COMPMID-675 - Reworked NEGEMMLowp interface/function
The new interface makes NEGEMMLowp able to work with ASYMM8 data types.
Implemented 2 new functions:
- NEGEMMLowpMatrixMultiplyCore
- NEGEMMLowpOutputStage
These functions should make the integration in android NN doable
For more information about GEMMLowp:
https://github.com/google/gemmlowp/blob/master/doc/low-precision.md
Change-Id: Ie2c775f45234f68ca53dba644b3a912b997fd890
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/95504
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
diff --git a/tests/validation/CPP/GEMMLowp.h b/tests/validation/CPP/GEMMLowp.h
index 2f903f2..c09d8f6 100644
--- a/tests/validation/CPP/GEMMLowp.h
+++ b/tests/validation/CPP/GEMMLowp.h
@@ -35,11 +35,11 @@
{
namespace reference
{
-SimpleTensor<int32_t> gemmlowp(const SimpleTensor<int8_t> &a, const SimpleTensor<int8_t> &b, SimpleTensor<int32_t> &c);
+template <typename T>
+SimpleTensor<int32_t> gemmlowp_matrix_multiply_core(const SimpleTensor<T> &a, const SimpleTensor<T> &b, int32_t a_offset, int32_t b_offset);
template <typename T>
-SimpleTensor<T> gemmlowp(const SimpleTensor<T> &a, const SimpleTensor<T> &b, SimpleTensor<T> &c,
- int32_t a_offset, int32_t b_offset, int32_t c_offset, int32_t c_mult_int, int32_t out_shift);
+SimpleTensor<uint8_t> gemmlowp_quantize_down_int32_to_uint8_scale(const SimpleTensor<T> &in, int32_t result_offset, int32_t result_mult_int, int32_t result_shift);
} // namespace reference
} // namespace validation
} // namespace test