MXNET-336 [Perl] Major Gluon update for Perl API.

[sergeykolychev]

Description

Major Gluon update towards parity with Python's API.
Miscellaneous bugfixes and improvements.
New Engine API.
Module::reshape moved to C++ backend.
Examples were updated to work on multi-gpu boxes.

Checklist

Essentials

Please feel free to remove inapplicable items for your PR.

• The PR title starts with [MXNET-$JIRA_ID], where $JIRA_ID refers to the relevant JIRA issue created (except PRs with tiny changes)
• Changes are complete (i.e. I finished coding on this PR)
• All changes have test coverage:
• Unit tests are added for small changes to verify correctness (e.g. adding a new operator)
• Nightly tests are added for complicated/long-running ones (e.g. changing distributed kvstore)
• Build tests will be added for build configuration changes (e.g. adding a new build option with NCCL)
• Code is well-documented:
• For user-facing API changes, API doc string has been updated.
• For new C++ functions in header files, their functionalities and arguments are documented.
• For new examples, README.md is added to explain the what the example does, the source of the dataset, expected performance on test set and reference to the original paper if applicable
• Check the API doc at http://mxnet-ci-doc.s3-accelerate.dualstack.amazonaws.com/PR-$PR_ID/$BUILD_ID/index.html
• To the my best knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change

Changes

Major Gluon update towards parity with Python's API.
Miscellaneous bugfixes and improvements.
New Engine API.
Module::reshape moved to C++ backend.
Examples were updated to work on multi-gpu boxes.

Comments

[sergeykolychev]

@tlby Robert, could you please review this PR. Thanks in advance.

[sergeykolychev]

@marcoabreu Hello there, there's a problem with caffe integration that errors out on every PR recently. Are you aware of this ? Is the fix in works ?

[tlby]

The retooling work here looks great, and the additions look solid.

[tlby]

This use of safemalloc() may be better as Newxc($1, av_len, dim_t). safemalloc() is not documented in perlapi, I think it just happens to be the default in most perl compilations.

[sergeykolychev]

Ok, thanks for this observation, I'll cleanup safemalloc usage up in future, need to verify that Newxc is present in all perls from 5.14 till 5.28

[tlby]

Yeah, it's been around, at least as far back as v5.8.8.