opencv/3rdparty/hal_rvv/hal_rvv.hpp
Liutong HAN 3fbaad36d7
Merge pull request #26624 from hanliutong:rvv-mean
Add RISC-V HAL implementation for meanStdDev #26624

`meanStdDev` benefits from the Universal Intrinsic backend of RVV, but we also found that the performance on the `8UC4` type is worse than the scalar version when there is a mask, and there is no optimization implementation on `32FC1`.

This patch implements `meanStdDev` function in RVV_HAL using native intrinsic, significantly optimizing the performance for `8UC1`, `8UC4` and `32FC1`.

This patch is tested on BPI-F3 for both gcc 14.2 and clang 19.1.
```
$ opencv_test_core --gtest_filter="*MeanStdDev*"
$ opencv_perf_core --gtest_filter="Size_MatType_meanStdDev*
```

![1734077611879](https://github.com/user-attachments/assets/71c85c9d-1db1-470d-81d1-bf546e27ad86)

### 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
- [ ] The PR is proposed to the proper branch
- [ ] There is a reference to the original bug report and related work
- [ ] There is accuracy test, performance test and test data in opencv_extra repository, if applicable
      Patch to opencv_extra has the same branch name.
- [ ] The feature is well documented and sample code can be built with the project CMake
2024-12-18 22:19:02 +03:00

28 lines
790 B
C++

// This file is part of OpenCV project.
// It is subject to the license terms in the LICENSE file found in the top-level directory
// of this distribution and at http://opencv.org/license.html.
#ifndef OPENCV_HAL_RVV_HPP_INCLUDED
#define OPENCV_HAL_RVV_HPP_INCLUDED
#include "opencv2/core/hal/interface.h"
#ifndef CV_HAL_RVV_071_ENABLED
# if defined(__GNUC__) && __GNUC__ == 10 && __GNUC_MINOR__ == 4 && defined(__THEAD_VERSION__) && defined(__riscv_v) && __riscv_v == 7000
# define CV_HAL_RVV_071_ENABLED 1
# else
# define CV_HAL_RVV_071_ENABLED 0
# endif
#endif
#if CV_HAL_RVV_071_ENABLED
#include "version/hal_rvv_071.hpp"
#endif
#if defined(__riscv_v) && __riscv_v == 1000000
#include "hal_rvv_1p0/merge.hpp" // core
#include "hal_rvv_1p0/mean.hpp" // core
#endif
#endif