Morph MORPH
SIGN IN SIGN UP
apache / mxnet UNCLAIMED

Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Scala, Go, Javascript and more

0 0 0 C++
Normal View History Raw
[Review Required] License Fixes based on Apache RAT Feedback Part 2 (#9482) * Some more licenses from RAT failures * minor edits * Two more files as per Apache RAT * Apache License to Makefiles and .proto files * Adding apache license to mxnet/cython/.pyx files
2018-01-21 19:44:53 -08:00
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.
integrate scala-package compiling into ROOT/Makefile
2016-02-21 14:12:16 +08:00
ROOTDIR = $(CURDIR)
[MXNET-16] Move mshadow/ps-lite/dlpack to 3rdparty #10132 (#10138) * move * update cmake/make * remove commment * update license * ps-lite move * update license * move mshadow for make * mshadow cmake * update license * update readme * Update Jenkinsfile * update license file with paths
2018-03-21 23:31:19 -07:00
TPARTYDIR = $(ROOTDIR)/3rdparty
integrate scala-package compiling into ROOT/Makefile
2016-02-21 14:12:16 +08:00
update mklml and mkl mac support (#7587)
2017-08-29 17:14:01 -07:00
ifeq ($(OS),Windows_NT)
UNAME_S := Windows
else
UNAME_S := $(shell uname -s)
Making MKL-DNN default on MXNet master (#13681) * mkldnn is default makefile and explicitly turn off for buidls * add endif * retrigger * retrigger * build mkldnn as static lib * update makefile to statically build mkldnn * build static mkldnn * fix static name * fix static name * update static for mac * rename mkldnn dep in ci * remove moving mkldnn dynamic lib * retrigger * remove commented code * retrigger * remove mkldnn dnaymic for unitest * retrigger * retrigger * force static for mkldnn lib * turn of mkldnn on arm builds * remove dynamic mkldnn bind * update jenkins to use only mkldnn * remove last flag * turn mkldnn by default on mac * move mkldnn files for GPU MKLDNN build * copy lib mxnet in gpu build * only link windows * add mkldnn.mk * try force linking * retrigger * retrigger * remove mkldnn dynanmic check * use ifndef * remove test mkldnn install * fix spacing * fix index * remove cp of mkldnn since statically linked * add libmkldnn.a to list of files to pack * include mkl_ml * add mkldnn to pack * add libiomp to ci pack * move static libs * fix typo * pack mkldnn * retrigger * add linux artifacts * move libmkldnn in gpu cmake build * move libmkldnn and libiomp5 on gpu workspace * move linked files * fix typo * fix typo * add artifacts for tensorrt * move mkldnn lib in scala build * move mkldnn lib on cpu scala * create dir for binding * rename libmkldnn in scala * move mklml dep in scala builds * move mkl to another linked folder * move libmkl to another dir * add libmklml * move mkldnn * move mkldnn on centos * specify new dynamic path * retrigger * remove mkldnn dynamic lib * remove moving mkldnn artifact * add ld path * retrigger * Revert "remove moving mkldnn artifact" This reverts commit 16cca196e9e1ad92db74f4e8a01b3b052076d268. * Revert "remove mkldnn dynamic lib" This reverts commit d51043622d4ef7fcb95aff6a3e84d91ab71b48c9. * update makefile * Revert RPATH change and trigger CI * correcting use-mkldnn flags for two tests * mkldnn default on linux for starters * reverting naming rules of pack_lib * adding mkldnn=0 flags to centos non mkldnn builds * adding mkldnn=0 flags to ubuntu gpu non mkldnn builds * removing mkldnn binary operation for ubuntu gpu cmake non mkldnn build * removing mkldnn binary operation for centos non-mkldnn unittest * adding explicit USE_MKLDNN=0 flags for clang builds * adding explicit USE_MKLDNN=0 flags for cpu ubuntu builds * removing mkldnn binaries from non mkldnn builds scala gpu * adding explicit flag mkldnn=0 for tensorrt gpu build * adding explicit flag mkldnn=0 for ubuntu cmake asan * adding centos cpu mkldnn tests to CI * adding CentOS GPU MKLDNN build and unittest * not keeping mkldnn default for mac os * setting mkldnn default for x86_64 only * running docs with mkldnn=0 flag * removing CentOS CPU Scala MKLDNN test * setting mkldnn default for x86_64 only * not making mkldn default on windows * removing Centos MKLDNN tests from CI * retrigger * retrigger * retrigger
2019-01-03 08:38:00 +05:30
UNAME_P := $(shell uname -p)
update mklml and mkl mac support (#7587)
2017-08-29 17:14:01 -07:00
endif
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
ifndef config
ifdef CXXNET_CONFIG
config = $(CXXNET_CONFIG)
else ifneq ("$(wildcard ./config.mk)","")
config = config.mk
add cpp doc
2015-08-21 23:49:07 -06:00
else
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
config = make/config.mk
endif
endif
ifndef DMLC_CORE
move dmlc-core & make" (#10231) +cmake amal examples update jenkins udpate amal.py to force include 3p update amal find_source logic remove log update license move nnvm misc nnvm updates fix perl make ; update whitelist add missing license header fix launcher.py update submodule name update amal expand file
2018-03-25 19:45:24 -07:00
DMLC_CORE = $(TPARTYDIR)/dmlc-core
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
endif
add doc for gluon, sym/nd contrib (#7284) * add doc for contrib * docs for gluon, sym/nd contrib
2017-08-02 16:19:41 -07:00
CORE_INC = $(wildcard $(DMLC_CORE)/include/*/*.h)
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
NNVM Refactor (#3194) * Init nnvm change * temp checkin * Move TShape to NNVM * Redirect Symbolic API to NNVM * Add Op Prop Adapter * Finish migrate in shape infer * Pass all symbolic test * temp commit * enable aux data * [EXEC] Basic version of exec for forward only * [EXEC] Enable most optimizations, still wait grad and context * fix legacy op with latest one * Update NNVM NodeRef * Adapt to newer interface * ALl registry of backop is complete * temp commit * Hack finish backward pass * [EXEC] One day pass * [EXEC] Pass all operator unittest * [EXEC] enable model parallel * Fully pass all legacy tests * Remove legacy symbolic code * update news * Make travis compile * Fix python3 * Update viz module to new json format
2016-09-01 21:17:33 -07:00
ifndef NNVM_PATH
[3rdparty] introduce tvm (#11130)
2018-06-02 17:58:25 -07:00
NNVM_PATH = $(TPARTYDIR)/tvm/nnvm
NNVM Refactor (#3194) * Init nnvm change * temp checkin * Move TShape to NNVM * Redirect Symbolic API to NNVM * Add Op Prop Adapter * Finish migrate in shape infer * Pass all symbolic test * temp commit * enable aux data * [EXEC] Basic version of exec for forward only * [EXEC] Enable most optimizations, still wait grad and context * fix legacy op with latest one * Update NNVM NodeRef * Adapt to newer interface * ALl registry of backop is complete * temp commit * Hack finish backward pass * [EXEC] One day pass * [EXEC] Pass all operator unittest * [EXEC] enable model parallel * Fully pass all legacy tests * Remove legacy symbolic code * update news * Make travis compile * Fix python3 * Update viz module to new json format
2016-09-01 21:17:33 -07:00
endif
Change Interface of NDArray & TBlob for DLPack Compatible (#6345) * Change Interface of NDArray & TBlob for DLPack Compatible Fix for cudnn operator Fix cpp tests * Update nnvm * Fix for MKL mem * Fix for windows macro * Bump up version number to 0.10.1 * Update NDArray Save&Load * trigger update * Add test for legacy data load * Use LegacyTShapeLoad * trigger update * Update tensor_blob.h
2017-05-29 23:37:08 -07:00
ifndef DLPACK_PATH
[MXNET-16] Move mshadow/ps-lite/dlpack to 3rdparty #10132 (#10138) * move * update cmake/make * remove commment * update license * ps-lite move * update license * move mshadow for make * mshadow cmake * update license * update readme * Update Jenkinsfile * update license file with paths
2018-03-21 23:31:19 -07:00
DLPACK_PATH = $(ROOTDIR)/3rdparty/dlpack
Change Interface of NDArray & TBlob for DLPack Compatible (#6345) * Change Interface of NDArray & TBlob for DLPack Compatible Fix for cudnn operator Fix cpp tests * Update nnvm * Fix for MKL mem * Fix for windows macro * Bump up version number to 0.10.1 * Update NDArray Save&Load * trigger update * Add test for legacy data load * Use LegacyTShapeLoad * trigger update * Update tensor_blob.h
2017-05-29 23:37:08 -07:00
endif
cuda support for linalg-functions, restructuring of linalg interfaces (#7147) * cuda support for linalg-functions, restructuring of linalg interfaces * incorporate newest mshadow * adjustments to linalg operators
2017-08-12 21:12:44 +02:00
ifndef AMALGAMATION_PATH
AMALGAMATION_PATH = $(ROOTDIR)/amalgamation
endif
debugging on ndarray delete
2015-09-15 08:53:03 +08:00
ifneq ($(USE_OPENMP), 1)
pass compilation, not tested
2015-09-04 01:15:31 +08:00
export NO_OPENMP = 1
endif
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
# use customized config file
include $(config)
MKL compile update to remove full mkl pack dependency for blas=mkl (#5036)
2017-02-15 19:54:43 -08:00
Fix USE_MKLDNN check in Makefile (#13775) * fix makefile * change make/config.mk * add comments * retrigger ci
2019-01-07 04:04:38 +08:00
ifndef USE_MKLDNN
Making MKL-DNN default on MXNet master (#13681) * mkldnn is default makefile and explicitly turn off for buidls * add endif * retrigger * retrigger * build mkldnn as static lib * update makefile to statically build mkldnn * build static mkldnn * fix static name * fix static name * update static for mac * rename mkldnn dep in ci * remove moving mkldnn dynamic lib * retrigger * remove commented code * retrigger * remove mkldnn dnaymic for unitest * retrigger * retrigger * force static for mkldnn lib * turn of mkldnn on arm builds * remove dynamic mkldnn bind * update jenkins to use only mkldnn * remove last flag * turn mkldnn by default on mac * move mkldnn files for GPU MKLDNN build * copy lib mxnet in gpu build * only link windows * add mkldnn.mk * try force linking * retrigger * retrigger * remove mkldnn dynanmic check * use ifndef * remove test mkldnn install * fix spacing * fix index * remove cp of mkldnn since statically linked * add libmkldnn.a to list of files to pack * include mkl_ml * add mkldnn to pack * add libiomp to ci pack * move static libs * fix typo * pack mkldnn * retrigger * add linux artifacts * move libmkldnn in gpu cmake build * move libmkldnn and libiomp5 on gpu workspace * move linked files * fix typo * fix typo * add artifacts for tensorrt * move mkldnn lib in scala build * move mkldnn lib on cpu scala * create dir for binding * rename libmkldnn in scala * move mklml dep in scala builds * move mkl to another linked folder * move libmkl to another dir * add libmklml * move mkldnn * move mkldnn on centos * specify new dynamic path * retrigger * remove mkldnn dynamic lib * remove moving mkldnn artifact * add ld path * retrigger * Revert "remove moving mkldnn artifact" This reverts commit 16cca196e9e1ad92db74f4e8a01b3b052076d268. * Revert "remove mkldnn dynamic lib" This reverts commit d51043622d4ef7fcb95aff6a3e84d91ab71b48c9. * update makefile * Revert RPATH change and trigger CI * correcting use-mkldnn flags for two tests * mkldnn default on linux for starters * reverting naming rules of pack_lib * adding mkldnn=0 flags to centos non mkldnn builds * adding mkldnn=0 flags to ubuntu gpu non mkldnn builds * removing mkldnn binary operation for ubuntu gpu cmake non mkldnn build * removing mkldnn binary operation for centos non-mkldnn unittest * adding explicit USE_MKLDNN=0 flags for clang builds * adding explicit USE_MKLDNN=0 flags for cpu ubuntu builds * removing mkldnn binaries from non mkldnn builds scala gpu * adding explicit flag mkldnn=0 for tensorrt gpu build * adding explicit flag mkldnn=0 for ubuntu cmake asan * adding centos cpu mkldnn tests to CI * adding CentOS GPU MKLDNN build and unittest * not keeping mkldnn default for mac os * setting mkldnn default for x86_64 only * running docs with mkldnn=0 flag * removing CentOS CPU Scala MKLDNN test * setting mkldnn default for x86_64 only * not making mkldn default on windows * removing Centos MKLDNN tests from CI * retrigger * retrigger * retrigger
2019-01-03 08:38:00 +05:30
ifneq ($(UNAME_S), Darwin)
ifneq ($(UNAME_S), Windows)
ifeq ($(UNAME_P), x86_64)
USE_MKLDNN=1
endif
endif
endif
endif
Revert "Refactor operators & MKLDNN (#8302)" (#9661) This reverts commit 2cc2aa2272881326c8a50c6204aedd71e1821c3f.
2018-01-31 21:09:49 -08:00
ifeq ($(USE_MKL2017), 1)
Refactor operators and add MKLDNN (#9677) * Remove MKL code. * Integrate MKLDNN. Update MXNet for MKLDNN. Enable MKLDNN Relu. Fix a compilation error. Change Makefile for MKLDNN. Remove infer storage in convolution. Update MXNet for MKLDNN. Support MKLDNN storage type in python. Update activation. Add MKLDNN base classes. Implement MKLDNN fully connected. Add MKLDNN convolution. Update MKLDNN interface in NDArray. MKLDNN convolution handle CreateMKLDNNData failure. Add another GetMKLDNNData in NDArray. Have mkldnn to define the data format. Create output MKLDNN memory explicitly for FC. Fix a bug in NDArray. Fix a bug in GetWeightDesc. Convert data layout if necessary in FC. remove unnecessary print in MKLDNN convolution. Add MKLDNN deconvolution. Add MKLDNNStream to manage primitives and memories. Use MKLDNNStream to register memory in NDArray. Use MKLDNNStream to manage resources in operators. Handle kAddTo in MKLDNN operators. Fix a bug in deconvolution. Fix bugs in NDArray. Revert "Fix bugs in NDArray." This reverts commit f5624a4aa9f9b9f9fe31f5e6cfa7a9752838fc4e. Fix a bug in NDArray. Fix a bug in NDArray. Reorder MKLDNN memory to default format in SetTBlob. Disable MKLDNN correctly. Fix a bug in activation. Reshape of NDArray supports MKLDNN. Fix a memory ref bug in NDArray. Reshape NDArray in MKLDNN FullyConnected. Fix data format conversion. Create MKLDNN NDArray in python. Support Slice for MKLDNN NDArray. Reduce the overhead of summing the result to the output array. Avoid unnecessary memory copy in NDArray. Fix a bug in data reordering. Fix a bug in NDArray. Don't hard code MKLDNN type. Support dilation in MKLDNN convolution. Fix a bug in sum results. Rewrite GetMKLDNNData. Add prepare_mkldnn.sh Enable MKLDNN activation. Fix a bug on FullyConnected. Handle 3 dims for MKLDNN NDArray. Fix a bug in MKLDNN FC. Support MKLDNN storage in KV store. Fix a bug in executor for non-default NDArray. Fix a link error in cast_storage.cc. Remove unnecessary function def Fall back to def storage if the type isn't supported by MKLDNN. Use NDArray for MKLDNN in python. Reshape output of MKLDNN convolution. Fix a bug in NDArray. Support more operations in MKLDNN NDArray. Fix a bug in deconvolution. Fix bugs in MKLDNN deconvolution. We still need to compute bias correctly. Have elemwise binary ops to fall to default for MKLDNN. Limit the cases that MKLDNN operations are called. Force the layout of mkldnn::memory from NDArray. Add MKLDNN softmax. Fix output storage type of MKLDNN softmax. Add MKLDNN sum. Fix a bug in elemwise sum. Fix a bug in MKLDNN softmax. Fix a bug in imperative. Clean up dispatch modes. Remove redundant code. MKLDNN Pooling Op integration MKLDNN Pooling Op integration add missing file fix mkldnn pooling op workspace issue handle workspace in MKLDNN pooling correctly. Use a non-MKLDNN op for testing. Allow to share arguments and their gradients between executors. Avoid using MKLDNN pooling when it's not supported. Support MKLDNN properly. Choose MKLDNN softmax more carefully. Fix a bug in MKLDNN pooling. Fall back if MKLDNN pooling isn't supported. Fix a bug in Slice of NDArray. Use int32 for workspace memory. Exclude MKLDNN act with tanh. Have two Reshape functions in NDArray. Copy data for NDArray with diff shapes. Add MKLDNN copy. Add MKLDNN version of elemwise_add. Add MKLDNN version of Flatten. add mkldnn surport for concat simplify MKLDNN Flatten. Enalbe MKLDNN deconvolution with bias. Fix a bug in CuDNN deconvolution. avoid using MKLDNNStorage when it's not defined. Remove ./cudnn_lrn-inl.h Fix for make lint. add mkldnn surport for concat fix the coding style for pr of mkldnn concat Only add input data for MKLDNN concat backward Remove unnecessary TODO. remove unnecessary __repr__ in MKLNDArray. better condition check for readability. Use macro when including mkldnn.hpp. Revert "Use CoreOpRunner for refactored Ops." This reverts commit a28586fc25950cc006cb317e26e0d17541ef0586. Fix a bug in test core. Limit MKLDNN ops being used. Fix complains from "make pylint" Move ContainStorage to common/utils.h Limit MKLDNN concat being used. Add license. Fix amalgamation Fix compilation error in mkldnn_ops-inl.h Fix a bug in deconvolution. Fix a bug in pooling. MKLDNN ops allocates temp mem. Fix a bug in pooling. Allocate align memory from temp space. Have parameter gradients stored in the default storage. Handle all cases in CopyFrom. Ensure NDArray returns memory with right memory descriptors. use auto to define memory in the operator. Use raw pointer for mkldnn memory. Move more code to mkldnn_base.cc Fix a compilation error. Address review comments. fix a bug in activation backward. Miss a macro in mkldnn_base.cc Fix a bug in data iterator in examples. Avoid memory allocation in ReshapeMKLDNN. Avoid memory allocation in storage cast. Fix a bug in cast storage. Handle sliced MKLDNN NDArray. Use memcpy if NDArray uses default format. Revert "Limit MKLDNN ops being used." This reverts commit 75e2ae570d03483868ec4ed8ed46015c7fa6c6fb. Enable mkldnn act backward has the same input layout. Fix a bug in mkldnn activation. Use MKLDNN sum in more cases. Improve perf of reorder. Avoid memory reorder in conv and deconv. Avoid unnecessary storage cast in fallback path. Revert "Use MKLDNN sum in more cases." This reverts commit 7a21ebca8bbe17fde49c3b1ca3f31b835a33afb8. Handle sliced ndarray in more cases. Fix a complain from make lint. Update Jenkins to test MKLDNN. debug compiling mkldnn. Use MKLDNN sum in more cases. Add mkldnn as a submodule. Compile with mkldnn in 3rdparty. Fix some coding styles. write the path to mkldnn lib in libmxnet.so. use rpath with $ORIGIN. Pack all lib files in Jenkins. pack and unpack mxnet with MKLDNN. Update Jenkinsfile Update Jenkinsfile Add mkldnn batch normalization Fix bugs in BN. Avoid memory allocation in MKLDNNCopy. only use MKLDNN BatchNorm for special cases. MKLDNN BatchNorm doesn't work well on the default layout. Add MKL-DNN based LRN Code Style Changes Fix a bug in BN. Fix a bug in LRN. Handle non-default storage in memory plan. Fix coding style. Fix a compilation error without mkldnn. Fix some coding styles for batch norm Improve forward of convolution. Add openmp and simd support to BN operator Retrieve MKLDNN Conv primitive based on signature. Retrieve Act primitive based on its signature. Fix a bug in pooling. Diable some MKLDNN activation and pooling. Cast MKLDNN storage with diff data type. Check if it's a view of NDArray. Reshaped and sliced arrays share the same chunks. Implement caching MKLDNN Act correctly. Fix a bug in check_consistency. Fix a potential bug when destroying NDArray. Fix bugs when allocating mem in NDArray. Fix coding style. Add micro when using mkldnn in ndarray. Fix a compilation error. Fix a bug in concat. Remove MKLDNNStorage. handle diff layouts in CopyFromToDnsImpl. Fallback correctly. Force weight grad to use default layout. Reorder weight arrays in (de)conv for faster inference. Avoid caching TBlob from NDArray. This commit may add some overhead of managing NDArray for each fallback. Fix a bug in Flatten. handle ndarray with def layout in mkldnn BN correctly. Align to page when mkldnn is enabled. Use default mem alloc for mkldnn. Reuse NDArrays. Support WriteInplace for sum. fix complains from "make lint". Avoid reallocation in NDArray. Handle weight arrays with special MKLDNN layouts. Remove unnecessary GetWeights. Fix compilation error without MKLDNN. Fix a bug in (de)conv for weight arrays. Fix a minor bug in MKLDNN conv. Fix a bug in MKLDNNOpSignature. Reimplement fallback for MKLDNN ops. Fix a bug in FallbackExecutor. Add params in hashcode. Invalidate data in outputs to accelerate. Fix a minor bug. Update mkldnn_base-inl.h Add primitive caching for Pooling forward computation Add hashcode in pooling parameters. Support NDArray copy with types unsupported by MKLDNN. Avoid using MKLDNN concat for negative dimension. Fix make lint complain. Disable mkldnn avg pooling for now. Fix a compile warning. Fix compile error when MKLDNN is disabled. OP primitive cache: use memory as signature for MKLDNN storage type Remove MKLDNN array in python. Disable Clang tests in Jenkins. Use mklml dockers to test mkldnn. Update MKLDNN repo to zhengda's mkldnn repo. Update MKLDNN repo to ashok's. Fix a bug in fallback. Change avg pooling algorithm to pooling_avg_include_padding Fix a code style in mkldnn pooling. Temp fix a bug in FC. Revert "Disable Clang tests in Jenkins." This reverts commit b4efa8f89592d30a27f9c30e2237e9420ac6749a. Rebase and Refactor deconv (#20) * rebase to Da,Zheng refactor branch Jan.14, add signature for mkldnn Deconv and modify classMKLDNNDeconvForward * fix make lint complains A simple way of caching BN inference. cache BN forward for both training and inference. Fix some minor problems in BN. Fix a bug in caching BN. force to build with avx2 in Jenkins. Remove the remaining MKLDNNStorageType Some minor updates in NDArray. a lot of updates to address comments. minor changes. * Use NNVM interface. Use NNVM interface for upsampling. Use NNVM interface for convolution. Use NNVM interface for deconvolution. Use NNVM interface for FullyConnected. Move NNVM interface to batch norm. Use NNVM interface for depthwise convolution. Use NNVM interface for softmax activation. Use NNVM interface for pooling. use NNVM interface for dropout. Use NNVM interface for activation. Use NNVM interface for CuDNN batch norm. Use NNVM interface for CuDNN pooling. Use NNVM interface for CuDNN softmax activation. Use NNVM interface for CuDNN activation. Use NNVM interface for CuDNN convolution. Use NNVM interface for CuDNN deconvolution. Move concat to nn/ Use NNVM interface for concat. Fix headers in concat. Move lrn to nn/. Use NNVM interface for LRN. Fix a compilation error in convolution. Fix a compilation error in activation. Fix coding style. Fix coding style for make lint. use enums in batch norm. Use CoreOpRunner for refactored Ops. Make FullyConnected stateless. Make upsampling stateless. Make pooling stateless. Make batchnorm stateless. Make SoftmaxActivation stateless. Fix a code style problem. pass amalgamation test for batch norm. pass amalgamation test for dropout. Get convolution ops from a function. Fix compilation errors for GPU. Fix thread local in diff platforms. Avoid using thread_local for non-CuDNN conv/deconv. Remove TODO in deconv. Fix a bug in batch norm. Fix a bug in fully connected. Don't set #inputs for backward convolution. Revert "Make pooling stateless." * revert modification in test_executor. * Fix a bug in FlattenStorageType. * Remove BN debug. * Remove remaining MXNET_USE_MKL2017 * Remove unused code in pooling. * Fixing bugs in gtests. * Fix lint errors. * a lot of minor updates to address comments. * Fix coding style in MKLDNN Pooling (#22) * revert the code change in the previous code refactor. * Fix a bug in pooling. * LRN coding style changes (#21) * LRN coding style change * Add const for local variables * Add req for LRN forward * rebase code * align API interface * revert modification in test_executor. * cast storage with MKLDNN properly. * Minor updates to address comments. * some minor updates. * Switch to the master branch of MKLDNN. * Minor updates to address comments. * Update activation.cc * Fix a bug in convert NDArray. * Add gluon model zoo tests. * Update GPU tests on model zoo. * Avoid using mobilenet for GPU tests with gluon models. mobilenet can't pass the test even without MKLDNN. * Update GPU tests on gluon. * change cmake to compile MKLDNN. * update cmake for MKLDNN. * Implement align myself. * Switch to intel/mkl-dnn. * Fix errors in align unittest. * Add unit test for LRN. * fix a compilation error. * use storage_type_assign to determine storage type. * avoid global pooling in mkldnn. There is a bug in global pooling in mkldnn. * compare all MKLDNN ops with native impls. add MXNET_MKLDNN_DEBUG to control the test. * Fix a bug in testing correctness. * print the name of buggy operator. * undo some modifications. * Fix a bug on reshaped array. * avoid testing outputs with NullOp. * turn on MKLDNN tests in Jenkins. * print each operator in MKLDNN tests. * rename test_gluon_model_zoo.py * Create hashcode for operator parameters properly. * Add USE_MKL2017 back. * Print warning messages. * move batchnorm tests to nnvm interface. * Delete batchnorm v1 tests. * Get inputs and outputs in batchnorm tests. * disable batchnorm tests for now. * Fix GPU tests on gluon model zoo. * Fix lint complains in tests. * Remove simd from openmp instructions in BatchNorm (#24) * Remove warnings. * Fix MKLDNN 1st compile failure issue (#23) * Fix compilation errors. * Remove ARCH_OPT in Jenkins. * Revert "avoid global pooling in mkldnn." This reverts commit f6efd342e64968cb848c9193d80e929968b8052c. * Move to the latest MKLDNN. This fixes the bug in global pooling. * WIP unit tests (#25) * WIP unit tests * some backward items initialized * Make more C++ unit tests work for batch norm (#28) * WIP unit tests * some backward items initialized * some backward items initialized * some backward items initialized * first unit test working * Working on types * backward types working for fp16 on first unit test * backward types working for fp16 on first unit test * backward types working for fp16 on first unit test * . * . * some tests working * fix input data * hangle gpu<->cpu for setting values * gpu working * gpu working * CAccessAsCPU class * Fix varying type in AccessAsCPU * starting to add channel axis tests * TestChannelAxisSimple * TestChannelAxisSimple * run bidirectional * run bidirectional * run bidirectional * CLEANUP * CLEANUP * .. * noaxis * .. * lint * revert * revert * Fix lint complains. * Fix a minor problem in Makefile. * fix GPU pooling. * Disable modelzoo inference tests. * update accuracy checks for MKLDNN. * Fix MKLDNN pooling for global pooling. * Fix Jenkins. * Fix a bug in Jenkins. * Fix Jenkins
2018-02-15 14:44:34 -08:00
$(warning "USE_MKL2017 is deprecated. We will switch to USE_MKLDNN.")
USE_MKLDNN=1
endif
ifeq ($(USE_MKLDNN), 1)
Implement mkldnn convolution fusion and quantization. (#12530) * Implement mkldnn convolution fusion. Implement mkldnn convolution quantization. * Fix lint * Fix performance regression caused by mkldnn fallback. * clean up include * Fix msbuild on openmp pragma. * Fix quantization test, allow to use original op names as exclude layer for quantization. * Fix unittest. * Fix unittest * fix lint * Add post quantize fusion * add test case * add head license in test case * Remove GetBoolHash() * Remove mkldnn fallback change. * Address Haibin's comments. * Add TIsMKLDNN for _sg_mkldnn_conv temporarily. * Address reminisce's comments. * Handle the case that inplace fail. * pass unit test. * Add symbol api get_backend_symbol() * Retrigger ci * update the test case * Check subgraph index. * Use index as FAvoidQuantizeInput's parameter. * Add mkldnn_hwigo support as quantizaiton needs. * Address KellenSunderland's comments. * Handle input order change after subgraph pass. * Fix ci test
2018-10-10 01:38:53 +08:00
MKLDNNROOT = $(ROOTDIR)/3rdparty/mkldnn/build/install
MKLROOT = $(ROOTDIR)/3rdparty/mkldnn/build/install
Refactor operators and add MKLDNN (#9677) * Remove MKL code. * Integrate MKLDNN. Update MXNet for MKLDNN. Enable MKLDNN Relu. Fix a compilation error. Change Makefile for MKLDNN. Remove infer storage in convolution. Update MXNet for MKLDNN. Support MKLDNN storage type in python. Update activation. Add MKLDNN base classes. Implement MKLDNN fully connected. Add MKLDNN convolution. Update MKLDNN interface in NDArray. MKLDNN convolution handle CreateMKLDNNData failure. Add another GetMKLDNNData in NDArray. Have mkldnn to define the data format. Create output MKLDNN memory explicitly for FC. Fix a bug in NDArray. Fix a bug in GetWeightDesc. Convert data layout if necessary in FC. remove unnecessary print in MKLDNN convolution. Add MKLDNN deconvolution. Add MKLDNNStream to manage primitives and memories. Use MKLDNNStream to register memory in NDArray. Use MKLDNNStream to manage resources in operators. Handle kAddTo in MKLDNN operators. Fix a bug in deconvolution. Fix bugs in NDArray. Revert "Fix bugs in NDArray." This reverts commit f5624a4aa9f9b9f9fe31f5e6cfa7a9752838fc4e. Fix a bug in NDArray. Fix a bug in NDArray. Reorder MKLDNN memory to default format in SetTBlob. Disable MKLDNN correctly. Fix a bug in activation. Reshape of NDArray supports MKLDNN. Fix a memory ref bug in NDArray. Reshape NDArray in MKLDNN FullyConnected. Fix data format conversion. Create MKLDNN NDArray in python. Support Slice for MKLDNN NDArray. Reduce the overhead of summing the result to the output array. Avoid unnecessary memory copy in NDArray. Fix a bug in data reordering. Fix a bug in NDArray. Don't hard code MKLDNN type. Support dilation in MKLDNN convolution. Fix a bug in sum results. Rewrite GetMKLDNNData. Add prepare_mkldnn.sh Enable MKLDNN activation. Fix a bug on FullyConnected. Handle 3 dims for MKLDNN NDArray. Fix a bug in MKLDNN FC. Support MKLDNN storage in KV store. Fix a bug in executor for non-default NDArray. Fix a link error in cast_storage.cc. Remove unnecessary function def Fall back to def storage if the type isn't supported by MKLDNN. Use NDArray for MKLDNN in python. Reshape output of MKLDNN convolution. Fix a bug in NDArray. Support more operations in MKLDNN NDArray. Fix a bug in deconvolution. Fix bugs in MKLDNN deconvolution. We still need to compute bias correctly. Have elemwise binary ops to fall to default for MKLDNN. Limit the cases that MKLDNN operations are called. Force the layout of mkldnn::memory from NDArray. Add MKLDNN softmax. Fix output storage type of MKLDNN softmax. Add MKLDNN sum. Fix a bug in elemwise sum. Fix a bug in MKLDNN softmax. Fix a bug in imperative. Clean up dispatch modes. Remove redundant code. MKLDNN Pooling Op integration MKLDNN Pooling Op integration add missing file fix mkldnn pooling op workspace issue handle workspace in MKLDNN pooling correctly. Use a non-MKLDNN op for testing. Allow to share arguments and their gradients between executors. Avoid using MKLDNN pooling when it's not supported. Support MKLDNN properly. Choose MKLDNN softmax more carefully. Fix a bug in MKLDNN pooling. Fall back if MKLDNN pooling isn't supported. Fix a bug in Slice of NDArray. Use int32 for workspace memory. Exclude MKLDNN act with tanh. Have two Reshape functions in NDArray. Copy data for NDArray with diff shapes. Add MKLDNN copy. Add MKLDNN version of elemwise_add. Add MKLDNN version of Flatten. add mkldnn surport for concat simplify MKLDNN Flatten. Enalbe MKLDNN deconvolution with bias. Fix a bug in CuDNN deconvolution. avoid using MKLDNNStorage when it's not defined. Remove ./cudnn_lrn-inl.h Fix for make lint. add mkldnn surport for concat fix the coding style for pr of mkldnn concat Only add input data for MKLDNN concat backward Remove unnecessary TODO. remove unnecessary __repr__ in MKLNDArray. better condition check for readability. Use macro when including mkldnn.hpp. Revert "Use CoreOpRunner for refactored Ops." This reverts commit a28586fc25950cc006cb317e26e0d17541ef0586. Fix a bug in test core. Limit MKLDNN ops being used. Fix complains from "make pylint" Move ContainStorage to common/utils.h Limit MKLDNN concat being used. Add license. Fix amalgamation Fix compilation error in mkldnn_ops-inl.h Fix a bug in deconvolution. Fix a bug in pooling. MKLDNN ops allocates temp mem. Fix a bug in pooling. Allocate align memory from temp space. Have parameter gradients stored in the default storage. Handle all cases in CopyFrom. Ensure NDArray returns memory with right memory descriptors. use auto to define memory in the operator. Use raw pointer for mkldnn memory. Move more code to mkldnn_base.cc Fix a compilation error. Address review comments. fix a bug in activation backward. Miss a macro in mkldnn_base.cc Fix a bug in data iterator in examples. Avoid memory allocation in ReshapeMKLDNN. Avoid memory allocation in storage cast. Fix a bug in cast storage. Handle sliced MKLDNN NDArray. Use memcpy if NDArray uses default format. Revert "Limit MKLDNN ops being used." This reverts commit 75e2ae570d03483868ec4ed8ed46015c7fa6c6fb. Enable mkldnn act backward has the same input layout. Fix a bug in mkldnn activation. Use MKLDNN sum in more cases. Improve perf of reorder. Avoid memory reorder in conv and deconv. Avoid unnecessary storage cast in fallback path. Revert "Use MKLDNN sum in more cases." This reverts commit 7a21ebca8bbe17fde49c3b1ca3f31b835a33afb8. Handle sliced ndarray in more cases. Fix a complain from make lint. Update Jenkins to test MKLDNN. debug compiling mkldnn. Use MKLDNN sum in more cases. Add mkldnn as a submodule. Compile with mkldnn in 3rdparty. Fix some coding styles. write the path to mkldnn lib in libmxnet.so. use rpath with $ORIGIN. Pack all lib files in Jenkins. pack and unpack mxnet with MKLDNN. Update Jenkinsfile Update Jenkinsfile Add mkldnn batch normalization Fix bugs in BN. Avoid memory allocation in MKLDNNCopy. only use MKLDNN BatchNorm for special cases. MKLDNN BatchNorm doesn't work well on the default layout. Add MKL-DNN based LRN Code Style Changes Fix a bug in BN. Fix a bug in LRN. Handle non-default storage in memory plan. Fix coding style. Fix a compilation error without mkldnn. Fix some coding styles for batch norm Improve forward of convolution. Add openmp and simd support to BN operator Retrieve MKLDNN Conv primitive based on signature. Retrieve Act primitive based on its signature. Fix a bug in pooling. Diable some MKLDNN activation and pooling. Cast MKLDNN storage with diff data type. Check if it's a view of NDArray. Reshaped and sliced arrays share the same chunks. Implement caching MKLDNN Act correctly. Fix a bug in check_consistency. Fix a potential bug when destroying NDArray. Fix bugs when allocating mem in NDArray. Fix coding style. Add micro when using mkldnn in ndarray. Fix a compilation error. Fix a bug in concat. Remove MKLDNNStorage. handle diff layouts in CopyFromToDnsImpl. Fallback correctly. Force weight grad to use default layout. Reorder weight arrays in (de)conv for faster inference. Avoid caching TBlob from NDArray. This commit may add some overhead of managing NDArray for each fallback. Fix a bug in Flatten. handle ndarray with def layout in mkldnn BN correctly. Align to page when mkldnn is enabled. Use default mem alloc for mkldnn. Reuse NDArrays. Support WriteInplace for sum. fix complains from "make lint". Avoid reallocation in NDArray. Handle weight arrays with special MKLDNN layouts. Remove unnecessary GetWeights. Fix compilation error without MKLDNN. Fix a bug in (de)conv for weight arrays. Fix a minor bug in MKLDNN conv. Fix a bug in MKLDNNOpSignature. Reimplement fallback for MKLDNN ops. Fix a bug in FallbackExecutor. Add params in hashcode. Invalidate data in outputs to accelerate. Fix a minor bug. Update mkldnn_base-inl.h Add primitive caching for Pooling forward computation Add hashcode in pooling parameters. Support NDArray copy with types unsupported by MKLDNN. Avoid using MKLDNN concat for negative dimension. Fix make lint complain. Disable mkldnn avg pooling for now. Fix a compile warning. Fix compile error when MKLDNN is disabled. OP primitive cache: use memory as signature for MKLDNN storage type Remove MKLDNN array in python. Disable Clang tests in Jenkins. Use mklml dockers to test mkldnn. Update MKLDNN repo to zhengda's mkldnn repo. Update MKLDNN repo to ashok's. Fix a bug in fallback. Change avg pooling algorithm to pooling_avg_include_padding Fix a code style in mkldnn pooling. Temp fix a bug in FC. Revert "Disable Clang tests in Jenkins." This reverts commit b4efa8f89592d30a27f9c30e2237e9420ac6749a. Rebase and Refactor deconv (#20) * rebase to Da,Zheng refactor branch Jan.14, add signature for mkldnn Deconv and modify classMKLDNNDeconvForward * fix make lint complains A simple way of caching BN inference. cache BN forward for both training and inference. Fix some minor problems in BN. Fix a bug in caching BN. force to build with avx2 in Jenkins. Remove the remaining MKLDNNStorageType Some minor updates in NDArray. a lot of updates to address comments. minor changes. * Use NNVM interface. Use NNVM interface for upsampling. Use NNVM interface for convolution. Use NNVM interface for deconvolution. Use NNVM interface for FullyConnected. Move NNVM interface to batch norm. Use NNVM interface for depthwise convolution. Use NNVM interface for softmax activation. Use NNVM interface for pooling. use NNVM interface for dropout. Use NNVM interface for activation. Use NNVM interface for CuDNN batch norm. Use NNVM interface for CuDNN pooling. Use NNVM interface for CuDNN softmax activation. Use NNVM interface for CuDNN activation. Use NNVM interface for CuDNN convolution. Use NNVM interface for CuDNN deconvolution. Move concat to nn/ Use NNVM interface for concat. Fix headers in concat. Move lrn to nn/. Use NNVM interface for LRN. Fix a compilation error in convolution. Fix a compilation error in activation. Fix coding style. Fix coding style for make lint. use enums in batch norm. Use CoreOpRunner for refactored Ops. Make FullyConnected stateless. Make upsampling stateless. Make pooling stateless. Make batchnorm stateless. Make SoftmaxActivation stateless. Fix a code style problem. pass amalgamation test for batch norm. pass amalgamation test for dropout. Get convolution ops from a function. Fix compilation errors for GPU. Fix thread local in diff platforms. Avoid using thread_local for non-CuDNN conv/deconv. Remove TODO in deconv. Fix a bug in batch norm. Fix a bug in fully connected. Don't set #inputs for backward convolution. Revert "Make pooling stateless." * revert modification in test_executor. * Fix a bug in FlattenStorageType. * Remove BN debug. * Remove remaining MXNET_USE_MKL2017 * Remove unused code in pooling. * Fixing bugs in gtests. * Fix lint errors. * a lot of minor updates to address comments. * Fix coding style in MKLDNN Pooling (#22) * revert the code change in the previous code refactor. * Fix a bug in pooling. * LRN coding style changes (#21) * LRN coding style change * Add const for local variables * Add req for LRN forward * rebase code * align API interface * revert modification in test_executor. * cast storage with MKLDNN properly. * Minor updates to address comments. * some minor updates. * Switch to the master branch of MKLDNN. * Minor updates to address comments. * Update activation.cc * Fix a bug in convert NDArray. * Add gluon model zoo tests. * Update GPU tests on model zoo. * Avoid using mobilenet for GPU tests with gluon models. mobilenet can't pass the test even without MKLDNN. * Update GPU tests on gluon. * change cmake to compile MKLDNN. * update cmake for MKLDNN. * Implement align myself. * Switch to intel/mkl-dnn. * Fix errors in align unittest. * Add unit test for LRN. * fix a compilation error. * use storage_type_assign to determine storage type. * avoid global pooling in mkldnn. There is a bug in global pooling in mkldnn. * compare all MKLDNN ops with native impls. add MXNET_MKLDNN_DEBUG to control the test. * Fix a bug in testing correctness. * print the name of buggy operator. * undo some modifications. * Fix a bug on reshaped array. * avoid testing outputs with NullOp. * turn on MKLDNN tests in Jenkins. * print each operator in MKLDNN tests. * rename test_gluon_model_zoo.py * Create hashcode for operator parameters properly. * Add USE_MKL2017 back. * Print warning messages. * move batchnorm tests to nnvm interface. * Delete batchnorm v1 tests. * Get inputs and outputs in batchnorm tests. * disable batchnorm tests for now. * Fix GPU tests on gluon model zoo. * Fix lint complains in tests. * Remove simd from openmp instructions in BatchNorm (#24) * Remove warnings. * Fix MKLDNN 1st compile failure issue (#23) * Fix compilation errors. * Remove ARCH_OPT in Jenkins. * Revert "avoid global pooling in mkldnn." This reverts commit f6efd342e64968cb848c9193d80e929968b8052c. * Move to the latest MKLDNN. This fixes the bug in global pooling. * WIP unit tests (#25) * WIP unit tests * some backward items initialized * Make more C++ unit tests work for batch norm (#28) * WIP unit tests * some backward items initialized * some backward items initialized * some backward items initialized * first unit test working * Working on types * backward types working for fp16 on first unit test * backward types working for fp16 on first unit test * backward types working for fp16 on first unit test * . * . * some tests working * fix input data * hangle gpu<->cpu for setting values * gpu working * gpu working * CAccessAsCPU class * Fix varying type in AccessAsCPU * starting to add channel axis tests * TestChannelAxisSimple * TestChannelAxisSimple * run bidirectional * run bidirectional * run bidirectional * CLEANUP * CLEANUP * .. * noaxis * .. * lint * revert * revert * Fix lint complains. * Fix a minor problem in Makefile. * fix GPU pooling. * Disable modelzoo inference tests. * update accuracy checks for MKLDNN. * Fix MKLDNN pooling for global pooling. * Fix Jenkins. * Fix a bug in Jenkins. * Fix Jenkins
2018-02-15 14:44:34 -08:00
export USE_MKLML = 1
MKL compile update to remove full mkl pack dependency for blas=mkl (#5036)
2017-02-15 19:54:43 -08:00
endif
[MXNET-16] Move mshadow/ps-lite/dlpack to 3rdparty #10132 (#10138) * move * update cmake/make * remove commment * update license * ps-lite move * update license * move mshadow for make * mshadow cmake * update license * update readme * Update Jenkinsfile * update license file with paths
2018-03-21 23:31:19 -07:00
include $(TPARTYDIR)/mshadow/make/mshadow.mk
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
include $(DMLC_CORE)/make/dmlc.mk
# all tge possible warning tread
Scalable build and warning fixes (#5396) * Scalable build and warning fixes * Minor adjustments * Scalable build and warning fixes * Minor adjustments * Remove unneeded link libraries * Revert Size() back to dim() * Make OpenMP required again * Trying to make windows openmp config work * Fixed build problem
2017-03-16 10:49:47 -07:00
WARNFLAGS= -Wall -Wsign-compare
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
CFLAGS = -DMSHADOW_FORCE_STREAM $(WARNFLAGS)
compile debug
2015-08-23 21:26:16 +08:00
Introduce a developer mode option in Makefile (#4364) this option will enable two addtionals for now: 1. Create symbol even for optimized build 2. Enforce warning as error while I am here, clean up all warnings so that I can get a clean build in developer mode.
2016-12-29 03:18:09 +08:00
ifeq ($(DEV), 1)
CFLAGS += -g -Werror
add -Werror to NVCCFLAGS (#6774) * add -Werror to NVCCFLAGS * add -Werror to NVCCFLAGS * RTLD_LOCAL instead of GLOBAL * fix device/host cross-call warnings
2017-06-22 15:24:38 -07:00
NVCCFLAGS += -Werror cross-execution-space-call
Introduce a developer mode option in Makefile (#4364) this option will enable two addtionals for now: 1. Create symbol even for optimized build 2. Enforce warning as error while I am here, clean up all warnings so that I can get a clean build in developer mode.
2016-12-29 03:18:09 +08:00
endif
compile debug
2015-08-23 21:26:16 +08:00
# CFLAGS for debug
change def of context, change pinned to special device
2015-09-19 14:32:28 -07:00
ifeq ($(DEBUG), 1)
Cures SEGV seen in test_symbol.py:test_zero_prop2 when DEBUG=1. (#6049)
2017-04-30 20:16:56 -07:00
CFLAGS += -g -O0
change def of context, change pinned to special device
2015-09-19 14:32:28 -07:00
else
Batch Norm rewrite without mshadow, 1D, 2D, 3D, float16, float32, float64 as well as operator gtest framework (#5936) * Batch Norm rewrite without mshadow as well as operator gtest framework * performance testing * lint fixes * use CUDNN for this test * remove superfluous omp define * Fix file names in comments * build, run, clean gtest works (although a test is failing) * CR comments * Adjust timing tests for more strenuous sample * Remove temp resource allocation * DeviceTensor3 added, forEachFast not yet converted * DeviceTensor3 version working * DeviceTensor3 working * . * Fix for use_global_stats * fixed bug with testing suite for double (Float64) * python unit tests working for batchnorm * python unit tests * Update documentation for mxnet.initializer.Mixed (#5937) * Update documentation for SVMOutput. (#5931) * Update documentation for SVMOutput. * Update doc for SVMOutput - fix formatting. * Adding install instruction for Ubuntu-CPU-Python (#5885) * edit ndarray API docs (#5806) * edit docs in broadcast_reduce_op * edit docs in broadcast_reduce_op * minor change * lint fix * fix * mx.nd.ones * mx.nd.repeat * mx.nd.reverse * add example in repeat * optimizer update * fix nanprod * fix optimizer_op api doc * fix reduce_op api doc * fix nd.ones api doc * mx.nd.repeat doc change * Update broadcast_reduce_op.h * Symbol docs fixes (#5930) * symbol docs minor formatting changes * deepcopy, infer_shape, infer_shape_partial docs modified * Few more small fixes * arithmetic functions fixes * some more modifications * changes after review * small change * grad function note added * More API Doc Edits (#5886) * edit activation doc * doc l2_normalization * edit MakeLoss doc * edit blockgrad doc * blockgrad fileline fix * edit MakeLoss doc cont. * doc change 'tensor' to 'multidimensional array' * l2normalization doc improve * makeloss doc improve, blockgrad doc improve * fix doc in activation, l2_normalization, make_loss * fix minor grammar * use .describe to avoid build failure. * Update documentation for mxnet.image.imdecode (#5957) * Update documentation for mxnet.image.imdecode * Update documentation for mxnet.image.imdecode (clarify that we need OpenCV and not the CV2 Python library) * Fix script by adding path to Dockerfile (#5958) * Clean install script * Add test for pip installations * Remove debug statements & comments * Make test runnable as script and from framework * Fix path to Dockerfiles * Putting failing cases at the end * Update doc for Custom operator. (#5875) * Update doc for Custom operator. * Update doc for Custom operator. * Fix formating in doc for Custom operator. * Fix formating in doc for Custom operator. * Minor change to ndarray.Custom documentation. * Minor edit in doc for Custom operator. * Minor change to doc for Custom operator. Data is 'NDArray-or-Symbol'. * Minor formatting change for Custom operator documentation. * For Custom operator doc, move example into ndarray_doc.py. * Minor change in Custom operator documentation * Improve the doc of pick + Update dmlc-core (#5946) * Add PickParam to fix the docstring and the initial value for axis * Update dmlc-core * Update dmlc-core * Image docs modified (#5973) * imageIter doc modified * edited imageiter * ADD missing Libri_sample.json, FIX minor bugs in speech_recognition example (#5962) * [KVStore] Add support for other data types (#5818) * Fix kvstore type * Fix lint * Parse inputs to DataDesc * Make module support dtype * Fix lint * Add default dtype in Comm * Fix lint * Revert rename * [cpp-package] Add C++ basic tutorial and build instruction (#5971) * Add C++ basic tutorial and build instruction * Remove binaries * Fix lint * Avoid sign-compare * Update documentation for mxnet.metric.np (#5977) * Getting rid of identity (#5935) * Activation ops (#5938) * [Ops] Add op: 'relu' * Add op: 'sigmoid' * Introduce 'kernel_launch_op' * Add tests and describe; move it to elemwise_unary_op * Fix GPU version * Convert caffe AbsVal to mx.symbol.abs in caffe converter (#5984) * Correction to LSTMCell docstring (#5986) * [Module] fix input_grads order (#5980) * fix input_grads order + update dmlc-core * set label to be optional * update env_var doc (#5964) * Adjusting make, Callback removed * batch norm gpu testing * Batch Norm rewrite without mshadow as well as operator gtest framework * performance testing * lint fixes * use CUDNN for this test * remove superfluous omp define * Fix file names in comments * build, run, clean gtest works (although a test is failing) * CR comments * Adjust timing tests for more strenuous sample * Remove temp resource allocation * rearrange source into cc and cu files * lint fixes * Trigger build * Use latest mshadow * temporarily revert channel position parameter field * Add more tests for batchnorm * Add more tests for batchnorm * test_operator_gpu working for all types * Compiles after AccReal * Compiles after AccReal * All tests working * All tests working * build, run, clean gtest works (although a test is failing) * vc++ requires explicit int type for omp for loop * Repair cpp-package * signed/unsigned fixed in cuda file * lint fixes in tests and cpp-package directories * more lint * use IsWriting() helper * Fall-through for unsupported MKL shapes/types * Fall-through for unsupported MKL shapes/types * cleaner mkl_off approach * Warning only whem MKL is requested * Warning only whem MKL is requested * lint * .. * python problem fixed * python problem fixed * Merge branch 'batchnorm' into batchnorm_pr # Conflicts: # src/operator/batch_norm.cc # src/operator/batch_norm.cu # tests/cpp/operator/batchnorm_test.cc * lint fix * lint fix * lint fix * lint fix * lint fix * Fix visual c++ compile problem * . * . * All unit tests pass again * lint fix * fix strange compile errors in CUDNN batchnorm header * FInish using flags instead of bools * lint * Fix timing pass count for forward pass * Fix R script install roxygen problem * code formatting, addition of doc strings is causing IDE to add spaces before the calls * removed commented * cr comments * Change back to compilable code * For CPU mode, store as invstd * move testing code around a little * lint fix * Use AccReal in some places to avoid fp16 problems * Fix minor invstd problem in cuda version * remove unused scale param * add permutation unit test, handle cudnn doesn't like 3D * . * lint * . * Remove mkl_off * lint fix and time cudnn when enabled
2017-05-15 20:27:28 -07:00
CFLAGS += -O3 -DNDEBUG=1
compile debug
2015-08-23 21:26:16 +08:00
endif
[3rdparty] introduce tvm (#11130)
2018-06-02 17:58:25 -07:00
CFLAGS += -I$(TPARTYDIR)/mshadow/ -I$(TPARTYDIR)/dmlc-core/include -fPIC -I$(NNVM_PATH)/include -I$(DLPACK_PATH)/include -I$(TPARTYDIR)/tvm/include -Iinclude $(MSHADOW_CFLAGS)
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
LDFLAGS = -pthread $(MSHADOW_LDFLAGS) $(DMLC_LDFLAGS)
[MXNET-703] TensorRT runtime integration (#11325) * [MXNET-703] TensorRT runtime integration Co-authored-by: Clement Fuji-Tsang <caenorst@hotmail.com> Co-authored-by: Kellen Sunderland <kellen.sunderland@gmail.com> * correctly assign self._optimized_symbol in executor * declare GetTrtCompatibleSubsets and ReplaceSubgraph only if MXNET_USE_TENSORRT * add comments in ReplaceSubgraph * Addressing Haibin's code review points * Check that shared_buffer is not empty when USE_TENSORRT is set * Added check that TensorRT binding is for inference only * Removed redundant decl. * WIP Refactored TRT integration and tests * Add more build guards, remove unused code * Remove ccache report * Remove redundant const in declaration * Clean Cmake TRT files * Remove TensorRT env var usage We don't want to use environment variables with TensorRT yet, the logic being that we want to try and have as much fwd compatiblity as possible when working on an experimental feature. Were we to add env vars they would have to be gaurenteed to work in the future until a major version change. Moving the functionality to a contrib call reduces this risk. * Use contrib optimize_graph instaed of bind * Clean up cycle detector * Convert lenet test to contrib optimize * Protect interface with trt build flag * Fix whitespace issues * Add another build guard to c_api * Move get_optimized_symbol to contrib area * Ignore gz files in test folder * Make trt optimization implicit * Remove unused declaration * Replace build guards with runtime errors * Change default value of TensorRT to off This is change applies to both TensorRT and non-TensorRT builds. * Warn user when TRT not active at runtime * Move TensorRTBind declaration, add descriptive errors * Test TensorRT graph execution, fix bugs * Fix lint and whitespace issues * Fix typo * Removed default value for set_use_tensorrt * Improved documentation and fixed spacing issues * Move static exec funcs to util files * Update comments to match util style * Apply const to loop element * Fix a few namespace issues * Make static funcs inline to avoid compiler warning * Remove unused inference code from lenet5_train * Add explicit trt contrib bind, update tests to use it * Rename trt bind call * Remove documentation that is not needed for trt * Reorder arguments, allow position calling
2018-08-10 02:38:04 -07:00
Enable C++ coverage (#12642) * Enable C++ coverage for make * Enable C++ test coverage for CMake * Add environment variable GCOV_PREFIX * Stash GCNO files * Enable test coverage for Scala * Fix -P not found error * Remove debug comments * Add linking with coverage * Trigger separate build * Add ignored files
2018-09-24 16:30:38 +02:00
ifeq ($(ENABLE_TESTCOVERAGE), 1)
CFLAGS += --coverage
LDFLAGS += --coverage
endif
[MXNET-703] TensorRT runtime integration (#11325) * [MXNET-703] TensorRT runtime integration Co-authored-by: Clement Fuji-Tsang <caenorst@hotmail.com> Co-authored-by: Kellen Sunderland <kellen.sunderland@gmail.com> * correctly assign self._optimized_symbol in executor * declare GetTrtCompatibleSubsets and ReplaceSubgraph only if MXNET_USE_TENSORRT * add comments in ReplaceSubgraph * Addressing Haibin's code review points * Check that shared_buffer is not empty when USE_TENSORRT is set * Added check that TensorRT binding is for inference only * Removed redundant decl. * WIP Refactored TRT integration and tests * Add more build guards, remove unused code * Remove ccache report * Remove redundant const in declaration * Clean Cmake TRT files * Remove TensorRT env var usage We don't want to use environment variables with TensorRT yet, the logic being that we want to try and have as much fwd compatiblity as possible when working on an experimental feature. Were we to add env vars they would have to be gaurenteed to work in the future until a major version change. Moving the functionality to a contrib call reduces this risk. * Use contrib optimize_graph instaed of bind * Clean up cycle detector * Convert lenet test to contrib optimize * Protect interface with trt build flag * Fix whitespace issues * Add another build guard to c_api * Move get_optimized_symbol to contrib area * Ignore gz files in test folder * Make trt optimization implicit * Remove unused declaration * Replace build guards with runtime errors * Change default value of TensorRT to off This is change applies to both TensorRT and non-TensorRT builds. * Warn user when TRT not active at runtime * Move TensorRTBind declaration, add descriptive errors * Test TensorRT graph execution, fix bugs * Fix lint and whitespace issues * Fix typo * Removed default value for set_use_tensorrt * Improved documentation and fixed spacing issues * Move static exec funcs to util files * Update comments to match util style * Apply const to loop element * Fix a few namespace issues * Make static funcs inline to avoid compiler warning * Remove unused inference code from lenet5_train * Add explicit trt contrib bind, update tests to use it * Rename trt bind call * Remove documentation that is not needed for trt * Reorder arguments, allow position calling
2018-08-10 02:38:04 -07:00
ifeq ($(USE_TENSORRT), 1)
CFLAGS += -I$(ROOTDIR) -I$(TPARTYDIR) -DONNX_NAMESPACE=$(ONNX_NAMESPACE) -DMXNET_USE_TENSORRT=1
LDFLAGS += -lprotobuf -pthread -lonnx -lonnx_proto -lnvonnxparser -lnvonnxparser_runtime -lnvinfer -lnvinfer_plugin
endif
# -L/usr/local/lib
proper debug flag (-g -G) for nvcc
2015-10-24 15:53:35 -07:00
ifeq ($(DEBUG), 1)
add -Werror to NVCCFLAGS (#6774) * add -Werror to NVCCFLAGS * add -Werror to NVCCFLAGS * RTLD_LOCAL instead of GLOBAL * fix device/host cross-call warnings
2017-06-22 15:24:38 -07:00
NVCCFLAGS += -std=c++11 -Xcompiler -D_FORCE_INLINES -g -G -O0 -ccbin $(CXX) $(MSHADOW_NVCCFLAGS)
proper debug flag (-g -G) for nvcc
2015-10-24 15:53:35 -07:00
else
add -Werror to NVCCFLAGS (#6774) * add -Werror to NVCCFLAGS * add -Werror to NVCCFLAGS * RTLD_LOCAL instead of GLOBAL * fix device/host cross-call warnings
2017-06-22 15:24:38 -07:00
NVCCFLAGS += -std=c++11 -Xcompiler -D_FORCE_INLINES -O3 -ccbin $(CXX) $(MSHADOW_NVCCFLAGS)
proper debug flag (-g -G) for nvcc
2015-10-24 15:53:35 -07:00
endif
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
add option for segfault logger (#7836) * add option for segfault logger * rename
2017-09-10 22:22:19 -07:00
# CFLAGS for segfault logger
ifeq ($(USE_SIGNAL_HANDLER), 1)
CFLAGS += -DMXNET_USE_SIGNAL_HANDLER=1
endif
Caffe without the patch, cpp-package fixed also with Caffe plugin (#5573) * PEP8 indentation fix * Remove executable flag * Remove executable flags * Fixing cpp-package including problems with caffe converter * Fix warnings * Remove need for caffe patch for caffe plugin * Ignore cmake paths, remove rebuildable op.h for cpp-package * cpp-package examples fixed. Makefile and CMake of op.h and examples added. * cpp-package examples fixed. Makefile and CMake of op.h and examples added. * Fix source file * Better example.mk * Turn off caffe by default * lint fixes * Trying to figure out how to fix submodules * nohub problem on travis so force retry * Edited test to just run make instead of obsolete make example * Fix cpp-package op.h bootstrapping with CMake * Build tweaking cpp-package * Lint fixes * op.h generator * FIx cpp-package for latest merge from master * Fix lint * static link * link whole static lib * Trigger another build attempt * Trigger another build attempt * Add caffe plugin support (disabled until dependencies can be added to build machine) * Add cufft library * win32 cufft library * don't link unit tests as static * Add include for rebuildable protobuf header caffe.pb.h
2017-03-30 20:13:36 -07:00
# Caffe Plugin
ifdef CAFFE_PATH
update mklml and mkl mac support (#7587)
2017-08-29 17:14:01 -07:00
CFLAGS += -DMXNET_USE_CAFFE=1
Caffe without the patch, cpp-package fixed also with Caffe plugin (#5573) * PEP8 indentation fix * Remove executable flag * Remove executable flags * Fixing cpp-package including problems with caffe converter * Fix warnings * Remove need for caffe patch for caffe plugin * Ignore cmake paths, remove rebuildable op.h for cpp-package * cpp-package examples fixed. Makefile and CMake of op.h and examples added. * cpp-package examples fixed. Makefile and CMake of op.h and examples added. * Fix source file * Better example.mk * Turn off caffe by default * lint fixes * Trying to figure out how to fix submodules * nohub problem on travis so force retry * Edited test to just run make instead of obsolete make example * Fix cpp-package op.h bootstrapping with CMake * Build tweaking cpp-package * Lint fixes * op.h generator * FIx cpp-package for latest merge from master * Fix lint * static link * link whole static lib * Trigger another build attempt * Trigger another build attempt * Add caffe plugin support (disabled until dependencies can be added to build machine) * Add cufft library * win32 cufft library * don't link unit tests as static * Add include for rebuildable protobuf header caffe.pb.h
2017-03-30 20:13:36 -07:00
endif
add travis
2015-07-03 17:06:23 -06:00
ifndef LINT_LANG
compatibility with opencv4 (#14313) * compatibility with opencv4 * update makefile * update Makefile * fix Makefile * fix lib path * update make.mk * add -lopencv_highgui * retrigger CI * update makefile * remove libs-L * Fix OpenCV linking order * try to fix the bug of linking static libraries
2019-03-07 10:10:02 +08:00
LINT_LANG = "all"
add travis
2015-07-03 17:06:23 -06:00
endif
Adjust MKLDNN compile flags order to resolve some corner case (#9936)
2018-03-16 13:55:27 +08:00
ifeq ($(USE_MKLDNN), 1)
CFLAGS += -DMXNET_USE_MKLDNN=1
CFLAGS += -DUSE_MKL=1
CFLAGS += -I$(ROOTDIR)/src/operator/nn/mkldnn/
ifneq ($(MKLDNNROOT), $(MKLROOT))
CFLAGS += -I$(MKLROOT)/include
LDFLAGS += -L$(MKLROOT)/lib
endif
Revert "Bumped minor version from 1.4.0 to 1.5.0 on master, updated License file" (#13558) * Revert "Chi_square_check for discrete distribution fix (#13543)" This reverts commit cf6e8cbd035bf315b3e8280416468a629c780d03. * Revert "Updated docs for randint operator (#13541)" This reverts commit e0ff3c36ee171386fef01fb86c54c343e4b04c14. * Revert "Simplifications and some fun stuff for the MNIST Gluon tutorial (#13094)" This reverts commit 8bbac827742c21607a863137792f03bd09847419. * Revert "Fix #13521 (#13537)" This reverts commit f6b4665995f8f8ff32862a029b2074475d8467eb. * Revert "Add a retry to qemu_provision (#13551)" This reverts commit f6f840110d74111f98c20eab5b08d64a46ebf0cd. * Revert "[MXNET-769] Use MXNET_HOME in a tempdir in windows to prevent access denied due t… (#13531)" This reverts commit bd8e0f8356676749ecae16ec38a366b4cc00bf15. * Revert "[MXNET-1249] Fix Object Detector Performance with GPU (#13522)" This reverts commit 1c8972c3c8f832519364916865541f48597581c7. * Revert "Fixing a 404 in the ubuntu setup doc (#13542)" This reverts commit cb0db290adcfd0fce956d02c234f81d453e41013. * Revert "Bumped minor version from 1.4.0 to 1.5.0 on master, updated License file (#13478)" This reverts commit 40db61908000ee86d21aac847ff2225807d6c168.
2018-12-06 17:44:06 +08:00
CFLAGS += -I$(MKLDNNROOT)/include
Revert "Feature/mkldnn static (#13628)" (#13638) This reverts commit 5bcf2bd6e8b48fa27bfcfdafd06401ec2d28978b.
2018-12-13 16:01:30 -08:00
LDFLAGS += -L$(MKLDNNROOT)/lib -lmkldnn -Wl,-rpath,'$${ORIGIN}'
Adjust MKLDNN compile flags order to resolve some corner case (#9936)
2018-03-16 13:55:27 +08:00
endif
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
# setup opencv
fix opencv local path
2015-10-02 11:52:44 +08:00
ifeq ($(USE_OPENCV), 1)
compatibility with opencv4 (#14313) * compatibility with opencv4 * update makefile * update Makefile * fix Makefile * fix lib path * update make.mk * add -lopencv_highgui * retrigger CI * update makefile * remove libs-L * Fix OpenCV linking order * try to fix the bug of linking static libraries
2019-03-07 10:10:02 +08:00
CFLAGS += -DMXNET_USE_OPENCV=1
fix Makefile (#14424)
2019-03-14 16:00:36 +08:00
ifneq ($(filter-out NONE, $(USE_OPENCV_INC_PATH)),)
compatibility with opencv4 (#14313) * compatibility with opencv4 * update makefile * update Makefile * fix Makefile * fix lib path * update make.mk * add -lopencv_highgui * retrigger CI * update makefile * remove libs-L * Fix OpenCV linking order * try to fix the bug of linking static libraries
2019-03-07 10:10:02 +08:00
CFLAGS += -I$(USE_OPENCV_INC_PATH)/include
fix Makefile (#14424)
2019-03-14 16:00:36 +08:00
ifeq ($(filter-out NONE, $(USE_OPENCV_LIB_PATH)),)
compatibility with opencv4 (#14313) * compatibility with opencv4 * update makefile * update Makefile * fix Makefile * fix lib path * update make.mk * add -lopencv_highgui * retrigger CI * update makefile * remove libs-L * Fix OpenCV linking order * try to fix the bug of linking static libraries
2019-03-07 10:10:02 +08:00
$(error Please add the path of OpenCV shared library path into `USE_OPENCV_LIB_PATH`, when `USE_OPENCV_INC_PATH` is not NONE)
endif
LDFLAGS += -L$(USE_OPENCV_LIB_PATH)
ifneq ($(wildcard $(USE_OPENCV_LIB_PATH)/libopencv_imgcodecs.*),)
LDFLAGS += -lopencv_imgcodecs
endif
ifneq ($(wildcard $(USE_OPENCV_LIB_PATH)/libopencv_highgui.*),)
LDFLAGS += -lopencv_highgui
endif
else
ifeq ("$(shell pkg-config --exists opencv4; echo $$?)", "0")
OPENCV_LIB = opencv4
else
OPENCV_LIB = opencv
endif
CFLAGS += $(shell pkg-config --cflags $(OPENCV_LIB))
LDFLAGS += $(shell pkg-config --libs-only-L $(OPENCV_LIB))
LDFLAGS += $(filter -lopencv_imgcodecs -lopencv_highgui, $(shell pkg-config --libs-only-l $(OPENCV_LIB)))
endif
LDFLAGS += -lopencv_imgproc -lopencv_core
add showme example and im2rec, modify io,=.md
2015-09-21 01:55:32 +08:00
BIN += bin/im2rec
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
else
compatibility with opencv4 (#14313) * compatibility with opencv4 * update makefile * update Makefile * fix Makefile * fix lib path * update make.mk * add -lopencv_highgui * retrigger CI * update makefile * remove libs-L * Fix OpenCV linking order * try to fix the bug of linking static libraries
2019-03-07 10:10:02 +08:00
CFLAGS += -DMXNET_USE_OPENCV=0
pass compilation, not tested
2015-09-04 01:15:31 +08:00
endif
rename narray->ndarray
2015-09-13 10:41:34 -07:00
ifeq ($(USE_OPENMP), 1)
simplify mac mkldnn build (#12724) * remove guard that prevent omp flag in mac * udpate doc for mac make build * update docs * update readme * set opencv to 1 in instructions * remove disable opencv line * update mac docs * fix indent
2018-10-16 09:30:39 -07:00
CFLAGS += -fopenmp
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
endif
add NNPACK support for high convolution inference perf (#3666) * add NNPACK support for high convolution inference perf * set USE_NNPACK to 0 * Fix header declaration * Fix input_size init value 1. data's shape is BxCxHxW, input_size is {width,height} 2. improve algorithm selection policy * Fix lint error
2016-11-10 12:19:46 +08:00
ifeq ($(USE_NNPACK), 1)
CFLAGS += -DMXNET_USE_NNPACK=1
LDFLAGS += -lnnpack
endif
Kernel operator tuning (#8686) * Refreshed branch bc_tune * local-build openmp as static * trigger * Somehow broadcast found its way back in, removed again * Trigger rebuild
2017-11-21 06:49:51 -08:00
ifeq ($(USE_OPERATOR_TUNING), 1)
CFLAGS += -DMXNET_USE_OPERATOR_TUNING=1
endif
enable use of lapack by default (#6704) * enable use of lapack by default * corrected search paths for lapack lib
2017-06-18 14:23:29 -07:00
# verify existence of separate lapack library when using blas/openblas/atlas
# switch off lapack support in case it can't be found
# issue covered with this
# - for Ubuntu 14.04 or lower, lapack is not automatically installed with openblas
# - for Ubuntu, installing atlas will not automatically install the atlas provided lapack library
Fix gperftools/jemalloc and lapack warning bug. (#11110) * Fix gperftools/jemalloc and lapack warning bug. * mend
2018-06-18 06:23:55 +08:00
# - for rhel7.2, try installing the package `lapack-static` via yum will dismiss this warning.
enable use of lapack by default (#6704) * enable use of lapack by default * corrected search paths for lapack lib
2017-06-18 14:23:29 -07:00
# silently switching lapack off instead of letting the build fail because of backward compatibility
Add BLAS3 and LAPACK routines (#6538) * Added linear algebra operators * more comments about style of wrapper interface * more appropriate fatal exit when lapack does not exist * more comments on row/col-major ordering * added config switch for lapack usage * switched lapack usage off by default
2017-06-13 14:13:20 -07:00
ifeq ($(USE_LAPACK), 1)
Refactor operators and add MKLDNN (#9677) * Remove MKL code. * Integrate MKLDNN. Update MXNet for MKLDNN. Enable MKLDNN Relu. Fix a compilation error. Change Makefile for MKLDNN. Remove infer storage in convolution. Update MXNet for MKLDNN. Support MKLDNN storage type in python. Update activation. Add MKLDNN base classes. Implement MKLDNN fully connected. Add MKLDNN convolution. Update MKLDNN interface in NDArray. MKLDNN convolution handle CreateMKLDNNData failure. Add another GetMKLDNNData in NDArray. Have mkldnn to define the data format. Create output MKLDNN memory explicitly for FC. Fix a bug in NDArray. Fix a bug in GetWeightDesc. Convert data layout if necessary in FC. remove unnecessary print in MKLDNN convolution. Add MKLDNN deconvolution. Add MKLDNNStream to manage primitives and memories. Use MKLDNNStream to register memory in NDArray. Use MKLDNNStream to manage resources in operators. Handle kAddTo in MKLDNN operators. Fix a bug in deconvolution. Fix bugs in NDArray. Revert "Fix bugs in NDArray." This reverts commit f5624a4aa9f9b9f9fe31f5e6cfa7a9752838fc4e. Fix a bug in NDArray. Fix a bug in NDArray. Reorder MKLDNN memory to default format in SetTBlob. Disable MKLDNN correctly. Fix a bug in activation. Reshape of NDArray supports MKLDNN. Fix a memory ref bug in NDArray. Reshape NDArray in MKLDNN FullyConnected. Fix data format conversion. Create MKLDNN NDArray in python. Support Slice for MKLDNN NDArray. Reduce the overhead of summing the result to the output array. Avoid unnecessary memory copy in NDArray. Fix a bug in data reordering. Fix a bug in NDArray. Don't hard code MKLDNN type. Support dilation in MKLDNN convolution. Fix a bug in sum results. Rewrite GetMKLDNNData. Add prepare_mkldnn.sh Enable MKLDNN activation. Fix a bug on FullyConnected. Handle 3 dims for MKLDNN NDArray. Fix a bug in MKLDNN FC. Support MKLDNN storage in KV store. Fix a bug in executor for non-default NDArray. Fix a link error in cast_storage.cc. Remove unnecessary function def Fall back to def storage if the type isn't supported by MKLDNN. Use NDArray for MKLDNN in python. Reshape output of MKLDNN convolution. Fix a bug in NDArray. Support more operations in MKLDNN NDArray. Fix a bug in deconvolution. Fix bugs in MKLDNN deconvolution. We still need to compute bias correctly. Have elemwise binary ops to fall to default for MKLDNN. Limit the cases that MKLDNN operations are called. Force the layout of mkldnn::memory from NDArray. Add MKLDNN softmax. Fix output storage type of MKLDNN softmax. Add MKLDNN sum. Fix a bug in elemwise sum. Fix a bug in MKLDNN softmax. Fix a bug in imperative. Clean up dispatch modes. Remove redundant code. MKLDNN Pooling Op integration MKLDNN Pooling Op integration add missing file fix mkldnn pooling op workspace issue handle workspace in MKLDNN pooling correctly. Use a non-MKLDNN op for testing. Allow to share arguments and their gradients between executors. Avoid using MKLDNN pooling when it's not supported. Support MKLDNN properly. Choose MKLDNN softmax more carefully. Fix a bug in MKLDNN pooling. Fall back if MKLDNN pooling isn't supported. Fix a bug in Slice of NDArray. Use int32 for workspace memory. Exclude MKLDNN act with tanh. Have two Reshape functions in NDArray. Copy data for NDArray with diff shapes. Add MKLDNN copy. Add MKLDNN version of elemwise_add. Add MKLDNN version of Flatten. add mkldnn surport for concat simplify MKLDNN Flatten. Enalbe MKLDNN deconvolution with bias. Fix a bug in CuDNN deconvolution. avoid using MKLDNNStorage when it's not defined. Remove ./cudnn_lrn-inl.h Fix for make lint. add mkldnn surport for concat fix the coding style for pr of mkldnn concat Only add input data for MKLDNN concat backward Remove unnecessary TODO. remove unnecessary __repr__ in MKLNDArray. better condition check for readability. Use macro when including mkldnn.hpp. Revert "Use CoreOpRunner for refactored Ops." This reverts commit a28586fc25950cc006cb317e26e0d17541ef0586. Fix a bug in test core. Limit MKLDNN ops being used. Fix complains from "make pylint" Move ContainStorage to common/utils.h Limit MKLDNN concat being used. Add license. Fix amalgamation Fix compilation error in mkldnn_ops-inl.h Fix a bug in deconvolution. Fix a bug in pooling. MKLDNN ops allocates temp mem. Fix a bug in pooling. Allocate align memory from temp space. Have parameter gradients stored in the default storage. Handle all cases in CopyFrom. Ensure NDArray returns memory with right memory descriptors. use auto to define memory in the operator. Use raw pointer for mkldnn memory. Move more code to mkldnn_base.cc Fix a compilation error. Address review comments. fix a bug in activation backward. Miss a macro in mkldnn_base.cc Fix a bug in data iterator in examples. Avoid memory allocation in ReshapeMKLDNN. Avoid memory allocation in storage cast. Fix a bug in cast storage. Handle sliced MKLDNN NDArray. Use memcpy if NDArray uses default format. Revert "Limit MKLDNN ops being used." This reverts commit 75e2ae570d03483868ec4ed8ed46015c7fa6c6fb. Enable mkldnn act backward has the same input layout. Fix a bug in mkldnn activation. Use MKLDNN sum in more cases. Improve perf of reorder. Avoid memory reorder in conv and deconv. Avoid unnecessary storage cast in fallback path. Revert "Use MKLDNN sum in more cases." This reverts commit 7a21ebca8bbe17fde49c3b1ca3f31b835a33afb8. Handle sliced ndarray in more cases. Fix a complain from make lint. Update Jenkins to test MKLDNN. debug compiling mkldnn. Use MKLDNN sum in more cases. Add mkldnn as a submodule. Compile with mkldnn in 3rdparty. Fix some coding styles. write the path to mkldnn lib in libmxnet.so. use rpath with $ORIGIN. Pack all lib files in Jenkins. pack and unpack mxnet with MKLDNN. Update Jenkinsfile Update Jenkinsfile Add mkldnn batch normalization Fix bugs in BN. Avoid memory allocation in MKLDNNCopy. only use MKLDNN BatchNorm for special cases. MKLDNN BatchNorm doesn't work well on the default layout. Add MKL-DNN based LRN Code Style Changes Fix a bug in BN. Fix a bug in LRN. Handle non-default storage in memory plan. Fix coding style. Fix a compilation error without mkldnn. Fix some coding styles for batch norm Improve forward of convolution. Add openmp and simd support to BN operator Retrieve MKLDNN Conv primitive based on signature. Retrieve Act primitive based on its signature. Fix a bug in pooling. Diable some MKLDNN activation and pooling. Cast MKLDNN storage with diff data type. Check if it's a view of NDArray. Reshaped and sliced arrays share the same chunks. Implement caching MKLDNN Act correctly. Fix a bug in check_consistency. Fix a potential bug when destroying NDArray. Fix bugs when allocating mem in NDArray. Fix coding style. Add micro when using mkldnn in ndarray. Fix a compilation error. Fix a bug in concat. Remove MKLDNNStorage. handle diff layouts in CopyFromToDnsImpl. Fallback correctly. Force weight grad to use default layout. Reorder weight arrays in (de)conv for faster inference. Avoid caching TBlob from NDArray. This commit may add some overhead of managing NDArray for each fallback. Fix a bug in Flatten. handle ndarray with def layout in mkldnn BN correctly. Align to page when mkldnn is enabled. Use default mem alloc for mkldnn. Reuse NDArrays. Support WriteInplace for sum. fix complains from "make lint". Avoid reallocation in NDArray. Handle weight arrays with special MKLDNN layouts. Remove unnecessary GetWeights. Fix compilation error without MKLDNN. Fix a bug in (de)conv for weight arrays. Fix a minor bug in MKLDNN conv. Fix a bug in MKLDNNOpSignature. Reimplement fallback for MKLDNN ops. Fix a bug in FallbackExecutor. Add params in hashcode. Invalidate data in outputs to accelerate. Fix a minor bug. Update mkldnn_base-inl.h Add primitive caching for Pooling forward computation Add hashcode in pooling parameters. Support NDArray copy with types unsupported by MKLDNN. Avoid using MKLDNN concat for negative dimension. Fix make lint complain. Disable mkldnn avg pooling for now. Fix a compile warning. Fix compile error when MKLDNN is disabled. OP primitive cache: use memory as signature for MKLDNN storage type Remove MKLDNN array in python. Disable Clang tests in Jenkins. Use mklml dockers to test mkldnn. Update MKLDNN repo to zhengda's mkldnn repo. Update MKLDNN repo to ashok's. Fix a bug in fallback. Change avg pooling algorithm to pooling_avg_include_padding Fix a code style in mkldnn pooling. Temp fix a bug in FC. Revert "Disable Clang tests in Jenkins." This reverts commit b4efa8f89592d30a27f9c30e2237e9420ac6749a. Rebase and Refactor deconv (#20) * rebase to Da,Zheng refactor branch Jan.14, add signature for mkldnn Deconv and modify classMKLDNNDeconvForward * fix make lint complains A simple way of caching BN inference. cache BN forward for both training and inference. Fix some minor problems in BN. Fix a bug in caching BN. force to build with avx2 in Jenkins. Remove the remaining MKLDNNStorageType Some minor updates in NDArray. a lot of updates to address comments. minor changes. * Use NNVM interface. Use NNVM interface for upsampling. Use NNVM interface for convolution. Use NNVM interface for deconvolution. Use NNVM interface for FullyConnected. Move NNVM interface to batch norm. Use NNVM interface for depthwise convolution. Use NNVM interface for softmax activation. Use NNVM interface for pooling. use NNVM interface for dropout. Use NNVM interface for activation. Use NNVM interface for CuDNN batch norm. Use NNVM interface for CuDNN pooling. Use NNVM interface for CuDNN softmax activation. Use NNVM interface for CuDNN activation. Use NNVM interface for CuDNN convolution. Use NNVM interface for CuDNN deconvolution. Move concat to nn/ Use NNVM interface for concat. Fix headers in concat. Move lrn to nn/. Use NNVM interface for LRN. Fix a compilation error in convolution. Fix a compilation error in activation. Fix coding style. Fix coding style for make lint. use enums in batch norm. Use CoreOpRunner for refactored Ops. Make FullyConnected stateless. Make upsampling stateless. Make pooling stateless. Make batchnorm stateless. Make SoftmaxActivation stateless. Fix a code style problem. pass amalgamation test for batch norm. pass amalgamation test for dropout. Get convolution ops from a function. Fix compilation errors for GPU. Fix thread local in diff platforms. Avoid using thread_local for non-CuDNN conv/deconv. Remove TODO in deconv. Fix a bug in batch norm. Fix a bug in fully connected. Don't set #inputs for backward convolution. Revert "Make pooling stateless." * revert modification in test_executor. * Fix a bug in FlattenStorageType. * Remove BN debug. * Remove remaining MXNET_USE_MKL2017 * Remove unused code in pooling. * Fixing bugs in gtests. * Fix lint errors. * a lot of minor updates to address comments. * Fix coding style in MKLDNN Pooling (#22) * revert the code change in the previous code refactor. * Fix a bug in pooling. * LRN coding style changes (#21) * LRN coding style change * Add const for local variables * Add req for LRN forward * rebase code * align API interface * revert modification in test_executor. * cast storage with MKLDNN properly. * Minor updates to address comments. * some minor updates. * Switch to the master branch of MKLDNN. * Minor updates to address comments. * Update activation.cc * Fix a bug in convert NDArray. * Add gluon model zoo tests. * Update GPU tests on model zoo. * Avoid using mobilenet for GPU tests with gluon models. mobilenet can't pass the test even without MKLDNN. * Update GPU tests on gluon. * change cmake to compile MKLDNN. * update cmake for MKLDNN. * Implement align myself. * Switch to intel/mkl-dnn. * Fix errors in align unittest. * Add unit test for LRN. * fix a compilation error. * use storage_type_assign to determine storage type. * avoid global pooling in mkldnn. There is a bug in global pooling in mkldnn. * compare all MKLDNN ops with native impls. add MXNET_MKLDNN_DEBUG to control the test. * Fix a bug in testing correctness. * print the name of buggy operator. * undo some modifications. * Fix a bug on reshaped array. * avoid testing outputs with NullOp. * turn on MKLDNN tests in Jenkins. * print each operator in MKLDNN tests. * rename test_gluon_model_zoo.py * Create hashcode for operator parameters properly. * Add USE_MKL2017 back. * Print warning messages. * move batchnorm tests to nnvm interface. * Delete batchnorm v1 tests. * Get inputs and outputs in batchnorm tests. * disable batchnorm tests for now. * Fix GPU tests on gluon model zoo. * Fix lint complains in tests. * Remove simd from openmp instructions in BatchNorm (#24) * Remove warnings. * Fix MKLDNN 1st compile failure issue (#23) * Fix compilation errors. * Remove ARCH_OPT in Jenkins. * Revert "avoid global pooling in mkldnn." This reverts commit f6efd342e64968cb848c9193d80e929968b8052c. * Move to the latest MKLDNN. This fixes the bug in global pooling. * WIP unit tests (#25) * WIP unit tests * some backward items initialized * Make more C++ unit tests work for batch norm (#28) * WIP unit tests * some backward items initialized * some backward items initialized * some backward items initialized * first unit test working * Working on types * backward types working for fp16 on first unit test * backward types working for fp16 on first unit test * backward types working for fp16 on first unit test * . * . * some tests working * fix input data * hangle gpu<->cpu for setting values * gpu working * gpu working * CAccessAsCPU class * Fix varying type in AccessAsCPU * starting to add channel axis tests * TestChannelAxisSimple * TestChannelAxisSimple * run bidirectional * run bidirectional * run bidirectional * CLEANUP * CLEANUP * .. * noaxis * .. * lint * revert * revert * Fix lint complains. * Fix a minor problem in Makefile. * fix GPU pooling. * Disable modelzoo inference tests. * update accuracy checks for MKLDNN. * Fix MKLDNN pooling for global pooling. * Fix Jenkins. * Fix a bug in Jenkins. * Fix Jenkins
2018-02-15 14:44:34 -08:00
ifeq ($(USE_BLAS),$(filter $(USE_BLAS),blas openblas atlas mkl))
Compile with LAPACK shared library (#11483) * Compile with LAPACK shared library * Add missing bracket * Reorder LAPACK paths
2018-07-03 01:56:54 +08:00
ifeq (,$(wildcard $(USE_LAPACK_PATH)/liblapack.a))
ifeq (,$(wildcard $(USE_LAPACK_PATH)/liblapack.so))
enable use of lapack by default (#6704) * enable use of lapack by default * corrected search paths for lapack lib
2017-06-18 14:23:29 -07:00
ifeq (,$(wildcard /lib/liblapack.a))
Compile with LAPACK shared library (#11483) * Compile with LAPACK shared library * Add missing bracket * Reorder LAPACK paths
2018-07-03 01:56:54 +08:00
ifeq (,$(wildcard /lib/liblapack.so))
enable use of lapack by default (#6704) * enable use of lapack by default * corrected search paths for lapack lib
2017-06-18 14:23:29 -07:00
ifeq (,$(wildcard /usr/lib/liblapack.a))
Compile with LAPACK shared library (#11483) * Compile with LAPACK shared library * Add missing bracket * Reorder LAPACK paths
2018-07-03 01:56:54 +08:00
ifeq (,$(wildcard /usr/lib/liblapack.so))
Locate liblapack.a in /usr/lib64 (#8073)
2017-12-09 01:16:03 +00:00
ifeq (,$(wildcard /usr/lib64/liblapack.a))
Compile with LAPACK shared library (#11483) * Compile with LAPACK shared library * Add missing bracket * Reorder LAPACK paths
2018-07-03 01:56:54 +08:00
ifeq (,$(wildcard /usr/lib64/liblapack.so))
enable use of lapack by default (#6704) * enable use of lapack by default * corrected search paths for lapack lib
2017-06-18 14:23:29 -07:00
USE_LAPACK = 0
CI docker revamp; Add Jetson, Raspberry and CentOS 7 build [MXNET-42][MXNET-43] (#9995) * Start dockerfile revamp * Fix lint * Fix lint * Fix lint * Add python unit test * Add new dependency * Add user creation * Fix file permission * Determine USER_ID automatically * REmove ENV command in dockerfile * Remove python nose timer, improve useradd * ENable nvidia docker * Add remaining unittests * Add CentOS 7 unittests * Add integration tests * Add TVM and LLVM to dockerfile * Add ARMv7 and ARMv8 to build * Fix amalgamation build * Improvements and android_arm64 fixes, missing removing pthread * Jetson fix (install unzip) * Build jetson with make until issue with libomp.so is resolved #10011 * Fix Amalgamation builds * Fix Jetson build by switching to cmake * Assign CentOS gpu test to gpu instance * Fix R builds * Assign jobs to right docker containers * Fix missing file permissions inside docker * Create homedir on centos * Enable lapack * Fix Lapack on Cent OS 7 * Disable MXNET_MKLDNN_DEBUG * Delete Dockerfiles * Last general refinements before finish * Remove docker_multiarch folder, superseeded by the new ci scripts * Address review comments * Fix Caffe Integrationstest * Fix deploy stage * Address review comments * Address review comments * Enable script to run on Mac
2018-03-09 23:33:10 +01:00
$(warning "USE_LAPACK disabled because libraries were not found")
enable use of lapack by default (#6704) * enable use of lapack by default * corrected search paths for lapack lib
2017-06-18 14:23:29 -07:00
endif
endif
endif
Add BLAS3 and LAPACK routines (#6538) * Added linear algebra operators * more comments about style of wrapper interface * more appropriate fatal exit when lapack does not exist * more comments on row/col-major ordering * added config switch for lapack usage * switched lapack usage off by default
2017-06-13 14:13:20 -07:00
endif
endif
Locate liblapack.a in /usr/lib64 (#8073)
2017-12-09 01:16:03 +00:00
endif
Compile with LAPACK shared library (#11483) * Compile with LAPACK shared library * Add missing bracket * Reorder LAPACK paths
2018-07-03 01:56:54 +08:00
endif
endif
endif
endif
enable use of lapack by default (#6704) * enable use of lapack by default * corrected search paths for lapack lib
2017-06-18 14:23:29 -07:00
# lapack settings.
ifeq ($(USE_LAPACK), 1)
ifneq ($(USE_LAPACK_PATH), )
LDFLAGS += -L$(USE_LAPACK_PATH)
endif
Refactor operators and add MKLDNN (#9677) * Remove MKL code. * Integrate MKLDNN. Update MXNet for MKLDNN. Enable MKLDNN Relu. Fix a compilation error. Change Makefile for MKLDNN. Remove infer storage in convolution. Update MXNet for MKLDNN. Support MKLDNN storage type in python. Update activation. Add MKLDNN base classes. Implement MKLDNN fully connected. Add MKLDNN convolution. Update MKLDNN interface in NDArray. MKLDNN convolution handle CreateMKLDNNData failure. Add another GetMKLDNNData in NDArray. Have mkldnn to define the data format. Create output MKLDNN memory explicitly for FC. Fix a bug in NDArray. Fix a bug in GetWeightDesc. Convert data layout if necessary in FC. remove unnecessary print in MKLDNN convolution. Add MKLDNN deconvolution. Add MKLDNNStream to manage primitives and memories. Use MKLDNNStream to register memory in NDArray. Use MKLDNNStream to manage resources in operators. Handle kAddTo in MKLDNN operators. Fix a bug in deconvolution. Fix bugs in NDArray. Revert "Fix bugs in NDArray." This reverts commit f5624a4aa9f9b9f9fe31f5e6cfa7a9752838fc4e. Fix a bug in NDArray. Fix a bug in NDArray. Reorder MKLDNN memory to default format in SetTBlob. Disable MKLDNN correctly. Fix a bug in activation. Reshape of NDArray supports MKLDNN. Fix a memory ref bug in NDArray. Reshape NDArray in MKLDNN FullyConnected. Fix data format conversion. Create MKLDNN NDArray in python. Support Slice for MKLDNN NDArray. Reduce the overhead of summing the result to the output array. Avoid unnecessary memory copy in NDArray. Fix a bug in data reordering. Fix a bug in NDArray. Don't hard code MKLDNN type. Support dilation in MKLDNN convolution. Fix a bug in sum results. Rewrite GetMKLDNNData. Add prepare_mkldnn.sh Enable MKLDNN activation. Fix a bug on FullyConnected. Handle 3 dims for MKLDNN NDArray. Fix a bug in MKLDNN FC. Support MKLDNN storage in KV store. Fix a bug in executor for non-default NDArray. Fix a link error in cast_storage.cc. Remove unnecessary function def Fall back to def storage if the type isn't supported by MKLDNN. Use NDArray for MKLDNN in python. Reshape output of MKLDNN convolution. Fix a bug in NDArray. Support more operations in MKLDNN NDArray. Fix a bug in deconvolution. Fix bugs in MKLDNN deconvolution. We still need to compute bias correctly. Have elemwise binary ops to fall to default for MKLDNN. Limit the cases that MKLDNN operations are called. Force the layout of mkldnn::memory from NDArray. Add MKLDNN softmax. Fix output storage type of MKLDNN softmax. Add MKLDNN sum. Fix a bug in elemwise sum. Fix a bug in MKLDNN softmax. Fix a bug in imperative. Clean up dispatch modes. Remove redundant code. MKLDNN Pooling Op integration MKLDNN Pooling Op integration add missing file fix mkldnn pooling op workspace issue handle workspace in MKLDNN pooling correctly. Use a non-MKLDNN op for testing. Allow to share arguments and their gradients between executors. Avoid using MKLDNN pooling when it's not supported. Support MKLDNN properly. Choose MKLDNN softmax more carefully. Fix a bug in MKLDNN pooling. Fall back if MKLDNN pooling isn't supported. Fix a bug in Slice of NDArray. Use int32 for workspace memory. Exclude MKLDNN act with tanh. Have two Reshape functions in NDArray. Copy data for NDArray with diff shapes. Add MKLDNN copy. Add MKLDNN version of elemwise_add. Add MKLDNN version of Flatten. add mkldnn surport for concat simplify MKLDNN Flatten. Enalbe MKLDNN deconvolution with bias. Fix a bug in CuDNN deconvolution. avoid using MKLDNNStorage when it's not defined. Remove ./cudnn_lrn-inl.h Fix for make lint. add mkldnn surport for concat fix the coding style for pr of mkldnn concat Only add input data for MKLDNN concat backward Remove unnecessary TODO. remove unnecessary __repr__ in MKLNDArray. better condition check for readability. Use macro when including mkldnn.hpp. Revert "Use CoreOpRunner for refactored Ops." This reverts commit a28586fc25950cc006cb317e26e0d17541ef0586. Fix a bug in test core. Limit MKLDNN ops being used. Fix complains from "make pylint" Move ContainStorage to common/utils.h Limit MKLDNN concat being used. Add license. Fix amalgamation Fix compilation error in mkldnn_ops-inl.h Fix a bug in deconvolution. Fix a bug in pooling. MKLDNN ops allocates temp mem. Fix a bug in pooling. Allocate align memory from temp space. Have parameter gradients stored in the default storage. Handle all cases in CopyFrom. Ensure NDArray returns memory with right memory descriptors. use auto to define memory in the operator. Use raw pointer for mkldnn memory. Move more code to mkldnn_base.cc Fix a compilation error. Address review comments. fix a bug in activation backward. Miss a macro in mkldnn_base.cc Fix a bug in data iterator in examples. Avoid memory allocation in ReshapeMKLDNN. Avoid memory allocation in storage cast. Fix a bug in cast storage. Handle sliced MKLDNN NDArray. Use memcpy if NDArray uses default format. Revert "Limit MKLDNN ops being used." This reverts commit 75e2ae570d03483868ec4ed8ed46015c7fa6c6fb. Enable mkldnn act backward has the same input layout. Fix a bug in mkldnn activation. Use MKLDNN sum in more cases. Improve perf of reorder. Avoid memory reorder in conv and deconv. Avoid unnecessary storage cast in fallback path. Revert "Use MKLDNN sum in more cases." This reverts commit 7a21ebca8bbe17fde49c3b1ca3f31b835a33afb8. Handle sliced ndarray in more cases. Fix a complain from make lint. Update Jenkins to test MKLDNN. debug compiling mkldnn. Use MKLDNN sum in more cases. Add mkldnn as a submodule. Compile with mkldnn in 3rdparty. Fix some coding styles. write the path to mkldnn lib in libmxnet.so. use rpath with $ORIGIN. Pack all lib files in Jenkins. pack and unpack mxnet with MKLDNN. Update Jenkinsfile Update Jenkinsfile Add mkldnn batch normalization Fix bugs in BN. Avoid memory allocation in MKLDNNCopy. only use MKLDNN BatchNorm for special cases. MKLDNN BatchNorm doesn't work well on the default layout. Add MKL-DNN based LRN Code Style Changes Fix a bug in BN. Fix a bug in LRN. Handle non-default storage in memory plan. Fix coding style. Fix a compilation error without mkldnn. Fix some coding styles for batch norm Improve forward of convolution. Add openmp and simd support to BN operator Retrieve MKLDNN Conv primitive based on signature. Retrieve Act primitive based on its signature. Fix a bug in pooling. Diable some MKLDNN activation and pooling. Cast MKLDNN storage with diff data type. Check if it's a view of NDArray. Reshaped and sliced arrays share the same chunks. Implement caching MKLDNN Act correctly. Fix a bug in check_consistency. Fix a potential bug when destroying NDArray. Fix bugs when allocating mem in NDArray. Fix coding style. Add micro when using mkldnn in ndarray. Fix a compilation error. Fix a bug in concat. Remove MKLDNNStorage. handle diff layouts in CopyFromToDnsImpl. Fallback correctly. Force weight grad to use default layout. Reorder weight arrays in (de)conv for faster inference. Avoid caching TBlob from NDArray. This commit may add some overhead of managing NDArray for each fallback. Fix a bug in Flatten. handle ndarray with def layout in mkldnn BN correctly. Align to page when mkldnn is enabled. Use default mem alloc for mkldnn. Reuse NDArrays. Support WriteInplace for sum. fix complains from "make lint". Avoid reallocation in NDArray. Handle weight arrays with special MKLDNN layouts. Remove unnecessary GetWeights. Fix compilation error without MKLDNN. Fix a bug in (de)conv for weight arrays. Fix a minor bug in MKLDNN conv. Fix a bug in MKLDNNOpSignature. Reimplement fallback for MKLDNN ops. Fix a bug in FallbackExecutor. Add params in hashcode. Invalidate data in outputs to accelerate. Fix a minor bug. Update mkldnn_base-inl.h Add primitive caching for Pooling forward computation Add hashcode in pooling parameters. Support NDArray copy with types unsupported by MKLDNN. Avoid using MKLDNN concat for negative dimension. Fix make lint complain. Disable mkldnn avg pooling for now. Fix a compile warning. Fix compile error when MKLDNN is disabled. OP primitive cache: use memory as signature for MKLDNN storage type Remove MKLDNN array in python. Disable Clang tests in Jenkins. Use mklml dockers to test mkldnn. Update MKLDNN repo to zhengda's mkldnn repo. Update MKLDNN repo to ashok's. Fix a bug in fallback. Change avg pooling algorithm to pooling_avg_include_padding Fix a code style in mkldnn pooling. Temp fix a bug in FC. Revert "Disable Clang tests in Jenkins." This reverts commit b4efa8f89592d30a27f9c30e2237e9420ac6749a. Rebase and Refactor deconv (#20) * rebase to Da,Zheng refactor branch Jan.14, add signature for mkldnn Deconv and modify classMKLDNNDeconvForward * fix make lint complains A simple way of caching BN inference. cache BN forward for both training and inference. Fix some minor problems in BN. Fix a bug in caching BN. force to build with avx2 in Jenkins. Remove the remaining MKLDNNStorageType Some minor updates in NDArray. a lot of updates to address comments. minor changes. * Use NNVM interface. Use NNVM interface for upsampling. Use NNVM interface for convolution. Use NNVM interface for deconvolution. Use NNVM interface for FullyConnected. Move NNVM interface to batch norm. Use NNVM interface for depthwise convolution. Use NNVM interface for softmax activation. Use NNVM interface for pooling. use NNVM interface for dropout. Use NNVM interface for activation. Use NNVM interface for CuDNN batch norm. Use NNVM interface for CuDNN pooling. Use NNVM interface for CuDNN softmax activation. Use NNVM interface for CuDNN activation. Use NNVM interface for CuDNN convolution. Use NNVM interface for CuDNN deconvolution. Move concat to nn/ Use NNVM interface for concat. Fix headers in concat. Move lrn to nn/. Use NNVM interface for LRN. Fix a compilation error in convolution. Fix a compilation error in activation. Fix coding style. Fix coding style for make lint. use enums in batch norm. Use CoreOpRunner for refactored Ops. Make FullyConnected stateless. Make upsampling stateless. Make pooling stateless. Make batchnorm stateless. Make SoftmaxActivation stateless. Fix a code style problem. pass amalgamation test for batch norm. pass amalgamation test for dropout. Get convolution ops from a function. Fix compilation errors for GPU. Fix thread local in diff platforms. Avoid using thread_local for non-CuDNN conv/deconv. Remove TODO in deconv. Fix a bug in batch norm. Fix a bug in fully connected. Don't set #inputs for backward convolution. Revert "Make pooling stateless." * revert modification in test_executor. * Fix a bug in FlattenStorageType. * Remove BN debug. * Remove remaining MXNET_USE_MKL2017 * Remove unused code in pooling. * Fixing bugs in gtests. * Fix lint errors. * a lot of minor updates to address comments. * Fix coding style in MKLDNN Pooling (#22) * revert the code change in the previous code refactor. * Fix a bug in pooling. * LRN coding style changes (#21) * LRN coding style change * Add const for local variables * Add req for LRN forward * rebase code * align API interface * revert modification in test_executor. * cast storage with MKLDNN properly. * Minor updates to address comments. * some minor updates. * Switch to the master branch of MKLDNN. * Minor updates to address comments. * Update activation.cc * Fix a bug in convert NDArray. * Add gluon model zoo tests. * Update GPU tests on model zoo. * Avoid using mobilenet for GPU tests with gluon models. mobilenet can't pass the test even without MKLDNN. * Update GPU tests on gluon. * change cmake to compile MKLDNN. * update cmake for MKLDNN. * Implement align myself. * Switch to intel/mkl-dnn. * Fix errors in align unittest. * Add unit test for LRN. * fix a compilation error. * use storage_type_assign to determine storage type. * avoid global pooling in mkldnn. There is a bug in global pooling in mkldnn. * compare all MKLDNN ops with native impls. add MXNET_MKLDNN_DEBUG to control the test. * Fix a bug in testing correctness. * print the name of buggy operator. * undo some modifications. * Fix a bug on reshaped array. * avoid testing outputs with NullOp. * turn on MKLDNN tests in Jenkins. * print each operator in MKLDNN tests. * rename test_gluon_model_zoo.py * Create hashcode for operator parameters properly. * Add USE_MKL2017 back. * Print warning messages. * move batchnorm tests to nnvm interface. * Delete batchnorm v1 tests. * Get inputs and outputs in batchnorm tests. * disable batchnorm tests for now. * Fix GPU tests on gluon model zoo. * Fix lint complains in tests. * Remove simd from openmp instructions in BatchNorm (#24) * Remove warnings. * Fix MKLDNN 1st compile failure issue (#23) * Fix compilation errors. * Remove ARCH_OPT in Jenkins. * Revert "avoid global pooling in mkldnn." This reverts commit f6efd342e64968cb848c9193d80e929968b8052c. * Move to the latest MKLDNN. This fixes the bug in global pooling. * WIP unit tests (#25) * WIP unit tests * some backward items initialized * Make more C++ unit tests work for batch norm (#28) * WIP unit tests * some backward items initialized * some backward items initialized * some backward items initialized * first unit test working * Working on types * backward types working for fp16 on first unit test * backward types working for fp16 on first unit test * backward types working for fp16 on first unit test * . * . * some tests working * fix input data * hangle gpu<->cpu for setting values * gpu working * gpu working * CAccessAsCPU class * Fix varying type in AccessAsCPU * starting to add channel axis tests * TestChannelAxisSimple * TestChannelAxisSimple * run bidirectional * run bidirectional * run bidirectional * CLEANUP * CLEANUP * .. * noaxis * .. * lint * revert * revert * Fix lint complains. * Fix a minor problem in Makefile. * fix GPU pooling. * Disable modelzoo inference tests. * update accuracy checks for MKLDNN. * Fix MKLDNN pooling for global pooling. * Fix Jenkins. * Fix a bug in Jenkins. * Fix Jenkins
2018-02-15 14:44:34 -08:00
ifeq ($(USE_BLAS),$(filter $(USE_BLAS),blas openblas atlas mkl))
enable use of lapack by default (#6704) * enable use of lapack by default * corrected search paths for lapack lib
2017-06-18 14:23:29 -07:00
LDFLAGS += -llapack
endif
CFLAGS += -DMXNET_USE_LAPACK
Add BLAS3 and LAPACK routines (#6538) * Added linear algebra operators * more comments about style of wrapper interface * more appropriate fatal exit when lapack does not exist * more comments on row/col-major ordering * added config switch for lapack usage * switched lapack usage off by default
2017-06-13 14:13:20 -07:00
endif
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
ifeq ($(USE_CUDNN), 1)
cudnn conv
2015-09-09 20:16:18 -06:00
CFLAGS += -DMSHADOW_USE_CUDNN=1
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
LDFLAGS += -lcudnn
endif
Runtime feature detection (#13549) * Prototype for runtime feature detection * Includes from diamond to quotes * Add CPU feature and BLAS flavour flags * Add BLAS flavour and CPU SSE and AVX flags * MXNET_USE_LAPACK * Fix C++ linting errors * Expose runtime feature detection in the public C API and in the Python API * Refactor Storage -> FeatureSet * Refine documentation * Add failure case * Fix pylint * Address CR comments
2019-01-21 16:45:07 +01:00
ifeq ($(use_blas), open)
CFLAGS += -DMXNET_USE_BLAS_OPEN=1
else ifeq ($(use_blas), atlas)
CFLAGS += -DMXNET_USE_BLAS_ATLAS=1
else ifeq ($(use_blas), mkl)
CFLAGS += -DMXNET_USE_BLAS_MKL=1
else ifeq ($(use_blas), apple)
CFLAGS += -DMXNET_USE_BLAS_APPLE=1
endif
[MXNET-372] Add build flag for USE_F16C in CMake and clarify flag in make (#10760) * add comments to makefile config * add f16c check in mxnet * update cmake * clarify * small updates * add message * add message * update msvc message * update mshadow * typo * only print message for MSVC if USE_F16C * improve build logic * update mshadow * remove def from amalgamation makefile * trigger CI
2018-05-01 21:18:23 -07:00
# whether to use F16C instruction set extension for fast fp16 compute on CPU
# if cross compiling you may want to explicitly turn it off if target system does not support it
ifndef USE_F16C
ifneq ($(OS),Windows_NT)
detected_OS := $(shell uname -s)
ifeq ($(detected_OS),Darwin)
F16C_SUPP = $(shell sysctl -a | grep machdep.cpu.features | grep F16C)
endif
ifeq ($(detected_OS),Linux)
F16C_SUPP = $(shell cat /proc/cpuinfo | grep flags | grep f16c)
endif
ifneq ($(strip $(F16C_SUPP)),)
USE_F16C=1
else
USE_F16C=0
endif
endif
# if OS is Windows, check if your processor and compiler support F16C architecture.
# One way to check if processor supports it is to download the tool
# https://docs.microsoft.com/en-us/sysinternals/downloads/coreinfo.
# If coreinfo -c shows F16C and compiler supports it,
# then you can set USE_F16C=1 explicitly to leverage that capability"
endif
GPerftools update, also include include/mxnet/*.h as sources for CLion (#8232) * GPROF update, also include include/mxnet/*.h as sources for CLionwq * Added FindGperftools.cmake * Add option USE_GPERFTOOLS * Add option USE_GPERFTOOLS * Add option USE_GPERFTOOLS * USE_GPERFTOOLS off by default for now * Add Apache license to FindGperftools.cmake * Update CMakeLists.txt Try to use GPerftools or JEmalloc by default * Update CMakeLists.txt Off by default for now * internal labeling * gperftools and jemalloc * gperftools and jemalloc on by default * Fixing the Caught error (#8199) * Temporarily disable some unit tests to fix the build (#8253) * Temporarily disable the following unit tests that have been causing build failures: test_rms: This can be re-enabled once https://github.com/apache/incubator-mxnet/issues/8230 is fixed. test_autograd_save_memory: This can be re-enabled once https://github.com/apache/incubator-mxnet/issues/8211 is fixed. * OMP num threads 0->1 * remove check * Update documentation links to point to mxnet.incubator.apache.org Update documentation links to point to mxnet.incubator.apache.org * add export to gluon (#8212) * add export * fix * add test * fix nnvm * fix * ReleaseFeedback: License Files (#8247) * Updating license Headers * License changes * Sequential aug (#8243) * add sequentialAug * add type for castaug * modify docs * Basic CPU Kernel OMP selection based upon whether GPU has been used (#7854) * Basic CPU Kernel OMP selection based upon whether GPU has been used * lint * Disabling the test_CSVIter for now (#7829) * Disabling the test_CSVIter for now This test causing random failure while running on windows. Disabling it for now till we fix it. An git hub issue has been created to track it. * Update test_io.py * Update test_io.py * Use OMP thread count as test in Kernel, set count for Kernel loop * lint * removed * Remove assert * Adjust DefaultOMPThreadsPerWorker * remove -1 from omp_cores * Trigger build * It is not clear why pylint claims that this is re-imported. It is not. This is not changed from master branch. Trying a different format. * lint * lint * Change getter/setter naming style * allow env override * check environment directly, since OMP_NUM_THREADS mnay have odd formatting (i.e. 3, 2"). * CR comments * Squashed commit of the following: commit ec704f1bf7709e1cd8a73ad2d4fa18dc62922012 Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 12:29:25 2017 -0700 Fix formatting commit 0218c49f37dbe787767936a22279764b0f219800 Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 12:21:48 2017 -0700 Splitting unary ops commit 9abbba14715088d41076397980cd5d3d49df68df Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 11:38:04 2017 -0700 split unary * Update mxnet_predict0.cc * Update mxnet_predict0.cc * fix oversight with bracket * Binary scatter working on CPU and GPU * return unchanged * This test case is BS. I can't even tell what's wrong on the CI build because so many errors coming from this test. * inconsequential cleanup * Update test_kvstore.py * Update CMakeLists.txt * Update CMakeLists.txt trigger build * force fail * remove forced error * test clean every make * Test * Copy Jenkinsfile from upstream/master to fix the build. * logic was reversed * Update threaded_engine.h Trigger build * Trigger rebuild * Trigger build * Trigger build * Multiplatform docker based builds (#7792) * Add dockerized multi-architecture build files * Add android arm64 build * Operators for sum(csr, axis=0) and sum(csr, axis=1) (#8174) * Add Infer storage for sparse slice operator * Remove unused files * Indentation fix and add gpu test for fallback * Change sum builtin to py_sum * Add sum_axis(csr,axis=0)=dense and sum(csr,axis=1)=dense operator * Documentation changes for sparse * Add fallback unittest for keepdims and exclude * PR review based changes : * Fix CHECK_NE * Change in_stype to int * Using const int instead of int * Initialize mid with the start * Generalizing * OMP num threads 0->1 * remove check
2017-10-13 19:24:38 -07:00
# gperftools malloc library (tcmalloc)
ifeq ($(USE_GPERFTOOLS), 1)
Fix static / dynamic linking of gperftools and jemalloc (#12714)
2018-10-02 20:07:42 +00:00
FIND_LIBFILEEXT=so
ifeq ($(USE_GPERFTOOLS_STATIC), 1)
FIND_LIBFILEEXT=a
endif
FIND_LIBFILE=$(wildcard $(USE_GPERFTOOLS_PATH)/libtcmalloc.$(FIND_LIBFILEEXT))
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
ifeq (,$(FIND_LIBFILE))
Fix static / dynamic linking of gperftools and jemalloc (#12714)
2018-10-02 20:07:42 +00:00
FIND_LIBFILE=$(wildcard /lib/libtcmalloc.$(FIND_LIBFILEEXT))
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
ifeq (,$(FIND_LIBFILE))
Fix static / dynamic linking of gperftools and jemalloc (#12714)
2018-10-02 20:07:42 +00:00
FIND_LIBFILE=$(wildcard /usr/lib/libtcmalloc.$(FIND_LIBFILEEXT))
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
ifeq (,$(FIND_LIBFILE))
Fix static / dynamic linking of gperftools and jemalloc (#12714)
2018-10-02 20:07:42 +00:00
FIND_LIBFILE=$(wildcard /usr/local/lib/libtcmalloc.$(FIND_LIBFILEEXT))
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
ifeq (,$(FIND_LIBFILE))
Fix static / dynamic linking of gperftools and jemalloc (#12714)
2018-10-02 20:07:42 +00:00
FIND_LIBFILE=$(wildcard /usr/lib64/libtcmalloc.$(FIND_LIBFILEEXT))
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
ifeq (,$(FIND_LIBFILE))
USE_GPERFTOOLS=0
endif
endif
endif
endif
endif
ifeq ($(USE_GPERFTOOLS), 1)
CFLAGS += -fno-builtin-malloc -fno-builtin-calloc -fno-builtin-realloc -fno-builtin-free
LDFLAGS += $(FIND_LIBFILE)
GPerftools update, also include include/mxnet/*.h as sources for CLion (#8232) * GPROF update, also include include/mxnet/*.h as sources for CLionwq * Added FindGperftools.cmake * Add option USE_GPERFTOOLS * Add option USE_GPERFTOOLS * Add option USE_GPERFTOOLS * USE_GPERFTOOLS off by default for now * Add Apache license to FindGperftools.cmake * Update CMakeLists.txt Try to use GPerftools or JEmalloc by default * Update CMakeLists.txt Off by default for now * internal labeling * gperftools and jemalloc * gperftools and jemalloc on by default * Fixing the Caught error (#8199) * Temporarily disable some unit tests to fix the build (#8253) * Temporarily disable the following unit tests that have been causing build failures: test_rms: This can be re-enabled once https://github.com/apache/incubator-mxnet/issues/8230 is fixed. test_autograd_save_memory: This can be re-enabled once https://github.com/apache/incubator-mxnet/issues/8211 is fixed. * OMP num threads 0->1 * remove check * Update documentation links to point to mxnet.incubator.apache.org Update documentation links to point to mxnet.incubator.apache.org * add export to gluon (#8212) * add export * fix * add test * fix nnvm * fix * ReleaseFeedback: License Files (#8247) * Updating license Headers * License changes * Sequential aug (#8243) * add sequentialAug * add type for castaug * modify docs * Basic CPU Kernel OMP selection based upon whether GPU has been used (#7854) * Basic CPU Kernel OMP selection based upon whether GPU has been used * lint * Disabling the test_CSVIter for now (#7829) * Disabling the test_CSVIter for now This test causing random failure while running on windows. Disabling it for now till we fix it. An git hub issue has been created to track it. * Update test_io.py * Update test_io.py * Use OMP thread count as test in Kernel, set count for Kernel loop * lint * removed * Remove assert * Adjust DefaultOMPThreadsPerWorker * remove -1 from omp_cores * Trigger build * It is not clear why pylint claims that this is re-imported. It is not. This is not changed from master branch. Trying a different format. * lint * lint * Change getter/setter naming style * allow env override * check environment directly, since OMP_NUM_THREADS mnay have odd formatting (i.e. 3, 2"). * CR comments * Squashed commit of the following: commit ec704f1bf7709e1cd8a73ad2d4fa18dc62922012 Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 12:29:25 2017 -0700 Fix formatting commit 0218c49f37dbe787767936a22279764b0f219800 Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 12:21:48 2017 -0700 Splitting unary ops commit 9abbba14715088d41076397980cd5d3d49df68df Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 11:38:04 2017 -0700 split unary * Update mxnet_predict0.cc * Update mxnet_predict0.cc * fix oversight with bracket * Binary scatter working on CPU and GPU * return unchanged * This test case is BS. I can't even tell what's wrong on the CI build because so many errors coming from this test. * inconsequential cleanup * Update test_kvstore.py * Update CMakeLists.txt * Update CMakeLists.txt trigger build * force fail * remove forced error * test clean every make * Test * Copy Jenkinsfile from upstream/master to fix the build. * logic was reversed * Update threaded_engine.h Trigger build * Trigger rebuild * Trigger build * Trigger build * Multiplatform docker based builds (#7792) * Add dockerized multi-architecture build files * Add android arm64 build * Operators for sum(csr, axis=0) and sum(csr, axis=1) (#8174) * Add Infer storage for sparse slice operator * Remove unused files * Indentation fix and add gpu test for fallback * Change sum builtin to py_sum * Add sum_axis(csr,axis=0)=dense and sum(csr,axis=1)=dense operator * Documentation changes for sparse * Add fallback unittest for keepdims and exclude * PR review based changes : * Fix CHECK_NE * Change in_stype to int * Using const int instead of int * Initialize mid with the start * Generalizing * OMP num threads 0->1 * remove check
2017-10-13 19:24:38 -07:00
endif
# jemalloc malloc library (if not using gperftools)
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
else
ifeq ($(USE_JEMALLOC), 1)
Fix static / dynamic linking of gperftools and jemalloc (#12714)
2018-10-02 20:07:42 +00:00
FIND_LIBFILEEXT=so
ifeq ($(USE_JEMALLOC_STATIC), 1)
FIND_LIBFILEEXT=a
endif
FIND_LIBFILE=$(wildcard $(USE_JEMALLOC_PATH)/libjemalloc.$(FIND_LIBFILEEXT))
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
ifeq (,$(FIND_LIBFILE))
Fix static / dynamic linking of gperftools and jemalloc (#12714)
2018-10-02 20:07:42 +00:00
FIND_LIBFILE=$(wildcard /lib/libjemalloc.$(FIND_LIBFILEEXT))
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
ifeq (,$(FIND_LIBFILE))
Fix static / dynamic linking of gperftools and jemalloc (#12714)
2018-10-02 20:07:42 +00:00
FIND_LIBFILE=$(wildcard /usr/lib/libjemalloc.$(FIND_LIBFILEEXT))
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
ifeq (,$(FIND_LIBFILE))
Fix static / dynamic linking of gperftools and jemalloc (#12714)
2018-10-02 20:07:42 +00:00
FIND_LIBFILE=$(wildcard /usr/local/lib/libjemalloc.$(FIND_LIBFILEEXT))
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
ifeq (,$(FIND_LIBFILE))
Fix static / dynamic linking of gperftools and jemalloc (#12714)
2018-10-02 20:07:42 +00:00
FIND_LIBFILE=$(wildcard /usr/lib/x86_64-linux-gnu/libjemalloc.$(FIND_LIBFILEEXT))
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
ifeq (,$(FIND_LIBFILE))
Fix static / dynamic linking of gperftools and jemalloc (#12714)
2018-10-02 20:07:42 +00:00
FIND_LIBFILE=$(wildcard /usr/lib64/libjemalloc.$(FIND_LIBFILEEXT))
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
ifeq (,$(FIND_LIBFILE))
USE_JEMALLOC=0
endif
endif
endif
endif
endif
endif
ifeq ($(USE_JEMALLOC), 1)
CFLAGS += -fno-builtin-malloc -fno-builtin-calloc -fno-builtin-realloc \
-fno-builtin-free -DUSE_JEMALLOC
LDFLAGS += $(FIND_LIBFILE)
endif
endif
GPerftools update, also include include/mxnet/*.h as sources for CLion (#8232) * GPROF update, also include include/mxnet/*.h as sources for CLionwq * Added FindGperftools.cmake * Add option USE_GPERFTOOLS * Add option USE_GPERFTOOLS * Add option USE_GPERFTOOLS * USE_GPERFTOOLS off by default for now * Add Apache license to FindGperftools.cmake * Update CMakeLists.txt Try to use GPerftools or JEmalloc by default * Update CMakeLists.txt Off by default for now * internal labeling * gperftools and jemalloc * gperftools and jemalloc on by default * Fixing the Caught error (#8199) * Temporarily disable some unit tests to fix the build (#8253) * Temporarily disable the following unit tests that have been causing build failures: test_rms: This can be re-enabled once https://github.com/apache/incubator-mxnet/issues/8230 is fixed. test_autograd_save_memory: This can be re-enabled once https://github.com/apache/incubator-mxnet/issues/8211 is fixed. * OMP num threads 0->1 * remove check * Update documentation links to point to mxnet.incubator.apache.org Update documentation links to point to mxnet.incubator.apache.org * add export to gluon (#8212) * add export * fix * add test * fix nnvm * fix * ReleaseFeedback: License Files (#8247) * Updating license Headers * License changes * Sequential aug (#8243) * add sequentialAug * add type for castaug * modify docs * Basic CPU Kernel OMP selection based upon whether GPU has been used (#7854) * Basic CPU Kernel OMP selection based upon whether GPU has been used * lint * Disabling the test_CSVIter for now (#7829) * Disabling the test_CSVIter for now This test causing random failure while running on windows. Disabling it for now till we fix it. An git hub issue has been created to track it. * Update test_io.py * Update test_io.py * Use OMP thread count as test in Kernel, set count for Kernel loop * lint * removed * Remove assert * Adjust DefaultOMPThreadsPerWorker * remove -1 from omp_cores * Trigger build * It is not clear why pylint claims that this is re-imported. It is not. This is not changed from master branch. Trying a different format. * lint * lint * Change getter/setter naming style * allow env override * check environment directly, since OMP_NUM_THREADS mnay have odd formatting (i.e. 3, 2"). * CR comments * Squashed commit of the following: commit ec704f1bf7709e1cd8a73ad2d4fa18dc62922012 Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 12:29:25 2017 -0700 Fix formatting commit 0218c49f37dbe787767936a22279764b0f219800 Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 12:21:48 2017 -0700 Splitting unary ops commit 9abbba14715088d41076397980cd5d3d49df68df Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 11:38:04 2017 -0700 split unary * Update mxnet_predict0.cc * Update mxnet_predict0.cc * fix oversight with bracket * Binary scatter working on CPU and GPU * return unchanged * This test case is BS. I can't even tell what's wrong on the CI build because so many errors coming from this test. * inconsequential cleanup * Update test_kvstore.py * Update CMakeLists.txt * Update CMakeLists.txt trigger build * force fail * remove forced error * test clean every make * Test * Copy Jenkinsfile from upstream/master to fix the build. * logic was reversed * Update threaded_engine.h Trigger build * Trigger rebuild * Trigger build * Trigger build * Multiplatform docker based builds (#7792) * Add dockerized multi-architecture build files * Add android arm64 build * Operators for sum(csr, axis=0) and sum(csr, axis=1) (#8174) * Add Infer storage for sparse slice operator * Remove unused files * Indentation fix and add gpu test for fallback * Change sum builtin to py_sum * Add sum_axis(csr,axis=0)=dense and sum(csr,axis=1)=dense operator * Documentation changes for sparse * Add fallback unittest for keepdims and exclude * PR review based changes : * Fix CHECK_NE * Change in_stype to int * Using const int instead of int * Initialize mid with the start * Generalizing * OMP num threads 0->1 * remove check
2017-10-13 19:24:38 -07:00
endif
[CI] Add MKLML build and test (#5419) * update * update * update * update * update * update * update * udpate * update * update * update * new docker * update * update * update * update * fix * update * update * update * update * update * update * update * update * update * update * add gpu test * update * udpate * fix * update * tiny * add back test_forward * use new pip * update
2017-03-16 15:01:51 -07:00
GPerftools update, also include include/mxnet/*.h as sources for CLion (#8232) * GPROF update, also include include/mxnet/*.h as sources for CLionwq * Added FindGperftools.cmake * Add option USE_GPERFTOOLS * Add option USE_GPERFTOOLS * Add option USE_GPERFTOOLS * USE_GPERFTOOLS off by default for now * Add Apache license to FindGperftools.cmake * Update CMakeLists.txt Try to use GPerftools or JEmalloc by default * Update CMakeLists.txt Off by default for now * internal labeling * gperftools and jemalloc * gperftools and jemalloc on by default * Fixing the Caught error (#8199) * Temporarily disable some unit tests to fix the build (#8253) * Temporarily disable the following unit tests that have been causing build failures: test_rms: This can be re-enabled once https://github.com/apache/incubator-mxnet/issues/8230 is fixed. test_autograd_save_memory: This can be re-enabled once https://github.com/apache/incubator-mxnet/issues/8211 is fixed. * OMP num threads 0->1 * remove check * Update documentation links to point to mxnet.incubator.apache.org Update documentation links to point to mxnet.incubator.apache.org * add export to gluon (#8212) * add export * fix * add test * fix nnvm * fix * ReleaseFeedback: License Files (#8247) * Updating license Headers * License changes * Sequential aug (#8243) * add sequentialAug * add type for castaug * modify docs * Basic CPU Kernel OMP selection based upon whether GPU has been used (#7854) * Basic CPU Kernel OMP selection based upon whether GPU has been used * lint * Disabling the test_CSVIter for now (#7829) * Disabling the test_CSVIter for now This test causing random failure while running on windows. Disabling it for now till we fix it. An git hub issue has been created to track it. * Update test_io.py * Update test_io.py * Use OMP thread count as test in Kernel, set count for Kernel loop * lint * removed * Remove assert * Adjust DefaultOMPThreadsPerWorker * remove -1 from omp_cores * Trigger build * It is not clear why pylint claims that this is re-imported. It is not. This is not changed from master branch. Trying a different format. * lint * lint * Change getter/setter naming style * allow env override * check environment directly, since OMP_NUM_THREADS mnay have odd formatting (i.e. 3, 2"). * CR comments * Squashed commit of the following: commit ec704f1bf7709e1cd8a73ad2d4fa18dc62922012 Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 12:29:25 2017 -0700 Fix formatting commit 0218c49f37dbe787767936a22279764b0f219800 Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 12:21:48 2017 -0700 Splitting unary ops commit 9abbba14715088d41076397980cd5d3d49df68df Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 11:38:04 2017 -0700 split unary * Update mxnet_predict0.cc * Update mxnet_predict0.cc * fix oversight with bracket * Binary scatter working on CPU and GPU * return unchanged * This test case is BS. I can't even tell what's wrong on the CI build because so many errors coming from this test. * inconsequential cleanup * Update test_kvstore.py * Update CMakeLists.txt * Update CMakeLists.txt trigger build * force fail * remove forced error * test clean every make * Test * Copy Jenkinsfile from upstream/master to fix the build. * logic was reversed * Update threaded_engine.h Trigger build * Trigger rebuild * Trigger build * Trigger build * Multiplatform docker based builds (#7792) * Add dockerized multi-architecture build files * Add android arm64 build * Operators for sum(csr, axis=0) and sum(csr, axis=1) (#8174) * Add Infer storage for sparse slice operator * Remove unused files * Indentation fix and add gpu test for fallback * Change sum builtin to py_sum * Add sum_axis(csr,axis=0)=dense and sum(csr,axis=1)=dense operator * Documentation changes for sparse * Add fallback unittest for keepdims and exclude * PR review based changes : * Fix CHECK_NE * Change in_stype to int * Using const int instead of int * Initialize mid with the start * Generalizing * OMP num threads 0->1 * remove check
2017-10-13 19:24:38 -07:00
# If not using tcmalloc or jemalloc, print a warning (user should consider installing)
ifneq ($(USE_GPERFTOOLS), 1)
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
ifneq ($(USE_JEMALLOC), 1)
GPerftools update, also include include/mxnet/*.h as sources for CLion (#8232) * GPROF update, also include include/mxnet/*.h as sources for CLionwq * Added FindGperftools.cmake * Add option USE_GPERFTOOLS * Add option USE_GPERFTOOLS * Add option USE_GPERFTOOLS * USE_GPERFTOOLS off by default for now * Add Apache license to FindGperftools.cmake * Update CMakeLists.txt Try to use GPerftools or JEmalloc by default * Update CMakeLists.txt Off by default for now * internal labeling * gperftools and jemalloc * gperftools and jemalloc on by default * Fixing the Caught error (#8199) * Temporarily disable some unit tests to fix the build (#8253) * Temporarily disable the following unit tests that have been causing build failures: test_rms: This can be re-enabled once https://github.com/apache/incubator-mxnet/issues/8230 is fixed. test_autograd_save_memory: This can be re-enabled once https://github.com/apache/incubator-mxnet/issues/8211 is fixed. * OMP num threads 0->1 * remove check * Update documentation links to point to mxnet.incubator.apache.org Update documentation links to point to mxnet.incubator.apache.org * add export to gluon (#8212) * add export * fix * add test * fix nnvm * fix * ReleaseFeedback: License Files (#8247) * Updating license Headers * License changes * Sequential aug (#8243) * add sequentialAug * add type for castaug * modify docs * Basic CPU Kernel OMP selection based upon whether GPU has been used (#7854) * Basic CPU Kernel OMP selection based upon whether GPU has been used * lint * Disabling the test_CSVIter for now (#7829) * Disabling the test_CSVIter for now This test causing random failure while running on windows. Disabling it for now till we fix it. An git hub issue has been created to track it. * Update test_io.py * Update test_io.py * Use OMP thread count as test in Kernel, set count for Kernel loop * lint * removed * Remove assert * Adjust DefaultOMPThreadsPerWorker * remove -1 from omp_cores * Trigger build * It is not clear why pylint claims that this is re-imported. It is not. This is not changed from master branch. Trying a different format. * lint * lint * Change getter/setter naming style * allow env override * check environment directly, since OMP_NUM_THREADS mnay have odd formatting (i.e. 3, 2"). * CR comments * Squashed commit of the following: commit ec704f1bf7709e1cd8a73ad2d4fa18dc62922012 Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 12:29:25 2017 -0700 Fix formatting commit 0218c49f37dbe787767936a22279764b0f219800 Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 12:21:48 2017 -0700 Splitting unary ops commit 9abbba14715088d41076397980cd5d3d49df68df Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 11:38:04 2017 -0700 split unary * Update mxnet_predict0.cc * Update mxnet_predict0.cc * fix oversight with bracket * Binary scatter working on CPU and GPU * return unchanged * This test case is BS. I can't even tell what's wrong on the CI build because so many errors coming from this test. * inconsequential cleanup * Update test_kvstore.py * Update CMakeLists.txt * Update CMakeLists.txt trigger build * force fail * remove forced error * test clean every make * Test * Copy Jenkinsfile from upstream/master to fix the build. * logic was reversed * Update threaded_engine.h Trigger build * Trigger rebuild * Trigger build * Trigger build * Multiplatform docker based builds (#7792) * Add dockerized multi-architecture build files * Add android arm64 build * Operators for sum(csr, axis=0) and sum(csr, axis=1) (#8174) * Add Infer storage for sparse slice operator * Remove unused files * Indentation fix and add gpu test for fallback * Change sum builtin to py_sum * Add sum_axis(csr,axis=0)=dense and sum(csr,axis=1)=dense operator * Documentation changes for sparse * Add fallback unittest for keepdims and exclude * PR review based changes : * Fix CHECK_NE * Change in_stype to int * Using const int instead of int * Initialize mid with the start * Generalizing * OMP num threads 0->1 * remove check
2017-10-13 19:24:38 -07:00
$(warning WARNING: Significant performance increases can be achieved by installing and \
enabling gperftools or jemalloc development packages)
allow custom path and static linking for custom mallocs in make (#12645) * allow custom path and static linking for custom mallocs * add config entry
2018-09-23 22:11:20 -07:00
endif
GPerftools update, also include include/mxnet/*.h as sources for CLion (#8232) * GPROF update, also include include/mxnet/*.h as sources for CLionwq * Added FindGperftools.cmake * Add option USE_GPERFTOOLS * Add option USE_GPERFTOOLS * Add option USE_GPERFTOOLS * USE_GPERFTOOLS off by default for now * Add Apache license to FindGperftools.cmake * Update CMakeLists.txt Try to use GPerftools or JEmalloc by default * Update CMakeLists.txt Off by default for now * internal labeling * gperftools and jemalloc * gperftools and jemalloc on by default * Fixing the Caught error (#8199) * Temporarily disable some unit tests to fix the build (#8253) * Temporarily disable the following unit tests that have been causing build failures: test_rms: This can be re-enabled once https://github.com/apache/incubator-mxnet/issues/8230 is fixed. test_autograd_save_memory: This can be re-enabled once https://github.com/apache/incubator-mxnet/issues/8211 is fixed. * OMP num threads 0->1 * remove check * Update documentation links to point to mxnet.incubator.apache.org Update documentation links to point to mxnet.incubator.apache.org * add export to gluon (#8212) * add export * fix * add test * fix nnvm * fix * ReleaseFeedback: License Files (#8247) * Updating license Headers * License changes * Sequential aug (#8243) * add sequentialAug * add type for castaug * modify docs * Basic CPU Kernel OMP selection based upon whether GPU has been used (#7854) * Basic CPU Kernel OMP selection based upon whether GPU has been used * lint * Disabling the test_CSVIter for now (#7829) * Disabling the test_CSVIter for now This test causing random failure while running on windows. Disabling it for now till we fix it. An git hub issue has been created to track it. * Update test_io.py * Update test_io.py * Use OMP thread count as test in Kernel, set count for Kernel loop * lint * removed * Remove assert * Adjust DefaultOMPThreadsPerWorker * remove -1 from omp_cores * Trigger build * It is not clear why pylint claims that this is re-imported. It is not. This is not changed from master branch. Trying a different format. * lint * lint * Change getter/setter naming style * allow env override * check environment directly, since OMP_NUM_THREADS mnay have odd formatting (i.e. 3, 2"). * CR comments * Squashed commit of the following: commit ec704f1bf7709e1cd8a73ad2d4fa18dc62922012 Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 12:29:25 2017 -0700 Fix formatting commit 0218c49f37dbe787767936a22279764b0f219800 Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 12:21:48 2017 -0700 Splitting unary ops commit 9abbba14715088d41076397980cd5d3d49df68df Author: Olivier <coolivie@amazon.com> Date: Mon Sep 25 11:38:04 2017 -0700 split unary * Update mxnet_predict0.cc * Update mxnet_predict0.cc * fix oversight with bracket * Binary scatter working on CPU and GPU * return unchanged * This test case is BS. I can't even tell what's wrong on the CI build because so many errors coming from this test. * inconsequential cleanup * Update test_kvstore.py * Update CMakeLists.txt * Update CMakeLists.txt trigger build * force fail * remove forced error * test clean every make * Test * Copy Jenkinsfile from upstream/master to fix the build. * logic was reversed * Update threaded_engine.h Trigger build * Trigger rebuild * Trigger build * Trigger build * Multiplatform docker based builds (#7792) * Add dockerized multi-architecture build files * Add android arm64 build * Operators for sum(csr, axis=0) and sum(csr, axis=1) (#8174) * Add Infer storage for sparse slice operator * Remove unused files * Indentation fix and add gpu test for fallback * Change sum builtin to py_sum * Add sum_axis(csr,axis=0)=dense and sum(csr,axis=1)=dense operator * Documentation changes for sparse * Add fallback unittest for keepdims and exclude * PR review based changes : * Fix CHECK_NE * Change in_stype to int * Using const int instead of int * Initialize mid with the start * Generalizing * OMP num threads 0->1 * remove check
2017-10-13 19:24:38 -07:00
endif
[CI] Add MKLML build and test (#5419) * update * update * update * update * update * update * update * udpate * update * update * update * new docker * update * update * update * update * fix * update * update * update * update * update * update * update * update * update * update * add gpu test * update * udpate * fix * update * tiny * add back test_forward * use new pip * update
2017-03-16 15:01:51 -07:00
[engine-refactor] stream manager and travis script
2015-09-08 23:29:03 +08:00
ifeq ($(USE_THREADED_ENGINE), 1)
CFLAGS += -DMXNET_USE_THREADED_ENGINE
endif
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
ifneq ($(ADD_CFLAGS), NONE)
CFLAGS += $(ADD_CFLAGS)
endif
ifneq ($(ADD_LDFLAGS), NONE)
LDFLAGS += $(ADD_LDFLAGS)
endif
[MXNET-303] Update Jetpack to version 3.2 (#10482)
2018-04-10 13:22:54 +00:00
ifeq ($(NVCC), NONE)
Use default nvcc when configured nvcc not present. (#10566)
2018-04-25 13:09:58 -07:00
# If NVCC has not been manually defined, use the CUDA_PATH bin dir.
[MXNET-303] Update Jetpack to version 3.2 (#10482)
2018-04-10 13:22:54 +00:00
ifneq ($(USE_CUDA_PATH), NONE)
NVCC=$(USE_CUDA_PATH)/bin/nvcc
endif
update build doc
2015-09-22 14:32:47 -04:00
endif
Use default nvcc when configured nvcc not present. (#10566)
2018-04-25 13:09:58 -07:00
# Guard against displaying nvcc info messages to users not using CUDA.
ifeq ($(USE_CUDA), 1)
# If NVCC is not at the location specified, use CUDA_PATH instead.
ifeq ("$(wildcard $(NVCC))","")
ifneq ($(USE_CUDA_PATH), NONE)
NVCC=$(USE_CUDA_PATH)/bin/nvcc
$(info INFO: nvcc was not found on your path)
$(info INFO: Using $(NVCC) as nvcc path)
else
$(warning WARNING: could not find nvcc compiler, the specified path was: $(NVCC))
endif
endif
endif
Changed make to support more gpu archs, multiple toolkits, reduce lib size. (#6588) * Updated make to support more gpu archs, tolerate multiple toolkit versions, reduce lib size. * Moved CUDA_ARCH setting to Makefile, removed from all make/*.mk files.
2017-06-07 10:12:32 -07:00
# Sets 'CUDA_ARCH', which determines the GPU architectures supported
# by the compiled kernels. Users can edit the KNOWN_CUDA_ARCHS list below
Fixed Makefile so a null CUDA_ARCH is treated like an unset one. (#7515)
2017-08-17 21:14:18 -07:00
# to remove archs they don't wish to support to speed compilation, or they can
# pre-set the CUDA_ARCH args in config.mk to a non-null value for full control.
Changed make to support more gpu archs, multiple toolkits, reduce lib size. (#6588) * Updated make to support more gpu archs, tolerate multiple toolkit versions, reduce lib size. * Moved CUDA_ARCH setting to Makefile, removed from all make/*.mk files.
2017-06-07 10:12:32 -07:00
#
# For archs in this list, nvcc will create a fat-binary that will include
# the binaries (SASS) for all architectures supported by the installed version
# of the cuda toolkit, plus the assembly (PTX) for the most recent such architecture.
# If these kernels are then run on a newer-architecture GPU, the binary will
# be JIT-compiled by the updated driver from the included PTX.
ifeq ($(USE_CUDA), 1)
Fixed Makefile so a null CUDA_ARCH is treated like an unset one. (#7515)
2017-08-17 21:14:18 -07:00
ifeq ($(CUDA_ARCH),)
add Volta and Turing arch (#13069) add 75 in Makefile
2018-11-21 10:48:59 +08:00
KNOWN_CUDA_ARCHS := 30 35 50 52 60 61 70 75
Changed make to support more gpu archs, multiple toolkits, reduce lib size. (#6588) * Updated make to support more gpu archs, tolerate multiple toolkit versions, reduce lib size. * Moved CUDA_ARCH setting to Makefile, removed from all make/*.mk files.
2017-06-07 10:12:32 -07:00
# Run nvcc on a zero-length file to check architecture-level support.
# Create args to include SASS in the fat binary for supported levels.
CUDA_ARCH := $(foreach arch,$(KNOWN_CUDA_ARCHS), \
update mklml and mkl mac support (#7587)
2017-08-29 17:14:01 -07:00
$(shell $(NVCC) -arch=sm_$(arch) -E --x cu /dev/null >/dev/null 2>&1 && \
echo -gencode arch=compute_$(arch),code=sm_$(arch)))
Changed make to support more gpu archs, multiple toolkits, reduce lib size. (#6588) * Updated make to support more gpu archs, tolerate multiple toolkit versions, reduce lib size. * Moved CUDA_ARCH setting to Makefile, removed from all make/*.mk files.
2017-06-07 10:12:32 -07:00
# Convert a trailing "code=sm_NN" to "code=[sm_NN,compute_NN]" to also
# include the PTX of the most recent arch in the fat-binaries for
# forward compatibility with newer GPUs.
CUDA_ARCH := $(shell echo $(CUDA_ARCH) | sed 's/sm_\([0-9]*\)$$/[sm_\1,compute_\1]/')
# Add fat binary compression if supported by nvcc.
COMPRESS := --fatbin-options -compress-all
CUDA_ARCH += $(shell $(NVCC) -cuda $(COMPRESS) --x cu /dev/null -o /dev/null >/dev/null 2>&1 && \
update mklml and mkl mac support (#7587)
2017-08-29 17:14:01 -07:00
echo $(COMPRESS))
Changed make to support more gpu archs, multiple toolkits, reduce lib size. (#6588) * Updated make to support more gpu archs, tolerate multiple toolkit versions, reduce lib size. * Moved CUDA_ARCH setting to Makefile, removed from all make/*.mk files.
2017-06-07 10:12:32 -07:00
endif
[Do not merge] Move to new CI (#8960) * Squashed commit of the following: commit ce53ad38a6ef260c7a5870faa58583b5fa2b1e30 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 10:53:09 2017 +0100 remove whitespace commit 71f1f8d5a71dd495a12594a4f01bf275fbded186 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 10:37:04 2017 +0100 Decrease timeout in Jenkinsfile commit ef88da18dd0f29b28ae15bb9840f3ff393e4e77f Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 10:35:44 2017 +0100 Cleanup and add comments commit ad59a48e4f68fe855029ba23a78bf9fa1abf1544 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 09:26:51 2017 +0100 Fix Integration Test Caffe-GPU to work on ubuntu16.04 with CUDA 8 commit fc7cf23e92a6bf48cc36968df274e55c4d47a17a Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 07:42:49 2017 +0100 Add verbose output to perl tests commit f076312de899fe859df3d03e74b3f97923c96a2c Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 06:40:31 2017 +0100 Adjust Jenkinsfile to ensure integration stage being run on nvidia-docker commit 9ed3cd0ba7411ad686bb0162e473684c7f689211 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 02:14:54 2017 +0100 Add nvidia dependencies to caffe_gpu commit 103225b70f734c92aa69416aa94eff3b24b28c5a Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 17:48:32 2017 +0100 Update GPU containers to ubuntu16.04 / CUDA 8 to fix OpenBLAS-hang 2 commit d1a12065826dabfbbe2102b375fb22a794eb5b3d Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 17:22:51 2017 +0100 Update GPU containers to ubuntu16.04 / CUDA 8 to fix OpenBLAS-hang commit b193874f0fd33a2d985fe430b9787d5cbf479662 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 14:39:36 2017 +0100 Fix typo commit c1e303468f494e28b8a4c625434a66c9b2c72282 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 14:25:43 2017 +0100 Fix typo commit 143fbe258e067d59efcf08ac948850b6de4cd9a0 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 14:18:44 2017 +0100 Fix unittest due to cuda commit f7255519bc96dcd25bfe83b89f8ec506af1caba6 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 14:00:48 2017 +0100 Make sure containers are run inside the right docker binary commit 7ae2c67ad91fcc2a7093f4d1bf417e6096bfa32e Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 13:43:56 2017 +0100 Fix typo commit 26d3a247ed502ea6f3ef25ded63ebe3f62ed9c72 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 13:42:34 2017 +0100 Trigger build commit aca7005ddb91ad9ce68fd4ee471664486e44a3fd Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 12:49:36 2017 +0100 Separated GPU-Build dockerfiles from runtime Dockerfiles commit 6091e5b171e3c50cecdfeacce04d21e2b3f65693 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 03:43:12 2017 +0100 Enable install of nvidia drivers in GPU Dockerfiles commit f0a3fe7474fb13d96bc84671e3402a8b33e71ccd Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 02:31:58 2017 +0100 Add cuda stub lib path to Makefile to fix libcuda.so.1 not found error commit 8233a3d032cdf77142f9e9c0424134da390f36fa Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 02:25:29 2017 +0100 Add softlink to fix libcuda.so.1 not found-error commit 8c3e50ae98a7de95476bc15dd8ca8c146003c1f1 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 01:44:49 2017 +0100 Fix libcuda.so.1 not found during build commit 2dbf35424060ec36d8cd4a084ddb3f1a976323cd Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Thu Nov 23 00:12:58 2017 +0100 Re-enable DEV-Flag in Jenkinsfile commit 6083f109c0811c310b7f22268c5d6556badf4426 Author: Pedro Larroy <pllarroy@amazon.com> Date: Wed Nov 22 14:38:37 2017 -0800 Fix uninitialized array warning commit 6e0dd21c121b275b341dbb9c304ce73ded64eae6 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 22:49:55 2017 +0100 Support CUDA-Builds on non-gpu-instances commit 5f53708e369c41ed77e76fd6623a9db6573f1b62 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 15:15:36 2017 +0100 Temp. disable dev compile mode to prevent warnings being treated as errors commit 5e5dff50af43ed7d618122d1c2f0379a1d7193fc Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 14:10:22 2017 +0100 Fix typo in Jenkins for windows commit a3d9f7483a693b0c513c677baaa5ba6621e722d1 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 14:07:18 2017 +0100 Fix tests/ci_build/with_the_same_user: line 27: /etc/sudoers.d/90-nopasswd-sudo: No such file or directory commit d76e342cfb76361ac166bc5e770bbef7721317d5 Merge: 5762ef88 0c0a5626 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 14:07:04 2017 +0100 Merge branch 'v0.12.0' of github.com:MXNetEdge/mabreu-incubator-mxnet into v0.12.0 commit 5762ef88b85754809fc98a01b149be5036737b64 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 13:46:57 2017 +0100 Add Kellens fix for C5 commit a93a9cd74ec362b4e3417e840169e3ce8b70cfba Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 11:40:12 2017 +0100 Add Jenkinsfile for windows commit 0c0a5626a614ac203cc7e93fbd5166e4f50267bb Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 11:40:12 2017 +0100 Add Jenkinsfile for windows commit 175d157171611778ee6a1e1804e4611e7ef4c136 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Mon Nov 20 19:56:18 2017 +0100 Remove failing test commit bb15df1464a02fbc9dff042b24e52b49187b9d55 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Mon Nov 20 17:25:29 2017 +0100 Disable unnecessary CUDA-archs commit c68265892256eb8756d2eba696ccc41766fbe3db Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Mon Nov 20 16:28:15 2017 +0100 Disable failing test test_operator:test_depthwise_convolution commit 07028474d10326029aa27afb81ff3eb3fa063b5f Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Mon Nov 20 15:01:08 2017 +0100 Increase git timeout commit 61d689424541801bad5a2ec5b46581810281d9a8 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Mon Nov 20 13:16:28 2017 +0100 Split up Jenkins-Tasks to appropriate machines * Remove temp file * Add separate compile job for MKLML-CPU * Fix typo * Revert Jenkinsfile * Remove MKL Experimental * Revert "Remove MKL Experimental" This reverts commit 5e0f20c67e872c1a5af7491610ab55d37f6fc5b8. * Disable test_gluon:test_dtype * Cleanup * Add sudo install to lint * Add comment about MKLML * Fix typo * Clean up dockerfiles regarding mklml * Enable failing tests * Restructure dockerfiles * Fix typo * Fix Dockerfiles * Update mklml library * MKLML fix * Try to fix failing MKLML-gpu build * Fix typo * Disable flaky test https://github.com/apache/incubator-mxnet/issues/8892 * Remove workspace from jenkinsfile * Revert "Remove workspace from jenkinsfile" This reverts commit c2cfe5db296eafb008f63a7ddb0c841ac3b8313e. * Misc changes Merge Jenkinsfiles Documented Dockerfiles and scripts Cleaned up test comments Restore Dockerfile.ubuntu1404_cuda75_cudnn5 * fix typo * Fix typo * Revert KNOWN_CUDA_ARCH changes * Change makefile comment * Remove Jenkinsfile_windows * Add myself to CONTRIBUTORS.md * Cleanup * Add CUDA_ARCH to Jenkinsfile * Fix typo * Fix typo * Fix typo * Fix typo * Add cuda_archs to ci_build * Add cuda_archs to ci_build * Fix typo and comment * Disable flaky test test_loss:test_ctc_loss_train #8892
2017-12-06 19:31:19 +01:00
$(info Running CUDA_ARCH: $(CUDA_ARCH))
Changed make to support more gpu archs, multiple toolkits, reduce lib size. (#6588) * Updated make to support more gpu archs, tolerate multiple toolkit versions, reduce lib size. * Moved CUDA_ARCH setting to Makefile, removed from all make/*.mk files.
2017-06-07 10:12:32 -07:00
endif
[kvstore] add ps-lite
2015-10-02 22:22:22 -04:00
# ps-lite
[MXNET-16] Move mshadow/ps-lite/dlpack to 3rdparty #10132 (#10138) * move * update cmake/make * remove commment * update license * ps-lite move * update license * move mshadow for make * mshadow cmake * update license * update readme * Update Jenkinsfile * update license file with paths
2018-03-21 23:31:19 -07:00
PS_PATH=$(ROOTDIR)/3rdparty/ps-lite
[kvstore] add ps-lite
2015-10-02 22:22:22 -04:00
DEPS_PATH=$(shell pwd)/deps
include $(PS_PATH)/make/ps.mk
ifeq ($(USE_DIST_KVSTORE), 1)
[kvstore] update ps-lite
2015-10-20 22:06:23 +00:00
CFLAGS += -DMXNET_USE_DIST_KVSTORE -I$(PS_PATH)/include -I$(DEPS_PATH)/include
[kvstore] add ps-lite
2015-10-02 22:22:22 -04:00
LIB_DEP += $(PS_PATH)/build/libps.a
[ps-lite] link libprotobuf.a and libzmq.a
2015-11-06 21:54:16 -05:00
LDFLAGS += $(PS_LDFLAGS_A)
[kvstore] add ps-lite
2015-10-02 22:22:22 -04:00
endif
[doc] new sphnix plugin (#6105) * update doc * rm * update * update ndarray * update mds * update * update * update * update * update * update * update image.md and others * update
2017-05-07 22:19:25 -07:00
.PHONY: clean all extra-packages test lint docs clean_all rcpplint rcppexport roxygen\
[OP] Add alias support (#3261)
2016-09-08 19:10:58 -07:00
cython2 cython3 cython cyclean
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
Caffe without the patch, cpp-package fixed also with Caffe plugin (#5573) * PEP8 indentation fix * Remove executable flag * Remove executable flags * Fixing cpp-package including problems with caffe converter * Fix warnings * Remove need for caffe patch for caffe plugin * Ignore cmake paths, remove rebuildable op.h for cpp-package * cpp-package examples fixed. Makefile and CMake of op.h and examples added. * cpp-package examples fixed. Makefile and CMake of op.h and examples added. * Fix source file * Better example.mk * Turn off caffe by default * lint fixes * Trying to figure out how to fix submodules * nohub problem on travis so force retry * Edited test to just run make instead of obsolete make example * Fix cpp-package op.h bootstrapping with CMake * Build tweaking cpp-package * Lint fixes * op.h generator * FIx cpp-package for latest merge from master * Fix lint * static link * link whole static lib * Trigger another build attempt * Trigger another build attempt * Add caffe plugin support (disabled until dependencies can be added to build machine) * Add cufft library * win32 cufft library * don't link unit tests as static * Add include for rebuildable protobuf header caffe.pb.h
2017-03-30 20:13:36 -07:00
all: lib/libmxnet.a lib/libmxnet.so $(BIN) extra-packages
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
Move legacy operators under the nn directory (#8858) * Move legacy operators to nn/ * Move CuDNN operators to nn/ * Update CuDNN algoreg. * fix amalgamation. * fix a path of a header file. * Fix the path for mkl header files. * Fix compilation errors in mkl. * Fix for the coding style. * Fix the path of the header files. * Fix headers in the cu files. * Compile cu files. * Update activation-inl.h * Update activation-inl.h
2017-11-30 10:56:21 -08:00
SRC = $(wildcard src/*/*/*/*.cc src/*/*/*.cc src/*/*.cc src/*.cc)
torch ndarray function backend
2016-01-07 22:52:33 -08:00
OBJ = $(patsubst %.cc, build/%.o, $(SRC))
Move legacy operators under the nn directory (#8858) * Move legacy operators to nn/ * Move CuDNN operators to nn/ * Update CuDNN algoreg. * fix amalgamation. * fix a path of a header file. * Fix the path for mkl header files. * Fix compilation errors in mkl. * Fix for the coding style. * Fix the path of the header files. * Fix headers in the cu files. * Compile cu files. * Update activation-inl.h * Update activation-inl.h
2017-11-30 10:56:21 -08:00
CUSRC = $(wildcard src/*/*/*/*.cu src/*/*/*.cu src/*/*.cu src/*.cu)
torch ndarray function backend
2016-01-07 22:52:33 -08:00
CUOBJ = $(patsubst %.cu, build/%_gpu.o, $(CUSRC))
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
move plugin rules to sub makefile
2016-01-12 21:22:09 -08:00
# extra operators
extra operator makefile fix
2015-12-23 15:07:58 -08:00
ifneq ($(EXTRA_OPERATORS),)
add ssd as example (#3674)
2016-11-07 15:10:50 -06:00
EXTRA_SRC = $(wildcard $(patsubst %, %/*.cc, $(EXTRA_OPERATORS)) $(patsubst %, %/*/*.cc, $(EXTRA_OPERATORS)))
fix caffe stream (#2984) * fix caffe stream * allow multiple extra operator * fix python 2/3 codec conflict
2016-08-10 18:33:41 -07:00
EXTRA_OBJ = $(patsubst %.cc, %.o, $(EXTRA_SRC))
add ssd as example (#3674)
2016-11-07 15:10:50 -06:00
EXTRA_CUSRC = $(wildcard $(patsubst %, %/*.cu, $(EXTRA_OPERATORS)) $(patsubst %, %/*/*.cu, $(EXTRA_OPERATORS)))
fix caffe stream (#2984) * fix caffe stream * allow multiple extra operator * fix python 2/3 codec conflict
2016-08-10 18:33:41 -07:00
EXTRA_CUOBJ = $(patsubst %.cu, %_gpu.o, $(EXTRA_CUSRC))
make ccsgd the default and add number of batch to predict
2015-12-16 23:38:36 -08:00
else
EXTRA_SRC =
EXTRA_OBJ =
EXTRA_CUSRC =
EXTRA_CUOBJ =
endif
torch ndarray function backend
2016-01-07 22:52:33 -08:00
# plugin
move plugin rules to sub makefile
2016-01-12 21:22:09 -08:00
PLUGIN_OBJ =
PLUGIN_CUOBJ =
include $(MXNET_PLUGINS)
torch ndarray function backend
2016-01-07 22:52:33 -08:00
Re-organize Scala maven build (#13626) * Re-organize scala maven build 1. Automatically detect which platform to build for scala. 2. Remove platform dependend submodules 3. Fix cyclic module dependencies 4. Fix scalatype style check 5. Now mvn can be executed in submodule 6. Maven build can be executed from any directory not only in root project 7. Checkin javah header file, and use verify task to detect native API changes 8. Improve incremental build performance 9. Remove unittest and integration-test profile, use proper task instead 10. Delete generated scala file during maven clean. * Redo maven deploy related tasks. 1. Removed maven release plugin. 2. Make maven build friendly to CI, allow cli override version. 3. Moved gpg signing to deploy stage. 4. Created a separeated deploy module. 5. Updated Makefile to new maven build change. 6. Remove unused nexus-staging-plugin 7. Added nightly and staging profile for CI. * Support mkldnn for Scala.
2019-01-08 17:25:53 -08:00
ifneq ($(UNAME_S), Windows)
split cpu & gpu version
2016-03-11 23:48:53 +08:00
ifeq ($(UNAME_S), Darwin)
NNVM Refactor (#3194) * Init nnvm change * temp checkin * Move TShape to NNVM * Redirect Symbolic API to NNVM * Add Op Prop Adapter * Finish migrate in shape infer * Pass all symbolic test * temp commit * enable aux data * [EXEC] Basic version of exec for forward only * [EXEC] Enable most optimizations, still wait grad and context * fix legacy op with latest one * Update NNVM NodeRef * Adapt to newer interface * ALl registry of backop is complete * temp commit * Hack finish backward pass * [EXEC] One day pass * [EXEC] Pass all operator unittest * [EXEC] enable model parallel * Fully pass all legacy tests * Remove legacy symbolic code * update news * Make travis compile * Fix python3 * Update viz module to new json format
2016-09-01 21:17:33 -07:00
WHOLE_ARCH= -all_load
NO_WHOLE_ARCH= -noall_load
Fix gcc5.3 + nvcc 8.0 (#2004)
2016-05-01 17:00:47 -07:00
else
NNVM Refactor (#3194) * Init nnvm change * temp checkin * Move TShape to NNVM * Redirect Symbolic API to NNVM * Add Op Prop Adapter * Finish migrate in shape infer * Pass all symbolic test * temp commit * enable aux data * [EXEC] Basic version of exec for forward only * [EXEC] Enable most optimizations, still wait grad and context * fix legacy op with latest one * Update NNVM NodeRef * Adapt to newer interface * ALl registry of backop is complete * temp commit * Hack finish backward pass * [EXEC] One day pass * [EXEC] Pass all operator unittest * [EXEC] enable model parallel * Fully pass all legacy tests * Remove legacy symbolic code * update news * Make travis compile * Fix python3 * Update viz module to new json format
2016-09-01 21:17:33 -07:00
WHOLE_ARCH= --whole-archive
NO_WHOLE_ARCH= --no-whole-archive
split cpu & gpu version
2016-03-11 23:48:53 +08:00
endif
endif
move plugin rules to sub makefile
2016-01-12 21:22:09 -08:00
# all dep
NNVM Refactor (#3194) * Init nnvm change * temp checkin * Move TShape to NNVM * Redirect Symbolic API to NNVM * Add Op Prop Adapter * Finish migrate in shape infer * Pass all symbolic test * temp commit * enable aux data * [EXEC] Basic version of exec for forward only * [EXEC] Enable most optimizations, still wait grad and context * fix legacy op with latest one * Update NNVM NodeRef * Adapt to newer interface * ALl registry of backop is complete * temp commit * Hack finish backward pass * [EXEC] One day pass * [EXEC] Pass all operator unittest * [EXEC] enable model parallel * Fully pass all legacy tests * Remove legacy symbolic code * update news * Make travis compile * Fix python3 * Update viz module to new json format
2016-09-01 21:17:33 -07:00
LIB_DEP += $(DMLC_CORE)/libdmlc.a $(NNVM_PATH)/lib/libnnvm.a
torch ndarray function backend
2016-01-07 22:52:33 -08:00
ALL_DEP = $(OBJ) $(EXTRA_OBJ) $(PLUGIN_OBJ) $(LIB_DEP)
NNVM Refactor (#3194) * Init nnvm change * temp checkin * Move TShape to NNVM * Redirect Symbolic API to NNVM * Add Op Prop Adapter * Finish migrate in shape infer * Pass all symbolic test * temp commit * enable aux data * [EXEC] Basic version of exec for forward only * [EXEC] Enable most optimizations, still wait grad and context * fix legacy op with latest one * Update NNVM NodeRef * Adapt to newer interface * ALl registry of backop is complete * temp commit * Hack finish backward pass * [EXEC] One day pass * [EXEC] Pass all operator unittest * [EXEC] enable model parallel * Fully pass all legacy tests * Remove legacy symbolic code * update news * Make travis compile * Fix python3 * Update viz module to new json format
2016-09-01 21:17:33 -07:00
add elementwise sym
2015-09-05 19:23:50 -06:00
ifeq ($(USE_CUDA), 1)
Change CUB submodule to track Nvidia CUB project. (#13322) * Change CUB submodule to track Nvidia CUB project. Directly change submodule URL will impact every developer. "git submodule update" won't work, developer has to use "git submodule sync" first. * retrigger CI
2019-03-31 16:38:15 -07:00
CFLAGS += -I$(ROOTDIR)/3rdparty/nvidia_cub
torch ndarray function backend
2016-01-07 22:52:33 -08:00
ALL_DEP += $(CUOBJ) $(EXTRA_CUOBJ) $(PLUGIN_CUOBJ)
Introduce the ENABLE_CUDA_RTC build option (#9428) When OFF the ENABLE_CUDA_RTC option skips building the CUDA runtime compilation feature (CudaModule) that requires access to the CUDA driver API. The MXNet shared library build with this option OFF has no direct dependency on the CUDA driver library (e.g. libcuda.so) and can be used on both GPU and CPU hosts. CPU-only hosts require no stub CUDA driver library.
2018-01-15 15:21:23 -05:00
LDFLAGS += -lcufft
ifeq ($(ENABLE_CUDA_RTC), 1)
LDFLAGS += -lcuda -lnvrtc
CFLAGS += -DMXNET_ENABLE_CUDA_RTC=1
endif
TVM bridge support to JIT NDArray Function by TVM (#9880) * TVM bridge support. Support wrap TVM compiled function as a NDArray function. * Testcases and CI to include TVM as dependency * address review comments * Add more comments, change to constexpr * change to log warn * update comment on the type code
2018-02-27 08:41:28 -08:00
# Make sure to add stubs as fallback in order to be able to build
[Do not merge] Move to new CI (#8960) * Squashed commit of the following: commit ce53ad38a6ef260c7a5870faa58583b5fa2b1e30 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 10:53:09 2017 +0100 remove whitespace commit 71f1f8d5a71dd495a12594a4f01bf275fbded186 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 10:37:04 2017 +0100 Decrease timeout in Jenkinsfile commit ef88da18dd0f29b28ae15bb9840f3ff393e4e77f Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 10:35:44 2017 +0100 Cleanup and add comments commit ad59a48e4f68fe855029ba23a78bf9fa1abf1544 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 09:26:51 2017 +0100 Fix Integration Test Caffe-GPU to work on ubuntu16.04 with CUDA 8 commit fc7cf23e92a6bf48cc36968df274e55c4d47a17a Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 07:42:49 2017 +0100 Add verbose output to perl tests commit f076312de899fe859df3d03e74b3f97923c96a2c Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 06:40:31 2017 +0100 Adjust Jenkinsfile to ensure integration stage being run on nvidia-docker commit 9ed3cd0ba7411ad686bb0162e473684c7f689211 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Sat Nov 25 02:14:54 2017 +0100 Add nvidia dependencies to caffe_gpu commit 103225b70f734c92aa69416aa94eff3b24b28c5a Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 17:48:32 2017 +0100 Update GPU containers to ubuntu16.04 / CUDA 8 to fix OpenBLAS-hang 2 commit d1a12065826dabfbbe2102b375fb22a794eb5b3d Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 17:22:51 2017 +0100 Update GPU containers to ubuntu16.04 / CUDA 8 to fix OpenBLAS-hang commit b193874f0fd33a2d985fe430b9787d5cbf479662 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 14:39:36 2017 +0100 Fix typo commit c1e303468f494e28b8a4c625434a66c9b2c72282 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 14:25:43 2017 +0100 Fix typo commit 143fbe258e067d59efcf08ac948850b6de4cd9a0 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 14:18:44 2017 +0100 Fix unittest due to cuda commit f7255519bc96dcd25bfe83b89f8ec506af1caba6 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 14:00:48 2017 +0100 Make sure containers are run inside the right docker binary commit 7ae2c67ad91fcc2a7093f4d1bf417e6096bfa32e Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 13:43:56 2017 +0100 Fix typo commit 26d3a247ed502ea6f3ef25ded63ebe3f62ed9c72 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 13:42:34 2017 +0100 Trigger build commit aca7005ddb91ad9ce68fd4ee471664486e44a3fd Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 12:49:36 2017 +0100 Separated GPU-Build dockerfiles from runtime Dockerfiles commit 6091e5b171e3c50cecdfeacce04d21e2b3f65693 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 03:43:12 2017 +0100 Enable install of nvidia drivers in GPU Dockerfiles commit f0a3fe7474fb13d96bc84671e3402a8b33e71ccd Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 02:31:58 2017 +0100 Add cuda stub lib path to Makefile to fix libcuda.so.1 not found error commit 8233a3d032cdf77142f9e9c0424134da390f36fa Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 02:25:29 2017 +0100 Add softlink to fix libcuda.so.1 not found-error commit 8c3e50ae98a7de95476bc15dd8ca8c146003c1f1 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Fri Nov 24 01:44:49 2017 +0100 Fix libcuda.so.1 not found during build commit 2dbf35424060ec36d8cd4a084ddb3f1a976323cd Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Thu Nov 23 00:12:58 2017 +0100 Re-enable DEV-Flag in Jenkinsfile commit 6083f109c0811c310b7f22268c5d6556badf4426 Author: Pedro Larroy <pllarroy@amazon.com> Date: Wed Nov 22 14:38:37 2017 -0800 Fix uninitialized array warning commit 6e0dd21c121b275b341dbb9c304ce73ded64eae6 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 22:49:55 2017 +0100 Support CUDA-Builds on non-gpu-instances commit 5f53708e369c41ed77e76fd6623a9db6573f1b62 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 15:15:36 2017 +0100 Temp. disable dev compile mode to prevent warnings being treated as errors commit 5e5dff50af43ed7d618122d1c2f0379a1d7193fc Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 14:10:22 2017 +0100 Fix typo in Jenkins for windows commit a3d9f7483a693b0c513c677baaa5ba6621e722d1 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 14:07:18 2017 +0100 Fix tests/ci_build/with_the_same_user: line 27: /etc/sudoers.d/90-nopasswd-sudo: No such file or directory commit d76e342cfb76361ac166bc5e770bbef7721317d5 Merge: 5762ef88 0c0a5626 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 14:07:04 2017 +0100 Merge branch 'v0.12.0' of github.com:MXNetEdge/mabreu-incubator-mxnet into v0.12.0 commit 5762ef88b85754809fc98a01b149be5036737b64 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 13:46:57 2017 +0100 Add Kellens fix for C5 commit a93a9cd74ec362b4e3417e840169e3ce8b70cfba Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 11:40:12 2017 +0100 Add Jenkinsfile for windows commit 0c0a5626a614ac203cc7e93fbd5166e4f50267bb Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Wed Nov 22 11:40:12 2017 +0100 Add Jenkinsfile for windows commit 175d157171611778ee6a1e1804e4611e7ef4c136 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Mon Nov 20 19:56:18 2017 +0100 Remove failing test commit bb15df1464a02fbc9dff042b24e52b49187b9d55 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Mon Nov 20 17:25:29 2017 +0100 Disable unnecessary CUDA-archs commit c68265892256eb8756d2eba696ccc41766fbe3db Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Mon Nov 20 16:28:15 2017 +0100 Disable failing test test_operator:test_depthwise_convolution commit 07028474d10326029aa27afb81ff3eb3fa063b5f Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Mon Nov 20 15:01:08 2017 +0100 Increase git timeout commit 61d689424541801bad5a2ec5b46581810281d9a8 Author: Marco de Abreu <marco.g.abreu+github@gmail.com> Date: Mon Nov 20 13:16:28 2017 +0100 Split up Jenkins-Tasks to appropriate machines * Remove temp file * Add separate compile job for MKLML-CPU * Fix typo * Revert Jenkinsfile * Remove MKL Experimental * Revert "Remove MKL Experimental" This reverts commit 5e0f20c67e872c1a5af7491610ab55d37f6fc5b8. * Disable test_gluon:test_dtype * Cleanup * Add sudo install to lint * Add comment about MKLML * Fix typo * Clean up dockerfiles regarding mklml * Enable failing tests * Restructure dockerfiles * Fix typo * Fix Dockerfiles * Update mklml library * MKLML fix * Try to fix failing MKLML-gpu build * Fix typo * Disable flaky test https://github.com/apache/incubator-mxnet/issues/8892 * Remove workspace from jenkinsfile * Revert "Remove workspace from jenkinsfile" This reverts commit c2cfe5db296eafb008f63a7ddb0c841ac3b8313e. * Misc changes Merge Jenkinsfiles Documented Dockerfiles and scripts Cleaned up test comments Restore Dockerfile.ubuntu1404_cuda75_cudnn5 * fix typo * Fix typo * Revert KNOWN_CUDA_ARCH changes * Change makefile comment * Remove Jenkinsfile_windows * Add myself to CONTRIBUTORS.md * Cleanup * Add CUDA_ARCH to Jenkinsfile * Fix typo * Fix typo * Fix typo * Fix typo * Add cuda_archs to ci_build * Add cuda_archs to ci_build * Fix typo and comment * Disable flaky test test_loss:test_ctc_loss_train #8892
2017-12-06 19:31:19 +01:00
# without full CUDA install (especially if run without nvidia-docker)
LDFLAGS += -L/usr/local/cuda/lib64/stubs
NCCL integration (#8294) * NCCL integration * Skipping NCCL test (since it requires NCCL library to be present and enabled in build) * Add Apache header to test_nccl.py * Fixes from review * Trigger CI * Removing API change for Pull * Fixes * Fix * Fix * Fix * Fix * Fix * Indentation fixes and importing unittest in test_nccl.py * sorted_key_attrs -> key_attrs * More fixes from review * Fix * Fix lint * Support for aggregation in NCCL * Fix typo * Fix missing logic * Move from CommNCCL to KVStoreNCCL * Fix * Moved nccl update to separate function * Add message about not supporting gradient compression * Fix lint * Trigger CI
2017-11-21 06:52:16 -08:00
ifeq ($(USE_NCCL), 1)
ifneq ($(USE_NCCL_PATH), NONE)
CFLAGS += -I$(USE_NCCL_PATH)/include
LDFLAGS += -L$(USE_NCCL_PATH)/lib
endif
LDFLAGS += -lnccl
CFLAGS += -DMXNET_USE_NCCL=1
else
CFLAGS += -DMXNET_USE_NCCL=0
endif
split cpu & gpu version
2016-03-11 23:48:53 +08:00
else
NCCL integration (#8294) * NCCL integration * Skipping NCCL test (since it requires NCCL library to be present and enabled in build) * Add Apache header to test_nccl.py * Fixes from review * Trigger CI * Removing API change for Pull * Fixes * Fix * Fix * Fix * Fix * Fix * Indentation fixes and importing unittest in test_nccl.py * sorted_key_attrs -> key_attrs * More fixes from review * Fix * Fix lint * Support for aggregation in NCCL * Fix typo * Fix missing logic * Move from CommNCCL to KVStoreNCCL * Fix * Moved nccl update to separate function * Add message about not supporting gradient compression * Fix lint * Trigger CI
2017-11-21 06:52:16 -08:00
CFLAGS += -DMXNET_USE_NCCL=0
add elementwise sym
2015-09-05 19:23:50 -06:00
endif
[WIP] New faster version of the RecordIO iterator (#7152) * Improved ImageRecordIter performance Added option for using libjpeg-turbo directly to decode images ImageRecordIter can now use .idx files generated by im2rec.py Added rec2idx.py utility to generate .idx files from .rec files When using IndexedRecordIO (.rec and .idx together) shuffle option performs global shuffling * Add ASF license header * Update dmlc-core to fix a bug on Windows * USE_TURBO_JPEG -> USE_LIBJPEG_TURBO * trigger test
2017-10-14 13:09:21 -07:00
ifeq ($(USE_LIBJPEG_TURBO), 1)
ifneq ($(USE_LIBJPEG_TURBO_PATH), NONE)
CFLAGS += -I$(USE_LIBJPEG_TURBO_PATH)/include
LDFLAGS += -L$(USE_LIBJPEG_TURBO_PATH)/lib
endif
LDFLAGS += -lturbojpeg
CFLAGS += -DMXNET_USE_LIBJPEG_TURBO=1
else
CFLAGS += -DMXNET_USE_LIBJPEG_TURBO=0
endif
[MXNET-862] Basic maven jenkins pipeline (#13450) * Jenkins Publish Nightly Maven Progress * Seperate Build, Test, and Deploy Stages with parallel
2019-01-08 11:31:35 -08:00
ifeq ($(CI), 1)
MAVEN_ARGS := -B
endif
NNVM Refactor (#3194) * Init nnvm change * temp checkin * Move TShape to NNVM * Redirect Symbolic API to NNVM * Add Op Prop Adapter * Finish migrate in shape infer * Pass all symbolic test * temp commit * enable aux data * [EXEC] Basic version of exec for forward only * [EXEC] Enable most optimizations, still wait grad and context * fix legacy op with latest one * Update NNVM NodeRef * Adapt to newer interface * ALl registry of backop is complete * temp commit * Hack finish backward pass * [EXEC] One day pass * [EXEC] Pass all operator unittest * [EXEC] enable model parallel * Fully pass all legacy tests * Remove legacy symbolic code * update news * Make travis compile * Fix python3 * Update viz module to new json format
2016-09-01 21:17:33 -07:00
# For quick compile test, used smaller subset
ALLX_DEP= $(ALL_DEP)
Define build target for mkldnn lib build to fix 'make clean USE_MKMLDNN=1' issue (#11090) * Define build target for mkldnn lib build to fix 'make clean USE_MKMLDNN=1' issue * fix create install dir and other minor issues * Fix GPU MKLDNN and cpp-package build failure * Fix issue to only link with full MKL when BLAS is mkl * simplify logic by removing MKLDNN_ROOT support and some renaming * retrigger Jenkins * retrigger Jenkins * retrigger Jenkins * retrigger Jenkins * retrigger Jenkins
2018-06-28 01:32:53 +08:00
build/src/%.o: src/%.cc | mkldnn
torch ndarray function backend
2016-01-07 22:52:33 -08:00
@mkdir -p $(@D)
Enable warning as error (#4451) * Engineering system enhencement Enable warning as error Enable Profiler code into build Speed up build a bit * Fix warning in scala package
2017-01-01 07:37:40 +08:00
$(CXX) -std=c++11 -c $(CFLAGS) -MMD -c $< -o $@
torch ndarray function backend
2016-01-07 22:52:33 -08:00
Define build target for mkldnn lib build to fix 'make clean USE_MKMLDNN=1' issue (#11090) * Define build target for mkldnn lib build to fix 'make clean USE_MKMLDNN=1' issue * fix create install dir and other minor issues * Fix GPU MKLDNN and cpp-package build failure * Fix issue to only link with full MKL when BLAS is mkl * simplify logic by removing MKLDNN_ROOT support and some renaming * retrigger Jenkins * retrigger Jenkins * retrigger Jenkins * retrigger Jenkins * retrigger Jenkins
2018-06-28 01:32:53 +08:00
build/src/%_gpu.o: src/%.cu | mkldnn
torch ndarray function backend
2016-01-07 22:52:33 -08:00
@mkdir -p $(@D)
Improve CCache handling (#13456) * Remove gitignore entries * Modify Makefile * Modify user permissions * Add new ccache wrapper function * Change PATH rewrite to a different one to resolve CUDA issues * Add ccache to gpu cmake * Enable ccache for every build * Set permissions for arm dockerfiles * Disable ccache for ASAN * Remove g++-8 ccache redirect * Update Android Dockerfiles for user permissions * Fix ASAN compiler typo * Remove sanity for speed * Move build dir creation in android armv8 * Revert "Remove sanity for speed" This reverts commit e8386a774dafe96337930b9cac36cb24fc36585e. * Add ccache for NVCC in Makefile
2018-12-14 20:54:51 +01:00
$(NVCC) $(NVCCFLAGS) $(CUDA_ARCH) -Xcompiler "$(CFLAGS)" --generate-dependencies -MT build/src/$*_gpu.o $< >build/src/$*_gpu.d
Improve build system to avoid calling pkg-config many times (#4111) * Improve build system to avoid calling pkg-config many times * support specify cuda arch
2016-12-07 15:49:40 +08:00
$(NVCC) -c -o $@ $(NVCCFLAGS) $(CUDA_ARCH) -Xcompiler "$(CFLAGS)" $<
torch ndarray function backend
2016-01-07 22:52:33 -08:00
move plugin rules to sub makefile
2016-01-12 21:22:09 -08:00
# A nvcc bug cause it to generate "generic/xxx.h" dependencies from torch headers.
# Use CXX to generate dependency instead.
torch ndarray function backend
2016-01-07 22:52:33 -08:00
build/plugin/%_gpu.o: plugin/%.cu
simplify makefile, automatically generate dependency
2015-09-07 18:12:23 -04:00
@mkdir -p $(@D)
[OP] 3d conv + 3d pool (#2585) * cudnn 3d conv & 3d pooling
2016-07-08 14:07:45 -07:00
$(CXX) -std=c++11 $(CFLAGS) -MM -MT build/plugin/$*_gpu.o $< >build/plugin/$*_gpu.d
Improve build system to avoid calling pkg-config many times (#4111) * Improve build system to avoid calling pkg-config many times * support specify cuda arch
2016-12-07 15:49:40 +08:00
$(NVCC) -c -o $@ $(NVCCFLAGS) $(CUDA_ARCH) -Xcompiler "$(CFLAGS)" $<
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
added mkldnn dependency for plugin compile target (#14274)
2019-03-01 16:26:21 -08:00
build/plugin/%.o: plugin/%.cc | mkldnn
make ccsgd the default and add number of batch to predict
2015-12-16 23:38:36 -08:00
@mkdir -p $(@D)
Enable warning as error (#4451) * Engineering system enhencement Enable warning as error Enable Profiler code into build Speed up build a bit * Fix warning in scala package
2017-01-01 07:37:40 +08:00
$(CXX) -std=c++11 -c $(CFLAGS) -MMD -c $< -o $@
make ccsgd the default and add number of batch to predict
2015-12-16 23:38:36 -08:00
fix caffe stream (#2984) * fix caffe stream * allow multiple extra operator * fix python 2/3 codec conflict
2016-08-10 18:33:41 -07:00
%_gpu.o: %.cu
make ccsgd the default and add number of batch to predict
2015-12-16 23:38:36 -08:00
@mkdir -p $(@D)
Improve CCache handling (#13456) * Remove gitignore entries * Modify Makefile * Modify user permissions * Add new ccache wrapper function * Change PATH rewrite to a different one to resolve CUDA issues * Add ccache to gpu cmake * Enable ccache for every build * Set permissions for arm dockerfiles * Disable ccache for ASAN * Remove g++-8 ccache redirect * Update Android Dockerfiles for user permissions * Fix ASAN compiler typo * Remove sanity for speed * Move build dir creation in android armv8 * Revert "Remove sanity for speed" This reverts commit e8386a774dafe96337930b9cac36cb24fc36585e. * Add ccache for NVCC in Makefile
2018-12-14 20:54:51 +01:00
$(NVCC) $(NVCCFLAGS) $(CUDA_ARCH) -Xcompiler "$(CFLAGS) -Isrc/operator" --generate-dependencies -MT $*_gpu.o $< >$*_gpu.d
Improve build system to avoid calling pkg-config many times (#4111) * Improve build system to avoid calling pkg-config many times * support specify cuda arch
2016-12-07 15:49:40 +08:00
$(NVCC) -c -o $@ $(NVCCFLAGS) $(CUDA_ARCH) -Xcompiler "$(CFLAGS) -Isrc/operator" $<
make ccsgd the default and add number of batch to predict
2015-12-16 23:38:36 -08:00
add doc for gluon, sym/nd contrib (#7284) * add doc for contrib * docs for gluon, sym/nd contrib
2017-08-02 16:19:41 -07:00
%.o: %.cc $(CORE_INC)
fix caffe stream (#2984) * fix caffe stream * allow multiple extra operator * fix python 2/3 codec conflict
2016-08-10 18:33:41 -07:00
@mkdir -p $(@D)
Enable warning as error (#4451) * Engineering system enhencement Enable warning as error Enable Profiler code into build Speed up build a bit * Fix warning in scala package
2017-01-01 07:37:40 +08:00
$(CXX) -std=c++11 -c $(CFLAGS) -MMD -Isrc/operator -c $< -o $@
fix caffe stream (#2984) * fix caffe stream * allow multiple extra operator * fix python 2/3 codec conflict
2016-08-10 18:33:41 -07:00
Set install path for libmxnet.so dynamic lib on Mac OS (#13629)
2018-12-12 20:55:24 -08:00
# Set install path for libmxnet.so on Mac OS
ifeq ($(UNAME_S), Darwin)
LDFLAGS += -Wl,-install_name,@rpath/libmxnet.so
endif
[OP] Fix load of scalar binary op
2016-01-08 08:09:11 -08:00
# NOTE: to statically link libmxnet.a we need the option
# --Wl,--whole-archive -lmxnet --Wl,--no-whole-archive
NNVM Refactor (#3194) * Init nnvm change * temp checkin * Move TShape to NNVM * Redirect Symbolic API to NNVM * Add Op Prop Adapter * Finish migrate in shape infer * Pass all symbolic test * temp commit * enable aux data * [EXEC] Basic version of exec for forward only * [EXEC] Enable most optimizations, still wait grad and context * fix legacy op with latest one * Update NNVM NodeRef * Adapt to newer interface * ALl registry of backop is complete * temp commit * Hack finish backward pass * [EXEC] One day pass * [EXEC] Pass all operator unittest * [EXEC] enable model parallel * Fully pass all legacy tests * Remove legacy symbolic code * update news * Make travis compile * Fix python3 * Update viz module to new json format
2016-09-01 21:17:33 -07:00
lib/libmxnet.a: $(ALLX_DEP)
[make] delete lib and bin, try to fix travis python
2015-10-08 23:37:04 -04:00
@mkdir -p $(@D)
simplify makefile, automatically generate dependency
2015-09-07 18:12:23 -04:00
ar crv $@ $(filter %.o, $?)
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
NNVM Refactor (#3194) * Init nnvm change * temp checkin * Move TShape to NNVM * Redirect Symbolic API to NNVM * Add Op Prop Adapter * Finish migrate in shape infer * Pass all symbolic test * temp commit * enable aux data * [EXEC] Basic version of exec for forward only * [EXEC] Enable most optimizations, still wait grad and context * fix legacy op with latest one * Update NNVM NodeRef * Adapt to newer interface * ALl registry of backop is complete * temp commit * Hack finish backward pass * [EXEC] One day pass * [EXEC] Pass all operator unittest * [EXEC] enable model parallel * Fully pass all legacy tests * Remove legacy symbolic code * update news * Make travis compile * Fix python3 * Update viz module to new json format
2016-09-01 21:17:33 -07:00
lib/libmxnet.so: $(ALLX_DEP)
update mklml and mkl mac support (#7587)
2017-08-29 17:14:01 -07:00
@mkdir -p $(@D)
$(CXX) $(CFLAGS) -shared -o $@ $(filter-out %libnnvm.a, $(filter %.o %.a, $^)) $(LDFLAGS) \
-Wl,${WHOLE_ARCH} $(filter %libnnvm.a, $^) -Wl,${NO_WHOLE_ARCH}
Compile MKLDNN in Mac. (#10115) * Compile MKLDNN in Mac. * A more general way of finding #cores. * find the right dyn library of mkldnn.
2018-03-14 19:20:47 -07:00
ifeq ($(USE_MKLDNN), 1)
ifeq ($(UNAME_S), Darwin)
Copy only one libmkldnn file (#10245) * Only copy libmkldnn.so * Update Jenkins. * use libmkldnn.so.0 * copy the right libmklml file.
2018-03-27 13:46:36 -07:00
install_name_tool -change '@rpath/libmklml.dylib' '@loader_path/libmklml.dylib' $@
install_name_tool -change '@rpath/libiomp5.dylib' '@loader_path/libiomp5.dylib' $@
install_name_tool -change '@rpath/libmkldnn.0.dylib' '@loader_path/libmkldnn.0.dylib' $@
Compile MKLDNN in Mac. (#10115) * Compile MKLDNN in Mac. * A more general way of finding #cores. * find the right dyn library of mkldnn.
2018-03-14 19:20:47 -07:00
endif
endif
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
Update makefile to depend on sub directories
2016-07-31 15:57:20 -07:00
$(PS_PATH)/build/libps.a: PSLITE
PSLITE:
[MXNET-472] ccache for docker builds (#11151) * [MXNET-472] Add ccache support to docker builds * Added ccache stages to all containers * Refactored ccache installs in docker images * Reformatted build.py * Added ccache install to android docker builds * Improved setting ccache directory and max cache size * Added ccache to cmake based docker builds * Removed unnessesary yum install from centos7 ccache build * Added default compilers for ccache for docker builds * Added README comment about ccache mapping into docker builds * Reverted not working ccache configurations * Added comments about ccache installation * Move install scripts * Update ubuntu_r.sh
2018-06-08 03:12:30 +02:00
$(MAKE) CXX="$(CXX)" DEPS_PATH="$(DEPS_PATH)" -C $(PS_PATH) ps
[kvstore] add ps-lite
2015-10-02 22:22:22 -04:00
Update makefile to depend on sub directories
2016-07-31 15:57:20 -07:00
$(DMLC_CORE)/libdmlc.a: DMLCCORE
DMLCCORE:
change recursive make commands per gnumake manual recommendation (#4983)
2017-02-11 23:12:15 -08:00
+ cd $(DMLC_CORE); $(MAKE) libdmlc.a USE_SSE=$(USE_SSE) config=$(ROOTDIR)/$(config); cd $(ROOTDIR)
api
2015-06-27 14:57:11 -06:00
Add NNVM .cc .h deps to Makfile (#6416) * Fix for NNVM build * Add deps to nnvm/include
2017-05-24 12:09:30 -07:00
NNVM_INC = $(wildcard $(NNVM_PATH)/include/*/*.h)
NNVM_SRC = $(wildcard $(NNVM_PATH)/src/*/*/*.cc $(NNVM_PATH)/src/*/*.cc $(NNVM_PATH)/src/*.cc)
$(NNVM_PATH)/lib/libnnvm.a: $(NNVM_INC) $(NNVM_SRC)
change recursive make commands per gnumake manual recommendation (#4983)
2017-02-11 23:12:15 -08:00
+ cd $(NNVM_PATH); $(MAKE) lib/libnnvm.a DMLC_CORE_PATH=$(DMLC_CORE); cd $(ROOTDIR)
NNVM Refactor (#3194) * Init nnvm change * temp checkin * Move TShape to NNVM * Redirect Symbolic API to NNVM * Add Op Prop Adapter * Finish migrate in shape infer * Pass all symbolic test * temp commit * enable aux data * [EXEC] Basic version of exec for forward only * [EXEC] Enable most optimizations, still wait grad and context * fix legacy op with latest one * Update NNVM NodeRef * Adapt to newer interface * ALl registry of backop is complete * temp commit * Hack finish backward pass * [EXEC] One day pass * [EXEC] Pass all operator unittest * [EXEC] enable model parallel * Fully pass all legacy tests * Remove legacy symbolic code * update news * Make travis compile * Fix python3 * Update viz module to new json format
2016-09-01 21:17:33 -07:00
bin/im2rec: tools/im2rec.cc $(ALLX_DEP)
add showme example and im2rec, modify io,=.md
2015-09-21 01:55:32 +08:00
$(BIN) :
[make] delete lib and bin, try to fix travis python
2015-10-08 23:37:04 -04:00
@mkdir -p $(@D)
[OP] 3d conv + 3d pool (#2585) * cudnn 3d conv & 3d pooling
2016-07-08 14:07:45 -07:00
$(CXX) $(CFLAGS) -std=c++11 -o $@ $(filter %.cpp %.o %.c %.a %.cc, $^) $(LDFLAGS)
add showme example and im2rec, modify io,=.md
2015-09-21 01:55:32 +08:00
Caffe without the patch, cpp-package fixed also with Caffe plugin (#5573) * PEP8 indentation fix * Remove executable flag * Remove executable flags * Fixing cpp-package including problems with caffe converter * Fix warnings * Remove need for caffe patch for caffe plugin * Ignore cmake paths, remove rebuildable op.h for cpp-package * cpp-package examples fixed. Makefile and CMake of op.h and examples added. * cpp-package examples fixed. Makefile and CMake of op.h and examples added. * Fix source file * Better example.mk * Turn off caffe by default * lint fixes * Trying to figure out how to fix submodules * nohub problem on travis so force retry * Edited test to just run make instead of obsolete make example * Fix cpp-package op.h bootstrapping with CMake * Build tweaking cpp-package * Lint fixes * op.h generator * FIx cpp-package for latest merge from master * Fix lint * static link * link whole static lib * Trigger another build attempt * Trigger another build attempt * Add caffe plugin support (disabled until dependencies can be added to build machine) * Add cufft library * win32 cufft library * don't link unit tests as static * Add include for rebuildable protobuf header caffe.pb.h
2017-03-30 20:13:36 -07:00
# CPP Package
ifeq ($(USE_CPP_PACKAGE), 1)
include cpp-package/cpp-package.mk
endif
Define build target for mkldnn lib build to fix 'make clean USE_MKMLDNN=1' issue (#11090) * Define build target for mkldnn lib build to fix 'make clean USE_MKMLDNN=1' issue * fix create install dir and other minor issues * Fix GPU MKLDNN and cpp-package build failure * Fix issue to only link with full MKL when BLAS is mkl * simplify logic by removing MKLDNN_ROOT support and some renaming * retrigger Jenkins * retrigger Jenkins * retrigger Jenkins * retrigger Jenkins * retrigger Jenkins
2018-06-28 01:32:53 +08:00
include mkldnn.mk
Change engine callback to old style; support multiple streams in Naive. Make PushSync inline; Use registry and env variable to select engine move tests into fine grained folders checkin gtest fix test
2015-09-12 16:49:36 -07:00
include tests/cpp/unittest.mk
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
Caffe without the patch, cpp-package fixed also with Caffe plugin (#5573) * PEP8 indentation fix * Remove executable flag * Remove executable flags * Fixing cpp-package including problems with caffe converter * Fix warnings * Remove need for caffe patch for caffe plugin * Ignore cmake paths, remove rebuildable op.h for cpp-package * cpp-package examples fixed. Makefile and CMake of op.h and examples added. * cpp-package examples fixed. Makefile and CMake of op.h and examples added. * Fix source file * Better example.mk * Turn off caffe by default * lint fixes * Trying to figure out how to fix submodules * nohub problem on travis so force retry * Edited test to just run make instead of obsolete make example * Fix cpp-package op.h bootstrapping with CMake * Build tweaking cpp-package * Lint fixes * op.h generator * FIx cpp-package for latest merge from master * Fix lint * static link * link whole static lib * Trigger another build attempt * Trigger another build attempt * Add caffe plugin support (disabled until dependencies can be added to build machine) * Add cufft library * win32 cufft library * don't link unit tests as static * Add include for rebuildable protobuf header caffe.pb.h
2017-03-30 20:13:36 -07:00
extra-packages: $(EXTRA_PACKAGES)
[cpptest] refactor
2015-09-23 20:26:29 +00:00
test: $(TEST)
move storage.h to include, simplify it
2015-06-14 17:36:33 -07:00
[CI] add script to build and run docker (#5325) * add ci_build.sh * test ci * test ci * fix lint
2017-03-09 23:25:01 -08:00
lint: cpplint rcpplint jnilint pylint
cpplint:
move dmlc-core & make" (#10231) +cmake amal examples update jenkins udpate amal.py to force include 3p update amal find_source logic remove log update license move nnvm misc nnvm updates fix perl make ; update whitelist add missing license header fix launcher.py update submodule name update amal expand file
2018-03-25 19:45:24 -07:00
3rdparty/dmlc-core/scripts/lint.py mxnet cpp include src plugin cpp-package tests \
CTC Loss Operator (#5834) * first commit of ctc loss interface * Code from WarpCTC without modification * rename folder to ctc_include * refactor warpctc code - remove reduction routines, use MShadow * add CTC tests against Torch WarpCTC * added gradient calculation * fixed lint * added warp ctc license * exclude warp ctc loss implementation from cpplint * get rid of Wreorder errors * fix input type signature to fix cpp package build, and rename "labels" to "label"
2017-05-03 09:40:17 +02:00
--exclude_path src/operator/contrib/ctc_include
[CI] add script to build and run docker (#5325) * add ci_build.sh * test ci * test ci * fix lint
2017-03-09 23:25:01 -08:00
pylint:
Fix RAT license checker which is broken in trunk (#12148) Remove obsolete CI folder
2018-08-13 23:10:37 +02:00
pylint --rcfile=$(ROOTDIR)/ci/other/pylintrc --ignore-patterns=".*\.so$$,.*\.dll$$,.*\.dylib$$" python/mxnet tools/caffe_converter/*.py
add travis
2015-07-03 17:06:23 -06:00
[doc] new sphnix plugin (#6105) * update doc * rm * update * update ndarray * update mds * update * update * update * update * update * update * update image.md and others * update
2017-05-07 22:19:25 -07:00
doc: docs
docs:
CI docker revamp; Add Jetson, Raspberry and CentOS 7 build [MXNET-42][MXNET-43] (#9995) * Start dockerfile revamp * Fix lint * Fix lint * Fix lint * Add python unit test * Add new dependency * Add user creation * Fix file permission * Determine USER_ID automatically * REmove ENV command in dockerfile * Remove python nose timer, improve useradd * ENable nvidia docker * Add remaining unittests * Add CentOS 7 unittests * Add integration tests * Add TVM and LLVM to dockerfile * Add ARMv7 and ARMv8 to build * Fix amalgamation build * Improvements and android_arm64 fixes, missing removing pthread * Jetson fix (install unzip) * Build jetson with make until issue with libomp.so is resolved #10011 * Fix Amalgamation builds * Fix Jetson build by switching to cmake * Assign CentOS gpu test to gpu instance * Fix R builds * Assign jobs to right docker containers * Fix missing file permissions inside docker * Create homedir on centos * Enable lapack * Fix Lapack on Cent OS 7 * Disable MXNET_MKLDNN_DEBUG * Delete Dockerfiles * Last general refinements before finish * Remove docker_multiarch folder, superseeded by the new ci scripts * Address review comments * Fix Caffe Integrationstest * Fix deploy stage * Address review comments * Address review comments * Enable script to run on Mac
2018-03-09 23:33:10 +01:00
make -C docs html
[doc] new sphnix plugin (#6105) * update doc * rm * update * update ndarray * update mds * update * update * update * update * update * update * update image.md and others * update
2017-05-07 22:19:25 -07:00
clean_docs:
make -C docs clean
[make] fix make doc warnings
2015-10-08 22:06:04 -04:00
add cpp doc
2015-08-21 23:49:07 -06:00
doxygen:
[DOC] Fix doxygen
2016-05-07 20:56:04 -07:00
doxygen docs/Doxyfile
add travis
2015-07-03 17:06:23 -06:00
[PYTHON] Optional Cython Module for Symbols (#3242) * [CYTHON] Checkin cython enhancement * fix lint * [DOC] Move common doc to base
2016-09-07 16:08:55 -07:00
# Cython build
[OP] Add alias support (#3261)
2016-09-08 19:10:58 -07:00
cython:
fix cython build (#6356) * fix cython * fix
2017-05-23 09:04:37 -07:00
cd python; python setup.py build_ext --inplace --with-cython
[OP] Add alias support (#3261)
2016-09-08 19:10:58 -07:00
[PYTHON] Optional Cython Module for Symbols (#3242) * [CYTHON] Checkin cython enhancement * fix lint * [DOC] Move common doc to base
2016-09-07 16:08:55 -07:00
cython2:
fix cython build (#6356) * fix cython * fix
2017-05-23 09:04:37 -07:00
cd python; python2 setup.py build_ext --inplace --with-cython
[PYTHON] Optional Cython Module for Symbols (#3242) * [CYTHON] Checkin cython enhancement * fix lint * [DOC] Move common doc to base
2016-09-07 16:08:55 -07:00
cython3:
fix cython build (#6356) * fix cython * fix
2017-05-23 09:04:37 -07:00
cd python; python3 setup.py build_ext --inplace --with-cython
[PYTHON] Optional Cython Module for Symbols (#3242) * [CYTHON] Checkin cython enhancement * fix lint * [DOC] Move common doc to base
2016-09-07 16:08:55 -07:00
cyclean:
rm -rf python/mxnet/*/*.so python/mxnet/*/*.cpp
[R] switch to roxygen, man is removed for now need to group by rdname
2015-10-15 21:52:22 -07:00
# R related shortcuts
rcpplint:
move dmlc-core & make" (#10231) +cmake amal examples update jenkins udpate amal.py to force include 3p update amal find_source logic remove log update license move nnvm misc nnvm updates fix perl make ; update whitelist add missing license header fix launcher.py update submodule name update amal expand file
2018-03-25 19:45:24 -07:00
3rdparty/dmlc-core/scripts/lint.py mxnet-rcpp ${LINT_LANG} R-package/src
[R] switch to roxygen, man is removed for now need to group by rdname
2015-10-15 21:52:22 -07:00
close# 3343 (#3421)
2016-09-30 17:48:29 -04:00
rpkg:
update Makefile
2015-11-04 11:29:47 -05:00
mkdir -p R-package/inst/libs
[R] im2rec in R. close #7273 (#7389)
2017-08-09 06:44:58 +00:00
cp src/io/image_recordio.h R-package/src
update Makefile
2015-11-04 11:29:47 -05:00
cp -rf lib/libmxnet.so R-package/inst/libs
Fix MXNet R package build (#13952) * fix mxnet r package build * add ci * remove mkldnn-gpu test for R * add minimal test for MKLDNN-R * pick mlp as minimal R test
2019-01-30 22:25:29 -08:00
if [ -e "lib/libmkldnn.so.0" ]; then \
cp -rf lib/libmkldnn.so.0 R-package/inst/libs; \
cp -rf lib/libiomp5.so R-package/inst/libs; \
cp -rf lib/libmklml_intel.so R-package/inst/libs; \
fi
update Makefile
2015-11-04 11:29:47 -05:00
mkdir -p R-package/inst/include
R-Package Makefile (#14068) * rpkg Makefile * rpkg makefile fix
2019-02-15 15:23:01 -05:00
cp -rl include/* R-package/inst/include
[R] add R test into jenkins; add rpkgtest into makefile (#6733) * simpify the makefile;add R test into jenkins;some basic test for R * fix GPU test * fix GPU test for R pkg
2017-06-28 16:01:23 -07:00
Rscript -e "if(!require(devtools)){install.packages('devtools', repo = 'https://cloud.r-project.org/')}"
Add back R tests and fix typo around R and perl tests (#13940) * Add back R tests and fix typo around R and perl tests * Fix permissions * Fix copy&paste mistake around roxygen and remove previous permission override
2019-01-21 20:01:17 +01:00
Rscript -e "if(!require(roxygen2)||packageVersion('roxygen2') < '6.1.1'){install.packages('roxygen2', repo = 'https://cloud.r-project.org/')}"
[R] add R test into jenkins; add rpkgtest into makefile (#6733) * simpify the makefile;add R test into jenkins;some basic test for R * fix GPU test * fix GPU test for R pkg
2017-06-28 16:01:23 -07:00
Rscript -e "library(devtools); library(methods); options(repos=c(CRAN='https://cloud.r-project.org/')); install_deps(pkg='R-package', dependencies = TRUE)"
remove legacy installation of Roxygen2 5.0 and add R-specific clean target (#12993) (#12998) * remove installation of legacy Roxygen2 vers. 5.0 * add R-specific clean target (#12993) * fixup! remove installation of legacy Roxygen2 vers. 5.0 * fixup! remove installation of legacy Roxygen2 vers. 5.0
2018-11-01 23:59:00 +01:00
cp R-package/dummy.NAMESPACE R-package/NAMESPACE
echo "import(Rcpp)" >> R-package/NAMESPACE
close# 3343 (#3421)
2016-09-30 17:48:29 -04:00
R CMD INSTALL R-package
remove legacy installation of Roxygen2 5.0 and add R-specific clean target (#12993) (#12998) * remove installation of legacy Roxygen2 vers. 5.0 * add R-specific clean target (#12993) * fixup! remove installation of legacy Roxygen2 vers. 5.0 * fixup! remove installation of legacy Roxygen2 vers. 5.0
2018-11-01 23:59:00 +01:00
Rscript -e "require(mxnet); mxnet:::mxnet.export('R-package'); warnings()"
rm R-package/NAMESPACE
R-Package Makefile (#14068) * rpkg Makefile * rpkg makefile fix
2019-02-15 15:23:01 -05:00
Rscript -e "devtools::document('R-package');warnings()"
[R] update Makefile. close #7675 (#7686)
2017-08-31 15:47:13 -04:00
R CMD INSTALL R-package
update Makefile
2015-11-04 11:29:47 -05:00
[R] add R test into jenkins; add rpkgtest into makefile (#6733) * simpify the makefile;add R test into jenkins;some basic test for R * fix GPU test * fix GPU test for R pkg
2017-06-28 16:01:23 -07:00
rpkgtest:
[MXNET-851] Test coverage metrics for R-package (#12391) * get test coverage metrics for r-package * update makefile to include test coverage, and fix test failures * write to json file
2018-09-13 07:42:37 -07:00
Rscript -e 'require(testthat);res<-test_dir("R-package/tests/testthat");if(!testthat:::all_passed(res)){stop("Test failures", call. = FALSE)}'
fix file lock issue (#13296)
2018-11-16 21:29:07 -08:00
Rscript -e 'res<-covr:::package_coverage("R-package");fileConn<-file(paste("r-package_coverage_",toString(runif(1)),".json"));writeLines(covr:::to_codecov(res), fileConn);close(fileConn)'
[R] add R test into jenkins; add rpkgtest into makefile (#6733) * simpify the makefile;add R test into jenkins;some basic test for R * fix GPU test * fix GPU test for R pkg
2017-06-28 16:01:23 -07:00
Add Scala package dev tools for deploy (#8498) * [scala] do not print op definition during compiling by default * [scala] add dev tools for changing artifactId * [scala] add scala make clean
2017-11-02 00:23:23 +08:00
scalaclean:
Re-organize Scala maven build (#13626) * Re-organize scala maven build 1. Automatically detect which platform to build for scala. 2. Remove platform dependend submodules 3. Fix cyclic module dependencies 4. Fix scalatype style check 5. Now mvn can be executed in submodule 6. Maven build can be executed from any directory not only in root project 7. Checkin javah header file, and use verify task to detect native API changes 8. Improve incremental build performance 9. Remove unittest and integration-test profile, use proper task instead 10. Delete generated scala file during maven clean. * Redo maven deploy related tasks. 1. Removed maven release plugin. 2. Make maven build friendly to CI, allow cli override version. 3. Moved gpg signing to deploy stage. 4. Created a separeated deploy module. 5. Updated Makefile to new maven build change. 6. Remove unused nexus-staging-plugin 7. Added nightly and staging profile for CI. * Support mkldnn for Scala.
2019-01-08 17:25:53 -08:00
(cd $(ROOTDIR)/scala-package && mvn clean)
[MXNET-1155] Add scala packageTest utility (#13046) * [MXNET-1155] Add scala packageTest utility * Clean up utility * Safe change directory in Makefile for scala * mvn install file instructions with details
2018-12-11 15:19:06 -08:00
integrate scala-package compiling into ROOT/Makefile
2016-02-21 14:12:16 +08:00
scalapkg:
Re-organize Scala maven build (#13626) * Re-organize scala maven build 1. Automatically detect which platform to build for scala. 2. Remove platform dependend submodules 3. Fix cyclic module dependencies 4. Fix scalatype style check 5. Now mvn can be executed in submodule 6. Maven build can be executed from any directory not only in root project 7. Checkin javah header file, and use verify task to detect native API changes 8. Improve incremental build performance 9. Remove unittest and integration-test profile, use proper task instead 10. Delete generated scala file during maven clean. * Redo maven deploy related tasks. 1. Removed maven release plugin. 2. Make maven build friendly to CI, allow cli override version. 3. Moved gpg signing to deploy stage. 4. Created a separeated deploy module. 5. Updated Makefile to new maven build change. 6. Remove unused nexus-staging-plugin 7. Added nightly and staging profile for CI. * Support mkldnn for Scala.
2019-01-08 17:25:53 -08:00
(cd $(ROOTDIR)/scala-package && mvn install -DskipTests)
scalainstall:
(cd $(ROOTDIR)/scala-package && mvn install)
integrate scala-package compiling into ROOT/Makefile
2016-02-21 14:12:16 +08:00
[MXNET-531] Integration Test for Scala (#11596) * Separate unit test and integration tests for Scala
2018-07-11 22:19:09 -07:00
scalaunittest:
Re-organize Scala maven build (#13626) * Re-organize scala maven build 1. Automatically detect which platform to build for scala. 2. Remove platform dependend submodules 3. Fix cyclic module dependencies 4. Fix scalatype style check 5. Now mvn can be executed in submodule 6. Maven build can be executed from any directory not only in root project 7. Checkin javah header file, and use verify task to detect native API changes 8. Improve incremental build performance 9. Remove unittest and integration-test profile, use proper task instead 10. Delete generated scala file during maven clean. * Redo maven deploy related tasks. 1. Removed maven release plugin. 2. Make maven build friendly to CI, allow cli override version. 3. Moved gpg signing to deploy stage. 4. Created a separeated deploy module. 5. Updated Makefile to new maven build change. 6. Remove unused nexus-staging-plugin 7. Added nightly and staging profile for CI. * Support mkldnn for Scala.
2019-01-08 17:25:53 -08:00
(cd $(ROOTDIR)/scala-package && mvn install)
[MXNET-531] Integration Test for Scala (#11596) * Separate unit test and integration tests for Scala
2018-07-11 22:19:09 -07:00
scalaintegrationtest:
Re-organize Scala maven build (#13626) * Re-organize scala maven build 1. Automatically detect which platform to build for scala. 2. Remove platform dependend submodules 3. Fix cyclic module dependencies 4. Fix scalatype style check 5. Now mvn can be executed in submodule 6. Maven build can be executed from any directory not only in root project 7. Checkin javah header file, and use verify task to detect native API changes 8. Improve incremental build performance 9. Remove unittest and integration-test profile, use proper task instead 10. Delete generated scala file during maven clean. * Redo maven deploy related tasks. 1. Removed maven release plugin. 2. Make maven build friendly to CI, allow cli override version. 3. Moved gpg signing to deploy stage. 4. Created a separeated deploy module. 5. Updated Makefile to new maven build change. 6. Remove unused nexus-staging-plugin 7. Added nightly and staging profile for CI. * Support mkldnn for Scala.
2019-01-08 17:25:53 -08:00
(cd $(ROOTDIR)/scala-package && mvn integration-test -DskipTests=false)
[MXNET-1155] Add scala packageTest utility (#13046) * [MXNET-1155] Add scala packageTest utility * Clean up utility * Safe change directory in Makefile for scala * mvn install file instructions with details
2018-12-11 15:19:06 -08:00
JNI code lint
2016-03-03 23:42:18 +08:00
jnilint:
Re-organize Scala maven build (#13626) * Re-organize scala maven build 1. Automatically detect which platform to build for scala. 2. Remove platform dependend submodules 3. Fix cyclic module dependencies 4. Fix scalatype style check 5. Now mvn can be executed in submodule 6. Maven build can be executed from any directory not only in root project 7. Checkin javah header file, and use verify task to detect native API changes 8. Improve incremental build performance 9. Remove unittest and integration-test profile, use proper task instead 10. Delete generated scala file during maven clean. * Redo maven deploy related tasks. 1. Removed maven release plugin. 2. Make maven build friendly to CI, allow cli override version. 3. Moved gpg signing to deploy stage. 4. Created a separeated deploy module. 5. Updated Makefile to new maven build change. 6. Remove unused nexus-staging-plugin 7. Added nightly and staging profile for CI. * Support mkldnn for Scala.
2019-01-08 17:25:53 -08:00
3rdparty/dmlc-core/scripts/lint.py mxnet-jnicpp cpp scala-package/native/src --exclude_path scala-package/native/src/main/native/org_apache_mxnet_native_c_api.h
JNI code lint
2016-03-03 23:42:18 +08:00
remove legacy installation of Roxygen2 5.0 and add R-specific clean target (#12993) (#12998) * remove installation of legacy Roxygen2 vers. 5.0 * add R-specific clean target (#12993) * fixup! remove installation of legacy Roxygen2 vers. 5.0 * fixup! remove installation of legacy Roxygen2 vers. 5.0
2018-11-01 23:59:00 +01:00
rclean:
$(RM) -r R-package/src/image_recordio.h R-package/NAMESPACE R-package/man R-package/R/mxnet_generated.R \
Pop up version number in ubuntu-r setup script (#4464) * Update install-mxnet-ubuntu-r.sh * improve Makefile and setup * improve Makefile and setup * update documents * add generated r pack to make clean
2017-01-04 09:12:49 -08:00
R-package/inst R-package/src/*.o R-package/src/*.so mxnet_*.tar.gz
remove legacy installation of Roxygen2 5.0 and add R-specific clean target (#12993) (#12998) * remove installation of legacy Roxygen2 vers. 5.0 * add R-specific clean target (#12993) * fixup! remove installation of legacy Roxygen2 vers. 5.0 * fixup! remove installation of legacy Roxygen2 vers. 5.0
2018-11-01 23:59:00 +01:00
ifneq ($(EXTRA_OPERATORS),)
clean: rclean cyclean $(EXTRA_PACKAGES_CLEAN)
Improve CCache handling (#13456) * Remove gitignore entries * Modify Makefile * Modify user permissions * Add new ccache wrapper function * Change PATH rewrite to a different one to resolve CUDA issues * Add ccache to gpu cmake * Enable ccache for every build * Set permissions for arm dockerfiles * Disable ccache for ASAN * Remove g++-8 ccache redirect * Update Android Dockerfiles for user permissions * Fix ASAN compiler typo * Remove sanity for speed * Move build dir creation in android armv8 * Revert "Remove sanity for speed" This reverts commit e8386a774dafe96337930b9cac36cb24fc36585e. * Add ccache for NVCC in Makefile
2018-12-14 20:54:51 +01:00
$(RM) -r build lib bin deps *~ */*~ */*/*~ */*/*/*~
Add scalaclean to make clean (#14322)
2019-03-04 13:26:21 -08:00
(cd scala-package && mvn clean) || true
change recursive make commands per gnumake manual recommendation (#4983)
2017-02-11 23:12:15 -08:00
cd $(DMLC_CORE); $(MAKE) clean; cd -
cd $(PS_PATH); $(MAKE) clean; cd -
cd $(NNVM_PATH); $(MAKE) clean; cd -
cuda support for linalg-functions, restructuring of linalg interfaces (#7147) * cuda support for linalg-functions, restructuring of linalg interfaces * incorporate newest mshadow * adjustments to linalg operators
2017-08-12 21:12:44 +02:00
cd $(AMALGAMATION_PATH); $(MAKE) clean; cd -
add ssd as example (#3674)
2016-11-07 15:10:50 -06:00
$(RM) -r $(patsubst %, %/*.d, $(EXTRA_OPERATORS)) $(patsubst %, %/*/*.d, $(EXTRA_OPERATORS))
$(RM) -r $(patsubst %, %/*.o, $(EXTRA_OPERATORS)) $(patsubst %, %/*/*.o, $(EXTRA_OPERATORS))
ndarray type support.
2015-12-28 23:13:08 -08:00
else
remove legacy installation of Roxygen2 5.0 and add R-specific clean target (#12993) (#12998) * remove installation of legacy Roxygen2 vers. 5.0 * add R-specific clean target (#12993) * fixup! remove installation of legacy Roxygen2 vers. 5.0 * fixup! remove installation of legacy Roxygen2 vers. 5.0
2018-11-01 23:59:00 +01:00
clean: rclean mkldnn_clean cyclean testclean $(EXTRA_PACKAGES_CLEAN)
$(RM) -r build lib bin *~ */*~ */*/*~ */*/*/*~
Add scalaclean to make clean (#14322)
2019-03-04 13:26:21 -08:00
(cd scala-package && mvn clean) || true
change recursive make commands per gnumake manual recommendation (#4983)
2017-02-11 23:12:15 -08:00
cd $(DMLC_CORE); $(MAKE) clean; cd -
cd $(PS_PATH); $(MAKE) clean; cd -
cd $(NNVM_PATH); $(MAKE) clean; cd -
cuda support for linalg-functions, restructuring of linalg interfaces (#7147) * cuda support for linalg-functions, restructuring of linalg interfaces * incorporate newest mshadow * adjustments to linalg operators
2017-08-12 21:12:44 +02:00
cd $(AMALGAMATION_PATH); $(MAKE) clean; cd -
ndarray type support.
2015-12-28 23:13:08 -08:00
endif
simplify makefile, automatically generate dependency
2015-09-07 18:12:23 -04:00
let `make clean` do clean_all to avoid frequently asked questions such as #1104 #1011 #958 #920 #721 #719
2015-12-29 22:16:18 -05:00
clean_all: clean
simplify makefile, automatically generate dependency
2015-09-07 18:12:23 -04:00
-include build/*.d
-include build/*/*.d
more torch ops
2016-01-09 02:11:39 -08:00
-include build/*/*/*.d
[WIP] Element wise op refactor (#3245) * [OPERATOR] Refactor Unary Ops * [OPERATOR] Refactor Binary Scalar Ops * Use alias
2016-09-10 23:41:04 -05:00
-include build/*/*/*/*.d
extra operator makefile fix
2015-12-23 15:07:58 -08:00
ifneq ($(EXTRA_OPERATORS),)
add ssd as example (#3674)
2016-11-07 15:10:50 -06:00
-include $(patsubst %, %/*.d, $(EXTRA_OPERATORS)) $(patsubst %, %/*/*.d, $(EXTRA_OPERATORS))
extra operator makefile fix
2015-12-23 15:07:58 -08:00
endif
Reference in New Issue Copy Permalink