- Added hash function inside of MKLDNN softmax op to be used as handle for primitives stroing in a
context

- Style fixes to softmax mkldnn op

- Fixes after review

- Coding style

- Fix to style

- style fixes

- style fix

- style fixes

- Fix to cody style check

- Rephrasing a comment

fix t obroken merge

Fixes to rebase

Conflicts:
	benchmark/fluid/models/machine_translation.py
	cmake/external/mkldnn.cmake
	paddle/fluid/operators/softmax_mkldnn_op.cc

- Bumped revision of MKL-DNN up to have softmax backward primitive

- Added choosing MKLDNN softmax grad operator

- First reuse of softmax backward

- Reinvented reusing for softmax

- Fix to crash in reinvented reuse

- Clang format fixes

- Clang format fixes

- Improved softmax mkldnn reuse mechanism

- clang format fixes

- Fix to broken merge

- Fix

fix mkldnn sum error

This reverts commit 4d8e8ee2, reversing
changes made to d6a9f005.

disable the LODTensor warning for now

Enhance Parallel Executor stable

Revert "MKLDNN layout: Support for sum operator"

* "add some api reference"

* "fix based on preview"

* "fix evaluator"

* "remove template doc"

* "add doc for exec"

* "add more changes"

* "fix based on preview"

* "chagne code format"

small thread-safety fix and doc improvements.

Add a ctest hung tool

move dist codes from  operaotrs/detail to operators/distributed

overlap rpc op memcpy in distributed training

Fix concat compile warning

Add bilinear interp supporting for uint8

Polish Non-Layer API

add url of cuda9.0_cudnn7_avx_mkl library

fix auc

try fix logical op doc

Transpiler doc

Add doc reference for Variable and Parameter