Under certain circumstances, we guarantee that a prepared model can be
executed successfully. In describing those circumstances, we
neglected to specify that operation input operands must have legal
values for the guarantee to hold. For example, the guarantee doesn't
hold if an ADD operation has an activation input that is not one of
the defined values; or if a RESHAPE operation has a shape input in
which two or more components are -1.
This change modifies the guarantee to apply only when operation input
operands have legal values. It also documents this guarantee for
burst execution.
Note that if an operation has an input operand that can be proven to
have an illegal value at preparation time (e.g., a constant value that
is illegal), model preparation might (but is not required to) fail for
that reason.
Bug: 135933040
Test: $ cd neuralnetworks ; mma
Change-Id: I8b421550dd89e4bbbdae899e7cb5e9e88a46d2fb
(cherry picked from commit 48544cc38a)
Bug: 121347610 document that NNAPI Execution inputs/outputs and HAL Request inputs/outputs must not be modified
Test: cd hardware/interfaces/neuralnetworks/1.0/vts/functional ; mma
Test: cd hardware/interfaces/neuralnetworks/1.2/vts/functional ; mma
Change-Id: Iac71d6d5ad92a90afd1b6babb7cfa128d7484c64
Fix Model::pools documentation.
Specify that an output cannot overlap with an output/input/constant.
Document circumstances under which a prepared model should be expected to execute correctly.
Bug: 87805381
Bug: 79888238
Bug: 77237715
Test: mma at tree root
Change-Id: Ibf3daacad79f1d7e6127acf3887bd98b7c6bdf4e
Merged-In: Ibf3daacad79f1d7e6127acf3887bd98b7c6bdf4e
IEvent was a synchronization primitive which caused some confusion
in the interface. Originally the event object was paired with an
asynchronous task, and the asynchronous task would signal this event
when the corresponding output was ready to be used.
In the case of IDevice::prepareModel, the function call would return an
IPreparedModel object that was not guaranteed to be prepared until the
runtime had returned from waiting on the corresponding event object.
The event object has been changed to two explicit callbacks--
IPreparedModelCallback and IExecutionCallback. Now,
IDevice::prepareModel no longer returns an unfinished IPreparedModel;
instead, it will pass the IPreparedModel object to the runtime through
IPreparedModelCallback::notify. When the runtime retreives the
IPreparedModel object, the asynchronous task has already finished
preparing the model.
The two callbacks are used for different purposes. Each has its own
version of notify to pass the data back to the runtime:
* IPreparedModelCallback::notify(ErrorStatus, IPreparedModel)
* IExecutionCallback::notify(ErrorStatus)
Bug: 63905942
Test: mm, vts, ml/nn/runtime/tests
Change-Id: I0c88cd262ba762e0af15e9da31ebe813a5d150b2
The NNAPI requires requests on a model to be asynchronously
processed. This CL implements a basic Event that can later
be used to block the runtime thread until the asynchronous
request has completed.
Bug: 63905942
Test: VtsHalNeuralnetworksV1_0TargetTest (32-bit, 64-bit) with sample driver enabled by cherry-pick
frameworks/ml/nn/runtime/test with and without sample driver enabled
Change-Id: Ie27a574aaaac312e7cbb731750f9c06278357a1c
The neuralnetworks hal was placed in
frameworks/ml/nn/hardware/interfaces while VTS tests were being
developed.
This CL moves it to hardware/interfaces and another CL removes
it from frameworks/ml/nn/hardware/interfaces.
VTS tests included in a sibling CL in this topic.
Bug: 63905942
Test: mm -j40
Change-Id: I0e6c84de72a763edbaef3e9b3063c3ecca216337
