mirrors/opencv
2
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2026-09-25 04:09:57 +03:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #30034 from asmorkalov:as/flip_alignment

Browse Source 
Take memory aligment into account in flip.
This commit is contained in:
Alexander Smorkalov
2026-09-22 13:12:36 +03:00
committed by GitHub
parent 52ba1b4afd 57a557ce30
commit 112ca450c5
1 changed files with 35 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files
modules/core/src/matrix_transform.cpp
+35 -21
View File
@@ -645,9 +645,7 @@ template<typename V> CV_ALWAYS_INLINE void flipHoriz_single( const uchar* src, s
int width_simd = width & -vlanes;
int height = size.height;
#if CV_STRONG_ALIGNMENT
CV_Assert(isAligned<sizeof(T)>(src, dst));
#endif
CV_DbgAssert(!CV_STRONG_ALIGNMENT || isAligned<sizeof(T)>(src, dst));
for( ; height--; src += sstep, dst += dstep )
{
@@ -812,14 +810,8 @@ CV_ALWAYS_INLINE void flipHoriz_vlanes_dispatch( const uchar* src, size_t sstep,
// SIMD flip for ESZ=24 (CV_64FC3)
CV_ALWAYS_INLINE void flipHoriz_24( const uchar* src, size_t sstep, uchar* dst, size_t dstep, Size size )
{
#if CV_STRONG_ALIGNMENT
// This kernel performs 64-bit scalar loads/stores, so require 8-byte alignment.
if (!isAligned<8>(((size_t)src)|((size_t)dst)|sstep|dstep))
{
flipHoriz_generic(src, sstep, dst, dstep, size, 24);
return;
}
#endif
CV_DbgAssert(!CV_STRONG_ALIGNMENT || isAligned<8>(((size_t)src)|((size_t)dst)|sstep|dstep));
const int lanes16 = 16;
int end = (int)(size.width * 24);
int width = (end + 1) / 2;
@@ -845,21 +837,43 @@ CV_ALWAYS_INLINE void flipHoriz_24( const uchar* src, size_t sstep, uchar* dst,
static void flipHoriz( const uchar* src, size_t sstep, uchar* dst, size_t dstep, Size size, size_t esz )
{
#if CV_SIMD || CV_SIMD_SCALABLE
#if CV_STRONG_ALIGNMENT
const size_t alignMark = ((size_t)src)|((size_t)dst)|sstep|dstep;
#else
const size_t alignMark = 0; // unaligned access is allowed: all the checks below fold to true
#endif
// SIMD-optimized dispatch
switch(esz)
{
case 1: flipHoriz_single<v_uint8>(src, sstep, dst, dstep, size); return; // CV_8UC1: 8-bit, 1 channel
case 2: flipHoriz_single<v_uint16>(src, sstep, dst, dstep, size); return; // CV_8UC2, CV_16UC1: 8-bit 2-channel or 16-bit 1-channel
case 3: flipHoriz_c3<v_uint8>(src, sstep, dst, dstep, size); return; // CV_8UC3: 8-bit, 3 channels
case 4: flipHoriz_single<v_uint32>(src, sstep, dst, dstep, size); return; // CV_8UC4, CV_16UC2, CV_32SC1, CV_32FC1: 8-bit 4-channel, 16-bit 2-channel, or 32-bit 1-channel
case 6: flipHoriz_c3<v_uint16>(src, sstep, dst, dstep, size); return; // CV_16UC3, CV_16SC3: 16-bit, 3 channels
case 8: flipHoriz_single<v_uint64>(src, sstep, dst, dstep, size); return; // CV_16UC4, CV_32SC2, CV_32FC2, CV_64FC1: 16-bit 4-channel, 32-bit 2-channel, or 64-bit 1-channel
case 12: flipHoriz_c3<v_uint32>(src, sstep, dst, dstep, size); return; // CV_32SC3, CV_32FC3: 32-bit, 3 channels
case 16: flipHoriz_vlanes_dispatch<16>(src, sstep, dst, dstep, size); return; // CV_32SC4, CV_32FC4, CV_64FC2: 32-bit 4-channel or 64-bit 2-channel
case 1: // CV_8UC1: 8-bit, 1 channel
flipHoriz_single<v_uint8>(src, sstep, dst, dstep, size); return;
case 2: // CV_8UC2, CV_16UC1: 8-bit 2-channel or 16-bit 1-channel
if (isAligned<2>(alignMark)) { flipHoriz_single<v_uint16>(src, sstep, dst, dstep, size); return; }
break;
case 3: // CV_8UC3: 8-bit, 3 channels
flipHoriz_c3<v_uint8>(src, sstep, dst, dstep, size); return;
case 4: // CV_8UC4, CV_16UC2, CV_32SC1, CV_32FC1: 8-bit 4-channel, 16-bit 2-channel, or 32-bit 1-channel
if (isAligned<4>(alignMark)) { flipHoriz_single<v_uint32>(src, sstep, dst, dstep, size); return; }
break;
case 6: // CV_16UC3, CV_16SC3: 16-bit, 3 channels
if (isAligned<2>(alignMark)) { flipHoriz_c3<v_uint16>(src, sstep, dst, dstep, size); return; }
break;
case 8: // CV_16UC4, CV_32SC2, CV_32FC2, CV_64FC1: 16-bit 4-channel, 32-bit 2-channel, or 64-bit 1-channel
if (isAligned<8>(alignMark)) { flipHoriz_single<v_uint64>(src, sstep, dst, dstep, size); return; }
break;
case 12: // CV_32SC3, CV_32FC3: 32-bit, 3 channels
if (isAligned<4>(alignMark)) { flipHoriz_c3<v_uint32>(src, sstep, dst, dstep, size); return; }
break;
case 16: // CV_32SC4, CV_32FC4, CV_64FC2: 32-bit 4-channel or 64-bit 2-channel
flipHoriz_vlanes_dispatch<16>(src, sstep, dst, dstep, size); return; // byte-wise access only
#if CV_SIMD128
case 24: flipHoriz_24(src, sstep, dst, dstep, size); return; // CV_64FC3: 64-bit, 3 channels
case 24: // CV_64FC3: 64-bit, 3 channels
if (isAligned<8>(alignMark)) { flipHoriz_24(src, sstep, dst, dstep, size); return; }
break;
#endif
case 32: flipHoriz_vlanes_dispatch<32>(src, sstep, dst, dstep, size); return; // CV_64FC4: 64-bit, 4 channels
case 32: // CV_64FC4: 64-bit, 4 channels
flipHoriz_vlanes_dispatch<32>(src, sstep, dst, dstep, size); return; // byte-wise access only
default:
break; // Fall through to generic implementation
}