aboutsummaryrefslogtreecommitdiff
path: root/src/core/NEON/kernels
AgeCommit message (Collapse)Author
2020-11-17COMPMID-3970: Failure when building with GCC < 6Georgios Pinitas
Address pre-N4387 tuple usage Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: Iefe6e08e27b8fe1e688d2ff9db8cb7e172b568f3 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4429 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-11-13COMPMID-3851: Fix regression on NEDepthwiseConvolutionLayerNativeKernelSang-Hoon Park
The exit condition of some for loops in quantized version of the kernel with depth_multiplier=1 is decided during compilation to fix performance issue. Change-Id: I849b3d63b2a2cf5eb374ae681898ae1c296fb4fe Signed-off-by: Sang-Hoon Park <sang-hoon.park@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4392 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-11-13COMPMID-3852: Fix NEReduction windowGeorgios Pinitas
ReductionOperations splits the kernel for scheduling on the X dimension when reduction axis is > 0. By setting the execution window to be unit one in the X dimension the execution was always restricted to a single thread. Alters the window to enable multi-threading Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: Idcbe2b78957678310bb8e895969f01de972d3667 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4389 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
2020-11-12COMPMID-3960: Mismatch on NEArithmeticSubtractionGeorgios Pinitas
Corner-case failure when both input shapes had unit shape on the X axis. Broadcasting was enabled leading to invalid window execution. Check is updated to cross-validate the presence of broadcasting by checking the X dimension in both input shapes. Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: I0b79542279e8d155d2661fddff9691d94a1f6855 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4391 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-11-12COMPMID-3776: Indirect GEMMGeorgios Pinitas
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: I51a1b0f098bc3a8c408c50c92221e4df3061e12c Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4343 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Sang-Hoon Park <sang-hoon.park@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-11-10COMPMID-3958: Fix build error with Werror=1Sang-Hoon Park
Remove unused variable in anonymouse namespace. Change-Id: Id9775cd7982f2a2ebf68f20e0c4e33013c3382a0 Signed-off-by: Sang-Hoon Park <sang-hoon.park@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4361 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-11-09COMPMID-3852: Fix complex multiplication remove padding performance regressionSheri Zhang
Signed-off-by: Sheri Zhang <sheri.zhang@arm.com> Change-Id: I2605baba63c9cca0370328860313b8ec09e04fb6 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4355 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-11-09COMPMID-2808: Add support for QASYMM8_SIGNED in NEROIAlignLayerSheri Zhang
Signed-off-by: Sheri Zhang <sheri.zhang@arm.com> Change-Id: Id4f4c96e1823a4b27886fee9baf70847172e619c Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4335 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-11-09COMPMID-3951 LargeGraph_FLOAT32_Rank4_25 CTS failures in Android Q in CL Fix1SiCong Li
* Fix CLSpaceToBatchLayerKernel and NESpaceToBatchLayerKernel validation errors by using the correctly calculated output tensor shape Signed-off-by: SiCong Li <sicong.li@arm.com> Change-Id: I21d61f870e6a23a2e38dcb95c939b0bf08082b6f Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4347 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-by: TeresaARM <teresa.charlinreyes@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-11-06COMPMID-3850: NEPooling regression for NHWCGeorgios Pinitas
Expand left-over loop to handle multiples of 8 for quantized data type during MaxPooling. Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: I1304d174c45d2c98247470ac8b4bb6752bbc03a6 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4339 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-11-03COMPMID-3638: Move NEON kernelsMichalis Spyrou
Signed-off-by: Michalis Spyrou <michalis.spyrou@arm.com> Change-Id: Ieed3e4bc8be7fef80c90c5094599b477a56fc473 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4285 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-10-30COMPMID-3926: Floor CTS failing in NeonMichele Di Giorgio
Depending n the value of `len`, the left-over loop might end up writing/reading out-of-bounds, therefore corrupting the memory. Change-Id: I1b0bb300f3e5ea668b585266e1aa6af7f93a5d1e Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4290 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Manuel Bottini <manuel.bottini@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-10-29COMPMID-3853: Decouple NEActivationLayerMichalis Spyrou
Decouple datatypes and remove Activation template. Binary size dropped by 25Kb. Signed-off-by: Michalis Spyrou <michalis.spyrou@arm.com> Change-Id: I32c207db124895fee25b56437f9495403315b867 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4217 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-10-29COMPMID-3827: Resize CTS failing in Neon after removing paddingManuel Bottini
Change-Id: I7cf27272e4e6e82b36a31a80ed47ae38fbbf9129 Signed-off-by: Manuel Bottini <manuel.bottini@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4269 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-10-20COMPMID-3637: Move utility headers from arm_compute to srcSang-Hoon Park
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: If9d6fa8c900b68c4b6fd373f2fc1f9abb83ea917 Signed-off-by: Michalis Spyrou <michalis.spyrou@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4145 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Sang-Hoon Park <sang-hoon.park@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-10-19COMPMID-3163: Remove padding from NEDepthwiseConvolutionLayerNativeKernelSang-Hoon Park
Change-Id: Ibbd6bee5c6a4ce4f212b207d17a65b9c33bcfa78 Signed-off-by: Sang-Hoon Park <sang-hoon.park@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4106 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-10-16COMPMID-3805: Fix SQRT non-zero output for zero inputSang-Hoon Park
- For AArch64, NEActivationLayerKernel uses vsqrt rather than vinvsqrt. - For non-AArch64, it masks values to ensure zero input results in zero output without producing NaN. - Test cases for FP16 and FP32's positive boundary values are added. Change-Id: Ic0104ee5d7045059c2e9bd052616a4a3b43a315d Signed-off-by: Sang-Hoon Park <sang-hoon.park@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4150 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
2020-10-14COMPMID-3172: Remove padding from NEGEMMMatrixMultiplyKernelMichele Di Giorgio
Template parameter has been removed, which reduces the binary size by: - ~4 kB for armv8.2a - ~12 kB for armv8a Change-Id: Ib499a18a4980a3ee7b201507b943f900adf20a73 Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4122 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Giorgio Arena <giorgio.arena@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-10-14COMPMID-3144: Remove padding from NEDirectConvolutionLayerKernelManuel Bottini
Change-Id: I22b907eebfbe037e6e1c7bf604172f4709a9cbed Signed-off-by: Manuel Bottini <manuel.bottini@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4082 Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>
2020-10-13COMPMID-3828: fix unsigned overflow in quantized reduce meanSang-Hoon Park
Change-Id: I9d3122b4858137d422548d1d417eb04a27ae9c7b Signed-off-by: Sang-Hoon Park <sang-hoon.park@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4143 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-by: TeresaARM <teresa.charlinreyes@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-10-09COMPMID-3794: Fix window loops causing performance regressionMichalis Spyrou
Signed-off-by: Michalis Spyrou <michalis.spyrou@arm.com> Change-Id: Id4d95c6ce5fed91bb079b8bfe1abceedefd20c97 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4117 Reviewed-by: Sheri Zhang <sheri.zhang@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-10-08COMPMID-3170: Remove padding in NEGEMMLowpMatrixMultiplyKernelmorgolock
Change-Id: Ie95442c6c6a145c1a45937b03cbd433bf08e36ab Signed-off-by: morgolock <pablo.tello@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4094 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-10-08COMPMID-3684: Use case data type decouplingGeorgios Pinitas
Decouples data types for NEFloorKernel Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: I6756300540bc5ef32a9990246eed8619a76855f2 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4084 Reviewed-by: Giorgio Arena <giorgio.arena@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-10-07COMPMID-3821: NEON Reduction op PROD failuresMichalis Spyrou
Signed-off-by: Michalis Spyrou <michalis.spyrou@arm.com> Change-Id: I8cfdd24c4e71a6a4be610ba67a75ad2943a43801 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4097 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-10-07COMPMID-3637: Move wrapper to srcGeorgios Pinitas
Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: I524b0c4b49c7a7035b7d078b9585d77b0d438e10 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4083 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-10-06COMPMID-3181: Remove padding from NEReductionOperationKernelSheri Zhang
COMPMID-3803: Remove padding from NEComplexPixelWiseMultiplicationKernel Signed-off-by: Sheri Zhang <sheri.zhang@arm.com> Change-Id: I309fc4ab62bacbca9203d2680a9d6d52f76f70e6 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4078 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Manuel Bottini <manuel.bottini@arm.com> Reviewed-by: Pablo Marquez <pablo.tello@arm.com>
2020-10-02COMPMID-3145: Remove padding from NEScaleKernelManuel Bottini
Change-Id: I530b12c6270d7dbeb3ef7af62484842ebcb65925 Signed-off-by: Manuel Bottini <manuel.bottini@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4000 Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
2020-10-02COMPMID-3183: Removed padding NEGEMMLowpReductionKernelmorgolock
Change-Id: Ibf7741ffdefcceb9683c919e79302fc35c36ea65 Signed-off-by: morgolock <pablo.tello@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4031 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
2020-09-30COMPMID-3802: Remove templates from NEDirectConvolutionLayerOutputStageKernelMichalis Spyrou
Removing bool template reduces the binary size by 20Kb. Signed-off-by: Michalis Spyrou <michalis.spyrou@arm.com> Change-Id: I652cea7d320a00b6c6e44cdacb61e77f3c10e56a Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4053 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-09-29COMPMID-3784 Add broadcast support to S32 NEPixelwiseMultiplicationSiCong Li
Signed-off-by: SiCong Li <sicong.li@arm.com> Change-Id: Ifae31c74eb46c561225394a387fc15332423bfa9 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4030 Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
2020-09-29COMPMID-3174: Remove padding from NEDirectConvolutionLayerOutputStageKernelMichalis Spyrou
Signed-off-by: Michalis Spyrou <michalis.spyrou@arm.com> Change-Id: I3c5cfe50e9cee30b66f4094da105d383c077aaf9 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4044 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-09-22COMPMID-3757: (u)int8: Don't select the 16-bit route on A53 for cases with ↵David Mansell
very few rows Also added 2D version of the 16-bit route, and altered the selection heuristic so that 2D mode will be used in cases where 1D mode won't thread well. Change-Id: I0057fde08456771dc0090ac51f50d82f8bb86044 Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3903 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-09-15COMPMID-3752: NEPermuteKernel does not support permutations2Michele Di Giorgio
Solves also: - COMPMID-3766: CTS Failures in Transpose Neon + FP16 Change-Id: I9d323f45f49cc0bce9e6329790bcf2f0eeec8572 Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3949 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Manuel Bottini <manuel.bottini@arm.com> Reviewed-by: Pablo Marquez <pablo.tello@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-09-10COMPMID-3159: Remove padding from NEPoolingLayerKernelMichalis Spyrou
Signed-off-by: Michalis Spyrou <michalis.spyrou@arm.com> Change-Id: Ib5b252e1b65794a8f360276d03ff94922e1991f8 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3946 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-09-10COMPMID-3583: Add S32 support to NEElementwiseDivisionGeorgios Pinitas
Division follows the flooring division approach where for example 5/2=2 while -5/2=-3 Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: I65756e0b31fe8d97f743a4c13dc5f96304722f75 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3929 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-09-09MLCE-229: Fixed requantization per channel in asm kernelmorgolock
Change-Id: Iaf1465f3144371e153ce123ac00da5cc092f77df Signed-off-by: morgolock <pablo.tello@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3939 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-09-09COMPMID-3581 Add S32 support to NEPixelWiseMultiplicationSiCong Li
* Add S32 support to NEPixelWiseMultiplication and NEPixelWiseMultiplicationKernel * Scale == 1/255 is not supported for S32, as on non-aarch64 the precision requirement is not met, and scale is a non-standard parameter anyway. * Fix the data types validation logics to also test for all invalid data type combinations. * Add validation tests for S32 NEON PixelWiseMultiplication * The wrap tolerance for ScaleOther (scale == 1/2^n) cases is set to 1 instead of 0 because the reference uses floating point division followed by rounding, which is isn't bit accurate. Change-Id: I28839afda7a4f98c985d1763620e08d98f740142 Signed-off-by: SiCong Li <sicong.li@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3923 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-09-08COMPMID-3151: Remove NEDepthwiseConvolutionLayer3x3KernelGeorgios Pinitas
Prefer NEDepthwiseConvolutionLayerNativeKernel as it has a native format of NHWC avoiding extra transformation to the NCHW domain. Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: If5d8de11691b8ef7f4c3816941f87417d0c8646b Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3930 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-09-07COMPMID-3155: Remove padding from NEGEMMLowpOffsetContributionKernelMichele Di Giorgio
Change-Id: I93c3b795cf6fe0b27008543b6671a3be0a965603 Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3916 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-09-07COMPMID-3748: Compiler issue with Bfloat16 on gcc8Georgios Pinitas
Treat bf16 memory on memset as raw memory by casting to void*. This hides the class-memaccess warning and is safe for the current class layout of arm_compute::bfloat16 Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: I5e242827d3737b4491d29abe7570eefee5b6edc1 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3928 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-09-07COMPMID-3580 Add S32 support to NEArithmeticSubtractionSiCong Li
* Fix convert policy validate logics and add missing validate test * Add S32 support to NEArithmeticSubtraction and NEArithmeticSubtractionKernel * Add S32 validation tests Change-Id: I1b6cb15b024613c202fe9f17747a83da43a5ddcf Signed-off-by: SiCong Li <sicong.li@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3908 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
2020-09-04COMPMID-3157: Remove padding from NEGEMMTranspose1xWKernelGian Marco Iodice
- Remove padding from NEGEMMTranspose1xWKernel - Extend test for validating zero padding requirement Change-Id: I9ce4ca95a500229b045dc140cfff21fdf7373700 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3920 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-09-03COMPMID-3143: Remove padding from NEGEMMInterleave4x4KernelGian Marco Iodice
- Remove padding from NEGEMMInterleave4x4Kernel - Extend test for validating zero padding requirement Change-Id: I94abc271e005f9dd6e1721b185631f55f598dbfd Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3915 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-08-28COMPMID-3504: Add support for BOOL in NEON comparison operatorsMichele Di Giorgio
Change-Id: I81b0c2482bc20b1ab5124ed6179bb94cbced7875 Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3869 Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-08-19MLCE-229: Support for negative shifts in asm kernelsmorgolock
Change-Id: I2c5e98aae7698963f106d7423df0e65cd00ee2a9 Signed-off-by: morgolock <pablo.tello@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3710 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Sheri Zhang <sheri.zhang@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-08-10COMPMID-3700: ArmNN OOB test failureSheri Zhang
Fix ArmNN compiling issue through removing defulat template value for offset_no_padding() and pooling2_nchw_maxpool_indices(). Signed-off-by: Sheri Zhang <sheri.zhang@arm.com> Change-Id: Ie7114d102b8533e757b8e841afcac1adfbcb3b54 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3697 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Pablo Marquez <pablo.tello@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-08-05COMPMID-3392: Collapse TensorMaps into a single TensorPackGeorgios Pinitas
Collapse InputTensorMap and OutputTensorMap to a single TensorPack mechanism. Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: Ie2fdfc6b07d84ad589169ec99ca64fcf45a00bec Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/253783 Tested-by: bsgcomp <bsgcomp@arm.com> Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com> Signed-off-by: Michalis Spyrou <michalis.spyrou@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3641 Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Reviewed-by: Sheri Zhang <sheri.zhang@arm.com>
2020-07-31COMPMID-3624: CTS failure on Resize quantized in Neon and CLMichele Di Giorgio
Allow computations with aligned corners when the tensors have width/height equal to 1. Change-Id: Ia01733f6c02e0740835b26a794b9a79fa35319b4 Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3634 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Sadik Armagan <sadik.armagan@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-07-28COMPMID-3575: Mixed preicision in NEInstanceNormalizationLayerKernelSang-Hoon Park
In order to fix the issue caused by the limited precision of FP16. mixed precision (float accumulator) is introduced to NEInstanceNormalizationLayerKernel. Since the reference kernel is doing the mixed precision, currently mixed preicision computation is default when it is called from NEInstanceNormalizationLayer. - Make NEInstanceNormalizationLayerKernel use kernel descriptor to enable mixed precision computation - NEInstanceNormalizationLayer is modified to use the descriptor Change-Id: I7766622d715df054e303f9b441380b15b51f02b2 Signed-off-by: Sang-Hoon Park <sang-hoon.park@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3589 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-07-24[ONCPUML-120]: Tweak of the launch heuristics for hybrid_u8u32_dot_16x4 kernelAleksandr Nikolaev
Hybrid kernel turns to be faster for qasymm8 than quantized_wrapper with interleaved. Signed-off-by: Aleksandr Nikolaev <aleksandr.nikolaev@arm.com> Change-Id: I200646aee6cdcabfe125b746c7d87bfa7d06e0fc Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3585 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
generated by cgit v1.2.1 (git 2.23.0) at 2024-06-12 18:09:29 +0000