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

Commit 9237ae88 authored by Miao Wang's avatar Miao Wang Committed by Michael Butler
Browse files

Fix the NNAPI HAL documentation about ADD and MUL 

  - ADD and MUL supports QUANT8_ASYMM since OMR1. It was a bug missing
  them in the HAL documentation.
  - Added the updated hash to current.txt for this ABI preserving
  change.

Bug: 75459529
Test: mm
Merged-In: I492a7431c0dbb3dc5967c586d080eb134e380bf2
Change-Id: I492a7431c0dbb3dc5967c586d080eb134e380bf2
(cherry picked from commit f6298402)
parent 7ed61354

current.txt

+1 −1
Original line number Diff line number Diff line
@@ -248,5 +248,5 @@ c8bc853546dd55584611def2a9fa1d99f657e3366c976d2f60fe6b8aa6d2cb87 android.hardwar 


# Future changes to HALs

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

088b30a9c9ce27bc955b08a03c38c208f8f65b51133053c7656c875479801b99 android.hardware.neuralnetworks@1.0::types

08ae9fc24f21f809e9b8501dfbc803662fcd6a8d8e1fb71d9dd7c0c4c6f5d556 android.hardware.neuralnetworks@1.0::types

neuralnetworks/1.0/types.hal

+2 −0
Original line number Diff line number Diff line
@@ -84,6 +84,7 @@ enum OperationType : int32_t { 
     *     output.dimension = {5, 4, 3, 2}

     *

     * Supported tensor types: {@link OperandType::TENSOR_FLOAT32}

     *                         {@link OperandType::TENSOR_QUANT8_ASYMM}

     * Supported tensor rank: up to 4

     *

     * Inputs:
@@ -645,6 +646,7 @@ enum OperationType : int32_t { 
     * input operands. It starts with the trailing dimensions, and works its way forward.

     *

     * Supported tensor types: {@link OperandType::TENSOR_FLOAT32}

     *                         {@link OperandType::TENSOR_QUANT8_ASYMM}

     * Supported tensor rank: up to 4

     *

     * Inputs: