From f4c2e4f8720cbb990d9be11af4fb87a4a4805da7 Mon Sep 17 00:00:00 2001
From: penghuiho <penghuihe@yeah.net>
Date: Fri, 23 Aug 2024 22:12:19 +0800
Subject: [PATCH] Merge pull request #26061 from penghuiho:fix-pow-bug
Fixed the simd bugs of iPow8u and iPow16u #26061
Add the following cases in opencv_perf_core:
* OCL_PowFixture_iPow.iPow/0, where GetParam() = (640x480, 8UC1)
* OCL_PowFixture_iPow.iPow/2, where GetParam() = (640x480, 16UC1)
iPow8u and iPow16u failed to call to simd accelerating while executing.
Fix the bug by changing the input type of iPow_SIMD function.
### 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
---
modules/core/perf/opencl/perf_arithm.cpp | 18 ++++++++++++++++++
modules/core/src/mathfuncs.cpp | 4 ++--
2 files changed, 20 insertions(+), 2 deletions(-)
diff --git a/modules/core/perf/opencl/perf_arithm.cpp b/modules/core/perf/opencl/perf_arithm.cpp
index 04d343a136..cb14aa92b1 100644
--- a/modules/core/perf/opencl/perf_arithm.cpp
+++ b/modules/core/perf/opencl/perf_arithm.cpp
@@ -688,6 +688,24 @@ OCL_PERF_TEST_P(PowFixture, Pow, ::testing::Combine(
SANITY_CHECK(dst, 1.5e-6, ERROR_RELATIVE);
}
+///////////// iPow ////////////////////////
+OCL_PERF_TEST_P(PowFixture, iPow, ::testing::Combine(
+ OCL_TEST_SIZES, OCL_PERF_ENUM(CV_8UC1, CV_8SC1,CV_16UC1,CV_16SC1,CV_32SC1)))
+{
+ const Size_MatType_t params = GetParam();
+ const Size srcSize = get<0>(params);
+ const int type = get<1>(params);
+
+ checkDeviceMaxMemoryAllocSize(srcSize, type);
+
+ UMat src(srcSize, type), dst(srcSize, type);
+ randu(src, 0, 100);
+ declare.in(src).out(dst);
+
+ OCL_TEST_CYCLE() cv::pow(src, 7.0, dst);
+
+ SANITY_CHECK_NOTHING();
+}
///////////// AddWeighted////////////////////////
typedef Size_MatType AddWeightedFixture;
diff --git a/modules/core/src/mathfuncs.cpp b/modules/core/src/mathfuncs.cpp
index 764d2d9b03..e1906fcc8f 100644
--- a/modules/core/src/mathfuncs.cpp
+++ b/modules/core/src/mathfuncs.cpp
@@ -791,7 +791,7 @@ struct iPow_SIMD
#if (CV_SIMD || CV_SIMD_SCALABLE)
template <>
-struct iPow_SIMD<uchar, int>
+struct iPow_SIMD<uchar, unsigned>
{
int operator() ( const uchar * src, uchar * dst, int len, int power )
{
@@ -871,7 +871,7 @@ struct iPow_SIMD<schar, int>
};
template <>
-struct iPow_SIMD<ushort, int>
+struct iPow_SIMD<ushort, unsigned>
{
int operator() ( const ushort * src, ushort * dst, int len, int power)
{