1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
|
#ifndef WS_MS_GRIDDER_H
#define WS_MS_GRIDDER_H
#include "msgridder.h"
#include "wstackinggridder.h"
#include "../structures/resources.h"
#include <casacore/casa/Arrays/Array.h>
#include <casacore/tables/Tables/ArrayColumn.h>
#include <aocommon/image.h>
#include <aocommon/lane.h>
#include <aocommon/multibanddata.h>
#include <complex>
#include <memory>
#include <thread>
namespace wsclean {
class WSMSGridder final : public MsGridder {
public:
typedef WStackingGridder<float> GridderType;
WSMSGridder(const Settings& settings, const Resources& resources,
MsProviderCollection& ms_provider_collection);
~WSMSGridder() noexcept final;
size_t GetNInversionPasses() const final { return _gridder->NPasses(); }
void StartInversion() final;
void StartInversionPass(size_t pass_index) final;
size_t GridMeasurementSet(const MsProviderCollection::MsData& ms_data) final;
void FinishInversionPass(size_t pass_index) final;
void FinishInversion() final;
size_t GetNPredictPasses() const final { return _gridder->NPasses(); }
void StartPredict(std::vector<aocommon::Image>&& images) final;
void StartPredictPass(size_t pass_index) final;
size_t PredictMeasurementSet(
const MsProviderCollection::MsData& ms_data) final;
void FinishPredictPass(size_t pass_index) final;
void FinishPredict() final;
std::vector<aocommon::Image> ResultImages() final {
if (IsComplex())
return {std::move(_realImage), std::move(_imaginaryImage)};
else
return {std::move(_realImage)};
}
void FreeImagingData() final { _gridder.reset(); }
size_t AntialiasingKernelSize() const { return _antialiasingKernelSize; }
size_t OverSamplingFactor() const { return _overSamplingFactor; }
bool HasNWSize() const { return NWWidth() != 0 || NWHeight() != 0; }
size_t NWWidth() const { return GetSettings().widthForNWCalculation; }
size_t NWHeight() const { return GetSettings().heightForNWCalculation; }
double NWFactor() const { return GetSettings().nWLayersFactor; }
private:
struct InversionWorkSample {
double uInLambda, vInLambda, wInLambda;
std::complex<float> sample;
};
struct PredictionWorkItem {
std::array<double, 3> uvw;
std::unique_ptr<std::complex<float>[]> data;
size_t data_desc_id;
size_t rowId;
};
void countSamplesPerLayer(MsProviderCollection::MsData& msData);
size_t GetSuggestedWGridSize() const final;
void startInversionWorkThreads(size_t maxChannelCount);
void finishInversionWorkThreads();
void workThreadPerSample(aocommon::Lane<InversionWorkSample>* workLane);
void predictCalcThread(aocommon::Lane<PredictionWorkItem>* input_lane,
aocommon::Lane<PredictionWorkItem>* output_lane,
const aocommon::MultiBandData* bands);
void predictWriteThread(
aocommon::Lane<PredictionWorkItem>* sampling_work_lane,
const MsProviderCollection::MsData* ms_data, GainMode gain_mode);
std::unique_ptr<GridderType> _gridder;
std::vector<aocommon::Lane<InversionWorkSample>> _inversionCPULanes;
std::vector<std::thread> _threadGroup;
size_t _antialiasingKernelSize, _overSamplingFactor;
const Resources _resources;
size_t _laneBufferSize;
aocommon::Image _realImage;
aocommon::Image _imaginaryImage;
};
} // namespace wsclean
#endif
|
