Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 2a970cb8 authored by David Gross's avatar David Gross
Browse files

Documentation improvements. 

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
parent 485c5cd1

current.txt

+2 −1
Original line number Diff line number Diff line
@@ -249,7 +249,8 @@ c8bc853546dd55584611def2a9fa1d99f657e3366c976d2f60fe6b8aa6d2cb87 android.hardwar 
# Future changes to HALs

5804ca86611d72e5481f022b3a0c1b334217f2e4988dad25730c42af2d1f4d1c android.hardware.neuralnetworks@1.0::IDevice

12e8dca4ab7d8aadd0ef8f1b438021938e2396139e85db2ed65783b08800aa52 android.hardware.neuralnetworks@1.0::IExecutionCallback

86b77e06da756a76aa3685be88765852dd982a86d8c90b8b4fc1130ed4184c8f android.hardware.neuralnetworks@1.0::types

574e8f1499436fb4075894dcae0b36682427956ecb114f17f1fe22d116a83c6b android.hardware.neuralnetworks@1.0::IPreparedModel

b07802a3433a9abec67a41525dc04eab9efb62760414e1d37e2f72cc5ae9cb8a android.hardware.neuralnetworks@1.0::types



# Documentation fixups for b/78135149

9e7a0b650d0e461ece2cfec0e1072abf8676f592b41a7fb48f01e88fc3c8f780 android.hardware.broadcastradio@1.0::types

neuralnetworks/1.0/IPreparedModel.hal

+6 −0
Original line number Diff line number Diff line
@@ -42,6 +42,12 @@ interface IPreparedModel { 
     * execute function. This callback must be provided with the ErrorStatus of

     * the execution.

     *

     * If the prepared model was prepared from a model wherein all

     * tensor operands have fully specified dimensions, and the inputs

     * to the function are valid, then the execution should launch

     * and complete successfully (ErrorStatus::NONE). There must be

     * no failure unless the device itself is in a bad state.

     *

     * Multiple threads can call the execute function on the same IPreparedModel

     * object concurrently with different requests.

     *

neuralnetworks/1.0/types.hal

+4 −2
Original line number Diff line number Diff line
@@ -1668,8 +1668,7 @@ struct Model { 
    vec<uint8_t> operandValues;



    /**

     * A collection of shared memory pools containing operand data that were

     * registered by the model.

     * A collection of shared memory pools containing operand values.

     *

     * An operand's value must be located here if and only if Operand::lifetime

     * equals OperandLifeTime::CONSTANT_REFERENCE.
@@ -1721,6 +1720,9 @@ struct RequestArgument { 
 * 1) Provides the input and output data to be used when executing the model.

 * 2) Specifies any updates to the input operand metadata that were left

 *    unspecified at model preparation time.

 *

 * An output must not overlap with any other output, with an input, or

 * with an operand of lifetime CONSTANT_REFERENCE.

 */

struct Request {

    /**

neuralnetworks/1.1/types.hal

+1 −2
Original line number Diff line number Diff line
@@ -404,8 +404,7 @@ struct Model { 
    vec<uint8_t> operandValues;



    /**

     * A collection of shared memory pools containing operand data that were

     * registered by the model.

     * A collection of shared memory pools containing operand values.

     *

     * An operand's value must be located here if and only if Operand::lifetime

     * equals OperandLifeTime::CONSTANT_REFERENCE.