- 05 Sep, 2019 1 commit
-
-
Amy Zhuang authored
* Use mkl-dnn v1.0 or v0.x depending on compilation flag. * Change cpu builder files. * Modify cmake files. Use mkldnn-v1.0 for DEX if NGRAPH_USE_MKLDNN_V1 is set to true, otherwise use mkldnn-v0.x. CODEGEN only builds with mkldnn-v1.0. * Implement mkldnn utility functions for mkldnn-v1.0. User mode scratchpad management for mkldnn-v1.0. * Query scratchpad size and allocate a buffer of max scratchpad size. * Do not create mkldnn::memory when query scratchpad size of Reorder. Modify mkldnn utility functions. Fix convolution_forward_init and inner_product_forward_init. Modify CPURuntimeContextCG. * Add user mode scratchpad to CODEGEN. * mkldnn-v1.0 splits LSTM states. Update Rnn/Lstm Op accordingly. * Address PR feedback: use MKLDNN_MAJOR_VERSION. * Modify cpu rnn fusion pass and related unit tests. * Change Rnn/Lstm arg types to Output. * Fix Lstm for CODEGEN. * Set native layout for Slice when input format is blocked. * Do not print scratchpad size. * Change external_mkldnn_v1.cmake. Fix a typo. * Add mkldnn_v1.patch for mkldnn-v1.0. * Address PR feedback. * Define MKLDNN_ERROR_MESSAGE. * Address PR feedback: change to NGRAPH_USE_LEGACY_MKLDNN. * Fix a bug. * Remove unused variable. * Fix compiler warnings. * Fix a bug for CODEGEN. * Move variable only needed for mkldnn-v0.20 inside #if. * Remove unused variables. * No in place Reshape rotation for blocked data layout with mkldnn-v1.0. * Modify mkldnn_v1.patch to force mkldnn to link to libiomp. * Fix style. * Change path for find_library and find_file. * Do not insert ConvertLayout before/after Quantize/DeQuantize for blocked data layout. * Write strides information to visualized graph. * Move variables only needed for mkldnn-v0 under #if. * Move more variables in rnn fusion. * Fix ConvertLayout constant folding for mkldnn-v1.0.
-
- 21 Aug, 2019 1 commit
-
-
Sang Ik Lee authored
* mkl-dnn: patch update 0.20.2 * mkl-dnn: Update version patch. * mkl-dnn: full version naming has changed.
-
- 19 Jul, 2019 1 commit
-
-
Sang Ik Lee authored
-
- 01 Jul, 2019 1 commit
-
-
Sang Ik Lee authored
-
- 18 Jun, 2019 2 commits
-
-
- 17 May, 2019 1 commit
-
-
Sang Ik Lee authored
-
- 02 May, 2019 1 commit
-
-
Sang Ik Lee authored
mklml on linux has no soname set. This will cause linking against the full path for mklml when using prebuilt mkldnn and mklml. (#2851)
-
- 25 Apr, 2019 1 commit
-
-
Sang Ik Lee authored
Workaround for gcc 4.8.2: Suppress mkl-dnn warning manually since pragma GCC diagnostic ignored is not working on gcc 4.8.2 (#2812)
-
- 20 Apr, 2019 1 commit
-
-
Sang Ik Lee authored
* CMAKE_CFG_INTDIR does not work at install time. Use CMAKE_INSTALL_CONFIG_NAME on Windows. * Rename NGRAPH_*_INSTALLSRC_DIRECTORY to NGRAPH_*_INSTALL_SRC_DIRECTORY
-
- 15 Mar, 2019 1 commit
-
-
Sang Ik Lee authored
* Copy and install all mkldnn libraries if versioning is enabled on linux. * Replace magic numbers with variables. * Support mkl-dnn library versioning on macos.
-
- 14 Mar, 2019 1 commit
-
-
Sang Ik Lee authored
* Set rpath for mkl-dnn. * Set library rpath to on Linux. * Use patchelf to set rpath of prebuilt mklml library. * Add patchelf to Linux Dockerfiles. * Revert "Add patchelf to Linux Dockerfiles." This reverts commit 1769505a866061552942e19467ddcc0dad0922e8. * Revert "Use patchelf to set rpath of prebuilt mklml library." This reverts commit 726f6553a0450520328607177d64baf48fa93dd2. * Copy cldnn runtime. * Copy mlsl libraries.
-
- 09 Mar, 2019 1 commit
-
-
Sang Ik Lee authored
* mkl-dnn prebuilt: Use IMPORTED_LINK_INTERFACE_LIBRARIES to ask to link mklml and omp runtime. * mkl-dnn prebuilt: fix IMPORTED libraries related issues. * Setup target libmkl for prebuilt mkl-dnn.
-
- 06 Mar, 2019 1 commit
-
-
Scott Cyphers authored
* Fix bug introduced by #2238 on r0.15 (#2515) * Fix bug introduced by #2238 * style * Avijit/add diag 4 distributed debugging (#2523) * Fix bug introduced by #2238 * Added a debug logging macro: NGRAPH_DIST_DEBUG which prints the timestamp and MPI rank for distributed AllReduceops * Fixed code formatting. * Moved the logging implementation to log.cpp * Fixed clang warning * Don't use namespace in header * Fix ifdef * disable warning * Fix files that had missing headers (#2539) (#2541) * Update mkl-dnn to v0.18. (#2543) * Update mkl-dnn to v0.18. * Update mkl-dnn patch file. * Fixed Centos build issues (gcc 4.8.5 doesn't implement std::put_time - hence the issue). Also fixed a logical error in the distributed setup for unit tests (#2545) * Migrate Klocwork issues (#2557) * Migrate Klocwork fixes * change to kick off another build * Fix crash by correcting index (#2561) * Update build script to fix Windows build issues and add some work tow… (#2531) * Update build script to fix Windows build issues and add some work towards making codegen work on Windows. * Use "git apply" instead of "patch". (#2532) Enable using pre-installed MKLDNN on Window. * Remove commented out lines.
-
- 27 Feb, 2019 2 commits
-
-
Sang Ik Lee authored
-
Scott Cyphers authored
* Add info about lib versions in an easy to find place * Review comments
-
- 26 Feb, 2019 1 commit
-
-
Pruthvi authored
* - MKLDNN would choose the algorithm which will potentially give best performance based on - convolution dimensions number of logical processors available. - (For auto-dispatching to work as intended, - use the same thread affinity settings when creating the convolution as when executing the convolution.) - The relationship between convolution sizes and the best performing algorithm is empirically based on performance observations * bump mkldnn version to V0.18-rc * Revert "- MKLDNN would choose the algorithm which will potentially give best performance based on" This reverts commit 904beb8ad8d4e829fbae5f38a803ea80a72b3ffd. * Update mkl-dnn patch for soversion removal.
-
- 25 Feb, 2019 1 commit
-
-
Sang Ik Lee authored
Update TBB build script for Windows. Fix typo. Fix incorrect omp lib name on Windows. Fix incorrect tbb.dll path on Windows. Make LIBRARY and ARCHIVE output directory consistent. Function missing on Windows. Update test::util::all_close() to fix compilation issue on Windows Export CPU_Executable on Windows. Change nbench path for unit-test on Windows. Change copy to copy_if_different. Install CPU backend on Windows. Disable tools test on Windows. Disable two failing unit test on Windows CPU. Fix incorrect CPU backend install path on Windows.
-
- 22 Feb, 2019 1 commit
-
-
Sang Ik Lee authored
-
- 19 Feb, 2019 1 commit
-
-
Sang Ik Lee authored
* Use git shallow clone if CMake >= 3.6. Remove BUILD_BYPRODUCTS. * Fix ONNX, protobuf integration.
-
- 15 Feb, 2019 1 commit
-
-
Sang Ik Lee authored
* Use uniform naming for external project prefixes and use INSTALL_DIR instead of SOURCE_DIR for referencing prebuilt external libraries. * TBB: Use single variable to share path across files.
-
- 12 Feb, 2019 1 commit
-
-
Sang Ik Lee authored
-
- 04 Feb, 2019 1 commit
-
-
Robert Kimball authored
* fix windows build * wip * mkldnn seems to build * address various errors building cpu backend with MSVC * wip * wip * Windows support. * Delete dependency of LLVM when building with MSVC. * Define EIGEN_HAS_CONSTEXPR when using MSVS. * Fix MSVC build errors. * Incorrect argument to 'decltype'. It is VC bug. Work around the error with rename the function into different name. * MINMAX issue in matmul_bias.cpp. * Correct TBB_LINK_LIBS on Windows. * Fix MSVC link errors. 1. redefine problems in cpu_builder.obj and convert_layout.obj. It is because cpu_builder.hpp contains an implicit implement of function runtime::cpu::Builder::build for cpu::op::ConvertLayout. The fix is deleting the registration item in cpu_builder.cpp and using REGISTER_CPU_OP_BUILDER in convert_layout.cpp. 2. Fix the dependent libraries path on Windows. It should be *.lib not *.dll when linking these libraries. * Set visibility for CPU backend to fix the MSVC linker error. MSVC complain that the .def file exceed the size limitatoin when using CMAKE_WINDOWS_EXPORT_ALL_SYMBOLS. All the functions with CPU_BACKEND_API are used by unit test or nbench. * Fix unit test build errors on Windows. * backend_unary_elementwise.in.cpp: Use all_close_f to test case BACKEDND sqrt * cpu_fustion.cpp: Fix 'NUM_STEPS' cannot be implicitly captured because no default capture mode has been specified * cpu_test.cpp: Use portable setenv and unsetenv from misc.hpp. * tools.cpp: Use portable fpopen from misc.hpp. * misc.hpp/misc.cpp: Add new files to host misc functions that Linux and Windows using different implementation. * Make Debug mode work with MSVC. * style * fix line ending
-
- 01 Feb, 2019 1 commit
-
-
Pruthvi authored
-
- 17 Jan, 2019 1 commit
-
-
Robert Kimball authored
* fix windows build * wip * mkldnn seems to build * address various errors building cpu backend with MSVC * wip * wip
-
- 08 Jan, 2019 1 commit
-
-
Sang Ik Lee authored
* Allow external mklml outside of prebuilt mkldnn install directory. * Limit prebuilt mkl-dnn support to Linux.
-
- 03 Jan, 2019 1 commit
-
-
Robert Kimball authored
* update licenses for 2019 * style
-
- 28 Dec, 2018 1 commit
-
-
Sang Ik Lee authored
* Forward nGraph's C/C++ compiler, build type and generator information to cmake based external projects. * Fix typo. * Pass generator related info properly. * Googletest is using DEBUG_POSTFIX * TBB uses postfix for debug lib. * Fix typo.
-
- 23 Dec, 2018 1 commit
-
-
Robert Kimball authored
* update build byproducts to support ninja * remove unused cmake code * more cmake cleanup * display error message if Ninja generator is requested * fix mkldnn ext project * revert onnx cmake file * revert protobuf cmake file * revert mlsl cmake file * more fixing
-
- 05 Dec, 2018 1 commit
-
-
Amy Zhuang authored
-
- 28 Nov, 2018 1 commit
-
-
Christian Convey authored
-
- 27 Nov, 2018 1 commit
-
-
Robert Kimball authored
-
- 21 Nov, 2018 1 commit
-
-
gaurides authored
* Update MKLDNN v0.16 -> v0.17 * Updated SHA1 for mac and win * Updated from 0.17-rc -> 0.17 * Patch to treat OMP as warning instead of error * Fixed compile error for clang * Commbined 2 patches into 1 file; renamed file to .patch * Corrected file permissions
-
- 15 Sep, 2018 1 commit
-
-
Jayaram Bobba authored
* Move to MKLDNN v0.16. Git hash picks up some bug fixes post v0.16 * Temporary workaround for an MKLDNN bug
-
- 23 Aug, 2018 1 commit
-
-
Nick Korovaiko authored
* revert MKLDNN ver to 0.14 * reinstate dep
-
- 22 Aug, 2018 1 commit
-
-
Robert Kimball authored
* add EXCLUDE_FROM_ALL to codegen library so we don't need to using conditionals in build. If codegen is needed it will be built. * update llvm dependencies
-
- 21 Aug, 2018 1 commit
-
-
Pruthvi authored
* - update MKLDNN version from v0.14 -> v0.15 * - added support to query tensor size from mkldnn desc for MKLDNN Op's - moved tensor size calculation to tensor_view_layout - populate mkldnn layout desc witk MKLDNN queried size in the cpu_layout pass * style fix * - Fixed failing test cases (cherry picked from commit d68d3738673c2081f458f9aa458c16361cc47657) * style fix * renamed tensor->logical_size() member function to tensor->size() * Addressed PR comments * - update mkldnn from v0.15 to v0.16 - update mkl SHA1 hash for lnx, max, win * style fix * Addressed Scott's PR comments * merge branch master with pruthvi/mkldnn_v0.15
-
- 26 Jun, 2018 1 commit
-
-
Robert Kimball authored
* cmake runs for interpreter * more updates towards building on windows
-
- 25 Jun, 2018 1 commit
-
-
Robert Kimball authored
* remove reference to ngraph core code from codegen. add stand-alone implementations of needed funcions * fixed potential pointer leak * clean up file_util * more file util cleanup, removing unused functions * interpreter works on mac * CPU and INTERPRETER build and pass unmit tests on macos * move get_directory to file_util * cleanup
-
- 19 Jun, 2018 1 commit
-
-
Robert Kimball authored
* fix mkldnn rpath * fix compile warning * close backends when exiting * set backend output directory of backends to the ngraph output directory * Aprocter/patch patch (#1119) * Move more rpath stuff inside if(NOT APPLE) * fix repatch problem with mkldnn library * add updated patch command for older versions of cmake
-