aboutsummaryrefslogtreecommitdiff
path: root/src/core/CL/cl_kernels
AgeCommit message (Collapse)Author
2024-04-30Add fp16 and integer data type support for ScatterNd in GpuGunes Bayir
Resolves: COMPMID-6899 Change-Id: I3743f2c9e5c21e1ec9f4c81d08c148666afad33a Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/11505 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-by: Sang Won Ha <sangwon.ha@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2024-04-25Add update/index/output (m+1)/2d/(m+n) support for CLScatterGunes Bayir
Resolves: COMPMID-6894, COMPMID-6896 Change-Id: I9d29fd3701a7e0f28d83f81a6c42a7234c2587c3 Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/11477 Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Ramy Elgammal <ramy.elgammal@arm.com> Dynamic-Fusion: Ramy Elgammal <ramy.elgammal@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2024-04-22Scatter GPU Kernel Implementation for 1D tensors.Mohammed Suhail Munshi
Resolves: [COMPMID-6891, COMPMID-6892] Change-Id: I5b094fff1bff4c4c59cc44f7d6beab0e40133d8e Signed-off-by: Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/11394 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>
2024-03-20Make Cpu/Gpu/Ref scalar/vectoral S32 division consistentGunes Bayir
- Neon(TM) implementation converts integers to float and performs the division because there is no vector integer division instructions. However, leftover loop still uses integer division, which makes results inconsistent depending on where we are in the tensor. - SVE path does it in integer domain. - OpenCL(TM) does it similar to Neon(TM) vector path. - Reference implementation does it in integer domain. These differences cause intermittent mismatches. This patch ensures all follow the same logic. On the other hand, the provided Neon(TM) implementation is faster than the Fp32 converted version. Resolves: COMPMID-6925 Change-Id: Ia12606d57f40a7d331b9b698f87fd4321496b275 Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/11316 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>
2024-01-18Fix divide-by-zero compilation errorViet-Hoa Do
* CONVERT_TO_TENSOR4D_STRUCT_NO_STEP is implemented and used in some CL kernels in the way that causes divide-by-zero issue. - Since the steps are all zeros, the issue might have been ignored by the compiler. Resolves: COMPMID-6795 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I0fb38fc62d63671b8abefa39b3d9b3ca6f49c7fe Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10967 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>
2023-12-22Fix nightly issue caused by gemm_reshaped_only_rhs_mmul kernelGunes Bayir
The issue appears when this kernel is used by convolution operators because the stride calculations consider only simple matrix multiplication. In conv2d triggered runs, Rhs does not have the same dimension as Lhs and Dst. Also, cases where Lhs and Dst are interpreted as 3d, where their X and Y dimensions (in convolution sense) are collapsed into one. Resolves: COMPMID-6764 Change-Id: If443e6eb8f7a5cca1acc58b37c598122a013e69b Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10913 Benchmark: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2023-12-14Fix validation error in CL generate proposals kernelGunes Bayir
This fix modifies some of the conversions done in the generate proposals kernel that causes DDK issues while compiling the kernel. The issues are mostly related to conversion from i64 to fp16, and it doesn't affect fp32. Firstly, type identifier size_t is converted into unsigned int. But, this alone was compiling but causing mismatches, even in older devices, where it was passing before. Therefore, the fp16 conversion delayed until vector construction where the integers are now converted to fp32, and then fp16. This, although may not be ideal, seems like the best solution. Resolves: COMPMID-6756 Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Change-Id: Iee61216c908fe51431985b80c3653fc32add4741 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10879 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-by: Pablo Marquez Tello <pablo.tello@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-11-15Fix device issue with CL softmaxViet-Hoa Do
* Performing the second pass in reverse order doesn't seem to work reliably in some specific devices. This patch introduces another approach to workaround the device issue. Resolves: COMPMID-6669 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I591f05ff06f8439ebe4d32093441ae871a292f4c Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10730 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>
2023-10-31[GPU] Update Reverse layer to allow negative axis and reversed axis orderAdnan AlSinan
- Adds option to use negative axis and inverted axis. - Adds validation tests for the above. Resolves COMPMID-6459 Change-Id: I88afd845d078f92c82ec8529ce7241fccd4c417e Signed-off-by: Adnan AlSinan <adnan.alsinan@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10523 Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-10-31Optimize CL softmaxViet-Hoa Do
* The new softmax implementation consists of only a single kernel. - There are 2 versions of softmax, one for the x dimension and one for any other dimensions. - Softmax kernel handles both native and quantized data type. Resolves: COMPMID-6447 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I4a9ae5bc63f78aebeaa85ee48a0d102c9c245eda Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10489 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>
2023-10-12Remove padding from CL comparison operatorViet-Hoa Do
* Add support for processing left-over vector to comparison kernel. * Combine native and quantized versions of CL comparison. Resolves: COMPMID-6424 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I31d43bdf0eab999cee6fa8144b5d8e921a1093e8 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10467 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>
2023-10-11Optimize CL reduction operationViet-Hoa Do
* Batch dimension is added to reduction operation. - All the dimensions higher than the batch dimension are collapsed so that the input and output tensors are always 3-4D. - CL kernel is called once instead of being repeatedly called to process each sliding window. Resolves: COMPMID-6443 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: Icd99939d52d3bb648f08537e5f52ef27e894061b Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10456 Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-10-05Optimize CLTranspose operatorJakub Sujak
* Transpose higher dimensional tensors (>2D) by collapsing higher dimensions into the third dimension thus avoiding multiple dispatches of the CL kernel * Maximize tile size without register spilling Resolves: COMPMID-6448 Change-Id: Iac094b8c428bdf319d9c28a8334cb55d58e2d14b Signed-off-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10443 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>
2023-09-29Implement Quantized Matmul T/T and T/Nt kernels using MMUL extensionGunes Bayir
Resolves: COMPMID-6476, COMPMID-6477 Change-Id: Ied37c269d5a108ff72f70e3ad932cf372bda5562 Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10346 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>
2023-09-28Implement Quantized Matmul Nt/T kernel using MMUL extensionGunes Bayir
Resolves: COMPMID-6474 Change-Id: Iaff5b512cf77975f2df02dcdf848711b13bf97a6 Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10341 Reviewed-by: Mohmun02 <MohammedSuhail.Munshi@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>
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>
2023-09-21Optimize the main loop in mat_mul_native_quantized_mmul_nt_ntGunes Bayir
Switch the loop unrolling order and reuse the pre-computed vectors Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Change-Id: I636c0530d6b21dae4dbb371c57d18b1f7c7246a8 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10355 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-09-18Implement Quantized MatMul kernel using MMUL extensionGunes Bayir
Resolves: COMPMID-6475 Change-Id: Ic867cdfff5d4391cb749a04bf7cc35cda63d3b71 Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10311 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-09-14Add skeleton of ClMatMulLowpNativeMMULKernelGunes Bayir
The skeleton code consists of modifications - to build the library with the quantized matmul kernel - refactoring of some common utilities - empty OpenCL Kernels for four configurations ([Lhs, Rhs] X [Nt, t]) - some validation tests and skeleton for functional tests Resolves: COMPMID-6473 Change-Id: Id8401f789d34277dceb1f91afd68c9c88275618a Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10273 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Benchmark: Arm Jenkins <bsgcomp@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>
2023-08-14Optimize CLReduce for Min/Max Axis=0Gunes Bayir
Resolves: COMPMID-6400 Change-Id: Id9935f9727f77a824afc75c35f044e3f5c173e0d Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10120 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-08-03Fix CL Tile operatorViet-Hoa Do
Resolves: COMPMID-6404 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I75aebe620567ed50817747589bbe8cfb63715a7b Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/10036 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: TeresaARM <teresa.charlinreyes@arm.com> Reviewed-by: Anitha Raj <Anitha.Raj@arm.com> Reviewed-by: Pablo Marquez Tello <pablo.tello@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-07-21Enable S64 output in CLArgMinMaxPablo Marquez Tello
Resolves MLCE-1089 Change-Id: I8b385ef8a00ec5de60299bc7a359766ba5417e68 Signed-off-by: Pablo Marquez Tello <pablo.tello@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9918 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-07-11Add Bias to MatMul Kernels and add support for use in Fully Connected LayerMohammed Suhail Munshi
Resolves: [COMPMID-6316] Signed-off-by: Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> Change-Id: I08e6bac9e6b46b76978da0dc6a48ccfe3dde5086 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9833 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>
2023-07-06Fix nightly failures in MatMulLowpNativeKernel when using bounded activation ↵Mohammed Suhail Munshi
functions - Added checks for supported activation functions in MatMulLowpKernel validate - Replaced incorrect float activation macro with quantized implementation in mat_mul_quantized Resolves: [COMPMID-6339] Signed-off-by: Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> Change-Id: I15661f14877f1d3305644e6473feb5482a67e773 Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/532858 Tested-by: bsgcomp <bsgcomp@arm.com> Reviewed-by: Pablo Tello <pablo.tello@arm.com> Comments-Addressed: bsgcomp <bsgcomp@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9855 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>
2023-07-05Rewrote CLArgMinMax for axis 0Pablo Marquez Tello
* Simpler implementation without stages for axis 0 * Removed considerable amount of code. Resolves COMPMID-6271 Change-Id: Ie8bcb2f0b55f87472f44b38872a23a922619a211 Signed-off-by: Pablo Marquez Tello <pablo.tello@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9849 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>
2023-06-29Implement FP32/16 MatMul Lhs T Rhs T/NT kernel using MMUL extensionGunes Bayir
Resolves: COMPMID-6196, COMPMID-6197 Change-Id: I22a1c32686eb70e7676c8b4d64a76dbaeb638cb3 Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9798 Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-06-26Use MatMul in fully connected layer with dynamic weights when supportedMohammed Suhail Munshi
- Use MatMul kernels in FC layer when using dynamic weights without broadcasting or bias. - Fix minor typo in IClMatMulNativeKernelConfig.h Partially Resolves : [COMPMID-6193] Signed-off-by: Mohammed Suhail Munshi <MohammedSuhail.Munshi@arm.com> Change-Id: Id494062b5b4f4e75ff9714c202dde941955afa52 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9797 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>
2023-06-23Implement FP32/FP16 MatMul NT/T kernel using the MMUL extensionRamy Elgammal
Resolves COMPMID-6195 Signed-off-by: ramy.elgammal@arm.com <ramy.elgammal@arm.com> Change-Id: I8e85fe73308ed84ebb142d6d6d1562b62dddfaa5 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9819 Reviewed-by: SiCong Li <sicong.li@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-06-23Fix doxygen warningsramy.elgammal@arm.com
Resolves: COMPMID-6312 Signed-off-by: ramy.elgammal@arm.com <ramy.elgammal@arm.com> Change-Id: I9f68ccd2edb8c4d03fec19e6b9c29609d4833342 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9806 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>
2023-06-19Implement FP32/FP16 MatMul NT/NT kernel using the MMUL extensionSiCong Li
Resolves COMPMID-6194 Signed-off-by: SiCong Li <sicong.li@arm.com> Change-Id: Ie45e2aa9533948b2e5235563cef1d3834494eccf Signed-off-by: SiCong Li <sicong.li@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9739 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>
2023-06-16Add 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>
2023-05-11Fix invalid vector length in CLViet-Hoa Do
Resolves: COMPMID-6252 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I97ddf8a6c83bc2621abc712094db6bc0fe3d97b1 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9620 Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2023-05-03Support multi-dimensional indices in the CL Gather Layer up to ↵Omar Al Khatib
four-dimensional output tensors Resolves [COMPMID-5775] Signed-off-by: Omar Al Khatib <omar.alkhatib@arm.com> Change-Id: I6f6c12ac08f0b0ad070ca5d715c531c2c3762c30 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9498 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>
2023-04-28Fix the gather layer indices checkViet-Hoa Do
* If the index is out-of-bound, both CPU and GPU implementations of the gather layer will output 0. Resolves: COMPMID-5964 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: Ib029b3acfb31452f2097c8c75448fb2697cfa332 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9487 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>
2023-04-27Add quantized CL MatMul kernel for LHS NT, RHS TJakub Sujak
Implement a native kernel for batched Matrix Multiplication for the quantized data types QASYMM8 and QASYMM8_SIGNED and with the MatMul attributes `adj_x = false, adj_y = true`. Resolves: COMPMID-5923 Change-Id: I477b2dd886edfe83beaba9efc7d6b05ed19f5da4 Signed-off-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9467 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>
2023-04-26Improve Winograd performance on OpenCLGian Marco Iodice
- Performs more output elements per work-item in the case of Fp16 computation in Winograd Input/Output transform Resolves COMPMID-6018 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Change-Id: If5e6f5182eff8c1f05a3505c437d0a997490f0bd Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9447 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-04-20Implement CL kernel for a native batched matmul Quantized - LHS transposed, ↵Omar Al Khatib
RHS transposed Resolves: [COMPMID-5924] Signed-off-by: Omar Al Khatib <omar.alkhatib@arm.com> Change-Id: I9ba657737eb1e3a096c8341ad4ad311571f8edeb Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9454 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>
2023-04-17Add quantized CL MatMul kernels for Lhs NT/T, Rhs NTGunes Bayir
Implement OpenCL kernels for batched Matrix Multiplication for the quantized data types QASYMM8 and QASYMM8_SIGNED. Quantized MatMul is supported with the following MatMul attributes: * adj_x = false, adj_y = false * adj_x = true, adj_y = false We consider native format kernels only. In other words, no reshaping of the operand matrices is done. Resolves: COMPMID-5921, COMPMID-5922 Change-Id: I99e0f68054a2bd635c60ec2641acc2e7ff398473 Signed-off-by: Omar Al Khatib <omar.alkhatib@arm.com> Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Signed-off-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9435 Reviewed-by: SiCong Li <sicong.li@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> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-04-03Add Cropping to CLBatchToSpaceOmar Al Khatib
- Deprecate dynamic block shape interface - Iterate over output window instead of input window for simpler implementation and better performance. - Add cropping support and cropping tests Resolves [COMPMID-5865] Signed-off-by: Omar Al Khatib <omar.alkhatib@arm.com> Change-Id: Ic67d44a6a39299ecdafc507f12e3dc5d517dfb62 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9385 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>
2023-03-29Add quantized support for unary elementwise in CPUViet-Hoa Do
* Add quantized unary elementwise in CPU using LUT. * Widen the input data range of the test suite. - Fix CPU exponential function overflow/underflow range. - Fix saturation issue of CL round operator. Resolves: COMPMID-5763 Signed-off-by: Viet-Hoa Do <viet-hoa.do@arm.com> Change-Id: I41445de2b4a33ec6b01e0ab701516c240c852d0b Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9367 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-by: Pablo Marquez Tello <pablo.tello@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-03-24Add Texture Pipe Support for Matmul Lhs T/NT Rhs T kernelsRamy Elgammal
Resolves: COMPMID-5952, COMPMID-5956 Change-Id: Idbd14538e7660792254072fa9631a6f03966f89b Signed-off-by: Ramy Elgammal <ramy.elgammal@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9371 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-03-24Add Texture Pipe Support for Matmul Lhs T/NT Rhs NT kernelsGunes Bayir
Resolves: COMPMID-5945, COMPMID-5954 Change-Id: I7b27021d21f8e08c4896f6b1f595a75125064f9e Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9356 Reviewed-by: Gian Marco Iodice <gianmarco.iodice@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> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-03-20Implement OpenCL MatMul for Lhs T Rhs T/NT FP32/16Gunes Bayir
- Implement opencl kernel for LHS transposed and RHS non-transposed - Implement opencl kernel for LHS transposed and RHS transposed - Add validation tests Resolves: COMPMID-5953, COMPMID-5955 Change-Id: I55589acbffe86c44e29807574975978a1ec09bad Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9345 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2023-03-17Implementation of RSQRT for quantized int8Ramy Elgammal
Resolves: COMPMID-5863 Change-Id: I9ff67face62826c1d335a6b941e8516be39bdac8 Signed-off-by: Ramy Elgammal <ramy.elgammal@arm.com> Reviewed-on: https://eu-gerrit-1.euhpc.arm.com/c/VisualCompute/ComputeLibrary/+/488768 Tested-by: bsgcomp <bsgcomp@arm.com> Comments-Addressed: bsgcomp <bsgcomp@arm.com> Reviewed-by: Gunes Bayir <gunes.bayir@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9225 Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-03-17Implement OpenCL MatMul for Lhs NT Rhs T/NT FP32/16Ramy Elgammal
- Implement ClNativeMatMulKernel class - Implement opencl kernel for LHS non-transposed and RHS non-transposed - Implement opencl kernel for LHS non-transposed and RHS transposed - Add test fixture and dataset for matmul - Implement transpose_tensor() for reference implementation to transpose high dimensional tensors Resolves: COMPMID-5944, COMPMID-5951 Co-authored-by: Gunes Bayir <gunes.bayir@arm.com> Co-authored-by: Ramy Elgammal <ramy.elgammal@arm.com> Change-Id: I1d5b8978f41be27baddb3153ade880472141573f Signed-off-by: Gunes Bayir <gunes.bayir@arm.com> Signed-off-by: Ramy Elgammal <ramy.elgammal@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9333 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-02-01Add Subtraction operator to Dynamic Fusion interfaceRamy Elgammal
Partially-Resolves: COMPMID-5518 Change-Id: I8358784815bcac461d50e384fa7bc96f476d3983 Signed-off-by: Ramy Elgammal <ramy.elgammal@arm.com> Signed-off-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9045 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Dynamic-Fusion: SiCong Li <sicong.li@arm.com> Benchmark: Arm Jenkins <bsgcomp@arm.com>
2023-01-31Add Multiplication operator (FP only) to Dynamic Fusion InterfaceJakub Sujak
Note: we use a separate test fixture for Multiplication op instead of reusing ElementwiseBinaryFixture to avoid exposing the internal enum ElementwiseOp to the public utils/TypePrinters.h as required by the data test case macros to print the test data. We also do not consider modifying the enum ArithmeticOp in the standard interface to include MUL without an implementation. Future work should consider refactoring this test fixture into the ElementwiseBinaryFixture to reduce the total number of fixtures/code duplication. Resolves: COMPMID-5779 Change-Id: I84207658ce0407095b028fca0ab7bfa2950255ec Signed-off-by: Jakub Sujak <jakub.sujak@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9013 Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> 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>
2023-01-18Revert "Update the heuristic for CLDepthwiseConvolutionNative kernel"Gian Marco Iodice
Resolves COMPMID-5813 Change-Id: I5ef6fe9fb6a54db18e41a71085896fd08bc08dbb Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8975 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>
2023-01-12Update the heuristic for CLDepthwiseConvolutionNative kernelGian Marco Iodice
- Use T_LOAD2D_INDIRECT macro instead of T_LOAD_NHWC_WITH_DILATION in the depthwise convolution opencl kernels - Update the heuristic for Arm® Mali™-G77 Resolves COMPMID-5716 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Change-Id: I32d375b220e04bf05f5d8f0af2231bde600f0665 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8930 Benchmark: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> 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>
generated by cgit v1.2.1 (git 2.23.0) at 2024-05-20 04:11:04 +0000