diff options
| author | Michalis Spyrou <michalis.spyrou@arm.com> | 2017-11-21 17:52:12 +0000 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-11-02 16:41:04 +0000 |
| commit | f3dfa279d536906dac3e618244b2c1d33e5ff28a (patch) | |
| tree | 6fdf1bf52ad5ce8fc33e18d5a011633c592b7958 /tests/validation/CPP/GEMMLowp.h | |
| parent | f202e50a8b89f143f74c393e33e0154817bd3c1d (diff) | |
| download | ComputeLibrary-f3dfa279d536906dac3e618244b2c1d33e5ff28a.tar.gz | |
COMPMID-632 Assembly: Integrate gemmlowp assembly version
Integrate generic gemmlowp assembly version for u8.
Change-Id: I17ed4494c25a132b2bac581febe1544e49b4f352
Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/110114
Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com <bsgcomp@arm.com>
Reviewed-by: Pablo Tello <pablo.tello@arm.com>
Diffstat (limited to 'tests/validation/CPP/GEMMLowp.h')
| -rw-r--r-- | tests/validation/CPP/GEMMLowp.h | 11 |
1 files changed, 7 insertions, 4 deletions
diff --git a/tests/validation/CPP/GEMMLowp.h b/tests/validation/CPP/GEMMLowp.h index ee33d8e0c0..6c72b56e7a 100644 --- a/tests/validation/CPP/GEMMLowp.h +++ b/tests/validation/CPP/GEMMLowp.h @@ -35,13 +35,16 @@ namespace validation { namespace reference { -SimpleTensor<int32_t> gemmlowp(const SimpleTensor<int8_t> &a, const SimpleTensor<int8_t> &b); - 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); +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, int32_t min = 0, int32_t max = 0); +template <typename T1, typename T2> +SimpleTensor<T1> gemmlowp_matrix_multiply_core(const SimpleTensor<T2> &a, const SimpleTensor<T2> &b, int32_t a_offset, int32_t b_offset); template <typename T> -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, int32_t min = 0, int32_t max = 0); +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); + +template <typename T1, typename T2> +SimpleTensor<T1> gemmlowp(const SimpleTensor<T2> &a, const SimpleTensor<T2> &b); template <typename T> SimpleTensor<uint8_t> gemmlowp_quantize_down_int32_to_uint8_scale(const SimpleTensor<T> &in, const SimpleTensor<T> &bias, int32_t result_offset, int32_t result_mult_int, int32_t result_shift, |