diff options
| author | Michele Di Giorgio <michele.digiorgio@arm.com> | 2017-07-11 15:00:52 +0100 |
|---|---|---|
| committer | Anthony Barbier <anthony.barbier@arm.com> | 2018-09-17 14:16:42 +0100 |
| commit | 81f0d15d6840a0ae8ef571114555a26da74c4a43 (patch) | |
| tree | a9eeda0a2b69961cd6a51d74e039bbed26a9b436 /arm_compute/core/NEON/NEFixedPoint.h | |
| parent | f70256bd46f03090281581c152bd17b4a50febcd (diff) | |
| download | ComputeLibrary-81f0d15d6840a0ae8ef571114555a26da74c4a43.tar.gz | |
COMPMID-444: Add support for QS8/QS16 NEON Arithmetic Add/Sub/Mul.
Change-Id: Ia482498688ca1884272b5062e3415e736e03d36f
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/80448
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Diffstat (limited to 'arm_compute/core/NEON/NEFixedPoint.h')
| -rw-r--r-- | arm_compute/core/NEON/NEFixedPoint.h | 8 |
1 files changed, 8 insertions, 0 deletions
diff --git a/arm_compute/core/NEON/NEFixedPoint.h b/arm_compute/core/NEON/NEFixedPoint.h index 09579f9120..50463b5efe 100644 --- a/arm_compute/core/NEON/NEFixedPoint.h +++ b/arm_compute/core/NEON/NEFixedPoint.h @@ -145,6 +145,14 @@ qint8x16_t vld1q_dup_qs8(const qint8_t *addr); */ qint16x8_t vld1q_dup_qs16(const qint16_t *addr); +/** Load two 16 bit fixed point vectors from memory (8x2 elements) + * + * @param[in] addr Memory address of the 16 bit fixed point vectors to load + * + * @return 16 bit fixed point vectors (8x2 elements) + */ +qint16x8x2_t vld2q_qs16(qint16_t *addr); + /** Store a single 8 bit fixed point vector to memory (8 elements) * * @param[in] addr Memory address where the 8 bit fixed point vector should be stored |