aboutsummaryrefslogtreecommitdiff
path: root/src/gpu
AgeCommit message (Expand)Author
2022-11-22Remove dynamic fusion prototype with tests and examplesSiCong Li
2022-11-14Optimize Transposed Convolution for CL backend (FP32/16)Gunes Bayir
2022-11-01Rework direct convolution heuristic on OpenCLGian Marco Iodice
2022-10-06Rework DepthwiseConvolution heuristic on OpenCLGian Marco Iodice
2022-10-06Improve start-up time in gemmlowp reshaped rhs only.Adnan AlSinan
2022-10-04Update GEMM reshaped rhs only heuristicGian Marco Iodice
2022-10-03Force CL kernel compilation with 64 registersViet-Hoa Do
2022-09-16Fix validation in validate_image2d_support_on_rhsGian Marco Iodice
2022-09-09Rework heuristic in ClConv2dGian Marco Iodice
2022-09-09Add a macro guard in all OpenCL kernels in gemmlowp.clGian Marco Iodice
2022-09-02Enable Winograd-based conv2d when IFM>=8 on GpuGian Marco Iodice
2022-08-17Revert "Fix performance regression in ClConv2D"Ramy Elgammal
2022-08-16Fix performance regression in ClConv2DGian Marco Iodice
2022-08-11Fix performance regression in Conv2D on OpenCLAdnan AlSinan
2022-08-11Disable unsafe FP optimizations in Winograd Output TransformGunes Bayir
2022-08-05Fix LeNet-f16 convolution regressionAdnan AlSinan
2022-07-22Add GemmLowp MMUL Reshaped Only Rhs Support for QASYMM8/QASYMM8_SIGNEDFreddie Liardet
2022-07-22Update ClConv2D heuristic to use direct convolutionAdnan AlSinan
2022-07-13Add Gemm MMUL Reshaped Only Rhs Support for FP32/FP16Gunes Bayir
2022-07-08Extended direct conv 2d interface for tuning the OpenCl kernelGian Marco Iodice
2022-06-28Fix OpenCL Winograd output transformGian Marco Iodice
2022-06-15Fix performance regression in Winograd Output Transform (OpenCL)Gian Marco Iodice
2022-05-26Disable unsafe FP optimizations causing accuracy issuesGunes Bayir
2022-05-11Fix inclusion guard for dynamic fusion moduleSiCong Li
2022-05-06Integrate Dynamic Fusion patchesSiCong Li
2022-04-19Add CLPool3d Int8 SupportMohammed Suhail Munshi
2022-04-14Enable dynamic cl tuning for dynamically fused kernelsSiCong Li
2022-04-14Include missing embedded headersSiCong Li
2022-04-13Add DirectConvolution2D kernel component for dynamic fusionGunes Bayir
2022-03-31Fix embedded kernel header inclusion for dynamic fusionGiorgio Arena
2022-03-15Implementation of ClPooling3dramelg01
2022-03-08Merge kernel prototype patchGiorgio Arena
2022-02-21Fix performance regression on Arm(R) Mali(TM)-G71Gian Marco Iodice
2022-02-11Improve start-up time for concatenation layersramelg01
2022-02-10Fix performance regression on the first layer of convolution-based modelGian Marco Iodice
2022-02-10Improve start-up time for winograd_output_transform_*_nhwcramelg01
2022-02-09Remove deprecated remap functions.Adnan AlSinan
2022-02-09Improve start-up time for winograd_input_transform_*_nhwcramelg01
2022-02-08Improve start-up time for winograd_filter_transform_*_nhwcramelg01
2022-02-02Revert "Rework gemm_mm_reshaped_only_rhs_ kernels with new macros"Ramy Elgammal
2022-01-25Rework gemm_mm_reshaped_only_rhs_ kernels with new macrosGian Marco Iodice
2022-01-21Fix heuristic in ClConv2DGian Marco Iodice
2022-01-12Enabled support for QASYMM8 in ClCastKernelPablo Marquez Tello
2021-12-25Add tests for FP Cpu Pooling where pool region is completely outside the inputSiCongLi
2021-12-23Rework gemm_reshape_lhs_ with new macrosAdnan AlSinan
2021-12-13Remove padding from ClDirectConv2dKernelAdnan AlSinan
2021-12-10Use #if directive instead of regular condition in CLDirectConv2DGiorgio Arena
2021-12-01Improve start-up direct convolution on OpenCLGian Marco Iodice
2021-11-26Rework gemm_reshape_rhs_(nt,t) with new macrosGian Marco Iodice
2021-11-20Improve start-up timer for GeMM (floating-point):ramelg01
generated by cgit v1.2.1 (git 2.23.0) at 2024-06-19 20:42:59 +0000