mirror/opencv
1
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2025-07-26 07:07:37 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Merge pull request #26863 from shyama7004:minor-change

Browse Source 
minor change
This commit is contained in:
Alexander Smorkalov 2025-02-03 08:37:57 +03:00 committed by GitHub
commit 3356b36d72
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
modules/imgproc/src/sumpixels.avx512_skx.hpp
View File

@ -385,7 +385,7 @@ __m512d IntegralCalculator < 3 > ::calculate_integral(const __m512i src_longs, c
// shifts data left by 3 and 6 qwords(lanes) and gets rolling sum in all lanes // shifts data left by 3 and 6 qwords(lanes) and gets rolling sum in all lanes
// Vertical LANES: 76543210 // Vertical LANES: 76543210
// src_longs : HGFEDCBA // src_longs : HGFEDCBA
// shit3lanes : + EDCBA // shift3lanes : + EDCBA
// shift6lanes : + BA // shift6lanes : + BA
// carry_over_idxs : + 65765765 (index position of result from previous iteration) // carry_over_idxs : + 65765765 (index position of result from previous iteration)
// = integral // = integral
@ -418,7 +418,7 @@ __m512d IntegralCalculator < 4 > ::calculate_integral(const __m512i src_longs, c
// shifts data left by 3 and 6 qwords(lanes) and gets rolling sum in all lanes // shifts data left by 3 and 6 qwords(lanes) and gets rolling sum in all lanes
// Vertical LANES: 76543210 // Vertical LANES: 76543210
// src_longs : HGFEDCBA // src_longs : HGFEDCBA
// shit4lanes : + DCBA // shift4lanes : + DCBA
// carry_over_idxs : + 76547654 (index position of result from previous iteration) // carry_over_idxs : + 76547654 (index position of result from previous iteration)
// = integral // = integral
__m512i shifted4lanes = _mm512_maskz_expand_epi64(0xF0, src_longs); __m512i shifted4lanes = _mm512_maskz_expand_epi64(0xF0, src_longs);
@ -464,6 +464,7 @@ void calculate_integral_avx512(const uchar *src, size_t _srcstep,
case 4: { case 4: {
IntegralCalculator< 4 > calculator; IntegralCalculator< 4 > calculator;
calculator.calculate_integral_avx512(src, _srcstep, sum, _sumstep, sqsum, _sqsumstep, width, height); calculator.calculate_integral_avx512(src, _srcstep, sum, _sumstep, sqsum, _sqsumstep, width, height);
break;
} }
} }
} }