platform/upstream/armnn.git
3 years agoIVGCVSW-5406 Update README.md with details of third party tools.
Colm Donelan [Wed, 14 Oct 2020 16:54:39 +0000 (17:54 +0100)]
IVGCVSW-5406 Update README.md with details of third party tools.

* Add Description, version and provenience to third party tools table.
* Update arm-security@arm.com email address.

Signed-off-by: Colm Donelan <Colm.Donelan@arm.com>
Change-Id: Id48f36ae2319688fafea9fc4757bc75b724b3d02

3 years agoIVGCVSW-5280 Switch tests/InferenceTest and derived tests over to cxxopts
James Ward [Mon, 12 Oct 2020 13:17:36 +0000 (14:17 +0100)]
IVGCVSW-5280 Switch tests/InferenceTest and derived tests over to cxxopts

* refactor AddCommandLineOptions() functions to allow checking of required options
* add CxxoptsUtils.hpp file for convenience functions
!referencetests:268500

Signed-off-by: James Ward <james.ward@arm.com>
Change-Id: Ica954b210b2981b7cd10995f0d75fcb2a2f7b443

3 years agoIVGCVSW-5335 Added Documentation for fast_math
Mike Kelly [Wed, 14 Oct 2020 10:48:21 +0000 (11:48 +0100)]
IVGCVSW-5335 Added Documentation for fast_math

 * Added Documentation for fast_math to CLBackendModelContext
 * Added Documentation for fast_math to NeonBackendModelContext

Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I43a0568ae6914e074a80130a051e5d9bb849f2ba

3 years agoIVGCVSW-5434 Remove boost/preprocessor.hpp
Jim Flynn [Tue, 13 Oct 2020 13:40:29 +0000 (14:40 +0100)]
IVGCVSW-5434 Remove boost/preprocessor.hpp

Change-Id: I51462d18ce0be4b88a23453cfdd16510f30dd1e3
Signed-off-by: Jim Flynn <jim.flynn@arm.com>
3 years agoIVGCVSW-4489 Remove remaining occurrence of boost::format
Matthew Sloyan [Mon, 12 Oct 2020 14:03:01 +0000 (15:03 +0100)]
IVGCVSW-4489 Remove remaining occurrence of boost::format

 * Replaced with fmt::format in Descriptors.cpp.
 * Removed remaining boost/format headers in ArmNN codebase.
 * Removed additional boost header in Network.cpp

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ib98b83bf4ec99ef98ce7a3635ec0dd478c3e43e1

3 years agoBugfix: Correctly parse the armnn_includes by separating with space
Francis Murtagh [Mon, 12 Oct 2020 15:45:49 +0000 (16:45 +0100)]
Bugfix: Correctly parse the armnn_includes by separating with space

 * Stops generate_wrap reading as -I/usr/local/include-I/usr/include
 * Allows generation using system headers

Signed-off-by: Francis Murtagh <francis.murtagh@arm.com>
Change-Id: I501c473b0f624f80ebccfade7b009a74bdcb2d0c

3 years agoIVGCVSW-5335 Documentation for fast_math
Mike Kelly [Thu, 8 Oct 2020 11:19:01 +0000 (12:19 +0100)]
IVGCVSW-5335 Documentation for fast_math

 * Changed documentation for fast_math to add warning about possibly reduction in precision.

Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: If954471efc6aef702e8490585571815d9e19b3fc

3 years agoLoad dynamic backends for YoloV3
Derek Lamberti [Thu, 17 Sep 2020 12:58:18 +0000 (13:58 +0100)]
Load dynamic backends for YoloV3

* Optional cmd option to dump optimized model to dot
* Optional cmd option to specify dynamic backends path
* input is now optional and must exist if given
* comparison files now optional and must exist if given

Change-Id: I1499c9eb715be3cacdba2c227e1a93dd997f355d
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com>
3 years agoIVGCVSW-5287 Switch tests/MultipleNetworksCifar10 over to cxxopts
James Ward [Mon, 5 Oct 2020 16:11:23 +0000 (17:11 +0100)]
IVGCVSW-5287 Switch tests/MultipleNetworksCifar10 over to cxxopts

Signed-off-by: James Ward <james.ward@arm.com>
Change-Id: Ia4ff69f2c8dd538ad7845053dc7a690f434c381c

3 years agoIVGCVSW-5279 Switch armnnQuantizer over to cxxopts
Matthew Sloyan [Thu, 8 Oct 2020 15:41:47 +0000 (16:41 +0100)]
IVGCVSW-5279 Switch armnnQuantizer over to cxxopts

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I79f12ba33b3ca58cdc4be531ffbf72fa20690792

3 years agoIVGCVSW-5291 Fix for Yolov3 producing 0s on Neon
David Monahan [Fri, 9 Oct 2020 11:30:04 +0000 (12:30 +0100)]
IVGCVSW-5291 Fix for Yolov3 producing 0s on Neon

Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: I9331d590e71fc479979c11be9dc750d6435f12bc

3 years agoIVGCVSW-5173 'Upgrade from v1.15 of tensorflow to v2.3'
Sadik Armagan [Fri, 9 Oct 2020 12:39:17 +0000 (13:39 +0100)]
IVGCVSW-5173 'Upgrade from v1.15 of tensorflow to v2.3'

* Fixed the tensorflow build issue.

Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I608c55fdf76ff4d436434c0a50b5368923d2ddb4

3 years agoIVGCVSW-5282 Switch tests/TfLiteMobilenetQuantized-Armnn over to cxxopts
Matthew Sloyan [Mon, 28 Sep 2020 11:58:14 +0000 (12:58 +0100)]
IVGCVSW-5282 Switch tests/TfLiteMobilenetQuantized-Armnn over to cxxopts

 * Required update to cxxopts v3.0 for unrecognised options to work.

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: If1803731474e42580d663c802a1f3ff240fadffe

3 years agoIVGCVSW-5363 Add Unmap layer and Unmap workload
Jim Flynn [Thu, 8 Oct 2020 10:42:30 +0000 (11:42 +0100)]
IVGCVSW-5363 Add Unmap layer and Unmap workload

Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Signed-off-by: Jim Flynn <jim.flynn@arm.com>
Change-Id: Ie5ecfa67e4763d0c058905592fe2e2fd7315f85c

3 years agoIVGCVSW-5286 Switch tests/ModelAccuracyTool-Armnn over to cxxopts
Matthew Sloyan [Tue, 6 Oct 2020 09:03:21 +0000 (10:03 +0100)]
IVGCVSW-5286 Switch tests/ModelAccuracyTool-Armnn over to cxxopts

 * Fixed two code errors which were causing compiler issues.
 * Added support for multiple input and output tensors.

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I298e4bf52a0b41349adab60295e0f22da3cc057b

3 years agoRemove Resize from list of layers that need padding in Neon
Teresa Charlin [Tue, 6 Oct 2020 11:11:39 +0000 (12:11 +0100)]
Remove Resize from list of layers that need padding in Neon

Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I054f0b71d4e9581c637fa09e40f6b661e58e39f3

3 years agoIVGCVSW-5405 Remove boost::make_iterator_range and boost::to_upper_copy
Matthew Sloyan [Tue, 6 Oct 2020 09:45:32 +0000 (10:45 +0100)]
IVGCVSW-5405 Remove boost::make_iterator_range and boost::to_upper_copy

 * Removed from ModelAccuracyTool-Armnn and ImageCSVFileGenerator
 * Fixed formatting in ImageCSVFileGenerator

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I09dfca27813582cc48f46d0507680368ed823a9c

3 years agoIVGCVSW-5362 Add Map layer and Map workload
Jim Flynn [Tue, 6 Oct 2020 09:14:50 +0000 (10:14 +0100)]
IVGCVSW-5362 Add Map layer and Map workload

Signed-off-by: Jim Flynn <jim.flynn@arm.com>
Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: Id2227c58809b84c7a7af61f7c0d88ad7d45ce558

3 years agoIVGCVSW-4488 Update cxxopts to version 3.0
Matthew Sloyan [Tue, 6 Oct 2020 15:06:07 +0000 (16:06 +0100)]
IVGCVSW-4488 Update cxxopts to version 3.0

 * Required to fix issue in TfLiteMobilenetQuantized-Armnn.cpp

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I3a465f62e3d656c9626113da6223e4fa26b535a6

3 years agoIVGCVSW-5213 Switch armnnConverter over to cxxopts
Colm Donelan [Tue, 6 Oct 2020 14:15:33 +0000 (15:15 +0100)]
IVGCVSW-5213 Switch armnnConverter over to cxxopts

* Swap out boost::program_options  with cxxopts.
* Remove const from argv declaration to conform to cxxopts.
* Clean up CLI handling around multiple inputs and shapes.
  Each input must be specified with its own -i param.
  Each corresponding -s parameter must be specified too.

Signed-off-by: Colm Donelan <Colm.Donelan@arm.com>
Change-Id: I01c2683dbca51b2322c98913731ac82a59e400a5

3 years agoIVGCVSW-5173 'Upgrade from v1.15 of tensorflow to v2.3'
Sadik Armagan [Thu, 1 Oct 2020 16:12:17 +0000 (17:12 +0100)]
IVGCVSW-5173 'Upgrade from v1.15 of tensorflow to v2.3'

* Updated the generate_tensorflow_protobuf script to search for .proto
  files in tensorflow directory instead of getting from a file
* Updated BuildGuideAndroidNDK.md to clone Tensorflow 2.3.1

Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I8d2b0f0a04327f930784648e9e8f7f25895d53d1

3 years agoIVGCVSW-5407 Failing Unittest
Mike Kelly [Mon, 5 Oct 2020 11:12:49 +0000 (12:12 +0100)]
IVGCVSW-5407 Failing Unittest

 * The number of error and warning messages will vary depending on the device but will always be greater than 0.

Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: Iee1d12e60d8c88c9116083d7274cea2ec23cb032

3 years agoUpdate ACL pin to fc2f6d0427e1d886fcccc68867d1af1ccd96608b
Teresa Charlin [Mon, 5 Oct 2020 08:53:18 +0000 (09:53 +0100)]
Update ACL pin to fc2f6d0427e1d886fcccc68867d1af1ccd96608b

* Set use_padding to false in neon workload

Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: Ia5367de331efe1d28dea4dfbefc0713720da81f9

3 years agoPyArmNN Updates
Éanna Ó Catháin [Thu, 10 Sep 2020 12:02:37 +0000 (13:02 +0100)]
PyArmNN Updates

* Updated setup.py to raise error on mandatory ext
* Updated examples section of main readme
* Added readme to img class
* Moved img class to new subdir

Change-Id: Iea5f6d87c97e571b8ca5636268231506538840c7
Signed-off-by: Éanna Ó Catháin <eanna.ocathain@arm.com>
Signed-off-by: Jakub Sujak <jakub.sujak@arm.com>
3 years agoIVGCVSW-5297 Remove boost::format from rest of ArmNN.
Colm Donelan [Wed, 9 Sep 2020 11:48:16 +0000 (12:48 +0100)]
IVGCVSW-5297 Remove boost::format from rest of ArmNN.

* Replacing calls to boost:format with fmt:format.
* TensorUtils.cpp added outputShape.reserve call.

Signed-off-by: Colm Donelan <Colm.Donelan@arm.com>
Change-Id: I4b2ed0f72039df824a2adca9309b8a9bbb158c5b

3 years agoIVGCVSW-5334 Remove remaining boost::numeric_cast from armnn
Matthew Sloyan [Wed, 23 Sep 2020 15:57:23 +0000 (16:57 +0100)]
IVGCVSW-5334 Remove remaining boost::numeric_cast from armnn

 * Floating point casts now use armnn::numeric_cast.
 * Also removed remaining header imports.

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I2d37847d67f164fc0a0ae17f34d49ff3d2210c30

3 years agoUpdate ACL pin to b84f9d34dbb857ad6113c0c89ad109498fa75fe5
Nikhil Raj [Fri, 2 Oct 2020 10:11:43 +0000 (11:11 +0100)]
Update ACL pin to b84f9d34dbb857ad6113c0c89ad109498fa75fe5

Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I0b8f698f20889eb09b7c881d7daadbd05f8db3c7

3 years agoIVGCVSW-4997 'Superfluous memcopy workloads'
Sadik Armagan [Tue, 29 Sep 2020 14:12:36 +0000 (15:12 +0100)]
IVGCVSW-4997 'Superfluous memcopy workloads'

* If Output Layer is already connected to MemCopy Layer do not insert
  CopyMemGenericWorkload.

Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I9f813be5a3de2bc62d16864edb3eeaf371ef48e0

3 years agoIVGCVSW-5306 Add floating point type checks to NumericCast.hpp
Matthew Sloyan [Wed, 23 Sep 2020 08:51:04 +0000 (09:51 +0100)]
IVGCVSW-5306 Add floating point type checks to NumericCast.hpp

 * Added Unit Tests to capture all combinations.

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I04db920a5f5f485dc00b2f16582cf7e0bbef3ef2

3 years agoIVGCVSW-5294 Remove boost::format armnn backends
James Ward [Thu, 10 Sep 2020 10:57:28 +0000 (11:57 +0100)]
IVGCVSW-5294 Remove boost::format armnn backends

 * replaced with fmt::format
 * one case required std:stringstream instead

Signed-off-by: James Ward <james.ward@arm.com>
Change-Id: Ife7c4cf5f143e43373f42edf6124158af132abc5

3 years agoIVGCVSW-5296 Remove boost::format armnn parsers
James Ward [Fri, 11 Sep 2020 16:32:44 +0000 (17:32 +0100)]
IVGCVSW-5296 Remove boost::format armnn parsers

 * replaced with fmt::format
 * one case required std::stringstream instead

Signed-off-by: James Ward <james.ward@arm.com>
Change-Id: Ica9a7eb4e7bed04aa03172058dd9e3d10efc8548

3 years agoIVGCVSW-5281 Switch tests/ImageCSVFileGenerator over to cxxopts
Matthew Sloyan [Fri, 25 Sep 2020 10:43:32 +0000 (11:43 +0100)]
IVGCVSW-5281 Switch tests/ImageCSVFileGenerator over to cxxopts

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I1b2ba4fcaebbafb70693b83b40b79db0071b4522

3 years agoUpdate ACL pin to de2e747de0a63933d0cfcb75739d9460ea7c2ff5
Nikhil Raj [Thu, 1 Oct 2020 09:55:05 +0000 (10:55 +0100)]
Update ACL pin to de2e747de0a63933d0cfcb75739d9460ea7c2ff5

Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Ie2371f743e2ae87617693052d3887a4e6732accc

3 years agoIVGCVSW-5285 Switch tests/ImageTensorGenerator over to cxxopts
Matthew Sloyan [Fri, 25 Sep 2020 16:22:00 +0000 (17:22 +0100)]
IVGCVSW-5285 Switch tests/ImageTensorGenerator over to cxxopts

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I9d43e347021634022a875fc4526baa438c467df0

3 years agoIVGCVSW-5283 Switch tests/profiling/gatordmock over to cxxopts
Matthew Sloyan [Wed, 30 Sep 2020 12:07:51 +0000 (13:07 +0100)]
IVGCVSW-5283 Switch tests/profiling/gatordmock over to cxxopts

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I4f06a414d6bd5f18c2ced882ac518052ed371cfc

3 years agoInclude layer GUID in SerializeToDot output
Rob Hughes [Fri, 28 Aug 2020 10:48:35 +0000 (11:48 +0100)]
Include layer GUID in SerializeToDot output

Change-Id: I1a6df60683cc51fcd9739b6dc98f1e722becf045
Signed-off-by: Robert Hughes <robert.hughes@arm.com>
3 years agoCOMPMID-3784 Fix 1 CTS MUL INT32 failure due to using SATURATE
Teresa Charlin [Tue, 29 Sep 2020 07:38:47 +0000 (08:38 +0100)]
COMPMID-3784 Fix 1 CTS MUL INT32 failure due to using SATURATE

* LargeGraph_TENSOR_INT32_Rank4/26

Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I9d07444db56e26c13a77bf022938644ed7953d6b

3 years agoIVGCVSW-5325 Fix non-channel per axis quantization
Finn Williams [Tue, 29 Sep 2020 18:54:00 +0000 (19:54 +0100)]
IVGCVSW-5325 Fix non-channel per axis quantization

Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: Ie0cf69b2cd76d6ecedab43d3d9ae267d23bbc052

3 years agoIVGCVSW-4519 Remove Boost Variant and apply_visitor variant
James Ward [Mon, 28 Sep 2020 10:56:35 +0000 (11:56 +0100)]
IVGCVSW-4519 Remove Boost Variant and apply_visitor variant

  * replace boost::variant with mapbox::util::variant
  * replace boost::apply_visitor with mapbox::util::apply_visitor
  * replace boost::get with mapbox::util::get

Signed-off-by: James Ward <james.ward@arm.com>
Change-Id: I38460cabbcd5e56d4d61151bfe3dcb5681ce696e

3 years agoIVGCVSW-5295 Remove boost::format from armnn profiling
Jan Eilers [Wed, 9 Sep 2020 18:11:16 +0000 (19:11 +0100)]
IVGCVSW-5295 Remove boost::format from armnn profiling

 * Replace all instances of boost::format with fmt::format from armnn/src/profiling

Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: I4722bdde52c740bc9bdce969128074a34a3ee75e

3 years agoRefactored Optimize(...) function to throw exceptions instead of returning null
Mike Kelly [Tue, 29 Sep 2020 19:50:35 +0000 (20:50 +0100)]
Refactored Optimize(...) function to throw exceptions instead of returning null

 * INetwork::Optimize(...) states that the function should throw an exception
   if it fails but the implementation in Network.cpp returned null in some
   scenarios instead. This has led to some confusion amongst users.

Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: I358d1293232c9464772aa0e39ab3355e3570c823

3 years agoUpdate ACL pin to d6d1b3682a2cdd54bae5498635b108a4b19a045a
Teresa Charlin [Tue, 29 Sep 2020 12:27:07 +0000 (13:27 +0100)]
Update ACL pin to d6d1b3682a2cdd54bae5498635b108a4b19a045a

Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I86f5848e895536334e3c6353e721ced871a192e0

3 years agoQuantization copy constructor
Derek Lamberti [Mon, 28 Sep 2020 15:11:50 +0000 (16:11 +0100)]
Quantization copy constructor

* Fix compiler implicit copy deprecation warning.
* Simplify copy operator by removing unnecessary equality comparison.
    Now just does pointer comparison instead.

Change-Id: I9ebe170c637c636919b9d8729a78449148b7f83e
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com>
3 years agoIVGCVSW-4519 Remove Boost Variant and apply_visitor variant
James Ward [Fri, 25 Sep 2020 10:43:21 +0000 (11:43 +0100)]
IVGCVSW-4519 Remove Boost Variant and apply_visitor variant

  * add mapbox/variant third party package

Signed-off-by: James Ward <james.ward@arm.com>
Change-Id: I181302780edd9dace40f158a11327316a12ce69a

3 years agoIVGCVSW-5295 Change fmt to be a header-only interface library
Jan Eilers [Fri, 25 Sep 2020 07:36:44 +0000 (08:36 +0100)]
IVGCVSW-5295 Change fmt to be a header-only interface library

 * Fix non-virtual-dtor warnings in fmt
 * Fix wrong fmt include in TfParser
 * Make fmt work in nn-driver
 * Make fmt a header-only interface library
 * Link fmt where necessary

Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: I9db7cd9a133a81262cbf79f04fb419ab97b88ea8

3 years agoUpdate ACL pin to d175ecea20f5c9e3bff07f7c5d3e4bb5519f7cb9
Nikhil Raj [Tue, 29 Sep 2020 13:34:34 +0000 (14:34 +0100)]
Update ACL pin to d175ecea20f5c9e3bff07f7c5d3e4bb5519f7cb9

Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Ib0b28b3840eb8c8d38ad32e1a572cd9a75ff8a82

3 years agoIVGCVSW-5301 Remove all boost::numeric_cast from armnn/src/profiling
Matthew Sloyan [Fri, 11 Sep 2020 09:14:57 +0000 (10:14 +0100)]
IVGCVSW-5301 Remove all boost::numeric_cast from armnn/src/profiling

 * Replaced with armnn/utility/NumericCast.hpp

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I14e89c00cd1dd607315113d5b8ca56c2e9f6c363

3 years agoIVGCVSW-5326 PyArmnn has hard dependencies on all parsers when using cmake
Francis Murtagh [Mon, 21 Sep 2020 14:27:00 +0000 (15:27 +0100)]
IVGCVSW-5326 PyArmnn has hard dependencies on all parsers when using cmake

Signed-off-by: Francis Murtagh <francis.murtagh@arm.com>
Change-Id: Ieffc1f3db548bc01ea0127a0d60659f0b0783ce3

3 years agoIVGCVSW-5325 Speed up the reference backend
Finn Williams [Thu, 17 Sep 2020 14:58:31 +0000 (15:58 +0100)]
IVGCVSW-5325 Speed up the reference backend

Change-Id: Id8bd0a0418be31d975b944b54bbacb25051ffb2e
Signed-off-by: Finn Williams <Finn.Williams@arm.com>
3 years agoIVGCVSW-4973 Enable QLstm projection unit tests on CL
Teresa Charlin [Fri, 25 Sep 2020 14:08:21 +0000 (15:08 +0100)]
IVGCVSW-4973 Enable QLstm projection unit tests on CL

 *Cosmetic changes on ClQLstmWorkload

Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I61f55343263e623aaae042d8dfe8c294540e98f1

3 years agoUpdate ACL pin to 840a72cc745c60eccbd26fe192b035ec68b2ee41
Nikhil Raj [Thu, 24 Sep 2020 16:58:34 +0000 (17:58 +0100)]
Update ACL pin to 840a72cc745c60eccbd26fe192b035ec68b2ee41

* Change tensor to non const to fix build error caused by ACL fix for QLSTM

Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I7ab0f644dfb3cb3cf21bda73028e9368f3354f4a

3 years agoAdd int32 and int64 ArgMax op support
Inki Dae [Thu, 10 Sep 2020 06:33:54 +0000 (15:33 +0900)]
Add int32 and int64 ArgMax op support

This patch adds int32 and int64 ArgMax op support.

Current ARMNN already has ArgMax op but not used, and
it doesn't support int64 output type.

So this patch adds a new type, Signed64, and also adds
ArgMinMax computation function for int64 type support.

In default, output tensor type of ArgMax op is int64 in case of
tensorflow lite model so this patch makes a proper function - ArgMax op
for int64 or int32 - to be called according to parsed output_type value.

With this patch, ARMNN supports both types - int64 and int32 - for
ArgMinMax op.

Changelog v1:
- Check if output data type of ArgMinMax op is valid or not.
- Use template function to support int32 and int64 types of ArgMinMax function.
- Keep using Signed32 as default data type of m_Output_Type.

Change-Id: I7a8e7e38dd9e5acc81464571d8b4d51378fc7f14
Signed-off-by: Inki Dae <inki.dae@samsung.com>
3 years agoUpdate ACL pint to 1643a45557fde79ee209f55c507860307ffe627c
Nikhil Raj [Wed, 23 Sep 2020 11:50:08 +0000 (12:50 +0100)]
Update ACL pint to 1643a45557fde79ee209f55c507860307ffe627c

Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Id8e240a711064c48e4bdb9b1236198647e365a7f

3 years agoIVGCVSW-5318 'Create a Neon/CL Workload Unit Test fast_math option enabled'
Sadik Armagan [Tue, 22 Sep 2020 13:35:19 +0000 (14:35 +0100)]
IVGCVSW-5318 'Create a Neon/CL Workload Unit Test fast_math option enabled'

* Unit test implemented to make sure it returns WINOGRAD
* Updated the enable-fast-math option in ExecuteNetwork to be consistent

Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Id64f114ae47966def69a9eef0770a4251ee56a41

3 years agoUpdate ACL pin to 34654b2d8dcaf268a9d1bf9e0cdb5ba548ced2b7
Nikhil Raj [Tue, 22 Sep 2020 12:07:49 +0000 (13:07 +0100)]
Update ACL pin to 34654b2d8dcaf268a9d1bf9e0cdb5ba548ced2b7

Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I3601db66ea270d6799627ac96ec95a9008d6942f

3 years agoAdded Security Issues section
Mike Kelly [Tue, 15 Sep 2020 12:37:38 +0000 (13:37 +0100)]
Added Security Issues section

 * Added Security Issues section with email address.

Signed-off-by: Mike Kelly <mike.kelly@arm.com>
Change-Id: If1ccfd3c7847cac0a09b6ec72fdfacba0011455e

3 years agoIVGCVSW-5303 Remove some boost::numeric_cast from rest of ArmNN
Matthew Sloyan [Mon, 14 Sep 2020 10:47:26 +0000 (11:47 +0100)]
IVGCVSW-5303 Remove some boost::numeric_cast from rest of ArmNN

 * Replaced with armnn/utility/NumericCast.hpp
 * Exclusions in TypeUtils.cpp and QuantizerVisitor.cpp
 * Excluded as requires float implementation in NumericCast.hpp

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I5c4c60e7028e1a51bf9379457278d253fd37bc70

3 years agoIVGCVSW-5300 Remove some boost::numeric_cast from armnn/backends
Matthew Sloyan [Wed, 9 Sep 2020 08:07:37 +0000 (09:07 +0100)]
IVGCVSW-5300 Remove some boost::numeric_cast from armnn/backends

 * Replaced with armnn/utility/NumericCast.hpp
 * Some exclusions in reference backend
 * Excluded as requires float implementation in NumericCast.hpp

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I9e4e9cd502c865452128fa04415fd6f250baa855

3 years agoIVGCVSW-5317 'Add enable_fast_math Option to ExecuteNetwork'
Sadik Armagan [Tue, 15 Sep 2020 16:17:08 +0000 (17:17 +0100)]
IVGCVSW-5317 'Add enable_fast_math Option to ExecuteNetwork'

Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I4eb3e27837aea926593d49f9ccea07bab8388d5b

3 years agoUpdate ACL pin to 16cf3c2318978580006633ea4d5eccdd8ba1194a
Nikhil Raj [Fri, 11 Sep 2020 09:19:42 +0000 (10:19 +0100)]
Update ACL pin to 16cf3c2318978580006633ea4d5eccdd8ba1194a

Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I6f1f66fdd2e49ca36d28fd2367f4b64dc6d38cca

3 years agoIVGCVSW-5305 AddBroadcastReshapeLayer as optimizer
Narumol Prangnawarat [Mon, 14 Sep 2020 15:12:44 +0000 (16:12 +0100)]
IVGCVSW-5305 AddBroadcastReshapeLayer as optimizer

 * Remove AddBroadcastReshapeLayer from TfLiteParser
 * Add AddBroadcastReshapeLayer as optimizer
 * AddBroadcastReshapeLayer optimizer unit tests
 * Load-scope dynamic tensor broadcasting unit tests

Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I3549e85b71b41cbd4d96c0f1ece7887acbca76d1

3 years agoMLECO-929 Add Object Detection sample application using the public ArmNN C++ API
Éanna Ó Catháin [Mon, 14 Sep 2020 16:36:49 +0000 (17:36 +0100)]
MLECO-929 Add Object Detection sample application using the public ArmNN C++ API

Change-Id: I14aa1b4b726212cffbefd6687203f93f936fa872
Signed-off-by: Éanna Ó Catháin <eanna.ocathain@arm.com>
3 years agoIVGCVSW-5302 Remove some boost::numeric_cast from parsers
Matthew Sloyan [Fri, 11 Sep 2020 15:17:48 +0000 (16:17 +0100)]
IVGCVSW-5302 Remove some boost::numeric_cast from parsers

 * Replaced with armnn/utility/NumericCast.hpp
 * Exclusions in armnnCaffeParser
 * Three excluded as requires float implementation in NumericCast.hpp

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ib468b606238694334a8319d0ed5db381ce37a915

3 years agoIVGCVSW-5157 'Pipe ModelOption through Network::LoadNetwork() to Workload factory'
Sadik Armagan [Mon, 14 Sep 2020 14:44:18 +0000 (15:44 +0100)]
IVGCVSW-5157 'Pipe ModelOption through Network::LoadNetwork() to Workload factory'

* Pass ModelOptions to WorkloadFactory
* Updated signature of CL and NEON Convolution2d workloads added FastMathEnabled param.

Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I536178be8e4dd4083489e69febadaf0feeba46d2

3 years agoIVGCVSW-5299 Remove some boost::numeric_cast from armnn/tests
Matthew Sloyan [Tue, 8 Sep 2020 11:00:32 +0000 (12:00 +0100)]
IVGCVSW-5299 Remove some boost::numeric_cast from armnn/tests

 * Replaced with armnn/utility/NumericCast.hpp
 * Removed combinations without float implementation in NumericCast.hpp

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ia4ec605f063cdb0071fff302ef48c610f9f9505e

3 years agoUpdate ACL pin to 5489394cdfe10afb469171cf521f26f923eeb2e2
Nikhil Raj [Thu, 10 Sep 2020 09:21:00 +0000 (10:21 +0100)]
Update ACL pin to 5489394cdfe10afb469171cf521f26f923eeb2e2

Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I05375f653f76084c64df47390acf1ec692cb16b6

3 years agoIVGCVSW-5245 Add Quantization operator=() function
Kevin May [Tue, 8 Sep 2020 14:50:18 +0000 (15:50 +0100)]
IVGCVSW-5245 Add Quantization operator=() function

* Add unit tests to check if Quantization info copied correctly

Signed-off-by: Kevin May <kevin.may@arm.com>
Change-Id: I7bb7bde5d97e82c57252c6d5131fbe21ad3096d2

3 years agoIVGCVSW-5156 Introduce ModelOptions to OptimizedNetwork
Sadik Armagan [Thu, 10 Sep 2020 12:37:32 +0000 (13:37 +0100)]
IVGCVSW-5156 Introduce ModelOptions to OptimizedNetwork

 * Introduced ModelOptions to IBackendInternal
 * Introduced ModelOptions to Network
 * Added FastMathEnabled parameter to Conv2d Validate function in CL and NEON
 * Added Optimizer tests

Signed-off-by: Ryan OShea <Ryan.OShea2@arm.com>
Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Ib54c1e82cb3d89a52756ed499cf91b6a7fdb2063

3 years agoIVGCVSW-5293 Remove boost::format from armnn/tests
James Ward [Mon, 7 Sep 2020 15:45:07 +0000 (16:45 +0100)]
IVGCVSW-5293 Remove boost::format from armnn/tests

 * Replaced boost::format with fmt::format

Signed-off-by: James Ward <james.ward@arm.com>
Change-Id: Icf5a6508e7be3d31bc063643491fc5e0607f21fa

3 years agoIVGCVSW-5197 Add support for 2nd input to ExpandDims of TfParser
Jan Eilers [Tue, 8 Sep 2020 07:57:40 +0000 (08:57 +0100)]
IVGCVSW-5197 Add support for 2nd input to ExpandDims of TfParser

 * ParseExpandDims did not support to pass the axis parameter as
   a second input tensor
 * Added related unit tests

Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: I8217950f0b42beaf5b9eaebdcad04267e4443ba3

3 years agofix cxxopts and ghc cross compilation issue
Vincent ABRIOU [Thu, 3 Sep 2020 11:02:43 +0000 (13:02 +0200)]
fix cxxopts and ghc cross compilation issue

While cross compiling using openembedded yocto environment,
CXXOPTS_INCLUDE and GHC_INCLUDE variables remains empty because the search
path point to the openembedded sysroot instead of the current directory.
To avoid this situation, NO_CMAKE_FIND_ROOT_PATH must be added to the
find_path call.

Signed-off-by: Vincent ABRIOU <vincent.abriou@st.com>
Change-Id: I3ba17a300d5ca7fbc58b6ec5c4eb15267dcee228

3 years agoUpdate ACL pin to 57f30a9309ff2e5e3b32731a785bf38b01d1fd69
Nikhil Raj [Wed, 9 Sep 2020 10:41:19 +0000 (11:41 +0100)]
Update ACL pin to 57f30a9309ff2e5e3b32731a785bf38b01d1fd69

Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: Ie34332830e52e9c5aa35e8641ffdf562f33e6b93

3 years agoIVGCVSW-5244 Load-scope dynamic tensor TfLite tests
Narumol Prangnawarat [Mon, 7 Sep 2020 13:05:22 +0000 (14:05 +0100)]
IVGCVSW-5244 Load-scope dynamic tensor TfLite tests

 * Infer tensor shapes at the beginning of Optimize function
 * Unit tests

Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I511f1228a12ebcad570e42a0c46d461ab9ccdc2c

3 years agoUpdate ACL pin to e8f05da5fb919aa209e1bf0e5c70dd15fff84b7f
Nikhil Raj [Tue, 8 Sep 2020 09:55:34 +0000 (10:55 +0100)]
Update ACL pin to e8f05da5fb919aa209e1bf0e5c70dd15fff84b7f

Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I5f91150aac74547219841ea4ebe453c99760ebfc

3 years agoUpdate ACL pin to ec4dee8c68a3d0f6d63db184bfb2f4589429778e
Teresa Charlin [Fri, 28 Aug 2020 14:13:05 +0000 (15:13 +0100)]
Update ACL pin to ec4dee8c68a3d0f6d63db184bfb2f4589429778e

 * Axis for LogSoftMax and SoftMax can be either positive or negative
Signed-off-by: Teresa Charlin <teresa.charlinreyes@arm.com>
Change-Id: I36b0507ad7600c0a98c3b8be3c0350045ee05b84
Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
3 years agoIVGCVSW-5261 Fix undefined reference to GetIdStatic()
David Monahan [Thu, 3 Sep 2020 08:02:52 +0000 (09:02 +0100)]
IVGCVSW-5261 Fix undefined reference to GetIdStatic()

 * Moved DynamicBackend tests to only build when ArmnnRef is enabled
   due to a dependency on them dynamically loading the ArmnnRef backend
   object

Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: Iee0480e7d0cf505bbb5c26629829d3d20fb60051

3 years agoMLECO-1226: update pyarmnn profiling helper utility
Nina Drozd [Fri, 28 Aug 2020 16:14:49 +0000 (17:14 +0100)]
MLECO-1226: update pyarmnn profiling helper utility

* fix key error in get_profiling_data
* make retrieval of inference measurements more generic

Signed-off-by: Nina Drozd <nina.drozd@arm.com>
Change-Id: I3fc147a6a93830b59e8b12f517be9f9c72370c09

3 years agoIVGCVSW-5256 Use CreateTensorHandle() function from TensorHandleFactory in the tests...
Finn Williams [Fri, 28 Aug 2020 12:17:05 +0000 (13:17 +0100)]
IVGCVSW-5256 Use CreateTensorHandle() function from TensorHandleFactory in the tests for layers  Q,R & T

Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I6fc613d31785298a0b7ed18f1abdd59bafed1e8e

3 years agoIVGCVSW-5231 Remove CreateTensorHandle in the test where there is NO_DEPRECATE_WARN
Keith Davis [Thu, 27 Aug 2020 14:38:12 +0000 (15:38 +0100)]
IVGCVSW-5231 Remove CreateTensorHandle in the test where there is NO_DEPRECATE_WARN

 * Done for all elementwise layers, Activation, BatchNorm, BatchToSpace

Signed-off-by: Keith Davis <keith.davis@arm.com>
Change-Id: Id1d15a0960233026aecf7a07e0d3f006e07e4abf

3 years agoIVGCVSW-5253 Use CreateTensorHandle() function from TensorHandleFactory in the tests...
Finn Williams [Thu, 27 Aug 2020 15:15:20 +0000 (16:15 +0100)]
IVGCVSW-5253 Use CreateTensorHandle() function from TensorHandleFactory in the tests for layers M-P

Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I324eee7d750e30f714e0d346b7da7b69866ff935

3 years agoIVGCVSW-5252 Use CreateTensorHandle() function from TensorHandleFactory in the tests...
Finn Williams [Thu, 27 Aug 2020 10:13:25 +0000 (11:13 +0100)]
IVGCVSW-5252 Use CreateTensorHandle() function from TensorHandleFactory in the tests for layers between G-L

Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I197351a479fb211787bd12a73c9618d2ded95898

3 years agoIVGCVSW-5249 Use CreateTensorHandle from ITensorHandleFactory in the test for all...
Keith Davis [Mon, 31 Aug 2020 07:32:55 +0000 (08:32 +0100)]
IVGCVSW-5249 Use CreateTensorHandle from ITensorHandleFactory in the test for all layers between C-D

Signed-off-by: Keith Davis <keith.davis@arm.com>
Change-Id: I9583adf50e67e63e73833f400d1c50fbff57f60c

3 years agoIVGCVSW-5250 Remove CreateTensorHandle in the test for layers between E-F
Finn Williams [Thu, 27 Aug 2020 09:37:36 +0000 (10:37 +0100)]
IVGCVSW-5250 Remove CreateTensorHandle in the test for layers between E-F

* Refactored Floor and FullyConnected tests

Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: Iad87254e638bdcb5d7b334b16ec87a0c981e48a0

3 years agoIVGCVSW-5257 'Remove CreateTensorHandle in the test for layers beginning with S'
Sadik Armagan [Fri, 28 Aug 2020 11:17:33 +0000 (12:17 +0100)]
IVGCVSW-5257 'Remove CreateTensorHandle in the test for layers beginning with S'

* Re-factored SplaceToDepth, Splitter, Stack and StridedSlice unit tests
  to use TensorHandleFactory for creating TensorHandles

Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Ib22bb09cd2120c02c548099eaa06db6e6f00b15e

3 years agoIVGCVSW-4979 'Remove CreateTensorHandle using WorkloadFactory in workload tests'
Sadik Armagan [Fri, 28 Aug 2020 09:38:53 +0000 (10:38 +0100)]
IVGCVSW-4979 'Remove CreateTensorHandle using WorkloadFactory in workload tests'

* Small refactor in unit tests using TensorHandleFactory to use reference instead of pointer

Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I1a702941890034a45029c014c8b11e185f45a807

3 years agoIVGCVSW-5257 'Remove CreateTensorHandle in the test for layers beginning with S'
Sadik Armagan [Thu, 27 Aug 2020 11:57:20 +0000 (12:57 +0100)]
IVGCVSW-5257 'Remove CreateTensorHandle in the test for layers beginning with S'

* Re-factored SoftmaxTestImpl to use TensorHandleFactory to create TensorHandles

Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I83559a89187bbed0d6f34ca589ea81c694bf5683

3 years agoIVGCVSW-5257 'Remove CreateTensorHandle in the test for layers beginning with S'
Sadik Armagan [Thu, 27 Aug 2020 11:46:36 +0000 (12:46 +0100)]
IVGCVSW-5257 'Remove CreateTensorHandle in the test for layers beginning with S'

* Re-factored SpaceToBatchNd tests to use TensorHandleFactory to create TensorHandles

Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: I096a6e30ecc97dd9b93b206157f16d912085703c

3 years agoExport Armnn cmake
Derek Lamberti [Tue, 25 Aug 2020 12:05:59 +0000 (13:05 +0100)]
Export Armnn cmake

Change-Id: I3c317e835aba5feafda240393ca26b6aebe8d655
Signed-off-by: Derek Lamberti <derek.lamberti@arm.com>
3 years agoIVGCVSW-5251 'Remove CreateTensorHandle for ArgMinMaxTestImpl'
Sadik Armagan [Thu, 27 Aug 2020 08:06:40 +0000 (09:06 +0100)]
IVGCVSW-5251 'Remove CreateTensorHandle for ArgMinMaxTestImpl'

* Refactored ArgMinMax tests to use TensorHandleFactory instead of WorkloadFactory

Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Ibff0f9370972f9a0a977c05275cb6168f8f88ae5

3 years agoIVGCVSW-5250 Remove CreateTensorHandle in the test for layers between E-F
Finn Williams [Wed, 26 Aug 2020 15:19:15 +0000 (16:19 +0100)]
IVGCVSW-5250 Remove CreateTensorHandle in the test for layers between E-F

* Added new test function to pass in the ITensorHandleFactory

Signed-off-by: Finn Williams <Finn.Williams@arm.com>
Change-Id: I9b2e9250200e092541e29796ec53cabd0b677acf

3 years agoIVGCVSW-5109 'Add SupportsInPlaceComputation to TensorHandleFactories'
Sadik Armagan [Tue, 25 Aug 2020 10:48:00 +0000 (11:48 +0100)]
IVGCVSW-5109 'Add SupportsInPlaceComputation to TensorHandleFactories'

* Added functionality to query if TensorHandleFactory supports InPlaceComputation

Signed-off-by: Sadik Armagan <sadik.armagan@arm.com>
Change-Id: Icf5bfc5f999fc5d03681dcb8cec88d921842458b

3 years agoUpdate Doxygen Project Version
Ryan OShea [Mon, 24 Aug 2020 17:03:23 +0000 (18:03 +0100)]
Update Doxygen Project Version

 * Change Project Version to 20.08

Signed-off-by: Ryan OShea <Ryan.OShea2@arm.com>
Change-Id: I02ceca1995bdefb98d73511748fb2142cd27969e

3 years agoMLECO-955: Added python object detection example for PyArmNN
Jakub Sujak [Wed, 17 Jun 2020 14:35:03 +0000 (15:35 +0100)]
MLECO-955: Added python object detection example for PyArmNN

Change-Id: I1344c027f4cc70520b7846b34dfbc2abf399d10a
Signed-off-by: Jakub Sujak <jakub.sujak@arm.com>
3 years agoMinor html changes to BuildGuideCrossCompilation.md to make it look better
Nikhil Raj [Fri, 21 Aug 2020 10:31:30 +0000 (11:31 +0100)]
Minor html changes to BuildGuideCrossCompilation.md to make it look better

Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I4213bf811d39898506c6d976089d90379a617056

3 years agoIVGCVSW-5200 Add import enabled optimizer options to PyArmNN
Narumol Prangnawarat [Fri, 21 Aug 2020 09:03:49 +0000 (10:03 +0100)]
IVGCVSW-5200 Add import enabled optimizer options to PyArmNN

Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: Ic7c288fd829d7f1f1ae2910c47fbccdd635def8f

3 years agoBugfix: Allow permutation of QuantizationDim
Francis Murtagh [Thu, 20 Aug 2020 14:38:29 +0000 (15:38 +0100)]
Bugfix: Allow permutation of QuantizationDim

Signed-off-by: Francis Murtagh <francis.murtagh@arm.com>
Change-Id: Ib98ec471e6fdd47600b7c62d0b4d19dd36e20cbd

3 years agoUpdate to provide resources to PyArmNN examples manually
Pavel Macenauer [Tue, 2 Jun 2020 11:54:59 +0000 (11:54 +0000)]
Update to provide resources to PyArmNN examples manually

Change-Id: I9ee751512abd5d4ec9faca499b5cea7c19028d22
Signed-off-by: Pavel Macenauer <pavel.macenauer@nxp.com>
3 years agoIVGCVSW-5208 Wrap SampleDynamicBackend in own namespace
Jan Eilers [Wed, 19 Aug 2020 13:14:36 +0000 (14:14 +0100)]
IVGCVSW-5208 Wrap SampleDynamicBackend in own namespace

 * to prevent name clashes with armnn

Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: I10b394d47e58cec45bacd489938ee048b7c9776e

3 years agoIVGCVSW-5012 Add importEnabled option for OptimizerOptions
Narumol Prangnawarat [Wed, 19 Aug 2020 13:39:07 +0000 (14:39 +0100)]
IVGCVSW-5012 Add importEnabled option for OptimizerOptions

 * Default importEnabled to false
 * Improve error messages

Signed-off-by: Narumol Prangnawarat <narumol.prangnawarat@arm.com>
Change-Id: I17f78986aa1d23e48b0844297a52029b1a9bbe3e

3 years agoUpdate ACL pin to their 20.08 release branch
Nikhil Raj [Wed, 19 Aug 2020 09:04:35 +0000 (10:04 +0100)]
Update ACL pin to their 20.08 release branch

Signed-off-by: Nikhil Raj <nikhil.raj@arm.com>
Change-Id: I4fab85d4660c6a36dd1f9e2d7a71a4c24e573a2f