In case of CVT reduced blanking, pixel clock calculation does not use h period estimates, it rather directly uses refresh rate and total vertical lines. This difference can lead to a minor mismatch between the pixel clocks calculated by v4l2-utils and the standards spreadsheet.
Cc: Hans Verkuil <[email protected]> Signed-off-by: Prashant Laddha <[email protected]> --- utils/v4l2-ctl/v4l2-ctl-modes.cpp | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/utils/v4l2-ctl/v4l2-ctl-modes.cpp b/utils/v4l2-ctl/v4l2-ctl-modes.cpp index d65cd75..7768998 100644 --- a/utils/v4l2-ctl/v4l2-ctl-modes.cpp +++ b/utils/v4l2-ctl/v4l2-ctl-modes.cpp @@ -139,6 +139,7 @@ bool calc_cvt_modeline(int image_width, int image_height, int active_h_pixel; int active_v_lines; int total_h_pixel; + int total_v_lines; int h_blank; int v_blank; @@ -229,6 +230,10 @@ bool calc_cvt_modeline(int image_width, int image_height, h_bp = h_blank / 2; h_fp = h_blank - h_bp - h_sync; + + pixel_clock = ((long long)total_h_pixel * HV_FACTOR * 1000000) + / h_period; + pixel_clock -= pixel_clock % CVT_PXL_CLK_GRAN; } else { /* Reduced blanking */ @@ -247,11 +252,12 @@ bool calc_cvt_modeline(int image_width, int image_height, if (vbi_lines < (CVT_RB_V_FPORCH + v_sync + CVT_MIN_V_BPORCH)) vbi_lines = CVT_RB_V_FPORCH + v_sync + CVT_MIN_V_BPORCH; - total_h_pixel = active_h_pixel + CVT_RB_H_BLANK; - h_blank = CVT_RB_H_BLANK; v_blank = vbi_lines; + total_h_pixel = active_h_pixel + h_blank; + total_v_lines = active_v_lines + v_blank; + h_sync = CVT_RB_H_SYNC; h_bp = h_blank / 2; @@ -259,11 +265,11 @@ bool calc_cvt_modeline(int image_width, int image_height, v_fp = CVT_RB_V_FPORCH; v_bp = v_blank - v_fp - v_sync; - } - pixel_clock = ((long long)total_h_pixel * HV_FACTOR * 1000000) - / h_period; - pixel_clock -= pixel_clock % CVT_PXL_CLK_GRAN; + pixel_clock = v_refresh * total_h_pixel * + (2 * total_v_lines + interlace) / 2; + pixel_clock -= pixel_clock % CVT_PXL_CLK_GRAN; + } cvt->standards = V4L2_DV_BT_STD_CVT; -- 1.9.1 -- To unsubscribe from this list: send the line "unsubscribe linux-media" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
