From 355fd5a44a2b740292406db61f038224cc8df88d Mon Sep 17 00:00:00 2001 From: Ingo Weyrich Date: Fri, 10 Jan 2020 14:25:51 +0100 Subject: [PATCH] Capture sharpening: add vectorization hints for clang --- rtengine/capturesharpening.cc | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/rtengine/capturesharpening.cc b/rtengine/capturesharpening.cc index e5bfde555..4beb4091a 100644 --- a/rtengine/capturesharpening.cc +++ b/rtengine/capturesharpening.cc @@ -131,6 +131,9 @@ inline void gauss3x3div (float** RESTRICT src, float** RESTRICT dst, float** RES const float c00 = kernel[1][1]; for (int i = 1; i < tileSize - 1; i++) { +#ifdef __clang__ + #pragma clang loop vectorize(assume_safety) +#endif for (int j = 1; j < tileSize - 1; j++) { const float val = c11 * (src[i - 1][j - 1] + src[i - 1][j + 1] + src[i + 1][j - 1] + src[i + 1][j + 1]) + c10 * (src[i - 1][j] + src[i][j - 1] + src[i][j + 1] + src[i + 1][j]) + @@ -151,6 +154,9 @@ inline void gauss5x5div (float** RESTRICT src, float** RESTRICT dst, float** RES for (int i = 2; i < tileSize - 2; ++i) { // I tried hand written SSE code but gcc vectorizes better +#ifdef __clang__ + #pragma clang loop vectorize(assume_safety) +#endif for (int j = 2; j < tileSize - 2; ++j) { const float val = c21 * ((src[i - 2][j - 1] + src[i - 2][j + 1]) + (src[i - 1][j - 2] + src[i - 1][j + 2]) + (src[i + 1][j - 2] + src[i + 1][j + 2]) + (src[i + 2][j - 1] + src[i + 2][j + 1])) + c20 * (src[i - 2][j] + src[i][j - 2] + src[i][j + 2] + src[i + 2][j]) + @@ -177,6 +183,9 @@ inline void gauss7x7div(float** RESTRICT src, float** RESTRICT dst, float** REST for (int i = 3; i < tileSize - 3; ++i) { // I tried hand written SSE code but gcc vectorizes better +#ifdef __clang__ + #pragma clang loop vectorize(assume_safety) +#endif for (int j = 3; j < tileSize - 3; ++j) { const float val = c31 * ((src[i - 3][j - 1] + src[i - 3][j + 1]) + (src[i - 1][j - 3] + src[i - 1][j + 3]) + (src[i + 1][j - 3] + src[i + 1][j + 3]) + (src[i + 3][j - 1] + src[i + 3][j + 1])) + c30 * (src[i - 3][j] + src[i][j - 3] + src[i][j + 3] + src[i + 3][j]) + @@ -199,6 +208,9 @@ inline void gauss3x3mult(float** RESTRICT src, float** RESTRICT dst, const int t const float c00 = kernel[1][1]; for (int i = 1; i < tileSize - 1; i++) { +#ifdef __clang__ + #pragma clang loop vectorize(assume_safety) +#endif for (int j = 1; j < tileSize - 1; j++) { const float val = c11 * (src[i - 1][j - 1] + src[i - 1][j + 1] + src[i + 1][j - 1] + src[i + 1][j + 1]) + c10 * (src[i - 1][j] + src[i][j - 1] + src[i][j + 1] + src[i + 1][j]) + @@ -220,6 +232,9 @@ inline void gauss5x5mult (float** RESTRICT src, float** RESTRICT dst, const int for (int i = 2; i < tileSize - 2; ++i) { // I tried hand written SSE code but gcc vectorizes better +#ifdef __clang__ + #pragma clang loop vectorize(assume_safety) +#endif for (int j = 2; j < tileSize - 2; ++j) { const float val = c21 * ((src[i - 2][j - 1] + src[i - 2][j + 1]) + (src[i - 1][j - 2] + src[i - 1][j + 2]) + (src[i + 1][j - 2] + src[i + 1][j + 2]) + (src[i + 2][j - 1] + src[i + 2][j + 1])) + c20 * (src[i - 2][j] + src[i][j - 2] + src[i][j + 2] + src[i + 2][j]) + @@ -246,6 +261,9 @@ inline void gauss7x7mult(float** RESTRICT src, float** RESTRICT dst, const int t for (int i = 3; i < tileSize - 3; ++i) { // I tried hand written SSE code but gcc vectorizes better +#ifdef __clang__ + #pragma clang loop vectorize(assume_safety) +#endif for (int j = 3; j < tileSize - 3; ++j) { const float val = c31 * ((src[i - 3][j - 1] + src[i - 3][j + 1]) + (src[i - 1][j - 3] + src[i - 1][j + 3]) + (src[i + 1][j - 3] + src[i + 1][j + 3]) + (src[i + 3][j - 1] + src[i + 3][j + 1])) + c30 * (src[i - 3][j] + src[i][j - 3] + src[i][j + 3] + src[i + 3][j]) +