Build Packages / Create release (push) Successful in 16s
Build Packages / build:rugnux:aarch64 (cross) (push) Successful in 8m27s
Build Packages / build:rugnux-tgz (x86_64) (push) Successful in 9m15s
Build Packages / build:viewer-tgz:cpu (push) Successful in 10m11s
Build Packages / build:viewer-tgz:cuda (push) Successful in 12m6s
Build Packages / build:rpm (rocky8_nocuda) (push) Successful in 15m44s
Build Packages / build:rpm (rocky9_nocuda) (push) Successful in 16m1s
Build Packages / build:windows:nocuda (push) Successful in 17m29s
Build Packages / build:windows:cuda (push) Successful in 19m58s
Build Packages / HDF5 consumer tests (DIALS, XDS) (push) Successful in 24m7s
Build Packages / build:rpm (ubuntu2404_nocuda) (push) Successful in 19m8s
Build Packages / build:rugnux:windows (push) Successful in 10m58s
Build Packages / build:rpm (ubuntu2204_nocuda) (push) Successful in 20m46s
Build Packages / Generate python client (push) Successful in 53s
Build Packages / build:rpm (rocky8_sls9) (push) Successful in 20m13s
Build Packages / Build documentation (push) Successful in 1m36s
Build Packages / build:rpm (rocky9_sls9) (push) Successful in 19m57s
Build Packages / build:rpm (rocky8) (push) Successful in 18m7s
Build Packages / build:rpm (rocky9) (push) Successful in 18m54s
Build Packages / build:rpm (ubuntu2204) (push) Successful in 19m32s
Build Packages / build:rpm (ubuntu2404) (push) Successful in 17m30s
Build Packages / Unit tests (push) Successful in 1h39m2s
* Fixed `jfjoch_broker` cancelling every data collection with a CUDA "out of memory" error after long operation: GPU memory no longer leaks with each collection. * Rugnux scales a rotation sweep until the per-frame scales settle instead of for a fixed three rounds, and says so when they did not - merged intensities, and the space group, resolution cut and frame rejection read off them, change accordingly; `--scaling-iterations` is now the cap on that loop (default 100). * Rugnux places every frame of a marCCD, SMV or miniCBF series at the spindle angle its own header states, so a series with missing frames, or with angles written modulo 360, is no longer read at the wrong geometry or refused. * Every rotation run writes two diagnostic files beside its reflections: `<prefix>_detector.jpg`, the detector projection with the pixel mask and the detected beam-stop shadow drawn on it, and `<prefix>_plot.txt`, one row per image. Reviewed-on: #82 Co-authored-by: Filip Leonarski <filip.leonarski@psi.ch>
68 lines
1.9 KiB
Plaintext
68 lines
1.9 KiB
Plaintext
// SPDX-FileCopyrightText: 2024 Filip Leonarski, Paul Scherrer Institute <filip.leonarski@psi.ch>
|
|
// SPDX-License-Identifier: GPL-3.0-only
|
|
|
|
#include <fstream>
|
|
#include <atomic>
|
|
|
|
#include "CUDAWrapper.h"
|
|
#include "JFJochException.h"
|
|
|
|
inline void cuda_err(cudaError_t val) {
|
|
if (val != cudaSuccess)
|
|
throw JFJochException(JFJochExceptionCategory::GPUCUDAError, cudaGetErrorString(val));
|
|
}
|
|
|
|
int32_t get_gpu_count() {
|
|
int device_count;
|
|
cudaError_t val = cudaGetDeviceCount(&device_count);
|
|
switch (val) {
|
|
case cudaSuccess:
|
|
return device_count;
|
|
case cudaErrorNoDevice:
|
|
case cudaErrorInsufficientDriver:
|
|
return 0;
|
|
default:
|
|
throw JFJochException(JFJochExceptionCategory::GPUCUDAError, cudaGetErrorString(val));
|
|
}
|
|
|
|
}
|
|
|
|
std::vector<std::string> get_gpu_names() {
|
|
std::vector<std::string> names;
|
|
const int32_t count = get_gpu_count();
|
|
names.reserve(count);
|
|
for (int32_t i = 0; i < count; i++) {
|
|
cudaDeviceProp prop{};
|
|
// A device that cannot be queried still exists and still gets work, so it is listed - just
|
|
// without a name. Losing the whole list over one unreadable device would be worse.
|
|
if (cudaGetDeviceProperties(&prop, i) == cudaSuccess)
|
|
names.emplace_back(prop.name);
|
|
else
|
|
names.emplace_back("unknown GPU");
|
|
}
|
|
return names;
|
|
}
|
|
|
|
void set_gpu(int32_t dev_id) {
|
|
auto dev_count = get_gpu_count();
|
|
|
|
// Ignore if no GPU present
|
|
if (dev_count > 0) {
|
|
if ((dev_id < 0) || (dev_id >= dev_count))
|
|
throw JFJochException(JFJochExceptionCategory::InputParameterInvalid, "Device ID cannot be negative");
|
|
|
|
cuda_err(cudaSetDevice(dev_id));
|
|
}
|
|
}
|
|
|
|
void pin_gpu() {
|
|
static std::atomic<uint32_t> counter{0};
|
|
auto dev_count = get_gpu_count();
|
|
if (dev_count > 0)
|
|
set_gpu(counter.fetch_add(1) % dev_count);
|
|
}
|
|
|
|
void cuda_clear_error() {
|
|
cudaGetLastError();
|
|
}
|