From 08c5a06e2b49df0d7912deedd6d26d2c603cfe58 Mon Sep 17 00:00:00 2001 From: Georgios Pinitas Date: Thu, 14 Dec 2017 17:53:39 +0000 Subject: COMPMID-750: Fix assembly kernel interfaces Assembly kernel interfaces were wrongly translating the layout of the input matrices. Boolean flags transform0 and transform1 do not match the actual interface of the gemm assembly code which expects transpose0 and transposed1. Change-Id: Ia4df65a533834647fa63e78e8c897924793949df Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/113410 Tested-by: BSG Visual Compute Jenkins server to access repositories on http://mpd-gerrit.cambridge.arm.com Reviewed-by: Pablo Tello --- arm_compute/core/NEON/kernels/arm64/NEGEMMLowpAArch64Kernel.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arm_compute/core/NEON/kernels/arm64/NEGEMMLowpAArch64Kernel.h') diff --git a/arm_compute/core/NEON/kernels/arm64/NEGEMMLowpAArch64Kernel.h b/arm_compute/core/NEON/kernels/arm64/NEGEMMLowpAArch64Kernel.h index a93df033de..3829d5e1d6 100644 --- a/arm_compute/core/NEON/kernels/arm64/NEGEMMLowpAArch64Kernel.h +++ b/arm_compute/core/NEON/kernels/arm64/NEGEMMLowpAArch64Kernel.h @@ -44,10 +44,11 @@ public: void run(const Window &window, const ThreadInfo &info) override; protected: - void internal_configure(const ITensor *input0, const ITensor *input1, ITensor *output, ITensor *workspace, float alpha, float beta, bool transform_0, bool transform_1) override; + void internal_configure(const ITensor *input0, const ITensor *input1, ITensor *output, ITensor *workspace, float alpha, float beta, bool is_transposed_0, bool is_transposed_1) override; private: - using NEGEMMLowpAArch64 = void(const ITensor *input0, const ITensor *input1, ITensor *output, ITensor *workspace, float alpha, float beta, bool transform_0, bool transform_1, const Window &window, + using NEGEMMLowpAArch64 = void(const ITensor *input0, const ITensor *input1, ITensor *output, ITensor *workspace, float alpha, float beta, bool is_transposed_0, + bool is_transposed_1, const Window &window, const ThreadInfo &info); NEGEMMLowpAArch64 *_func; }; -- cgit v1.2.1