Madan mohan Manokar 28a1d0dddb Merge pull request #29242 from amd:fast_gemm_simd ...

Optimized gemm implementation with Universal SIMD #29242

- vectorized GEMMSingleMul and GEMMBlockMul

### Pull Request Readiness Checklist

See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request

- [x] I agree to contribute to the project under Apache 2 License.
- [x] To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
- [x] The PR is proposed to the proper branch
- [ ] There is a reference to the original bug report and related work
- [x] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
      Patch to opencv_extra has the same branch name.
- [x] The feature is well documented and sample code can be built with the project CMake

2026-07-01 13:47:43 +03:00

..

cuda

cuda: add compatibility layer for depreciated vector types

2025-09-01 13:25:50 +03:00

opencl

core: implement OpenCL kernel for UMat::diag to avoid aliasing races

2026-05-07 09:18:29 +08:00

parallel

core: restore parallel name on failure attempt

2025-09-19 03:25:40 +00:00

utils

Merge pull request #28935 from molhamfetnah:fix/21960-unicode-temp-path

2026-06-28 14:20:45 +03:00

algorithm.cpp

compatibility: keep Ptr<FileStorage> stubs till OpenCV 5.0

2022-12-16 00:47:44 +00:00

alloc.cpp

Partially back-port #25075 to 4.x

2024-03-05 12:15:39 +03:00

arithm_ipp.hpp

Revert "Disable IPP with AVX512 in cv::compare because of performance regression"

2026-03-02 01:47:14 -08:00

arithm.cpp

Merge pull request #27972 from asmorkalov:as/dst_mat_type

2025-11-10 08:00:17 +03:00

arithm.dispatch.cpp

Merge pull request #23109 from seanm:misc-warnings

2023-10-06 13:33:21 +03:00

arithm.simd.hpp

Fixed cv::mul overflow for U16 type.

2026-02-27 10:04:59 +03:00

array.cpp

Merge pull request #29132 from hcy11123323:4.x

2026-06-10 13:11:27 +03:00

async.cpp

Merge pull request #23736 from seanm:c++11-simplifications

2024-01-19 16:53:08 +03:00

batch_distance.cpp

Partially back-port #25075 to 4.x

2024-03-05 12:15:39 +03:00

bindings_utils.cpp

Merge pull request #26602 from shyama7004:minor-fix

2025-01-10 15:02:18 +03:00

buffer_area.cpp

core: remove unnecessary pointer cleanup in BufferArea

2022-07-24 11:58:17 +03:00

bufferpool.impl.hpp

Utilize CV_UNUSED macro

2018-09-07 20:33:52 +09:00

channels.cpp

Merge pull request #25598 from asmorkalov:as/tables_range_check_core

2024-05-17 10:48:40 +03:00

check.cpp

Merge pull request #25981 from Kumataro:fix25971

2024-08-09 14:22:02 +03:00

command_line_parser.cpp

Partially back-port #25075 to 4.x

2024-03-05 12:15:39 +03:00

conjugate_gradient.cpp

opencv: Use cv::AutoBuffer<>::data()

2018-07-04 19:11:29 +03:00

convert_c.cpp

Add static integer casting to get rid of the warning, safer

2026-03-13 22:25:08 +05:30

convert_scale.dispatch.cpp

Deprecated convertTypeStr and made new variant that also takes the buffer size

2023-04-26 09:48:15 -04:00

convert_scale.simd.hpp

Merge pull request #27006 from hanliutong:rvv-fix-ui-1024

2025-03-04 16:49:59 +03:00

convert.dispatch.cpp

Merge pull request #27972 from asmorkalov:as/dst_mat_type

2025-11-10 08:00:17 +03:00

convert.hpp

core: Rename cv::float16_t to cv::hfloat (#25217)

2024-03-21 23:44:19 +03:00

convert.simd.hpp

Drop commented out convertTo impl with IPP.

2025-04-07 14:18:37 +03:00

copy.cpp

Merge pull request #27318 from Kumataro:fix27298

2025-11-17 10:58:20 +03:00

count_non_zero.dispatch.cpp

Merge remote-tracking branch 'origin/3.4' into merge-3.4

2023-04-21 10:55:04 +03:00

count_non_zero.simd.hpp

core: use void* in countNonZero signature per team review

2026-03-09 14:20:43 +08:00

cuda_gpu_mat_nd.cpp

Merge pull request #26472 from chacha21:gpumatnd_step

2024-12-17 17:26:14 +03:00

cuda_gpu_mat.cpp

Added no-imp placeholder when HAVE_CUDA is false

2024-08-01 10:00:31 +02:00

cuda_host_mem.cpp

build: fix warnings

2019-03-05 14:47:04 +03:00

cuda_info.cpp

Merge pull request #27636 from jmackay2:cuda_13

2025-08-11 12:15:24 +03:00

cuda_stream.cpp

cuda: add python bindings to allow GpuMat and Stream objects to be initialized from raw pointers

2023-05-22 11:02:04 +03:00

datastructs.cpp

Partially back-port #25075 to 4.x

2024-03-05 12:15:39 +03:00

directx.cpp

bugfix convertFromD3D11Texture2D

2022-03-03 07:21:53 +09:00

directx.inc.hpp

Merge pull request #13972 from Mainvooid:add_cuda_support_for_D3D11_interop

2019-03-24 18:34:09 +03:00

downhill_simplex.cpp

Fix spelling typos

2019-12-27 12:46:53 +00:00

dxt.cpp

Merge pull request #27612 from eplankin:icv_update_2022.2

2025-08-25 18:14:03 +03:00

gl_core_3_1.cpp

build: unreachable code after CV_Error() (part 2)

2018-04-24 16:03:40 +03:00

gl_core_3_1.hpp

…

glob.cpp

Partially back-port #25075 to 4.x

2024-03-05 12:15:39 +03:00

hal_internal.cpp

Merge pull request #28203 from AdwaithBatchu:fix-macos-lapack-warnings

2025-12-18 13:53:56 +03:00

hal_internal.hpp

core: include version.hpp in cvdef.h, fix precomp.hpp usage

2021-02-16 11:10:45 +00:00

hal_replacement.hpp

core: fix lut_data data type

2025-10-16 21:19:12 +09:00

has_non_zero.dispatch.cpp

Merge pull request #22947 from chacha21:hasNonZero

2023-06-09 13:37:20 +03:00

has_non_zero.simd.hpp

Merge pull request #25598 from asmorkalov:as/tables_range_check_core

2024-05-17 10:48:40 +03:00

intel_gpu_gemm.inl.hpp

core(ocl): buffer bounds in intelblas_gemm_buffer_NT

2021-09-10 12:10:41 +00:00

kmeans.cpp

Merge pull request #27596 from pratham-mcw:core-kmeans-loop-unroll

2025-09-23 15:48:10 +03:00

lapack.cpp

core: fix numerical instability and out-of-bounds store in VBLAS SIMD

2026-05-26 17:38:10 +08:00

lda.cpp

Merge remote-tracking branch 'upstream/3.4' into merge-3.4

2019-04-01 18:11:55 +03:00

logger.cpp

Merge pull request #27154 from kinchungwong:logging_callback_simple_c

2025-03-30 16:17:07 +03:00

lpsolver.cpp

Merge remote-tracking branch 'origin/3.4' into merge-3.4

2023-06-20 09:56:57 +03:00

lut.cpp

Merge pull request #29250 from amd:fast_lut8u_simd

2026-06-18 17:29:18 +03:00

lut.dispatch.cpp

Merge pull request #29250 from amd:fast_lut8u_simd

2026-06-18 17:29:18 +03:00

lut.simd.hpp

Merge pull request #29250 from amd:fast_lut8u_simd

2026-06-18 17:29:18 +03:00

mathfuncs_core.dispatch.cpp

Moved IPP math functions from core module to HAL.

2026-06-15 21:53:10 +03:00

mathfuncs_core.simd.hpp

Merge pull request #29283 from pratham-mcw:exp-neon-opt

2026-06-17 11:50:58 +03:00

mathfuncs.cpp

Merge pull request #29109 from akretz:fix-issue-23644

2026-05-24 10:53:48 +03:00

mathfuncs.hpp

core: move const tables outside of dispatched code

2018-08-08 17:54:54 +03:00

matmul.dispatch.cpp

Backport some of C API removal in core module implementation.

2024-11-14 11:24:00 +03:00

matmul.simd.hpp

Merge pull request #29242 from amd:fast_gemm_simd

2026-07-01 13:47:43 +03:00

matrix_c.cpp

Partially back-port #25075 to 4.x

2024-03-05 12:15:39 +03:00

matrix_decomp.cpp

Add semicolons after CV_INSTRUMENT macros

2018-09-14 06:45:31 +09:00

matrix_expressions.cpp

Backport some of C API removal in core module implementation.

2024-11-14 11:24:00 +03:00

matrix_iterator.cpp

core: include version.hpp in cvdef.h, fix precomp.hpp usage

2021-02-16 11:10:45 +00:00

matrix_operations.cpp

parallelize sort_ using parallel_for_

2026-05-31 11:07:12 +08:00

matrix_sparse.cpp

Remove useless C headers

2025-01-13 16:34:28 +01:00

matrix_transform.cpp

Merge pull request #29172 from hcy11123323:op

2026-06-24 09:05:14 +03:00

matrix_wrap.cpp

Merge pull request #28242 from vpisarev:fix_input_array_std_vector

2025-12-23 14:25:07 +03:00

matrix.cpp

Force step to be ptrdiff_t in resize

2026-03-30 10:48:00 +02:00

mean.dispatch.cpp

Move IPP minMaxIdx to HAL.

2025-03-24 09:21:22 +03:00

mean.simd.hpp

Merge pull request #29339 from amd:fast_mean_simd

2026-06-23 13:13:15 +03:00

merge.dispatch.cpp

Merge pull request #25598 from asmorkalov:as/tables_range_check_core

2024-05-17 10:48:40 +03:00

merge.simd.hpp

Merge pull request #23980 from hanliutong:rewrite-core

2023-08-11 08:33:33 +03:00

minmax.dispatch.cpp

core: Optimize minMaxIdx/minMaxLoc

2026-06-18 16:33:50 +00:00

minmax.simd.hpp

Merge pull request #29379 from amd:fix_warning_minmaxloc

2026-06-25 15:51:33 +03:00

norm.dispatch.cpp

Fixed out-of-bound access to function table in cv::norm for HORM_HAMING.

2025-06-23 15:42:23 +03:00

norm.simd.hpp

Expand char to 32 bit of float

2026-05-23 10:17:16 +02:00

ocl_disabled.impl.hpp

Merge pull request #24918 from opencv-pushbot:gitee/alalek/core_convertfp16_replacement

2024-01-26 12:56:52 +03:00

ocl.cpp

added source comments to mark TSAN false positives. see: #27638

2025-08-15 01:17:54 +02:00

opengl.cpp

Merge pull request #28981 from Srujan-rai:fix/opengl-extensions-memory-leak

2026-06-10 20:26:55 +03:00

out.cpp

core: Rename cv::float16_t to cv::hfloat (#25217)

2024-03-21 23:44:19 +03:00

ovx.cpp

Fixed compilation on windows with openvx

2020-01-06 06:32:56 +03:00

parallel_impl.cpp

parallel_impl.cpp: fix assembler syntax for powerpc*-darwin

2026-02-18 04:35:41 +08:00

parallel_impl.hpp

core: fix pthreads performance

2018-01-26 04:09:11 +00:00

parallel.cpp

Merge pull request #28122 from asmorkalov:as/dynamic_openmp

2025-12-05 14:13:19 +03:00

pca.cpp

Add semicolons after CV_INSTRUMENT macros

2018-09-14 06:45:31 +09:00

persistence_base64_encoding.cpp

Merge pull request #28907 from chacha21:more_autobuffer

2026-06-16 18:51:37 +03:00

persistence_base64_encoding.hpp

Merge pull request #23109 from seanm:misc-warnings

2023-10-06 13:33:21 +03:00

persistence_impl.hpp

Merge pull request #26434 from dkurt:dk/int64_file_storage_4.x

2024-11-11 14:13:33 +03:00

persistence_json.cpp

Merge pull request #27587 from fengyuentau:4x/core/filestorage_json_support_backslash

2025-07-29 12:04:55 +03:00

persistence_types.cpp

Merge pull request #28907 from chacha21:more_autobuffer

2026-06-16 18:51:37 +03:00

persistence_xml.cpp

Merge pull request #26434 from dkurt:dk/int64_file_storage_4.x

2024-11-11 14:13:33 +03:00

persistence_yml.cpp

Merge pull request #28620 from mvanhorn:osc/28619-fix-yaml-parsekey-empty-key-oob

2026-03-12 10:37:24 +03:00

persistence.cpp

Merge pull request #29368 from tonuonu:fix-filestorage-read-bigint-29363

2026-06-23 15:38:59 +03:00

persistence.hpp

Remove useless C headers

2025-01-13 16:34:28 +01:00

precomp.hpp

Merge pull request #13879 from chacha21:REDUCE_SUM2

2023-04-28 20:42:52 +03:00

rand.cpp

Merge pull request #26259 from Kumataro:fix26258

2024-10-08 15:55:00 +03:00

reduce.dispatch.cpp

Merge pull request #28782 from kjg0724:reduce-simd-optimization

2026-04-21 12:24:38 +03:00

reduce.simd.hpp

Merge pull request #28782 from kjg0724:reduce-simd-optimization

2026-04-21 12:24:38 +03:00

softfloat.cpp

Merge pull request #23109 from seanm:misc-warnings

2023-10-06 13:33:21 +03:00

split.dispatch.cpp

Merge pull request #25598 from asmorkalov:as/tables_range_check_core

2024-05-17 10:48:40 +03:00

split.simd.hpp

Merge pull request #23109 from seanm:misc-warnings

2023-10-06 13:33:21 +03:00

stat_c.cpp

core: rework code locality

2021-03-02 23:24:28 +00:00

stat.dispatch.cpp

Add semicolons after CV_INSTRUMENT macros

2018-09-14 06:45:31 +09:00

stat.hpp

Split stat.cpp into smaller pieces

2018-02-12 14:14:08 +03:00

stat.simd.hpp

Fixed several cases of unaligned pointer cast

2024-12-02 16:15:23 +03:00

stl.cpp

core: get rid of built-in String type

2018-08-27 14:42:18 +03:00

sum.dispatch.cpp

New HAL entry for cv::sum and IPP adoption.

2025-05-23 12:35:11 +03:00

sum.simd.hpp

Merge pull request #25598 from asmorkalov:as/tables_range_check_core

2024-05-17 10:48:40 +03:00

system.cpp

Merge pull request #28935 from molhamfetnah:fix/21960-unicode-temp-path

2026-06-28 14:20:45 +03:00

tables.cpp

removed duplicated Hershey fonts

2014-08-15 12:09:16 +04:00

trace.cpp

Merge pull request #21937 from Kumataro:4.x-fix-21911

2022-05-13 17:32:05 +00:00

types.cpp

Improve precision of RotatedRect::points

2026-01-08 09:52:35 +01:00

umatrix.cpp

core: implement OpenCL kernel for UMat::diag to avoid aliasing races

2026-05-07 09:18:29 +08:00

umatrix.hpp

core(ocl): fix deadlock in UMatDataAutoLock

2018-01-16 17:33:06 +03:00

va_intel.cpp

Fallback to vaCreateImage + vaPutImage/vaGetImage when vaDeriveImage fails

2022-01-31 17:12:37 -05:00

va_wrapper.impl.hpp

build: made environment access a separate feature

2024-10-30 18:37:22 +03:00