aboutsummaryrefslogtreecommitdiff
path: root/src/cpu/operators
AgeCommit message (Collapse)Author
2023-12-12Winograd changes to enable fp16 in armv8a multi_isa buildsPablo Marquez Tello
* Changes in filelist.json: moved fp16 code from common to fp16 * Replaced the guard __ARM_FEATURE_FP16_VECTOR_ARITHMETIC with ENABLE_FP16_KERNELS. * Resolves COMPMID-6755 Change-Id: I4da1c53d3f9e4734e5e67125265ab4e3fc0dcbe4 Signed-off-by: Pablo Marquez Tello <pablo.tello@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10865 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-12-06Revert "thread_local _custom_scheduler"Pablo Marquez Tello
This reverts commit ded5b182675e3166e947a8eb637b5b1e925816ab. Resolves COMPMID-6735 Signed-off-by: Pablo Marquez Tello <pablo.tello@arm.com> Change-Id: I9b69ca1ec80a671171d3f52081c4b8c61a676617 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10838 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: <felixjohnny.thomasmathibalan@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-12-05Optimize CpuSoftmaxKernel for axis=0Gunes Bayir
Implement a single kernel instead of having two consecutive ones. In the previous setup, one kernel was calculating the maximum value in the axis, and this maximum was being subtracted from each data while calculating the softmax, i.e. softmax(x_i) = exp(x_i - max) / sum_i( exp(x_i - max) ) This patch integrates these two stages into a single kernel for Neon™ for all data types. This will save some memory because we don't need to hold the max values in a separate auxiliary tensor. It also introduces some other optimizations that will ease memory pressure when the data type is float/half, by using the dst tensor as temporary storage for already exponentiated inputs. It removes the references to SVE and SVE2 implementations, and most of the associated files; but, it leaves the implementations as these may be used in the future. Resolves: COMPMID-6500 Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Change-Id: Icff9976d1214c4c6cbe15a62ca60b8a77d3784cc Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10688 Reviewed-by: SiCong Li <sicong.li@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-11-24thread_local _custom_schedulerDavid Svantesson
Resolves ONCPUML-1331 This patch adds an option to make _custom_scheduler thread_local to support usage of multiple schedulers handled outside of ACL. It also adds num_threads() function to Scheduler which reverts to querying CPUInfo if no scheduler has been set. Change-Id: Iff706165d8d091895331a5bb3a76f6cabe048912 Signed-off-by: David Svantesson-Yeung <david.svantesson-yeung@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10748 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-11-15Fix various coverity issuesSiCong Li
Resolves COMPMID-6677 Signed-off-by: SiCong Li <sicong.li@arm.com> Change-Id: I99bf2385f6edc0836faacb31f5c66ed4fb051e40 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10729 Benchmark: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2023-11-10Fix CpuGemmConv2d int8 segfaultSiCong Li
Bypass importation of memory of the original weights into the reinterpreted_weights auxiliary tensor if other weight transformation path is selected (which would've freed the original weights and its tensor info) Resolves COMPMID-6635 Signed-off-by: SiCong Li <sicong.li@arm.com> Change-Id: Ib8a345c3ac542bc3745d6a67db822b55df37e827 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10698 Benchmark: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Anitha Raj <Anitha.Raj@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2023-11-08Optimize CpuGemmConv2d start-up timeSiCong Li
When weight has no holes, we can replace CpuWeightsReshapeKernel with: - Collapse by reinterpreting weight's 3 spatial dimensions - Perform CpuTranspose For more details see the documentation in src/cpu/operators/CpuGemmConv2d.cpp This is one optimization since the CpuTranspose is better performing than CpuWeightsReshapeKernel A second optimization is to fuse this transpose with other weight transformations (e.g. pretranspose_B_array in CpuGemmAssemblyDispatch) However this second optimization depends on how the underlying gemm methods (the fall back path: CpuGemmMatrixMultiplyKernel or the assembly path: CpuGemmAssemblyDispatch) chooses to fuse the transpose. Therefore, this patch moves the transpose down from CpuGemmConv2d, to the individual gemm operators where the fusion decision needs to be made, by passing an extra "transpose_b" flag to CpuGemm New transpose_b flag in different scopes (they are all the same, but with different names because pretranspose_b has a different meaning in GemmAssemblyDispatch): GEMMInfo::pretranspose_B -> AsmGemmInfo::transpose_b New auxilliary tensors holding the transposed b result: - CpuGemm optimized path: CpuGemmAssemblyDispatch::PrePretransposedB - CpuGemm fallback path: CpuGemm::PreTransposedRHS Note that this patch does not yet have the second optimization (COMPMID-6595), but it prepares for it. Relates to COMPMID-6595 Resolves COMPMID-6499 Change-Id: I999a2da9da4b2b15369a3cc06d7872c86e0190ea Signed-off-by: SiCong Li <sicong.li@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10526 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Anitha Raj <Anitha.Raj@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-10-02Optimize CL and Neon Winograd testsGunes Bayir
Several test optimizations have been introduced into Winograd tests for Gpu and Cpu backends. The testing strategy has been detailed as a comment header in the test design files. In summary - Very large shapes in the nightly are made smaller - If the underlying kernel is the same for different data types, we only need to stress some key aspects of the kernels (e.g. read/write lengths in case of fp32/fp16). - In case the underlying kernel is the same (OpenCL), Fp16 is tested on a subset of the shapes - In Cpu, there is no need to test every combination for both NCHW and NHWC as we just permute the inputs and use NHWC kernels anyways - All activations does not need to be tested for each and every shape Resolves: COMPMID-6464 Change-Id: Ie25fded85c65b9c7386dc21b23f9b695b1e77b07 Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10393 Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-09-28Apply clang-format on repositoryFelix Thomasmathibalan
Code is formatted as per a revised clang format configuration file(not part of this delivery). Version 14.0.6 is used. Exclusion List: - files with .cl extension - files that are not strictly C/C++ (e.g. Android.bp, Sconscript ...) And the following directories - compute_kernel_writer/validation/ - tests/ - include/ - src/core/NEON/kernels/convolution/ - src/core/NEON/kernels/arm_gemm/ - src/core/NEON/kernels/arm_conv/ - data/ There will be a follow up for formatting of .cl files and the files under tests/ and compute_kernel_writer/validation/. Signed-off-by: Felix Thomasmathibalan <felixjohnny.thomasmathibalan@arm.com> Change-Id: Ib7eb1fcf4e7537b9feaefcfc15098a804a3fde0a Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10391 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com>
2023-09-20Fix the validation issue in AddMulAdd fused kernelGunes Bayir
Resolves: COMPMID-6558 Change-Id: I015d504aaa9b8a1a232b01e49ab373d415ea1de9 Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10340 Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Reviewed-by: TeresaARM <teresa.charlinreyes@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2023-09-15Remove deprecated support for BF16 in CpuCastAdnan AlSinan
Resolves : [COMPMID-6212] Signed-off-by: Omar Al Khatib <omar.alkhatib@arm.com> Signed-off-by: Adnan AlSinan <adnan.alsinan@arm.com> Change-Id: I29bbd9a3d96af462faf7f0ee13b9849f75e05356 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10319 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
2023-09-15Fix include dependencies for mass reformatting patchGunes Bayir
This patch fixes some include dependencies in certain files that caused build failures in https://review.mlplatform.org/c/ml/ComputeLibrary/+/10287. It also circumvents some clang-format glitches. Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Change-Id: I8e9d3307edd2d1afd17c685c9bc9429624130e5a Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10313 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: <felixjohnny.thomasmathibalan@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-09-04Remove legacy PostOps codeJakub Sujak
PostOps was the experimental interface for Dynamic Fusion. It is now replaced by the new Dynamic Fusion interface with code generation using the Compute Kernel Writer. Resolves: COMPMID-6190 Change-Id: I813b48facef2fd6f3aee332588886b4f9b3d33d8 Signed-off-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10219 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-08-23Update CpuGemmConv2d and CpuFlatten to use CpuReshape operatorAnitha Raj
- Following CpuReshapeKernel Optimizations, update the CpuGemmConv2D and CpuFlatten to use CpuReshape operator instead of CpuReshapeKernel - Minor changes to comment in NEReorgLayerKernel.h Resolves COMPMID-6504 Signed-off-by: Anitha Raj <anitha.raj@arm.com> Change-Id: Ib6ee1fdc313d91249f9fe41c81e73324031c1ff4 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10186 Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-08-22Optimize CpuReshapeKernelAnitha Raj
Resolves COMPMID-5279 Change-Id: Id9b007eed62c200702bbfcc83b94dab7b5de1714 Signed-off-by: Anitha Raj <anitha.raj@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9962 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-08-17Fix various static check issuesViet-Hoa Do
Resolves: COMPMID-6495 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I916829222a6211fa096a833a2afc5fab5eb34ea4 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10143 Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-08-07Document the Conv2D heuristicGian Marco Iodice
- Add a new section in the documentation to describe how the conv2D heuristic works on Arm® Cortex®-based CPUs and Arm® Mali™-based GPUs - Add CKW_UNUSED in compute_kernel_writer/src/cl/CLTile.cpp to avoid the compilation error due to an unused variable - Remove FFT from the list of algorithms to be selected by the CPU Conv2d heuristic. Resolves COMPMID-6163 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Change-Id: I51384d7749451b2562642683e8b2429a355166bb Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10065 Benchmark: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-07-28Retain back-compatibility for arm_compute/core/Types.hSiCong Li
* Some symbols have been moved from core/Types.h. This patch retains back compatibility so that the user can still include this header for those symbols * A new header core/CoreTypes.h is created to avoid circular dependency. This header includes essential small types that are used across functions * Move all function info types into function_info folder for easier tracking Resolves COMPMID-6330 Related to https://review.mlplatform.org/c/ml/ComputeLibrary/+/9757 Signed-off-by: SiCong Li <sicong.li@arm.com> Change-Id: I4739175c2d4d184a9bc8e28b881b497fab03ca60 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9979 Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2023-07-19Add support for input S64/U64 in CpuCastKernelPablo Marquez Tello
* The kernel now supports the following conversions: S64 -> F32 U64 -> F32 * Resolves MLCE-1089 Change-Id: I277cf58b78d919fde25947520d2056e1412c7f82 Signed-off-by: Pablo Marquez Tello <pablo.tello@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9935 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-07-10Do not include headers necessary for logging when logging is disabledMatthew Bentham
Speeds up compilation by 30% for some files when logging is disabled. Signed-off-by: Matthew Bentham <Matthew.Bentham@arm.com> Change-Id: Ia479bd50a80616a34e33ead13db8558f8dbaa1aa Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/534480 Tested-by: bsgcomp <bsgcomp@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Comments-Addressed: bsgcomp <bsgcomp@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9880 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-07-04Depthwise channel pre-multiplicationMichael Tyler
Resolves: COMPMID-6337 Change-Id: Ie9097b3f56e8071426c621386a5988bd7f7e8ef2 Signed-off-by: Michael Tyler <michael.tyler@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9852 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-06-23Address the issues with the ACL coverage pipeline failures related to matmul.Renato Arantes
Signed-off-by: Renato Arantes <renato.arantes@arm.com> Change-Id: I98de659d1289c930e366727d4799f0dacc8121ab Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9782 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-06-16Add Fused Activation to OpenCL MatMulMohammed Suhail Munshi
- Added fused activation to MatMul function interface - Added fused activation to CL backend - Includes tests for supported Activation Functions in MatMul Resolves: [COMPMID-6192] Signed-off-by: Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> Change-Id: Ie103212b600b60699eaf6a6394d609e6e1f5aba6 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/522465 Comments-Addressed: bsgcomp <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Tested-by: bsgcomp <bsgcomp@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9714 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-06-15Break up Utils.h a bit to reduce unused code being included everywhereMatthew Bentham
Move some maths-related things from Utils.h to new Math.h header in utils/math. Move some routines used for Tensor shape validation to Validate.h Change-Id: I8ce89fe03ec3ae1b61d1a80c282b8b91eea0cfb3 Signed-off-by: Matthew Bentham <Matthew.Bentham@arm.com> Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/524783 Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Tested-by: Viet-Hoa Do <viet-hoa.do@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9743 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-06-15Break up arm_compute/core/Types.h a bitMatthew Bentham
Split some of the larger types with inlined code into their own header files, so that the implementation of them needn't be included everywhere. Change-Id: Id3ec2d42efbd33cedb55705a5a24e1b90c8b7a01 Signed-off-by: Matthew Bentham <Matthew.Bentham@arm.com> Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/524782 Tested-by: bsgcomp <bsgcomp@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Comments-Addressed: bsgcomp <bsgcomp@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9757 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Pablo Marquez Tello <pablo.tello@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-05-10Re-enable dyanmic weights in Neon™ depthwise convolutionRamy Elgammal
- Call Neon™ depthwise convolution validation inside in its configure() method. Resolves: COMPMID-6188 Signed-off-by: Ramy Elgammal <ramy.elgammal@arm.com> Change-Id: Ib2ae4d995ff2bbc92ce4496d4ab93cf09113e3e9 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9594 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-05-05Connect CLMatMul function to quantized kernels and resolve NE BatchMatMul ↵Jakub Sujak
int_8 failures * Adapt the CLMatMul function and ClMatMul operator to use quantized kernels. * Add function-level tests. Resolves: COMPMID-5929 and COMPMID-5811 Change-Id: I5348cdcf07b8074c138e04dfef0a73399377accd Signed-off-by: Jakub Sujak <jakub.sujak@arm.com> Signed-off-by: Omar Al Khatib <omar.alkhatib@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9575 Reviewed-by: Mohmun02 <MohammedSuhail.Munshi@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2023-05-05Disable dynamic weights in unsupported operatorsViet-Hoa Do
Resolves: COMPMID-6185 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: Icfd9d177083ecdf41dc13e5b2ae982ff67492f8a Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9577 Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-05-03Fix im2col for fast-maths mode with padding.Renato Arantes
Following the investigation proposed by ONCPUML-1193, padding is implemented in im2col when the input channel is not a multiple of blocks requested by the weight format. Partially resolves: ONCPUML-1193 Signed-off-by: Renato Arantes <renato.arantes@arm.com> Change-Id: I350c7a1b2dcae63f8d94f5b6f1f86e948eab1f09 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9508 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-05-03Fix CPU MatMul broadcast detectionViet-Hoa Do
Resolves: COMPMID-6155 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: Ie651be65404b0b737464d7a79ebcc58475863ba0 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9555 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-05-02Fix fully connected and matmul mismatchesViet-Hoa Do
* There is an issue with quantized fully connected and matmul when the lower bound of bounded ReLU is negative. * Use int32_t for the calculation of min/max quantized value rather than PixelValue to avoid this issue. Partially resolves: COMPMID-5996 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I7b22e9d56a2441fc6a4c5c4e627f57d6e00d6ff1 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9502 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-04-26Integrate multi-threaded pretranspose_B_arraySiCong Li
This is required for the case where rhs (B) is dynamic and needs to be pretransposed in every run. In a multi-threaded setting, this means the previously single-threaded pretranspose_B_array would become the bottleneck Resolves COMPMID-5896 Signed-off-by: SiCong Li <sicong.li@arm.com> Change-Id: Id508c46992188a0f76a505152931d4955d04c16d Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9455 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-04-19Add quantized support for CPU MatMulViet-Hoa Do
Resolves: COMPMID-5899 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I89d96e292c3492ba9b1900a3e5683f9dcd11dfc6 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9440 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-04-14Fix dynamic weights for CPU connected layerViet-Hoa Do
Resolves: COMPMID-5995 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I707b8918bebee7e70d4de5207ef555c806e7a305 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9405 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-04-13Implement MatMul Function and Operator with Floating Point support for CPUMohammed Suhail Munshi
- Implements MatMul function and operator for floating point datatype FP16/FP32 - Includes support for transposing dynamic tensors prior to matrix multiplication. - Adds tests for 2D/3D/4D+ tensors in MatMul with F32/F16 datatype (with all combinations of transposed/not-transposed tensors) - Updates fixture to allow for testing fused activation in MatMul - Adds tests for matmul with and without fused activation Resolved: [COMPMID-5898] Signed-off-by: Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> Change-Id: Iefa84b26dd723c9a51e6c3f91023152c6c31ace2 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9411 Reviewed-by: SiCong Li <sicong.li@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-03-21Add dynamic weights for CPU fully connected layerViet-Hoa Do
Resolves: COMPMID-5917 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I073067b490f2a1b96b81a037ea431c9a2e5c7503 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9322 Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-03-13[ONCPUML-1174] Allow src/weights mismatch for fixed formatJonathan Deakin
Without this, we have to pass in weights to be NHWC, even if they are in fact blocked/interleaved for consumption by a fixed format kernel. Signed-off-by: Jonathan Deakin <jonathan.deakin@arm.com> Change-Id: I9ee8720a21a16b17816dbecf6308e1668ddda59c Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9285 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-03-03NEGEMMLowpMatrixMultiplyCore should be configured for optimized int8 kernel.Ethan Doe
Currently the validation routine incorrectly prevents optimized INT8 Gemm kernel from being used if the input is QASYMM8 and output type is S32. This change allows QASYMM8 input and S32 output types to leverage optimized kernel. Signed-off-by: Ethan Doe <yidoe@amazon.com> Change-Id: I65b060f522795db07d6d4df86fb7c6ddd1c626d4 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9250 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-by: Pablo Marquez Tello <pablo.tello@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-03-01Add support for kernel indices in MaxpoolAdnan AlSinan
- Add a max pooling implementation that returns kernel indices. - Add a parameter in pooling info object to pick kernel indices impl. - Add validation tests. Resolves: [ONCPUML-1187] Signed-off-by: Adnan AlSinan <adnan.alsinan@arm.com> Change-Id: I485ef1604f676ee14d5f7f62d33699e49c38e4d3 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9192 Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-02-01Add new operator AddMulAdd for Neon™ backend for Float/Quantized typesGunes Bayir
This is a fused operator that merges Add + Mul + Add [+ Relu-based-Activation] layers and have an intermediate output after the first Add. It's supported for FP16/32/QASYMM8/QASYMM8_SIGNED data types. The subsequent Add and Mul are intended for scaling and the coefficients only have one dimension (per channel). The inputs are - input1 : nD tensor [X, Y, Z, W, ..] - input2 : nD tensor [X, Y, Z, W, ..] - add_coef : 1D tensor [X] - mul_coef : 1D tensor [X] The outputs are - out1 : nD tensor (intermediate output) [X, Y, Z, W, ..] - out2 : nD tensor (final output) [X, Y, Z, W, ..] The operation can be summarized as follows: out1 <- input1 + input2 out2 <- Act(out1 * mul_coef + add_coef) The activation function can be Identity, Relu, Bounded Relu or Lower/Upper Bounded Relu. The intermediate output can be skipped by providing a nullptr. The reason of providing this operator is to be able to fuse in case of Residual network patterns and save computations by reducing memory back and forward. Resolves: COMPMID-5463 Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Change-Id: I8ef577aa623b036e9a9f655cc088493fd19a6109 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9055 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-02-01Remove fixed format strides hackJonathan Deakin
- Remove hack in CpuGemmAssemblyDispatch.cpp which tried to guess strides for fixed format kernels. Instead, expect that strides will have been correctly set on weights externally - Update fixed format test fixtures to set the strides - If the fixed format uses fast math mode, then weights should be of type BFLOAT16. Change the validation logic to accept this. Resolves: [ONCPUML-1131] Co-authored-by: Milos Puzovic <Milos.Puzovic@arm.com> Change-Id: I0f18d8b86b0f639be25fd122fa06a591e90645f2 Signed-off-by: Jonathan Deakin <jonathan.deakin@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8985 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-01-18Add broadcast batched matmul validation casesSiCong Li
Related to: COMPMID-5660 Signed-off-by: SiCong Li <sicong.li@arm.com> Change-Id: I2314c8b21acc638402c77080d59db2f3fed58fe2 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8911 Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-by: Mohmun02 <MohammedSuhail.Munshi@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-01-11Deprecated BF16 support in DepthConvertPablo Marquez Tello
* Removed BF16 validation tests for DepthConvert * Revert back to using inline assembly to convert to/from BF16 * Resolves COMPMID-5800 Change-Id: I803b2ad19ead297417f780c97c5b724cca6b394c Signed-off-by: Pablo Marquez Tello <pablo.tello@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8929 Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2022-11-01Updateable weights in depthwise convolutionMilos Puzovic
Check whether weights are defined as constant, if they are not constant then do not pack them if they are already packed so that they can be updated. Signed-off-by: Milos Puzovic <Milos.Puzovic@arm.com> Change-Id: I73447e31e3660b05f8f40e04ea4ea2003eb9b802 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8539 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2022-11-01Add check for Batch Matmul in GemmAssemblyDispatchMohammed Suhail Munshi
Relates to : COMPMID-5507 Change-Id: Ia2c4ea153ac2524ffa6b2a9c10f3a0318a8a67a1 Signed-off-by: Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8509 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2022-10-20Update reinterpret tensor as 1D for CPU addViet-Hoa Do
* Use the same implementation as other layers. Resolves: COMPMID-5108 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I5a50259b398b71ca1f61b5ee8daa539bf8263fac Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8501 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2022-10-12Optimize Neon™ Logistic ActivationMohammed Suhail Munshi
- Use a 1d execution window to improve memory access pattern. Resolves: [COMPMID-5465] Signed-off-by: Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> Change-Id: Ida30669ffa06eb002ca43a6edf15e25a6eaad2f6 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8344 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2022-10-12Adding documentation section explaining how BF16 is usedRamy Elgammal
Resolves: COMPMID-5494 Signed-off-by: Ramy Elgammal <ramy.elgammal@arm.com> Change-Id: I8f512745855b8ca21181a9ab21323bfff6aeb866 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/458884 Tested-by: bsgcomp <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Comments-Addressed: bsgcomp <bsgcomp@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8391 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Pablo Marquez Tello <pablo.tello@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2022-10-07Optimize Neon™ SUB operator by squashing execution windowJakub Sujak
Resolves: COMPMID-5462 Change-Id: I2c7151c8faf4016cc33592fff04d492d7cbc8fd6 Signed-off-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8366 Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2022-09-16Optimize Quantized/Integer Bilinear Scale for Neon™Gunes Bayir
This patch introduces several performance optimizations regarding the Bilinear Scale operator with REPLICATE Border mode. Changes apply only to NHWC. This patch - Reduces the memory footprint by disabling precomputation of indices and weights when they're not used - Rewrites the kernels for QASYMM8/QASYMM8_SIGNED/U8(Uint8) - Adds S8(Int8) Bilinear Scale for Border mode REPLICATE - Removes Bilinear Scale SVE kernels for Quantized and Integer types and adjust the heuristics to choose the Neon™ implementation - Adds new test cases where the input and output of the Bilinear Scale operator have different quantization scale and offset Resolves: COMPMID-5453, COMPMID-5454 Change-Id: I3d251e76e0c6978fd5a0a1795ec62ab536bec93c Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8250 Reviewed-by: SiCong Li <sicong.li@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
generated by cgit v1.2.1 (git 2.23.0) at 2024-06-28 15:07:35 +0000