is held at the card. * jfjoch_broker: Summation of more than 256 frames per image is split between the card and the host (`summation_mode` is then `hybrid`), and live Rugnux runs on such collections; pedestal runs take a receiver of their own. * jfjoch_broker (experimental): `stills_downsample` analyses 1 of every N serial images (chosen by a hash; the first 2000 all; the others written without analysis; reset to 1 by `/initialize`); `max_spot_count` left out means 250 for per-image indexing and up to 1000 for live stills; the live status gives the merge before the model work, the fitted resolution and the anomalous peak height of each unexplained blob. * jfjoch_broker: Without a `pcie` entry the broker runs on a mock of the card instead of the HLS simulation; its spot, ROI and azimuthal results are zero then. * Web frontend: Redesigned Rugnux live page; Vite 8, TypeScript 6, Mol* 5.13 and dependency security updates. * Tools: `jfjoch_fpga_stills_test` measures live stills on the cards with a stored dataset; `jfjoch_fpga_test` loses the HLS-simulation option. * Other: FFTW no longer fetched or linked (the CPU FFT indexer keeps the name `FFTW`); PCIe driver now GPL-2.0-or-later and third-party notices complete; several crash fixes. Reviewed-on: #98 Co-authored-by: Filip Leonarski <filip.leonarski@psi.ch>
59 lines
2.6 KiB
Metal
59 lines
2.6 KiB
Metal
// SPDX-FileCopyrightText: 2026 Filip Leonarski, Paul Scherrer Institute <filip.leonarski@psi.ch>
|
|
// SPDX-License-Identifier: GPL-3.0-only
|
|
|
|
// The side-chain field's direct sums (SideChainFieldMetal.h), compiled when first used from this source, which
|
|
// CMakeLists.txt copies into the binary after bragg_integration/BraggExp.h - with fast math off and no a*b+c
|
|
// contracted, as on the host, so that every operation below rounds as EvaluatorCPU::Contribution's does.
|
|
|
|
#include <metal_stdlib>
|
|
using namespace metal;
|
|
|
|
#pragma clang fp contract(off)
|
|
|
|
struct Atom { // FieldAtom
|
|
float x, y, z, b, occ;
|
|
int element;
|
|
};
|
|
|
|
struct Params {
|
|
uint rows, n_ops, first;
|
|
};
|
|
|
|
// SideChainField.cpp's SinCos2Pi(): sin and cos of 2 pi v, v taken to [-1/2, 1/2] by its nearest integer, the half
|
|
// angle by its Taylor series, then the double-angle formulas.
|
|
static void sincos_2pi(float v, thread float &sn, thread float &cs) {
|
|
const float x = 3.14159265f * (v - rint(v));
|
|
const float x2 = x * x;
|
|
const float s = x * (1 + x2 * (-1.f / 6 + x2 * (1.f / 120 + x2 * (-1.f / 5040 + x2 * (1.f / 362880 + x2 * (-1.f / 39916800))))));
|
|
const float c = 1 + x2 * (-1.f / 2 + x2 * (1.f / 24 + x2 * (-1.f / 720 + x2 * (1.f / 40320 + x2 * (-1.f / 3628800 + x2 * (1.f / 479001600))))));
|
|
sn = 2 * s * c;
|
|
cs = c * c - s * s;
|
|
}
|
|
|
|
// One thread per row m and state first + s: the state's atoms in order, each over the operators in order.
|
|
kernel void contribution(device const float *hx [[buffer(0)]], device const float *hy [[buffer(1)]],
|
|
device const float *hz [[buffer(2)]], device const float *ht [[buffer(3)]],
|
|
device const float *f0 [[buffer(4)]], device const float *stol2 [[buffer(5)]],
|
|
device const Atom *atoms [[buffer(6)]], device const int *start [[buffer(7)]],
|
|
constant Params &p [[buffer(8)]], device float2 *out [[buffer(9)]],
|
|
uint2 gid [[thread_position_in_grid]]) {
|
|
const uint m = gid.x, s = gid.y;
|
|
if (m >= p.rows)
|
|
return;
|
|
const float st = stol2[m];
|
|
float re = 0, im = 0;
|
|
for (int a = start[p.first + s]; a < start[p.first + s + 1]; a++) {
|
|
const Atom at = atoms[a];
|
|
const float fa = f0[uint(at.element) * p.rows + m] * at.occ * BraggExp(-at.b * st);
|
|
for (uint k = 0; k < p.n_ops; k++) {
|
|
const uint o = k * p.rows + m;
|
|
const float v = hx[o] * at.x + hy[o] * at.y + hz[o] * at.z + ht[o];
|
|
float sn, cs;
|
|
sincos_2pi(v, sn, cs);
|
|
re += fa * cs;
|
|
im += fa * sn;
|
|
}
|
|
}
|
|
out[s * p.rows + m] = float2(re, im);
|
|
}
|