aboutsummaryrefslogtreecommitdiff
path: root/src/core/CL/gemm
AgeCommit message (Collapse)Author
2021-04-16Add GEMM heuristic for Mali-G78Gian Marco Iodice
- Replace std::map with a basic container with std::array Change-Id: I76f53ca61676ca0e5136ce61a3f3adb10e22b4c3 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5441 Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2021-02-25Avoid division with zeroGeorgios Pinitas
- Check pixel alignment before calculating the padding extension for CLGEMMReshapeRHSKernel - Add missing control flow in NEWinogradLayer to avoid division with zero for N_Blocks Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: I53fa432e0a4c463840917d81bedeb64a29bb07a4 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5177 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Manuel Bottini <manuel.bottini@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2021-02-23Avoid division with zero on CLGEMM lhs/rhs configurationGeorgios Pinitas
Resolves COMPMID-4285 Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: I414e3ba352ed86345d6355d9ae366ea92e641842 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/5153 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2021-01-22Update Fp16 heuristic for Mali-G76 and Mali-G77Gian Marco Iodice
Change-Id: I199c10574d0d84737adbc73829f60a220c7088b4 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4897 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Giorgio Arena <giorgio.arena@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-12-03Rename the files and classes required by the OpenCL GEMM heuristicSiCong Li
All existing kernel type selection heuristics CLGEMMKernelSelection<Architecture> are renamed to CLGEMMDefaultType<Architecture> All existing kernel configuration heuristics CLGEMM<KernelType>KernelConfiguration<Architecture> are renamed to CLGEMMDefaultConfig<KernelType><Architecture> This refactoring is required to make room for tuner-based heuristics Resolves COMPMID-3842 Change-Id: I2c9f1029ad67f1e2808c79871698fc4486d45306 Signed-off-by: SiCong Li <sicong.li@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4639 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>
2020-12-01Update default C++ standard to C++14Georgios Pinitas
(3RDPARTY_UPDATE) Resolves: COMPMID-3849 Signed-off-by: Georgios Pinitas <georgios.pinitas@arm.com> Change-Id: I6369f112337310140e2d6c8e79630cd11138dfa0 Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4544 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-11-23COMPMID-4018: Fix heuristic fallback for CLGEMMReshapedRHSOnly forGian Marco Iodice
Mali-G52 - Missing fallback in case of export to cl_image Change-Id: I5bb3013fd1350628f16e4709c4bb31999fece22d Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4531 Reviewed-by: Pablo Marquez Tello <pablo.tello@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-11-18COMPMID-3980: Fix F16 regression on Mali-G76Gian Marco Iodice
- The issue was related to the wrong N0 used in CLGEMMReshaped heuristic Change-Id: Iecb010f0de383169c9fb8543faf15b2f93ba2ed8 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4464 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-11-18COMPMID-3981: Fix missing fallback for export to cl_imageGian Marco Iodice
- Fix missing fallback in the CLGEMMReshaped heuristic on Mali-G77 Change-Id: I0a243c7ed153216966d0809a3b3348f030a845eb Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4463 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-11-17COMPMID-3975: Update CLGEMM heuristic for Mali-G52 F16Gian Marco Iodice
- Update the GEMM heuristic for Mali-G52 - Fix the GEMV heuristic on Mali-G76 Change-Id: I895153d02d09e73c51ecf158e2ae43f81f6d7b05 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4425 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-3959: Update Mali-G52 heuristic for CLGEMM - F32Gian Marco Iodice
- Add heuristic in CLGEMMKernelSelection - Add heuristic in CLGEMMReshapedRHSOnly - Add heuristic in CLGEMMReshaped Change-Id: Ibaa13398f7a5976418a0ab1b6696ace09cc480fa Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4366 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-11-03COMPMID-3939: Update GEMM heuristic Mali-G77Gian Marco Iodice
- Update heuristic for GEMM reshaped RHS only - Fix left-over block size in CLGEMMMatrixMultiplyReshapedOlyRHSKernel Change-Id: I34c738821ed2e4a537da4a15058eec164cb6b61f Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4305 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-10-30COMPMID-3930: Update CLGEMM heuristic for fp16. Mali-G76Gian Marco Iodice
- Since the GEMM kernel can now work without padding, the heuristic requires to be fine-tuned to exploit this feature - The heuristic affects Mali-G76 FP16 only Change-Id: Ia430627f02131ad956ce2219b80c83c8e7cabaf2 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4284 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: SiCong Li <sicong.li@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>
2020-10-07COMPMID-3674: Update heuristic for Mali-G77Gian Marco Iodice
- Update the heuristic (m==1) for CLGEMMReshapedOnlyRHS Change-Id: I216c158f2802d3d331e23e0d9eb0127107ec8af0 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4092 Reviewed-by: Giorgio Arena <giorgio.arena@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Georgios Pinitas <georgios.pinitas@arm.com>
2020-09-24COMPMID-3673: Update heuristic for GEMM FP16 on Mali G76Gian Marco Iodice
- Update heuristic for CLGEMMReshapedKernel - FP16 - Update heuristic for CLGEMMReshapedOnlyRHSKernel - FP16 Change-Id: I35aa73e59d8c2d1bc6b2dd318fd8eeb3e42c27a4 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4026 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-09-21COMPMID-3787: Update heuristic for m = 1 on CLGEMMReshapedOnlyRHSKernelGian Marco Iodice
- The change affects Mali-G71 GPUs and should improve the performance of GEMM in case of m = 1 Change-Id: I6b0e217e93fe468ec1325a5da74684811519c42f Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4002 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: SiCong Li <sicong.li@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-08-28COMPMID-3670 Extend cl image support to f16 in CLGEMMReshapeRHSMatrixKernelSiCong Li
Change-Id: Ic0569fe9ed99e61084b601ce84ddc7ef288d1899 Signed-off-by: SiCong Li <sicong.li@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3852 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-08-07COMPMID-3683: Fix performance regression on Mali-G76 (Fully connected)Gian Marco Iodice
COMPMID-3682: Fix performance regression on Mali-G76 (Convolution) Updated the heuristic for GEMMReshapedOnlYRHS for Mali-G76 in order to take into account small workload cases Change-Id: I99fccbd0e94e4e21c0d1b88e23f02af06ef16ee9 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3689 Reviewed-by: SiCong Li <sicong.li@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-07-21COMPMID-3326; Update heuristic for GEMMReshaped and GEMMReshapedOnlyRHSGian Marco Iodice
- Update the heuristic for Arm Mali-G77 (F32) in order to use the OpenCL image2d object on GEMM Change-Id: Ife6736a22ec2a114368bb338908f0c5f239dfad6 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3593 Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michele Di Giorgio <michele.digiorgio@arm.com>
2020-07-15COMPMID-3326: Update heuristic for GEMMReshaped and GEMMReshapedOnlyRHSGian Marco Iodice
- Update the heuristic for Arm Mali-G76 (F32) in order to use the OpenCL image2d object on GEMM - Create utility function to validate the support for image2d Change-Id: I0913ac5f27fd07992b0ac188af753a2abeb034ca Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3559 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-07-09COMPMID-3325: Add support in gemm_tuner for cl_imageManuel Bottini
Change-Id: I78f815005516ca0e83366bab017884530c1d2e86 Signed-off-by: Manuel Bottini <manuel.bottini@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3518 Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-07-09COMPMID-3324: Adjusting capitalization of Arm copyright claim to reflect Arm ↵Michele Di Giorgio
preferred presentation Change-Id: Ib7dcfcbb24b408999dfae366b9da396485aacf78 Signed-off-by: Michele Di Giorgio <michele.digiorgio@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3525 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-05-07COMPMID-3459: Updated heuristic in CLGEMMReshapedOnlyRHSKernelGian Marco Iodice
Change-Id: Ib958649d0193c148f45d72404da21d3f02dc5183 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3154 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Michalis Spyrou <michalis.spyrou@arm.com>
2020-05-06COMPMID-3459: Update the heuristic for GEMV for F16 on Mali-G77Gian Marco Iodice
Change-Id: I10020cd5c88ec1926087531c4c4ce0a1cfaf43c9 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3151 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2020-05-05COMPMID-3428: Fixed Mali-G77 heuristic for GEMMReshapedOnlyRHS - f16Gian Marco Iodice
Change-Id: Ie1f021235edf3d60171a252d519c53df2d7fb0c6 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3146 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-04-20COMPMID-3304: Update OpenCL GEMM heuristic for Int8Gian Marco Iodice
Change-Id: I6b7ff678d8d0437a1639db2ff602ea1cdb155464 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/3056 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-04-08COMPMID-3112: Reworking heuristic for CLGEMM - part1Gian Marco Iodice
The new heuristic only affects the floating point execution Change-Id: Ia6edc14ab1bdda4cee31b7afb096d0305d99b809 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/2942 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2020-02-26COMPMID-3202: Add support for Valhall architecture in GEMMGian Marco Iodice
Change-Id: I2cd0b5ee0ae2e3c65a04c7be70478b0452e55132 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/2730 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2019-10-04COMPMID-2572: Update the heuristic in CLGEMM for FP16Gian Marco Iodice
Change-Id: Ia7f248d72bf2690a8a4dae3f2e2afc983109bd6e Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/2035 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Giorgio Arena <giorgio.arena@arm.com>
2019-09-30COMPMID-2571: Add mixed-precision support in CLGEMMReshaped for FP16Gian Marco Iodice
Change-Id: I5ba90d4de4594ed784c7230aa6b10503be67c001 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/1991 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2019-09-26COMPMID-2571: Add support for FP16 in CLGEMMReshaped - part 1Gian Marco Iodice
Change-Id: I8adb8850cc5ade49ebc1dbf63401f03d5ecad708 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/1983 Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
2019-09-23COMPMID-2668: Add GEMM/GEMMLowp heuristic for Mali-G51Gian Marco Iodice
Change-Id: I920ee5ffeff86806b9ac0620250dba98d1d081ba Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/1955 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2019-07-18COMPMID-2096: Refactor the CLGEMMLowp function selection (heuristic)Gian Marco Iodice
Change-Id: I15a8b39e0354d3b6686ed4cc8c361782c0512037 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/1410 Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com> Reviewed-by: VidhyaSudhan Loganathan <vidhyasudhan.loganathan@arm.com>
2019-04-16COMPMID-2110: Enable CLGEMMLowpMatrixMultiplyReshapeOnlyRHSKernel in CLGEMMLowpGian Marco Iodice
Change-Id: Ic32c803c3e2a067de10a7e46c85c962a970957b6 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/969 Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
2019-04-08COMPMID-2097: Implement a heuristic to dispatch CLGEMMReshapedOnlyRHS kernel ↵Gian Marco Iodice
from CLGEMM Change-Id: I4170a80647b02501aa669e2c0347ddc39888ee76 Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com> Reviewed-on: https://review.mlplatform.org/c/928 Reviewed-by: Giuseppe Rossini <giuseppe.rossini@arm.com> Comments-Addressed: Arm Jenkins <bsgcomp@arm.com> Tested-by: Arm Jenkins <bsgcomp@arm.com>
generated by cgit v1.2.1 (git 2.23.0) at 2024-05-23 12:55:52 +0000