diff mbox

v4l2-dv-timing: avoid rounding twice in gtf hblank calc

Message ID 1433492011-28939-1-git-send-email-prladdha@cisco.com (mailing list archive)
State New, archived
Headers show

Commit Message

Prashant Laddha June 5, 2015, 8:13 a.m. UTC 
Currently, in gtf hblank calculations, the rounding is used twice,
one at intermediate division and one at final state where hblank
is rounded to nearest multiple of twice cell granularity. This
error got introduced in commit d7ed5a3, where it missed combining
the rounding step. Correcting the same in this patch.

Cc: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Prashant Laddha <prladdha@cisco.com>
---
 drivers/media/v4l2-core/v4l2-dv-timings.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
diff mbox

Patch

diff --git a/drivers/media/v4l2-core/v4l2-dv-timings.c b/drivers/media/v4l2-core/v4l2-dv-timings.c
index 7e15749..0d849fc 100644
--- a/drivers/media/v4l2-core/v4l2-dv-timings.c
+++ b/drivers/media/v4l2-core/v4l2-dv-timings.c
@@ -586,20 +586,22 @@  bool v4l2_detect_gtf(unsigned frame_height,
 
 		num = ((image_width * GTF_D_C_PRIME * (u64)hfreq) -
 		      ((u64)image_width * GTF_D_M_PRIME * 1000));
-		den = hfreq * (100 - GTF_D_C_PRIME) + GTF_D_M_PRIME * 1000;
+		den = (hfreq * (100 - GTF_D_C_PRIME) + GTF_D_M_PRIME * 1000) *
+		      (2 * GTF_CELL_GRAN);
 		h_blank = div_u64((num + (den >> 1)), den);
+		h_blank *= (2 * GTF_CELL_GRAN);
 	} else {
 		u64 num;
 		u32 den;
 
 		num = ((image_width * GTF_S_C_PRIME * (u64)hfreq) -
 		      ((u64)image_width * GTF_S_M_PRIME * 1000));
-		den = hfreq * (100 - GTF_S_C_PRIME) + GTF_S_M_PRIME * 1000;
+		den = (hfreq * (100 - GTF_S_C_PRIME) + GTF_S_M_PRIME * 1000) *
+		      (2 * GTF_CELL_GRAN);
 		h_blank = div_u64((num + (den >> 1)), den);
+		h_blank *= (2 * GTF_CELL_GRAN);
 	}
 
-	h_blank = ((h_blank + GTF_CELL_GRAN) / (2 * GTF_CELL_GRAN)) *
-		  (2 * GTF_CELL_GRAN);
 	frame_width = image_width + h_blank;
 
 	pix_clk = (image_width + h_blank) * hfreq;