All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jacek Anaszewski <j.anaszewski@samsung.com>
To: linux-media@vger.kernel.org
Cc: s.nawrocki@samsung.com, andrzej.p@samsung.com,
	Jacek Anaszewski <j.anaszewski@samsung.com>,
	Kyungmin Park <kyungmin.park@samsung.com>
Subject: [PATCH v2 6/9] s5p-jpeg: Prevent erroneous downscaling for Exynos3250 SoC
Date: Fri, 11 Jul 2014 17:19:47 +0200	[thread overview]
Message-ID: <1405091990-28567-7-git-send-email-j.anaszewski@samsung.com> (raw)
In-Reply-To: <1405091990-28567-1-git-send-email-j.anaszewski@samsung.com>

JPEG codec on Exynos3250 SoC produces broken raw image if
a JPEG is decoded to YUV420 format and downscaling by
more then 2 is applied. Prevent this by asserting downscale
ratio to 2.

Signed-off-by: Jacek Anaszewski <j.anaszewski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
---
 drivers/media/platform/s5p-jpeg/jpeg-core.c |   18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/drivers/media/platform/s5p-jpeg/jpeg-core.c b/drivers/media/platform/s5p-jpeg/jpeg-core.c
index 2491ef8..1ef004b 100644
--- a/drivers/media/platform/s5p-jpeg/jpeg-core.c
+++ b/drivers/media/platform/s5p-jpeg/jpeg-core.c
@@ -1317,12 +1317,16 @@ static int exynos4_jpeg_get_output_buffer_size(struct s5p_jpeg_ctx *ctx,
 	return w * h * fmt_depth >> 3;
 }
 
+static int exynos3250_jpeg_try_downscale(struct s5p_jpeg_ctx *ctx,
+				   struct v4l2_rect *r);
+
 static int s5p_jpeg_s_fmt(struct s5p_jpeg_ctx *ct, struct v4l2_format *f)
 {
 	struct vb2_queue *vq;
 	struct s5p_jpeg_q_data *q_data = NULL;
 	struct v4l2_pix_format *pix = &f->fmt.pix;
 	struct v4l2_ctrl *ctrl_subs;
+	struct v4l2_rect scale_rect;
 	unsigned int f_type;
 
 	vq = v4l2_m2m_get_vq(ct->fh.m2m_ctx, f->type);
@@ -1382,6 +1386,20 @@ static int s5p_jpeg_s_fmt(struct s5p_jpeg_ctx *ct, struct v4l2_format *f)
 		ct->crop_rect.width = pix->width;
 		ct->crop_rect.height = pix->height;
 	}
+
+	/*
+	 * Prevent downscaling to YUV420 format by more than 2
+	 * for Exynos3250 SoC as it produces broken raw image
+	 * in such cases.
+	 */
+	if (ct->mode == S5P_JPEG_DECODE &&
+	    f_type == FMT_TYPE_CAPTURE &&
+	    ct->jpeg->variant->version == SJPEG_EXYNOS3250 &&
+	    pix->pixelformat == V4L2_PIX_FMT_YUV420 &&
+	    ct->scale_factor > 2) {
+		scale_rect.width = ct->out_q.w / 2;
+		scale_rect.height = ct->out_q.h / 2;
+		exynos3250_jpeg_try_downscale(ct, &scale_rect);
 	}
 
 	return 0;
-- 
1.7.9.5


  parent reply	other threads:[~2014-07-11 15:20 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-11 15:19 [PATCH v2 0/9] Add support for Exynos3250 SoC to the s5p-jpeg driver Jacek Anaszewski
2014-07-11 15:19 ` [PATCH v2 1/9] s5p-jpeg: Add support for Exynos3250 SoC Jacek Anaszewski
2014-07-11 15:19 ` [PATCH v2 2/9] s5p-jpeg: return error immediately after get_byte fails Jacek Anaszewski
2014-07-11 15:19 ` [PATCH v2 3/9] s5p-jpeg: Adjust jpeg_bound_align_image to Exynos3250 needs Jacek Anaszewski
2014-07-11 15:19 ` [PATCH v2 4/9] s5p-jpeg: fix g_selection op Jacek Anaszewski
2014-07-11 15:19 ` [PATCH v2 5/9] s5p-jpeg: Assure proper crop rectangle initialization Jacek Anaszewski
2014-07-11 15:19 ` Jacek Anaszewski [this message]
2014-07-11 15:19 ` [PATCH v2 7/9] s5p-jpeg: add chroma subsampling adjustment for Exynos3250 Jacek Anaszewski
2014-07-11 15:19 ` [PATCH v2 8/9] Documentation: devicetree: Document sclk-jpeg clock for exynos3250 SoC Jacek Anaszewski
2014-07-14  9:56   ` Mark Rutland
2014-07-22 11:42     ` Sylwester Nawrocki
2014-07-22 11:48       ` Arnd Bergmann
2014-07-22 14:18         ` Sylwester Nawrocki
2014-07-22 14:44           ` Arnd Bergmann
2014-07-22 14:55             ` Sylwester Nawrocki
2014-07-11 15:19 ` [PATCH v2 9/9] ARM: dts: exynos3250: add JPEG codec device node Jacek Anaszewski
2014-07-11 16:40 ` [PATCH v2 0/9] Add support for Exynos3250 SoC to the s5p-jpeg driver Sylwester Nawrocki

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1405091990-28567-7-git-send-email-j.anaszewski@samsung.com \
    --to=j.anaszewski@samsung.com \
    --cc=andrzej.p@samsung.com \
    --cc=kyungmin.park@samsung.com \
    --cc=linux-media@vger.kernel.org \
    --cc=s.nawrocki@samsung.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.