diff options
author | Georgios Pinitas <georgios.pinitas@arm.com> | 2019-12-09 18:35:57 +0000 |
---|---|---|
committer | Georgios Pinitas <georgios.pinitas@arm.com> | 2019-12-11 11:16:42 +0000 |
commit | f2cdce30ca782cac7caebc43c7e67caf677b7358 (patch) | |
tree | ab38fbe929b537a783344d1f5f1803574101c222 /arm_compute/core/NEON/NEMath.inl | |
parent | 1d8977771274b713c1e237e7dcaea7b61a84a86a (diff) | |
download | ComputeLibrary-f2cdce30ca782cac7caebc43c7e67caf677b7358.tar.gz |
COMPMID-2841: Enable aarch32 builds
Decouples the execution state from the architecture.
Now architectures can be set as (armv7a, armv8a, etc) and execution
state using the `estate` flag with the following options (auto, 32, 64).
Change-Id: Ie7f757b3565495a39c7e20fb350a72fd9c5a2a4f
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com>
Reviewed-on: https://review.mlplatform.org/c/2438
Reviewed-by: Pablo Marquez <pablo.tello@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'arm_compute/core/NEON/NEMath.inl')
-rw-r--r-- | arm_compute/core/NEON/NEMath.inl | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/arm_compute/core/NEON/NEMath.inl b/arm_compute/core/NEON/NEMath.inl index a3601f6a25..179f1b6299 100644 --- a/arm_compute/core/NEON/NEMath.inl +++ b/arm_compute/core/NEON/NEMath.inl @@ -434,7 +434,7 @@ inline float16x8_t vexpq_f16(float16x8_t x) const float32x4_t x_high = vcvt_f32_f16(vget_high_f16(x)); const float32x4_t x_low = vcvt_f32_f16(vget_low_f16(x)); - const float16x8_t res = vcvt_high_f16_f32(vcvt_f16_f32(vexpq_f32(x_low)), vexpq_f32(x_high)); + const float16x8_t res = vcombine_f16(vcvt_f16_f32(vexpq_f32(x_low)), vcvt_f16_f32(vexpq_f32(x_high))); return res; } @@ -444,7 +444,7 @@ inline float16x8_t vlogq_f16(float16x8_t x) const float32x4_t x_high = vcvt_f32_f16(vget_high_f16(x)); const float32x4_t x_low = vcvt_f32_f16(vget_low_f16(x)); - const float16x8_t res = vcvt_high_f16_f32(vcvt_f16_f32(vlogq_f32(x_low)), vlogq_f32(x_high)); + const float16x8_t res = vcombine_f16(vcvt_f16_f32(vlogq_f32(x_low)), vcvt_f16_f32(vlogq_f32(x_high))); return res; } |