Functions — github.com/clMathLibraries/clFFT

CodeHub MCPcopy Create free account

hub / github.com/clMathLibraries/clFFT / functions

Functions2,530 in github.com/clMathLibraries/clFFT

⨍ Functions 2,530◇ Types & classes 165

↓ 1 callersMethodGetPosition

src/library/generator.stockham.cpp:2029

↓ 1 callersMethodGetRadices

src/library/generator.stockham.cpp:356

↓ 1 callersMethodGetRadix

src/library/generator.stockham.cpp:2030

↓ 1 callersFunctionMD5_Final

src/library/md5sum.c:252

↓ 1 callersFunctionMD5_Init

src/library/md5sum.c:207

↓ 1 callersFunctionMD5_Update

src/library/md5sum.c:218

↓ 1 callersFunctionOffsetCalcLeadingDimensionBatched

generating string for calculating offset within sqaure transpose kernels (genTransposeKernelLeadingDimensionBatched)

src/library/generator.transpose.cpp:48

↓ 1 callersMethodPrint

src/statTimer/statisticalTimer.CPU.cpp:358

↓ 1 callersFunctionReadKernelFromFile

src/library/generator.stockham.cpp:261

↓ 1 callersMethodReserve

The caller can pre-allocate memory, to improve performance. nEvents is an approximate value for how many seperate events the caller will think they wi

src/statTimer/statisticalTimer.CPU.cpp:162

↓ 1 callersMethodReset

src/statTimer/statisticalTimer.CPU.cpp:138

↓ 1 callersMethodSetEnvelope

This routine will query the OpenCL context for it's devices and their hardware limitations, which we synthesize into a hardware "envelope". We only qu

src/library/plan.cpp:4651

↓ 1 callersMethodSetGrouping

src/library/generator.stockham.cpp:2033

↓ 1 callersMethodSetNextPass

src/library/generator.stockham.cpp:2032

↓ 1 callersMethodStop

src/statTimer/statisticalTimer.CPU.cpp:205

↓ 1 callersFunctionSwap_OffsetCalc

generating string for calculating offset within swap kernels (genSwapKernel)

src/library/generator.transpose.cpp:66

↓ 1 callersMethod__str__

src/scripts/perf/fftPerformanceTesting.py:72

↓ 1 callersMethod_step

src/scripts/perf/fftPerformanceTesting.py:217

↓ 1 callersMethod_step

src/scripts/perf/measurePerformance.py:486

↓ 1 callersMethodcalcFlops

src/statTimer/statisticalTimer.GPU.h:84

↓ 1 callersMethodchange_hermitian_to_real

/ strides and distance are those of the output (that is, the new real buffer)

src/tests/buffer.h:738

↓ 1 callersMethodchange_real_to_hermitian

/ strides and distance are those of the output (that is, the new hermitian buffer)

src/tests/buffer.h:722

↓ 1 callersFunctioncheckTimeOutPut

src/scripts/perf/measurePerformance.py:144

↓ 1 callersMethodcheck_memory_boundaries

src/tests/buffer_memory.h:82

↓ 1 callersFunctionclfftErrorStatusAsString

src/library/private.h:169

↓ 1 callersFunctionclfftInitBinaryCache

src/library/fft_binary_lookup.cpp:82

↓ 1 callersFunctionclfftInitRequestLibNoMemAlloc

src/library/fft_binary_lookup.cpp:69

↓ 1 callersFunctioncommatize

T could be 32-bit or 64-bit

src/statTimer/statisticalTimer.GPU.cpp:32

↓ 1 callersMethodcreatePlan

src/library/repo.cpp:312

↓ 1 callersMethoddeleteData

src/library/repo.h:62

↓ 1 callersMethoddeletePlan

src/library/repo.cpp:349

↓ 1 callersFunctiondiscoverCLPlatforms

src/callback-client/openCL.misc.cpp:357

↓ 1 callersMethodenqueue

src/library/enqueue.cpp:533

↓ 1 callersMethodenter

src/library/lock.h:79

↓ 1 callersMethodexclusive_create

src/library/fft_binary_lookup.cpp:117

↓ 1 callersMethodexecute

src/tests/fftw_transform.h:119

↓ 1 callersFunctionfactor

src/scripts/perf/measurePerformance.py:333

↓ 1 callersMethodfinalizeVariant

src/library/fft_binary_lookup.cpp:268

↓ 1 callersFunctionfloats_are_about_equal

src/tests/buffer.h:38

↓ 1 callersMethodfound

src/library/fft_binary_lookup.cpp:428

↓ 1 callersFunctiongenSwapKernelGeneral

swap lines. a more general kernel generator. this function accepts any ratio in theory. But in practice we restrict it to 1:2, 1:3, 1:5 and 1:10 ratio

src/library/generator.transpose.cpp:920

↓ 1 callersFunctiongenTransposeKernel

src/library/generator.transpose.gcn.cpp:352

↓ 1 callersFunctiongenTransposeKernelLeadingDimensionBatched

generate transpose kernel with square 2d matrix of row major with blocks along the leading dimension aka leading dimension batched Below is a matrix(r

src/library/generator.transpose.cpp:2756

↓ 1 callersFunctiongenTransposePrototype

src/library/generator.transpose.gcn.cpp:206

↓ 1 callersFunctiongenTransposePrototype

src/library/generator.transpose.cpp:162

↓ 1 callersFunctiongenTwiddleMath

Small snippet of code that multiplies the twiddle factors into the butterfiles. It is only emitted if the plan tells the generator that it wants the

src/library/generator.transpose.gcn.cpp:176

↓ 1 callersFunctiongenerate235Radices

src/scripts/perf/performanceUtility.py:33

↓ 1 callersMethodgenerateKernel

src/library/generator.stockham.cpp:4677

↓ 1 callersMethodgenerateKernel

src/library/generator.copy.cpp:621

↓ 1 callersMethodgenerateKernel

OpenCL does not take unicode strings as input, so this routine returns only ASCII strings Feed this generator the FFTPlan, and it returns the generate

src/library/generator.transpose.gcn.cpp:1139

↓ 1 callersFunctionget235RadicesNumberInRange

(minimum, maximum)

src/scripts/perf/measurePerformance.py:426

↓ 1 callersMethodgetAverageTime

src/statTimer/statisticalTimer.CPU.cpp:293

↓ 1 callersMethodgetMean

src/statTimer/statisticalTimer.CPU.cpp:251

↓ 1 callersMethodgetMinimumTime

src/statTimer/statisticalTimer.CPU.cpp:302

↓ 1 callersMethodgetProgram

src/library/fft_binary_lookup.cpp:663

↓ 1 callersFunctiongetSingleBinaryFromProgram

src/library/fft_binary_lookup.cpp:451

↓ 1 callersMethodgetStdDev

src/statTimer/statisticalTimer.CPU.cpp:285

↓ 1 callersFunctiongetStorageMode

src/library/fft_binary_lookup.cpp:251

↓ 1 callersMethodgetUniqueID

This function's purpose is to provide a mapping from a 'friendly' human readable text string to an index into internal data structures.

src/statTimer/statisticalTimer.CPU.cpp:231

↓ 1 callersMethodgetWorkSizes

src/library/generator.copy.cpp:582

↓ 1 callersFunctionget_next_num_with_1235_factors

src/scripts/perf/measurePerformance.py:386

↓ 1 callersMethodincrease_allocated_memory

/ N.B. memory will be cleared after this

src/tests/buffer_memory.h:114

↓ 1 callersMethodinitParams

src/library/generator.stockham.cpp:4476

↓ 1 callersMethodinitParams

src/library/generator.copy.cpp:512

↓ 1 callersMethodinitParams

src/library/generator.transpose.gcn.cpp:1026

↓ 1 callersFunctioninitializeCL

src/callback-client/openCL.misc.cpp:409

↓ 1 callersMethodinput_layout

src/tests/cl_transform.h:548

↓ 1 callersFunctionisPrime

src/scripts/perf/measurePerformance.py:306

↓ 1 callersFunctionis_numeric_type

src/scripts/perf/fftPerformanceTesting.py:152

↓ 1 callersMethodleave

src/library/lock.h:101

↓ 1 callersFunctionmake_directory

src/library/fft_binary_lookup.cpp:675

↓ 1 callersFunctionmax_mem_available_on_cl_device

src/tests/test_constants.cpp:85

↓ 1 callersFunctionmax_memory_size_for_one_buffer

the CPU can have a lot of memory hanging off of it. in these cases and when CPU is the cl device, huge amounts of memory might be dedicated to each bu

src/tests/accuracy_test_random.cpp:80

↓ 1 callersFunctionmax_problem_size

(exe, layout, precision, device)

src/scripts/perf/fftPerformanceTesting.py:117

↓ 1 callersFunctionmd5sum

src/library/md5sum.c:305

↓ 1 callersMethodnumber_of_data_points

src/tests/buffer.h:878

↓ 1 callersMethodnumber_of_data_points_single_batch

src/tests/buffer.h:867

↓ 1 callersMethodoutput_layout

src/tests/cl_transform.h:554

↓ 1 callersMethodplaceness

src/tests/cl_transform.h:458

↓ 1 callersFunctionplotFromDataFile

src/scripts/perf/plotPerformance.py:69

↓ 1 callersFunctionpow235

src/library/plan.cpp:38

↓ 1 callersFunctionprettyPrintCLPlatforms

src/client/openCL.misc.cpp:214

↓ 1 callersMethodpruneOutliers

src/statTimer/statisticalTimer.CPU.cpp:318

↓ 1 callersFunctionrandom_supported_problem_size

src/tests/accuracy_test_random.cpp:113

↓ 1 callersMethodread_plan_from_file

src/tests/cl_transform.h:1159

↓ 1 callersMethodrefresh_plan

src/tests/cl_transform.h:395

↓ 1 callersMethodreleaseResources

src/library/repo.cpp:40

↓ 1 callersMethodretrieveDeviceAndDriverInfo

src/library/fft_binary_lookup.cpp:698

↓ 1 callersFunctionrunTest

/////////////////////////////////////////////////////////////////////////// Run a simple test for CUDA FFT ///////////////////////////////////////////

src/cuFFT-client/cuFFT-client.cpp:113

↓ 1 callersFunctionselectAction

src/library/plan.cpp:401

↓ 1 callersMethodsetName

src/library/lock.h:74

↓ 1 callersMethodsetNormalize

src/statTimer/statisticalTimer.CPU.cpp:180

↓ 1 callersMethodsetProgram

src/library/fft_binary_lookup.cpp:668

↓ 1 callersMethodset_input_precallback_userdatatype

src/tests/cl_transform.h:672

↓ 1 callersMethodsetclProgram

src/library/repo.cpp:203

↓ 1 callersFunctionsize_of_one_point

src/tests/accuracy_test_random.cpp:58

↓ 1 callersFunctionsplit1D_for_inplace

src/library/plan.cpp:69

↓ 1 callersFunctionvalidate_number_for_1235

src/scripts/perf/measurePerformance.py:364

↓ 1 callersMethodvariantRaw

src/library/fft_binary_lookup.cpp:239

↓ 1 callersMethodwrite_plan_to_file

src/tests/cl_transform.h:1149

← previous next →201–300 of 2,530, ranked by callers