MNN/source/backend/opencl/core/runtime/OpenCLWrapper.cpp

646 lines
28 KiB
C++
Raw Normal View History

2019-04-17 10:49:11 +08:00
//
// OpenCLWrapper.cpp
// MNN
//
// Created by MNN on 2019/02/28.
// Copyright © 2018, Alibaba Group Holding Limited
//
2019-12-27 22:16:57 +08:00
#include "backend/opencl/core/runtime/OpenCLWrapper.hpp"
2020-12-15 14:12:35 +08:00
#ifdef WIN32
#include <libloaderapi.h>
#else
2019-04-17 10:49:11 +08:00
#include <dlfcn.h>
2020-11-05 16:41:56 +08:00
#endif
2019-04-17 10:49:11 +08:00
#include <memory>
#include <string>
#include <vector>
2021-04-28 18:02:10 +08:00
#include <mutex>
2019-04-17 10:49:11 +08:00
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
#ifdef MNN_USE_LIB_WRAPPER
2019-04-17 10:49:11 +08:00
namespace MNN {
bool OpenCLSymbols::LoadOpenCLLibrary() {
if (handle_ != nullptr) {
return true;
}
2022-12-30 15:18:58 +08:00
static const std::vector<std::string> gOpencl_library_paths = {
#if defined(__APPLE__) || defined(__MACOSX)
"libOpenCL.so", "/System/Library/Frameworks/OpenCL.framework/OpenCL"
#elif defined(__ANDROID__)
"libOpenCL.so",
"libGLES_mali.so",
"libmali.so",
#if defined(__aarch64__)
// Qualcomm Adreno
"/system/vendor/lib64/libOpenCL.so",
"/system/lib64/libOpenCL.so",
// Mali
"/system/vendor/lib64/egl/libGLES_mali.so",
"/system/lib64/egl/libGLES_mali.so",
#else
// Qualcomm Adreno
"/system/vendor/lib/libOpenCL.so", "/system/lib/libOpenCL.so",
// Mali
"/system/vendor/lib/egl/libGLES_mali.so", "/system/lib/egl/libGLES_mali.so",
// other
"/system/vendor/lib/libPVROCL.so", "/data/data/org.pocl.libs/files/lib/libpocl.so"
#endif
#elif defined(__linux__)
"/usr/lib/libOpenCL.so",
"/usr/local/lib/libOpenCL.so",
"/usr/local/lib/libpocl.so",
"/usr/lib64/libOpenCL.so",
"/usr/lib32/libOpenCL.so",
"libOpenCL.so"
/*
* 0: System32, 1: SysWOW64
* --------------------------------------
* | Real CPU / | x64 | x86 |
* | / Target CPU | | |
* --------------------------------------
* | x64 | 0 / 1 | 1 |
* --------------------------------------
* | x86 | Error | 0 |
* --------------------------------------
* 0 / 1: 0 if OpenCL.dll (System32, 64bit on x64), otherwise 1 (SysWOW64, 32bit compatible on 64bit OS)
*/
#elif defined(_WIN64)
"C:/Windows/System32/OpenCL.dll",
"C:/Windows/SysWOW64/OpenCL.dll"
#elif defined(_WIN32)
"C:/Windows/SysWOW64/OpenCL.dll",
"C:/Windows/System32/OpenCL.dll"
#endif
};
2019-04-17 10:49:11 +08:00
for (const auto &opencl_lib : gOpencl_library_paths) {
if (LoadLibraryFromPath(opencl_lib)) {
return true;
}
}
return false;
}
bool OpenCLSymbols::UnLoadOpenCLLibrary() {
if (handle_ != nullptr) {
2020-12-15 14:12:35 +08:00
#if defined(WIN32)
2020-11-05 16:41:56 +08:00
if (FreeLibrary(handle_) == 0) {
#else
2019-04-17 10:49:11 +08:00
if (dlclose(handle_) != 0) {
2020-11-05 16:41:56 +08:00
#endif
2019-04-17 10:49:11 +08:00
return false;
}
handle_ = nullptr;
return true;
}
return true;
}
bool OpenCLSymbols::isError() {
return mIsError;
}
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
bool OpenCLSymbols::isSvmError() {
return mSvmError;
}
2022-12-30 15:18:58 +08:00
bool OpenCLSymbols::isPropError() {
return mPropError;
}
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
2019-04-17 10:49:11 +08:00
bool OpenCLSymbols::LoadLibraryFromPath(const std::string &library_path) {
2020-12-15 14:12:35 +08:00
#if defined(WIN32)
2020-11-05 16:41:56 +08:00
handle_ = LoadLibraryA(library_path.c_str());
if (handle_ == nullptr) {
return false;
}
#define MNN_LOAD_FUNCTION_PTR(func_name) func_name = reinterpret_cast<func_name##Func>(GetProcAddress(handle_, #func_name)); \
if(func_name == nullptr){ \
mIsError = true; \
}
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
#define MNN_LOAD_SVM_PTR(func_name) func_name = reinterpret_cast<func_name##Func>(GetProcAddress(handle_, #func_name)); \
if(func_name == nullptr){ \
mSvmError = true; \
}
2022-12-30 15:18:58 +08:00
#define MNN_LOAD_PROP_PTR(func_name) func_name = reinterpret_cast<func_name##Func>(GetProcAddress(handle_, #func_name)); \
if(func_name == nullptr){ \
mPropError = true; \
}
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
2020-11-05 16:41:56 +08:00
#else
2019-04-17 10:49:11 +08:00
handle_ = dlopen(library_path.c_str(), RTLD_NOW | RTLD_LOCAL);
if (handle_ == nullptr) {
return false;
}
#define MNN_LOAD_FUNCTION_PTR(func_name) func_name = reinterpret_cast<func_name##Func>(dlsym(handle_, #func_name)); \
if(func_name == nullptr){ \
mIsError = true; \
}
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
#define MNN_LOAD_SVM_PTR(func_name) func_name = reinterpret_cast<func_name##Func>(dlsym(handle_, #func_name)); \
if(func_name == nullptr){ \
mSvmError = true; \
}
2022-12-30 15:18:58 +08:00
#define MNN_LOAD_PROP_PTR(func_name) func_name = reinterpret_cast<func_name##Func>(dlsym(handle_, #func_name)); \
if(func_name == nullptr){ \
mPropError = true; \
}
2020-11-05 16:41:56 +08:00
#endif
2019-04-17 10:49:11 +08:00
MNN_LOAD_FUNCTION_PTR(clGetPlatformIDs);
MNN_LOAD_FUNCTION_PTR(clGetPlatformInfo);
MNN_LOAD_FUNCTION_PTR(clBuildProgram);
MNN_LOAD_FUNCTION_PTR(clEnqueueNDRangeKernel);
MNN_LOAD_FUNCTION_PTR(clSetKernelArg);
MNN_LOAD_FUNCTION_PTR(clReleaseKernel);
MNN_LOAD_FUNCTION_PTR(clCreateProgramWithSource);
MNN_LOAD_FUNCTION_PTR(clCreateBuffer);
MNN_LOAD_FUNCTION_PTR(clCreateImage2D);
MNN_LOAD_FUNCTION_PTR(clRetainKernel);
MNN_LOAD_FUNCTION_PTR(clCreateKernel);
MNN_LOAD_FUNCTION_PTR(clGetProgramInfo);
MNN_LOAD_FUNCTION_PTR(clFlush);
MNN_LOAD_FUNCTION_PTR(clFinish);
MNN_LOAD_FUNCTION_PTR(clReleaseProgram);
MNN_LOAD_FUNCTION_PTR(clRetainContext);
MNN_LOAD_FUNCTION_PTR(clGetContextInfo);
MNN_LOAD_FUNCTION_PTR(clCreateProgramWithBinary);
MNN_LOAD_FUNCTION_PTR(clCreateCommandQueue);
MNN_LOAD_FUNCTION_PTR(clReleaseCommandQueue);
MNN_LOAD_FUNCTION_PTR(clEnqueueMapBuffer);
MNN_LOAD_FUNCTION_PTR(clEnqueueMapImage);
MNN_LOAD_FUNCTION_PTR(clRetainProgram);
MNN_LOAD_FUNCTION_PTR(clGetProgramBuildInfo);
MNN_LOAD_FUNCTION_PTR(clEnqueueReadBuffer);
MNN_LOAD_FUNCTION_PTR(clEnqueueWriteBuffer);
2022-12-30 15:18:58 +08:00
MNN_LOAD_FUNCTION_PTR(clEnqueueCopyBuffer);
2019-04-17 10:49:11 +08:00
MNN_LOAD_FUNCTION_PTR(clWaitForEvents);
MNN_LOAD_FUNCTION_PTR(clReleaseEvent);
MNN_LOAD_FUNCTION_PTR(clCreateContext);
MNN_LOAD_FUNCTION_PTR(clCreateContextFromType);
MNN_LOAD_FUNCTION_PTR(clReleaseContext);
MNN_LOAD_FUNCTION_PTR(clRetainCommandQueue);
MNN_LOAD_FUNCTION_PTR(clEnqueueUnmapMemObject);
MNN_LOAD_FUNCTION_PTR(clRetainMemObject);
MNN_LOAD_FUNCTION_PTR(clReleaseMemObject);
MNN_LOAD_FUNCTION_PTR(clGetDeviceInfo);
MNN_LOAD_FUNCTION_PTR(clGetDeviceIDs);
MNN_LOAD_FUNCTION_PTR(clRetainEvent);
MNN_LOAD_FUNCTION_PTR(clGetKernelWorkGroupInfo);
MNN_LOAD_FUNCTION_PTR(clGetEventInfo);
MNN_LOAD_FUNCTION_PTR(clGetEventProfilingInfo);
2021-03-03 16:02:51 +08:00
MNN_LOAD_FUNCTION_PTR(clGetMemObjectInfo);
2019-04-17 10:49:11 +08:00
MNN_LOAD_FUNCTION_PTR(clGetImageInfo);
MNN_LOAD_FUNCTION_PTR(clEnqueueCopyImage);
2019-12-27 22:16:57 +08:00
MNN_LOAD_FUNCTION_PTR(clEnqueueReadImage);
MNN_LOAD_FUNCTION_PTR(clEnqueueWriteImage);
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
2022-12-30 15:18:58 +08:00
MNN_LOAD_PROP_PTR(clCreateCommandQueueWithProperties);
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
MNN_LOAD_SVM_PTR(clSVMAlloc);
MNN_LOAD_SVM_PTR(clSVMFree);
MNN_LOAD_SVM_PTR(clEnqueueSVMMap);
MNN_LOAD_SVM_PTR(clEnqueueSVMUnmap);
MNN_LOAD_SVM_PTR(clSetKernelArgSVMPointer);
2019-04-17 10:49:11 +08:00
#undef MNN_LOAD_FUNCTION_PTR
return true;
}
2021-04-28 18:02:10 +08:00
OpenCLSymbolsOperator* OpenCLSymbolsOperator::createOpenCLSymbolsOperatorSingleInstance() {
2022-12-30 15:18:58 +08:00
static std::once_flag sFlagInitSymbols;
static OpenCLSymbolsOperator* gInstance = nullptr;
2021-04-28 18:02:10 +08:00
std::call_once(sFlagInitSymbols, [&]() {
gInstance = new OpenCLSymbolsOperator;
});
return gInstance;
}
2019-04-17 10:49:11 +08:00
std::shared_ptr<OpenCLSymbols> OpenCLSymbolsOperator::gOpenclSymbols;
OpenCLSymbols *OpenCLSymbolsOperator::getOpenclSymbolsPtr() {
return gOpenclSymbols.get();
}
OpenCLSymbolsOperator::OpenCLSymbolsOperator() {
#ifdef LOG_VERBOSE
MNN_PRINT("start OpenCLSymbolsOperator !\n");
#endif
if (gOpenclSymbols.get() == nullptr) {
gOpenclSymbols.reset(new OpenCLSymbols());
} else {
#ifdef LOG_VERBOSE
MNN_PRINT(" OpenCLSymbols already now !\n");
#endif
}
if (false == gOpenclSymbols->LoadOpenCLLibrary()) {
gOpenclSymbols.reset();
}
#ifdef LOG_VERBOSE
MNN_PRINT("end OpenCLSymbolsOperator !\n");
#endif
}
OpenCLSymbolsOperator::~OpenCLSymbolsOperator() {
#ifdef LOG_VERBOSE
MNN_PRINT("start ~OpenCLSymbolsOperator !\n");
#endif
if (nullptr == gOpenclSymbols.get()) {
return;
}
gOpenclSymbols.get()->UnLoadOpenCLLibrary();
#ifdef LOG_VERBOSE
MNN_PRINT("end ~OpenCLSymbolsOperator !\n");
#endif
}
} // namespace MNN
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clGetPlatformIDs(cl_uint num_entries, cl_platform_id *platforms, cl_uint *num_platforms) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clGetPlatformIDs;
MNN_CHECK_NOTNULL(func);
return func(num_entries, platforms, num_platforms);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clGetPlatformInfo(cl_platform_id platform, cl_platform_info param_name, size_t param_value_size,
2019-04-17 10:49:11 +08:00
void *param_value, size_t *param_value_size_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clGetPlatformInfo;
MNN_CHECK_NOTNULL(func);
return func(platform, param_name, param_value_size, param_value, param_value_size_ret);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clGetDeviceIDs(cl_platform_id platform, cl_device_type device_type, cl_uint num_entries, cl_device_id *devices,
2019-04-17 10:49:11 +08:00
cl_uint *num_devices) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clGetDeviceIDs;
MNN_CHECK_NOTNULL(func);
return func(platform, device_type, num_entries, devices, num_devices);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clGetDeviceInfo(cl_device_id device, cl_device_info param_name, size_t param_value_size, void *param_value,
2019-04-17 10:49:11 +08:00
size_t *param_value_size_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clGetDeviceInfo;
MNN_CHECK_NOTNULL(func);
return func(device, param_name, param_value_size, param_value, param_value_size_ret);
}
2020-11-05 16:41:56 +08:00
cl_context CL_API_CALL clCreateContext(const cl_context_properties *properties, cl_uint num_devices, const cl_device_id *devices,
2019-04-17 10:49:11 +08:00
void(CL_CALLBACK *pfn_notify)(const char *, const void *, size_t, void *), void *user_data,
cl_int *errcode_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clCreateContext;
MNN_CHECK_NOTNULL(func);
return func(properties, num_devices, devices, pfn_notify, user_data, errcode_ret);
}
2020-11-05 16:41:56 +08:00
cl_context CL_API_CALL clCreateContextFromType(const cl_context_properties *properties, cl_device_type device_type,
2019-04-17 10:49:11 +08:00
void(CL_CALLBACK *pfn_notify)(const char *, const void *, size_t, void *),
void *user_data, cl_int *errcode_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clCreateContextFromType;
MNN_CHECK_NOTNULL(func);
return func(properties, device_type, pfn_notify, user_data, errcode_ret);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clRetainContext(cl_context context) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clRetainContext;
MNN_CHECK_NOTNULL(func);
return func(context);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clReleaseContext(cl_context context) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clReleaseContext;
MNN_CHECK_NOTNULL(func);
return func(context);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clGetContextInfo(cl_context context, cl_context_info param_name, size_t param_value_size, void *param_value,
2019-04-17 10:49:11 +08:00
size_t *param_value_size_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clGetContextInfo;
MNN_CHECK_NOTNULL(func);
return func(context, param_name, param_value_size, param_value, param_value_size_ret);
}
2020-11-05 16:41:56 +08:00
cl_program CL_API_CALL clCreateProgramWithSource(cl_context context, cl_uint count, const char **strings, const size_t *lengths,
2019-04-17 10:49:11 +08:00
cl_int *errcode_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clCreateProgramWithSource;
MNN_CHECK_NOTNULL(func);
return func(context, count, strings, lengths, errcode_ret);
}
2020-11-05 16:41:56 +08:00
cl_program CL_API_CALL clCreateProgramWithBinary(cl_context context, cl_uint count, const cl_device_id * device_list, const size_t* length, const unsigned char ** buffer, cl_int * binary_status, cl_int * errcode_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clCreateProgramWithBinary;
MNN_CHECK_NOTNULL(func);
return func(context, count, device_list, length, buffer, binary_status, errcode_ret);
}
cl_int CL_API_CALL clGetProgramInfo(cl_program program, cl_program_info param_name, size_t param_value_size, void *param_value,
2019-04-17 10:49:11 +08:00
size_t *param_value_size_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clGetProgramInfo;
MNN_CHECK_NOTNULL(func);
return func(program, param_name, param_value_size, param_value, param_value_size_ret);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clGetProgramBuildInfo(cl_program program, cl_device_id device, cl_program_build_info param_name,
2019-04-17 10:49:11 +08:00
size_t param_value_size, void *param_value, size_t *param_value_size_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clGetProgramBuildInfo;
MNN_CHECK_NOTNULL(func);
return func(program, device, param_name, param_value_size, param_value, param_value_size_ret);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clRetainProgram(cl_program program) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clRetainProgram;
MNN_CHECK_NOTNULL(func);
return func(program);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clReleaseProgram(cl_program program) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clReleaseProgram;
MNN_CHECK_NOTNULL(func);
return func(program);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clBuildProgram(cl_program program, cl_uint num_devices, const cl_device_id *device_list, const char *options,
2019-04-17 10:49:11 +08:00
void(CL_CALLBACK *pfn_notify)(cl_program program, void *user_data), void *user_data) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clBuildProgram;
MNN_CHECK_NOTNULL(func);
return func(program, num_devices, device_list, options, pfn_notify, user_data);
}
2020-11-05 16:41:56 +08:00
cl_kernel CL_API_CALL clCreateKernel(cl_program program, const char *kernelName, cl_int *errcode_ret) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clCreateKernel;
MNN_CHECK_NOTNULL(func);
return func(program, kernelName, errcode_ret);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clRetainKernel(cl_kernel kernel) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clRetainKernel;
MNN_CHECK_NOTNULL(func);
return func(kernel);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clReleaseKernel(cl_kernel kernel) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clReleaseKernel;
MNN_CHECK_NOTNULL(func);
return func(kernel);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clSetKernelArg(cl_kernel kernel, cl_uint arg_index, size_t arg_size, const void *arg_value) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clSetKernelArg;
MNN_CHECK_NOTNULL(func);
return func(kernel, arg_index, arg_size, arg_value);
}
2020-11-05 16:41:56 +08:00
cl_mem CL_API_CALL clCreateBuffer(cl_context context, cl_mem_flags flags, size_t size, void *host_ptr, cl_int *errcode_ret) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clCreateBuffer;
MNN_CHECK_NOTNULL(func);
return func(context, flags, size, host_ptr, errcode_ret);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clRetainMemObject(cl_mem memobj) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clRetainMemObject;
MNN_CHECK_NOTNULL(func);
return func(memobj);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clReleaseMemObject(cl_mem memobj) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clReleaseMemObject;
MNN_CHECK_NOTNULL(func);
return func(memobj);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clGetImageInfo(cl_mem image, cl_image_info param_name, size_t param_value_size, void *param_value,
2019-04-17 10:49:11 +08:00
size_t *param_value_size_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clGetImageInfo;
MNN_CHECK_NOTNULL(func);
return func(image, param_name, param_value_size, param_value, param_value_size_ret);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clRetainCommandQueue(cl_command_queue command_queue) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clRetainCommandQueue;
MNN_CHECK_NOTNULL(func);
return func(command_queue);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clReleaseCommandQueue(cl_command_queue command_queue) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clReleaseCommandQueue;
MNN_CHECK_NOTNULL(func);
return func(command_queue);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clEnqueueReadBuffer(cl_command_queue command_queue, cl_mem buffer, cl_bool blocking_read, size_t offset,
2019-04-17 10:49:11 +08:00
size_t size, void *ptr, cl_uint num_events_in_wait_list, const cl_event *event_wait_list,
cl_event *event) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clEnqueueReadBuffer;
MNN_CHECK_NOTNULL(func);
return func(command_queue, buffer, blocking_read, offset, size, ptr, num_events_in_wait_list, event_wait_list,
event);
}
2022-12-30 15:18:58 +08:00
cl_int CL_API_CALL
clEnqueueCopyBuffer(cl_command_queue command_queue,
cl_mem src_buffer,
cl_mem dst_buffer,
size_t src_offset,
size_t dst_offset,
size_t size,
cl_uint num_events_in_wait_list,
const cl_event * event_wait_list,
cl_event * event) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clEnqueueCopyBuffer;
MNN_CHECK_NOTNULL(func);
return func(command_queue, src_buffer, dst_buffer, src_offset, dst_offset, size, num_events_in_wait_list, event_wait_list, event);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clEnqueueWriteBuffer(cl_command_queue command_queue, cl_mem buffer, cl_bool blocking_write, size_t offset,
2019-04-17 10:49:11 +08:00
size_t size, const void *ptr, cl_uint num_events_in_wait_list,
const cl_event *event_wait_list, cl_event *event) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clEnqueueWriteBuffer;
MNN_CHECK_NOTNULL(func);
return func(command_queue, buffer, blocking_write, offset, size, ptr, num_events_in_wait_list, event_wait_list,
event);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clEnqueueReadImage(cl_command_queue command_queue, cl_mem cl_image, cl_bool is_block, const size_t * origin, const size_t * region, size_t row_pitch,
2019-12-27 22:16:57 +08:00
size_t slice_pitch, void * ptr, cl_uint num_events_in_wait_list, const cl_event * event_wait_list, cl_event * event) CL_API_SUFFIX__VERSION_1_0 {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clEnqueueReadImage;
MNN_CHECK_NOTNULL(func);
return func(command_queue, cl_image, is_block, origin, region, row_pitch, slice_pitch, ptr, num_events_in_wait_list, event_wait_list,
event);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clEnqueueWriteImage(cl_command_queue command_queue, cl_mem cl_image, cl_bool is_block, const size_t * origin, const size_t * region, size_t row_pitch,
2019-12-27 22:16:57 +08:00
size_t slice_pitch, const void * ptr, cl_uint num_events_in_wait_list, const cl_event * event_wait_list, cl_event * event) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clEnqueueWriteImage;
MNN_CHECK_NOTNULL(func);
return func(command_queue, cl_image, is_block, origin, region, row_pitch, slice_pitch, ptr, num_events_in_wait_list, event_wait_list,
event);
}
2020-11-05 16:41:56 +08:00
void* CL_API_CALL clEnqueueMapBuffer(cl_command_queue command_queue, cl_mem buffer, cl_bool blocking_map, cl_map_flags map_flags,
2019-04-17 10:49:11 +08:00
size_t offset, size_t size, cl_uint num_events_in_wait_list, const cl_event *event_wait_list,
cl_event *event, cl_int *errcode_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clEnqueueMapBuffer;
MNN_CHECK_NOTNULL(func);
return func(command_queue, buffer, blocking_map, map_flags, offset, size, num_events_in_wait_list, event_wait_list,
event, errcode_ret);
}
2020-11-05 16:41:56 +08:00
void* CL_API_CALL clEnqueueMapImage(cl_command_queue command_queue, cl_mem image, cl_bool blocking_map, cl_map_flags map_flags,
2019-04-17 10:49:11 +08:00
const size_t *origin, const size_t *region, size_t *image_row_pitch, size_t *image_slice_pitch,
cl_uint num_events_in_wait_list, const cl_event *event_wait_list, cl_event *event,
cl_int *errcode_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clEnqueueMapImage;
MNN_CHECK_NOTNULL(func);
return func(command_queue, image, blocking_map, map_flags, origin, region, image_row_pitch, image_slice_pitch,
num_events_in_wait_list, event_wait_list, event, errcode_ret);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clEnqueueUnmapMemObject(cl_command_queue command_queue, cl_mem memobj, void *mapped_ptr,
2019-04-17 10:49:11 +08:00
cl_uint num_events_in_wait_list, const cl_event *event_wait_list, cl_event *event) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clEnqueueUnmapMemObject;
MNN_CHECK_NOTNULL(func);
return func(command_queue, memobj, mapped_ptr, num_events_in_wait_list, event_wait_list, event);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clGetKernelWorkGroupInfo(cl_kernel kernel, cl_device_id device, cl_kernel_work_group_info param_name,
2019-04-17 10:49:11 +08:00
size_t param_value_size, void *param_value, size_t *param_value_size_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clGetKernelWorkGroupInfo;
MNN_CHECK_NOTNULL(func);
return func(kernel, device, param_name, param_value_size, param_value, param_value_size_ret);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clGetEventProfilingInfo(cl_event event, cl_profiling_info param_name, size_t param_value_size, void *param_value,
2019-04-17 10:49:11 +08:00
size_t *param_value_size_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clGetEventProfilingInfo;
MNN_CHECK_NOTNULL(func);
return func(event, param_name, param_value_size, param_value, param_value_size_ret);
}
2021-03-03 16:02:51 +08:00
cl_int CL_API_CALL clGetMemObjectInfo(cl_mem memobj, cl_mem_info param_name, size_t param_value_size, void *param_value,
size_t *param_value_size_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clGetMemObjectInfo;
MNN_CHECK_NOTNULL(func);
return func(memobj, param_name, param_value_size, param_value, param_value_size_ret);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clEnqueueNDRangeKernel(cl_command_queue command_queue, cl_kernel kernel, cl_uint work_dim,
2019-04-17 10:49:11 +08:00
const size_t *global_work_offset, const size_t *global_work_size,
const size_t *local_work_size, cl_uint num_events_in_wait_list,
const cl_event *event_wait_list, cl_event *event) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clEnqueueNDRangeKernel;
MNN_CHECK_NOTNULL(func);
return func(command_queue, kernel, work_dim, global_work_offset, global_work_size, local_work_size,
num_events_in_wait_list, event_wait_list, event);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clWaitForEvents(cl_uint num_events, const cl_event *event_list) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clWaitForEvents;
MNN_CHECK_NOTNULL(func);
return func(num_events, event_list);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clRetainEvent(cl_event event) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clRetainEvent;
MNN_CHECK_NOTNULL(func);
return func(event);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clReleaseEvent(cl_event event) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clReleaseEvent;
MNN_CHECK_NOTNULL(func);
return func(event);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clGetEventInfo(cl_event event, cl_event_info param_name, size_t param_value_size, void *param_value,
2019-04-17 10:49:11 +08:00
size_t *param_value_size_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clGetEventInfo;
MNN_CHECK_NOTNULL(func);
return func(event, param_name, param_value_size, param_value, param_value_size_ret);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clFlush(cl_command_queue command_queue) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clFlush;
MNN_CHECK_NOTNULL(func);
return func(command_queue);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clFinish(cl_command_queue command_queue) {
2019-04-17 10:49:11 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clFinish;
MNN_CHECK_NOTNULL(func);
return func(command_queue);
}
2020-11-05 16:41:56 +08:00
cl_mem CL_API_CALL clCreateImage2D(cl_context context, cl_mem_flags flags, const cl_image_format *image_format, size_t imageWidth,
2019-04-17 10:49:11 +08:00
size_t imageHeight, size_t image_row_pitch, void *host_ptr, cl_int *errcode_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clCreateImage2D;
MNN_CHECK_NOTNULL(func);
return func(context, flags, image_format, imageWidth, imageHeight, image_row_pitch, host_ptr, errcode_ret);
}
2020-11-05 16:41:56 +08:00
cl_command_queue CL_API_CALL clCreateCommandQueue(cl_context context, cl_device_id device, cl_command_queue_properties properties,
2019-04-17 10:49:11 +08:00
cl_int *errcode_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clCreateCommandQueue;
MNN_CHECK_NOTNULL(func);
return func(context, device, properties, errcode_ret);
}
2020-11-05 16:41:56 +08:00
cl_int CL_API_CALL clEnqueueCopyImage(cl_command_queue queue,
2019-04-17 10:49:11 +08:00
cl_mem src_image,
cl_mem dst_image,
const size_t * src_origin,
const size_t * dst_origin,
const size_t * region,
cl_uint num_events_in_wait_list,
const cl_event * event_wait_list ,
cl_event * event) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clEnqueueCopyImage;
MNN_CHECK_NOTNULL(func);
return func(queue, src_image, dst_image, src_origin, dst_origin, region, num_events_in_wait_list, event_wait_list, event);
}
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
// clCreateCommandQueueWithProperties wrapper
2022-12-30 15:18:58 +08:00
cl_command_queue CL_API_CALL clCreateCommandQueueWithProperties(cl_context context, cl_device_id device, const cl_queue_properties *properties, cl_int *errcode_ret) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clCreateCommandQueueWithProperties;
MNN_CHECK_NOTNULL(func);
return func(context, device, properties, errcode_ret);
}
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
// clSVMAlloc wrapper, use OpenCLWrapper function.
2022-02-18 11:30:27 +08:00
void* CL_API_CALL clSVMAlloc(cl_context context, cl_mem_flags flags, size_t size, cl_uint align) {
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clSVMAlloc;
MNN_CHECK_NOTNULL(func);
return func(context, flags, size, align);
}
// clSVMFree wrapper, use OpenCLWrapper function.
2022-02-18 11:30:27 +08:00
void CL_API_CALL clSVMFree(cl_context context, void *buffer) {
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clSVMFree;
MNN_CHECK_NOTNULL(func);
func(context, buffer);
}
// clEnqueueSVMMap wrapper, use OpenCLWrapper function.
2022-02-18 11:30:27 +08:00
cl_int CL_API_CALL clEnqueueSVMMap(cl_command_queue command_queue, cl_bool blocking, cl_map_flags flags, void *host_ptr,
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
size_t size, cl_uint num_events_in_wait_list, const cl_event *event_wait_list, cl_event *event) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clEnqueueSVMMap;
MNN_CHECK_NOTNULL(func);
return func(command_queue, blocking, flags, host_ptr, size, num_events_in_wait_list, event_wait_list, event);
}
// clEnqueueSVMUnmap wrapper, use OpenCLWrapper function.
2022-02-18 11:30:27 +08:00
cl_int CL_API_CALL clEnqueueSVMUnmap(cl_command_queue command_queue, void *host_ptr, cl_uint num_events_in_wait_list,
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
const cl_event *event_wait_list, cl_event *event) {
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clEnqueueSVMUnmap;
MNN_CHECK_NOTNULL(func);
return func(command_queue, host_ptr, num_events_in_wait_list, event_wait_list, event);
}
// clSetKernelArgSVMPointer wrapper, use OpenCLWrapper function.
2022-02-18 11:30:27 +08:00
cl_int CL_API_CALL clSetKernelArgSVMPointer(cl_kernel kernel, cl_uint index, const void *host_ptr) {
[MNN:Sync] Sync internal github Commits: 8148ae75c 弗人 bugfix 14cb8ec7f 弗人 [Converter:Bugfix] bugfix for onnx depthwise convtranspose 476fbcd90 雁行 [MNN:Feature] Open AVX cast and bugfix for contentCFG. 5e26b9fd3 雁行 [Test:Feature] Add android test. 37e147b25 雁行 [MNN:Bugfix] Bugfix for floordiv. 144c185f5 tianbu.xsw hangxing fix hiai b4fd429d6 tianbu.xsw updateCacheFile bugfix -- update cache size d4ba572a8 雁行 [MNN:Bugfix] Support int8 in AVX2 and some Bugfix. 43061f07e xiaying [MNN:Bugfix] Fix bug for module mode run part of model 398cc5ab6 tianhang.yth refactor demo 736380600 xiaying [Express:Bugfix] Fix memory leak for copy branch b8dab0a27 tianhang.yth MNNFloat2Int8 sizeQuad=0 crash fix 94b95bfed ghz [BugFix]1.Better method for fast pack valid check 6a921f85e xiaying [Converter:Bugfix] Fix bug for Fuseconsttosubgraph 5f77ae889 tianhang.yth numThread bugfix a807ef879 tianhang.yth add createSession(configs, runtimeinfo) API, add pymnn demo, pymnn logcat bugfix ad05409d3 xiaying [MNN:Bugfix] Fix bug for StaticModule's sizecompute overflow, add error print for module mode 9d81b8299 xiaying [MNN:Bugfix] Fix bug for Unique op for output size = 1 03b15e9af xiaying [Test:Feature] Add MatMulBConst Test, Fix bug for single Convert c944a76ee tianhang.yth add auto backend and getSessionInfo @tianbu 91fa7267b ghz [BugFix]1.fix the error in eP check bf0041f77 ghz [BugFix]1.Fix the logic error in eP check. 2.Fix the sp align error 693871672 雁行 [CPU:Bugfix] rm adrp instruction for clang compiler bug. 1b8f6b3d8 ghz 1.Fix the wronly use of r13 in arm32 version. 2.Fix the missing callee register save and restore process. feb7ecc4c 弗人 modify log of python offline quant 040c04811 ghz [BufFix]1.replace platform-related regs. 2.fix the same problem in arm32 version 609f37db8 弗人 add log for python quant, python convert 5511dd30a ghz [BugFix]1.Add testcases in SparseConv to check all functional code branch. 2. Fix the bug in "MNNPackC4ForMatMul_A.S" in arm64, which is caused by the missing check of eReal parameter. a93ff9280 tianhang.yth add tf.Unique op support 9729ff773 allen.lk [Bugfix] Fix one arm32 instruction syntax that clang works but gcc DOES NOT work. use index instruction instead. 297c1ad14 雁行 [Expr:Bugfix] bugfix for tensor content used by shape compute. ef8c369e3 弗人 catch exception 07c2dd670 弗人 add dependence to setup, base64 encode url, add time log 177e590c1 弗人 [Python:Feature] add aliyun log for python quant tool 40a7928cf allen.lk [Debug:Sparse] 1.Add group parameter in torchscript converter. 2. Stop split running to avoid memory corruption when check failed in TransformGroupConvolution 3. fix Op split issue in TransformGroupConvolution 3bdea84a1 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. c3c6fbdbd allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. bc590eee4 雁行 [Converter:Bugfix] bugfix for onnx instancenormalization convert. d8918593f tianhang.yth add auto backend and getSessionInfo @tianbu 83a198ed7 杭行 update d0dd3e09b 杭行 update 99540202e xiaying [Converter:Optimize] Opt the tensor convert insert 333d8db82 allen.lk [Debug:Sparse] Fix All platform-register r9 / x18 issue on arm32 and arm64. db5994672 杭行 merge 6293de7b8 tianbu.xsw fix pymnn updateCacheFile 5c2e11cb1 tianbu.xsw do updateCache in createSession 6e7641ff4 tianbu.xsw do not limit cacheFile for a model 5287a65e4 tianbu.xsw bugfix 52ba53a91 tianbu.xsw revert pymnn api 60284d830 tianbu.xsw bugfix 6d8077490 tianbu.xsw rename updateCacheFile api params 3cb172710 tianhang.yth updateCacheFile API size default value is 0 c5b69aabf tianbu.xsw updateCacheFile python api fix 5d5da7aa5 tianbu.xsw reflector code 5707877a4 雁行 [MNN:Speed] Speedup for softmax in x86 and arm. 2a211825c tianbu.xsw reflector code for updateCacheFile 76db3a835 tianbu.xsw [Cache Feature]: Add updateCacheFile API for increment cache b06b0fd43 allen.lk [Debug:Sparse] Fix and warning one kind of segmentfault cause by memory corruption when resize ConvolutionWinograd. Avoid to use some registers as arm restriction. e68bfa495 雁行 [Converter:Feature] Add UUID when model convert. a9cb935dc xiaying [MNN:Speed] Support c4nhwc for more fastblit 019f40353 xiaying [Converter:Refractor] Reduce memory used by MNNConvert(bert from 5G -> 1G) d2a6d3d05 xiaying [MNN:Bugfix] Fix bug for identity output not find 604d0801b xiaying [Converter:Bugfix] Fix bug for FuseGeLu 4bada2367 xiaying [MNN:Refractor] SegmentMean rewrite as segment 82070e708 xiaying [MNN:Bugfix] Fix bug for GeometryBinary e8ea4266e xiaying Fix bug for ShapeTensorConvert compute for dim = 1 error 1f1cf1991 xiaying [Tools:Bugfix] Fix system compability for fastTestOnnx 6f422efe2 xiaying [Tools:Bugfix] Remove color for checkDir for easy to dump 968f7ec88 xiaying [MNN:Speed] Support turn broadcast binary to loop 3e7aaf46f xiaying [MNN:Refractor] Set Convolution1x1Strassen support variable input/output ptr 1f65ab163 xiaying [MNN:Bugfix] Fix bug for mini mnn can't convert model d65953d47 xiaying [MNN:Bugfix] Fix bug for armv7a - android-14 + ARM82 8b68be45c xiaying [MNN:Feature] Add segment 8a8f264f5 xiaying [Vulkan:Bugfix] Remove unuseful print 025bb0fda xiaying [Converter:Bugfix] Fix bug for oneof don't support 43900251e tianbu.xsw enable setCacheFile python API ebfb05c74 tianbu.xsw [Metal Feature] support metallib obtain from walle transfer task 9665c0a79 弗人 add check for path in json file c66fef224 xiaying [Converter:Bugfix] Fix bug for oneof don't support 42f192852 xiaying [MNN:Bugfix] Fix bug for not set output / saveTensor into origin Schedule's outputs 1b95354ff 雁行 [Feature]: Support shape compute for SetDiff1D, and null input for Prod. 83966d043 xiaying [Test:Feature] Add test for static module 42d1be933 xiaying [Converter:Bugfix] Fix bug for mnn convert and static model add more outputs for origin model 9067531c3 xiaying [Converter:Refractor] formatLicence 99558bed9 xiaying [Converter:Bugfix] Count the op for unuseful and controlflow 4f6da0fa7 allen.lk [Feature:GRUMultiOutput] fix multi output dimension type c6b219bce xiaying [Converter:Feature] Turn torch converter to object dd4e68a37 xiaying [Converter:Feature] Support dump supported ops 80b6a60a3 xiaying [Converter:Info] If has output name, print output name instead of computed 015278fc3 xiaying [MNN:Refractor] Revert IfModule's debug info 23ac967c4 xiaying Don't transform for multi-input convolution/deconvolution b02b0d4de xiaying Fix bug for multi-input for conv1d 254d8b1d4 xiaying Fix bug for Conv1dSqueezeMove for multi input convolution 1d d47d0b9ca xiaying Fix bug for CPURaster's fuse nc4hw4 357c5bd33 xiaying Fix ConvBiasAdd for conv's inputs op > 1 55b1f0c9c xiaying [Converter:Bugfix] Don't transform for multi-input convolution/deconvolution 1902a30f5 xiaying [Converter:Bugfix] Fix bug for Conv1dSqueezeMove for multi input convolution 1d c23fe617b xiaying [MNN:Bugfix] Fix bug for multi-input for conv1d 8ff018426 xiaying [MNN:Bugfix] Fix bug for CPURaster's fuse nc4hw4 d4e8cd602 xiaying [Converter:Bugfix] Fix ConvBiasAdd for conv's inputs op > 1 846266b42 tianbu.xsw return when program and tune both nullptr fd67c76a9 xiaying [Converter:Bugfix] DepthwiseConvWeightMerge only valid for tflite e77a242c4 xiaying [Converter:Feature] Support tflite's half pixel be054c377 tianbu.xsw [OpenCL Bugfix] do not rewrite cache when binary program is produced 51e65aa35 xiaying [Converter:Feature] Support tflite for fp16 and multi-input convolution 1ccdfdeb5 tianbu.xsw redefine svm macro name 31234d372 tianbu.xsw [OpenCL SVM] add macro for only use wrapper d739e35da xiaying [MNN:Bugfix] Fix compile bug for grid op 24ab13c79 Joker feat(arm82): add GridSample op support in arm82 backend, AVX(by xiaying) 7b142978e xiaying [AVX512:Speed] Optimize for e <= 8 5f6febe7b tianbu.xsw code refactor 998d91b57 xiaying [Express:Speed] Merge submodule for speed 22c89146f tianhang.yth fix alpha div by zero bug and arm server compile bug 8f829a170 tianbu.xsw [OpenCL Pad] unify conv/deconv pad computing 4a28f603e xiaying [Express:Speed] Shared Const for All Submodule c74cf28f3 xiaying [MNN:Refractor] Seperate Const init and schedule 2a1eebb7a xiaying [Tools:Bugfix] Fix bug for modelTest.py count size 72f04008c xiaying [MNN:Refractor] Delete unuseful const op 1e735d03c xiaying [Converter:Bugfix] Fix bug for static module gen 4dfadbc6e xiaying [MNN:Refractor] Rewrite const init mode 1fcf0417a xiaying [MNN:Bugfix] Fix bug for deconvolutin multi-input for multi-batch 41d429cfd xiaying [Train:Bugfix] Revert convert NCHW for mnistTrain f947a5f01 xiaying [Test:Feature] Add testTrain dad59b6f6 tianbu.xsw move realize code from Backend.hpp to Tensor.cpp cf4473ad1 xiaying [Train:Bugfix] Support pad for GeometryPoolGrad 91ab13734 xiaying [MNN:Bugfix] Fix compile bug for avx512 742e80f47 xiaying [MNN:Refractor] Opt the logic for checknan judge 12543b841 xiaying [ARM82:Bugfix] Fix compile bug for ios 3a2b0a49f xiaying [ARM82:Speed] Opt Pack / Unpack for armv8 c0f1995cd xiaying [ARM82:Speed] Opt MNNPackC8FP16 and MNNUnpackC8FP16 by asm e0fc77dcf xiaying [MNN:Speed] Fix bug for DeconvolutionWithStride for C4HW4, open it 584bec578 xiaying [MNN:Bugfix] Fix bug for format set error for onnx d5bd4148d xiaying [MNN:Bugfix] Fix bug for format set error for onnx b00265841 xiaying [MNN:Bugfix] Fix bug for SparseConvolutionTiledExecutor bb09188ac xiaying [Test:Bugfix] Fix bug for run into sparse auto 426d1babd xiaying [MNN:Refractor] Small bugfix for Group convolution and pack 7d0ea1c46 tianbu.xsw [testModel Feature] support testModel.out input resize 4169c54ce xiaying [MNN:Bugfix] Fix bug for checkNAN for origin 412a82222 xiaying [Test:Bugfix] Fix bug for CheckNAN's error of matmul 319b1d425 xiaying [MNN:Bugfix] Fix bug for multi-batch for ConvInt8 050b728a6 xiaying [Test:Bugfix] Use NCHW for ConvInt8Test 7db3423a1 xiaying [OpenCL:Bugfix] Fix bug for opencl::image,opencl::buffer for C4HW4 adcec6a7f xiaying [Vulkan:Bugfix] Fix bug for invalid tensor size limit d2a7cf4e9 xiaying [Vulkan:Bugfix] Fix bug for onCopyBuffer of nc4hw4 557bebdd3 xiaying [MNN:Bugfix] Fix bug for BF16-ARM32 bbe186649 tianbu.xsw [Update AUTO mode]: fix MNN_FORWARD_AUTO choose priority 6deb23439 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size b137590e4 xiaying [MNN:Bugfix] Fix bug for GeometryBinary don't care about NC4HW4 same size 7003558ea xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case b5f8cae5a xiaying [Converter:Bugfix] Fix bug for onnx pad for serveral case 29b09e125 xiaying [MNN:Bugfix] Fix bug for arm64-bf16 42ce00770 xiaying [MNN:Bugfix] Fix bug for ARM64 - float a2d89fc18 雁行 [Converter:Feature] Support Binary Unary for Torch. 7f1c0deb1 xiaying [MNN:Bugfix] Fix bug for Raster for Int8 8335a6f18 tianbu.xsw [OpenCL Shared Memory] modify data_format method b359e031b xiaying [ARM82:Bugfix] Fix bug for arm82 and speed up pack / unpack c8 24bf3fc88 雁行 [Convert:Feature] Support LayerNormFuse without gamma beta. 3e629624b xiaying [MNN:Bugfix] Fix bug for float - armv7a 2b7908ec7 tianbu.xsw modify workItemSize 3cee0d413 xiaying [MNN:Bugfix] test wrong clear 9cbbfb998 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 2d7a44484 xiaying [MNN:Bugfix] fix compile bug for c++ < 14 eb7d0cb53 xiaying [Test:Bugfix] Don't test for NC4HW4 directly 7b40ca8d1 xiaying [MNN:Bugfix] Fix bug for ConvolutionGroup 2694d8a91 xiaying [MNN:Bugfix] Fix bug for CPUGridSample f89af60f6 xiaying [MNN:Bugfix] Fix compile bug for arm a151abcdd xiaying [MNN:Bugfix] Fix bug for convert for int8 / int16 b254dbe61 雁行 [MNN:Bugfix] Bugfix for Conv onClone. d08150631 xiaying [MNN:Bugfix] Fix bug for fast rcnn e5568a0df xiaying [MNN:Bugfix] Fix bug for CPURaster treat NC4HW4 fast blit 128318933 雁行 [Raster:Bugfix] bugfix for Raster merge onResize. 03caacbea xiaying [MNN:Bugfix] fix bug for CPUDeconvolution and Convolution1x1Strassen for iw != ow e1e3c245c xiaying [MNN:Bugfix] Fix bug for ConvolutionWinograd 2524cbc6d xiaying [MNN:Bugfix] Fix bug for CPUSoftmax 44ec79b8f xiaying [MNN:Bugfix] Fix bug for CPUConvolutionDepthwise / Scale / DeconvolutionDW 21ae956ce xiaying [MNN:Bugfix] Fix bug for Multi-Batch-TiledExecutor 09a5069c7 xiaying [MNN:Speed] Add offset for src and dst 6776c6784 xiaying [MNN:Bugfix] Fix bug for trainable model cc83ae30b xiaying [MNN:Bugfix] Fix bug for trainable model
2021-07-29 11:46:59 +08:00
auto func = MNN::OpenCLSymbolsOperator::getOpenclSymbolsPtr()->clSetKernelArgSVMPointer;
MNN_CHECK_NOTNULL(func);
return func(kernel, index, host_ptr);
}
#endif //MNN_USE_LIB_WRAPPER