* cherry picked

* "cherry picked platform"

* "add comment"

* "fix ci"

* - Rebase of conv reuse

- clag formatter fixes

- Fix to conv reuse

- Yet another fix

- Fix

- Fix

- clagn format

* - comment update

* move to platform

* "move init from framework to platform"

* "remove used init"

* "fix ci"

* "fix ci"

* "fix generic"

* "fix ci"

* "fix ci"

* "fix ci"

* "disable fragile test"

* Make paddle no longer depend on boost

* Update enforce.h

- 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

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

* 1. Create buddy allocator in each places before NcclBcast the variables
2. Check the memory usage of ALL gpus rather than the first one

* 1. Make NCCLGroupGuard guards only the ncclBcast part, which avoid ncclGroupEnd blocking the exception throwing
2. NOTE the usage of NCCLGroupGuard

* Remove the memory usage check of gpus

* Fix code style

In cupti samples, only cuptiFlush is used.
I can't find any places calling cuptiFinalize and
this API can error out as not_implemented in some
cuda installation.