Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1578016 > unrolled thread

[PATCH 0/4] [media] Fixes for colorspace logic in exynos-gsc and s5p-mfc drivers

Back to article view | Back to linux.kernel

Contents

  [PATCH 0/4] [media] Fixes for colorspace logic in exynos-gsc and s5p-mfc drivers Thibault Saunier <thibault.saunier@osg.samsung.com> - 2017-02-10 00:00 +0100
    [PATCH 4/4] [media] s5p-mfc: Always check and set 'v4l2_pix_format:field' field Thibault Saunier <thibault.saunier@osg.samsung.com> - 2017-02-10 00:00 +0100
      Re: [PATCH 4/4] [media] s5p-mfc: Always check and set  'v4l2_pix_format:field' field kbuild test robot <lkp@intel.com> - 2017-02-10 10:40 +0100
    [PATCH 2/4] [media] exynos-gsc: Respect userspace colorspace setting Thibault Saunier <thibault.saunier@osg.samsung.com> - 2017-02-10 00:00 +0100

#1578016 — [PATCH 0/4] [media] Fixes for colorspace logic in exynos-gsc and s5p-mfc drivers

Hello,

This patchset fixes a few issues on the colorspace logic for the exynos-gsc
and s5p-mfc drivers.

We now handle the colorspace in those drivers, and make sure to respect user setting if
possible.

We also now set the 'v4l2_pix_format:field' if userspace passed ANY, avoiding GStreamer
spamming error at us about the driver not following the standard.

Best regards,

Thibault Saunier


Javier Martinez Canillas (1):
  [media] exynos-gsc: Use 576p instead 720p as a threshold for
    colorspaces

Thibault Saunier (3):
  [media] exynos-gsc: Respect userspace colorspace setting
  [media] s5p-mfc: Set colorspace in VIDIO_{G,TRY}_FMT
  [media] s5p-mfc: Always check and set 'v4l2_pix_format:field' field

 drivers/media/platform/exynos-gsc/gsc-core.c | 22 +++++++++++----
 drivers/media/platform/s5p-mfc/s5p_mfc_dec.c | 42 ++++++++++++++++++++++++++--
 2 files changed, 57 insertions(+), 7 deletions(-)

-- 
2.11.1

[toc] | [next] | [standalone]

#1578017 — [PATCH 4/4] [media] s5p-mfc: Always check and set 'v4l2_pix_format:field' field

It is required by the standard that the field order is set by the
driver.

Signed-off-by: Thibault Saunier <thibault.saunier@osg.samsung.com>
---
 drivers/media/platform/s5p-mfc/s5p_mfc_dec.c | 23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

diff --git a/drivers/media/platform/s5p-mfc/s5p_mfc_dec.c b/drivers/media/platform/s5p-mfc/s5p_mfc_dec.c
index 960d6c7052bd..dfb21b4aee10 100644
--- a/drivers/media/platform/s5p-mfc/s5p_mfc_dec.c
+++ b/drivers/media/platform/s5p-mfc/s5p_mfc_dec.c
@@ -345,7 +345,6 @@ static int vidioc_g_fmt(struct file *file, void *priv, struct v4l2_format *f)
 		   rectangle. */
 		pix_mp->width = ctx->buf_width;
 		pix_mp->height = ctx->buf_height;
-		pix_mp->field = V4L2_FIELD_NONE;
 		pix_mp->num_planes = 2;
 		/* Set pixelformat to the format in which MFC
 		   outputs the decoded frame */
@@ -369,7 +368,6 @@ static int vidioc_g_fmt(struct file *file, void *priv, struct v4l2_format *f)
 		   so width and height have no meaning */
 		pix_mp->width = 0;
 		pix_mp->height = 0;
-		pix_mp->field = V4L2_FIELD_NONE;
 		pix_mp->plane_fmt[0].bytesperline = ctx->dec_src_buf_size;
 		pix_mp->plane_fmt[0].sizeimage = ctx->dec_src_buf_size;
 		pix_mp->pixelformat = ctx->src_fmt->fourcc;
@@ -379,6 +377,14 @@ static int vidioc_g_fmt(struct file *file, void *priv, struct v4l2_format *f)
 		mfc_debug(2, "%s-- with error\n", __func__);
 		return -EINVAL;
 	}
+
+	if (pix_mp->field == V4L2_FIELD_ANY) {
+		pix_mp->field = V4L2_FIELD_NONE;
+	} else if (pix_mp->field != V4L2_FIELD_NONE) {
+		mfc_err("Not supported field order(%d)\n", pix_mp->field);
+		return -EINVAL;
+	}
+
 	mfc_debug_leave();
 	return 0;
 }
@@ -389,6 +395,19 @@ static int vidioc_try_fmt(struct file *file, void *priv, struct v4l2_format *f)
 	struct s5p_mfc_dev *dev = video_drvdata(file);
 	struct v4l2_pix_format_mplane *pix_mp = &f->fmt.pix_mp;
 	struct s5p_mfc_fmt *fmt;
+	enum v4l2_field field;
+
+	field = f->fmt.pix.field;
+	if (field == V4L2_FIELD_ANY) {
+		field = V4L2_FIELD_NONE;
+	} else if (V4L2_FIELD_NONE != field) {
+		mfc_debug("Not supported field order(%d)\n", pix_mp->field);
+		return -EINVAL;
+	}
+
+	/* V4L2 specification suggests the driver corrects the format struct
+	 * if any of the dimensions is unsupported */
+	f->fmt.pix.field = field;
 
 	mfc_debug(2, "Type is %d\n", f->type);
 	if (f->type == V4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE) {
-- 
2.11.1

[toc] | [prev] | [next] | [standalone]

#1578332 — Re: [PATCH 4/4] [media] s5p-mfc: Always check and set 'v4l2_pix_format:field' field

Hi Thibault,

[auto build test WARNING on linuxtv-media/master]
[also build test WARNING on v4.10-rc7 next-20170209]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]

url:    https://github.com/0day-ci/linux/commits/Thibault-Saunier/exynos-gsc-Use-576p-instead-720p-as-a-threshold-for-colorspaces/20170210-113658
base:   git://linuxtv.org/media_tree.git master
reproduce:
        # apt-get install sparse
        make ARCH=x86_64 allmodconfig
        make C=1 CF=-D__CHECK_ENDIAN__


sparse warnings: (new ones prefixed by >>)

   include/linux/compiler.h:253:8: sparse: attribute 'no_sanitize_address': unknown attribute
   drivers/media/platform/s5p-mfc/s5p_mfc_dec.c:404:17: sparse: Expected ) in function call
   drivers/media/platform/s5p-mfc/s5p_mfc_dec.c:404:17: sparse: got pix_mp
>> drivers/media/platform/s5p-mfc/s5p_mfc_dec.c:404:17: sparse: incompatible types for operation (>=)
   drivers/media/platform/s5p-mfc/s5p_mfc_dec.c:404:17:    left side has type int extern [signed] [addressable] [toplevel] mfc_debug_level
   drivers/media/platform/s5p-mfc/s5p_mfc_dec.c:404:17:    right side has type char static *<noident>
   In file included from drivers/media/platform/s5p-mfc/s5p_mfc_dec.c:28:0:
   drivers/media/platform/s5p-mfc/s5p_mfc_dec.c: In function 'vidioc_try_fmt':
   drivers/media/platform/s5p-mfc/s5p_mfc_debug.h:25:23: warning: comparison between pointer and integer
      if (mfc_debug_level >= level)   \
                          ^
   drivers/media/platform/s5p-mfc/s5p_mfc_dec.c:404:3: note: in expansion of macro 'mfc_debug'
      mfc_debug("Not supported field order(%d)\n", pix_mp->field);
      ^~~~~~~~~
   drivers/media/platform/s5p-mfc/s5p_mfc_debug.h:25:23: warning: comparison with string literal results in unspecified behavior [-Waddress]
      if (mfc_debug_level >= level)   \
                          ^
   drivers/media/platform/s5p-mfc/s5p_mfc_dec.c:404:3: note: in expansion of macro 'mfc_debug'
      mfc_debug("Not supported field order(%d)\n", pix_mp->field);
      ^~~~~~~~~
   drivers/media/platform/s5p-mfc/s5p_mfc_dec.c:404:48: error: expected ')' before 'pix_mp'
      mfc_debug("Not supported field order(%d)\n", pix_mp->field);
                                                   ^
   drivers/media/platform/s5p-mfc/s5p_mfc_debug.h:26:32: note: in definition of macro 'mfc_debug'
       printk(KERN_DEBUG "%s:%d: " fmt, \
                                   ^~~
   In file included from include/linux/printk.h:6:0,
                    from include/linux/kernel.h:13,
                    from include/linux/clk.h:16,
                    from drivers/media/platform/s5p-mfc/s5p_mfc_dec.c:14:
   include/linux/kern_levels.h:4:18: warning: format '%s' expects a matching 'char *' argument [-Wformat=]
    #define KERN_SOH "\001"  /* ASCII Start Of Header */
                     ^
   include/linux/kern_levels.h:14:20: note: in expansion of macro 'KERN_SOH'
    #define KERN_DEBUG KERN_SOH "7" /* debug-level messages */
                       ^~~~~~~~
   drivers/media/platform/s5p-mfc/s5p_mfc_debug.h:26:11: note: in expansion of macro 'KERN_DEBUG'
       printk(KERN_DEBUG "%s:%d: " fmt, \
              ^~~~~~~~~~
   drivers/media/platform/s5p-mfc/s5p_mfc_dec.c:404:3: note: in expansion of macro 'mfc_debug'
      mfc_debug("Not supported field order(%d)\n", pix_mp->field);
      ^~~~~~~~~
   include/linux/kern_levels.h:4:18: warning: format '%d' expects a matching 'int' argument [-Wformat=]
    #define KERN_SOH "\001"  /* ASCII Start Of Header */
                     ^
   include/linux/kern_levels.h:14:20: note: in expansion of macro 'KERN_SOH'
    #define KERN_DEBUG KERN_SOH "7" /* debug-level messages */
                       ^~~~~~~~
   drivers/media/platform/s5p-mfc/s5p_mfc_debug.h:26:11: note: in expansion of macro 'KERN_DEBUG'
       printk(KERN_DEBUG "%s:%d: " fmt, \
              ^~~~~~~~~~
   drivers/media/platform/s5p-mfc/s5p_mfc_dec.c:404:3: note: in expansion of macro 'mfc_debug'
      mfc_debug("Not supported field order(%d)\n", pix_mp->field);
      ^~~~~~~~~

vim +404 drivers/media/platform/s5p-mfc/s5p_mfc_dec.c

   388		mfc_debug_leave();
   389		return 0;
   390	}
   391	
   392	/* Try format */
   393	static int vidioc_try_fmt(struct file *file, void *priv, struct v4l2_format *f)
   394	{
   395		struct s5p_mfc_dev *dev = video_drvdata(file);
   396		struct v4l2_pix_format_mplane *pix_mp = &f->fmt.pix_mp;
   397		struct s5p_mfc_fmt *fmt;
   398		enum v4l2_field field;
   399	
   400		field = f->fmt.pix.field;
   401		if (field == V4L2_FIELD_ANY) {
   402			field = V4L2_FIELD_NONE;
   403		} else if (V4L2_FIELD_NONE != field) {
 > 404			mfc_debug("Not supported field order(%d)\n", pix_mp->field);
   405			return -EINVAL;
   406		}
   407	
   408		/* V4L2 specification suggests the driver corrects the format struct
   409		 * if any of the dimensions is unsupported */
   410		f->fmt.pix.field = field;
   411	
   412		mfc_debug(2, "Type is %d\n", f->type);

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation

[toc] | [prev] | [next] | [standalone]

#1578018 — [PATCH 2/4] [media] exynos-gsc: Respect userspace colorspace setting

If the colorspace is specified by userspace we should respect
it and not reset it ourself if we can support it.

Signed-off-by: Thibault Saunier <thibault.saunier@osg.samsung.com>
---
 drivers/media/platform/exynos-gsc/gsc-core.c | 25 +++++++++++++++++--------
 1 file changed, 17 insertions(+), 8 deletions(-)

diff --git a/drivers/media/platform/exynos-gsc/gsc-core.c b/drivers/media/platform/exynos-gsc/gsc-core.c
index 2beb43401987..63bb4577827d 100644
--- a/drivers/media/platform/exynos-gsc/gsc-core.c
+++ b/drivers/media/platform/exynos-gsc/gsc-core.c
@@ -445,10 +445,14 @@ int gsc_try_fmt_mplane(struct gsc_ctx *ctx, struct v4l2_format *f)
 
 	pix_mp->num_planes = fmt->num_planes;
 
-	if (pix_mp->width > 720 && pix_mp->height > 576) /* HD */
-		pix_mp->colorspace = V4L2_COLORSPACE_REC709;
-	else /* SD */
-		pix_mp->colorspace = V4L2_COLORSPACE_SMPTE170M;
+	if (pix_mp->colorspace != V4L2_COLORSPACE_REC709 &&
+		pix_mp->colorspace != V4L2_COLORSPACE_SMPTE170M &&
+		pix_mp->colorspace != V4L2_COLORSPACE_DEFAULT) {
+		if (pix_mp->width > 720 && pix_mp->height > 576) /* HD */
+		  pix_mp->colorspace = V4L2_COLORSPACE_REC709;
+		else /* SD */
+		  pix_mp->colorspace = V4L2_COLORSPACE_SMPTE170M;
+	  }
 
 	for (i = 0; i < pix_mp->num_planes; ++i) {
 		struct v4l2_plane_pix_format *plane_fmt = &pix_mp->plane_fmt[i];
@@ -492,12 +496,17 @@ int gsc_g_fmt_mplane(struct gsc_ctx *ctx, struct v4l2_format *f)
 	pix_mp->height		= frame->f_height;
 	pix_mp->field		= V4L2_FIELD_NONE;
 	pix_mp->pixelformat	= frame->fmt->pixelformat;
-	if (pix_mp->width > 720 && pix_mp->height > 576) /* HD */
-		pix_mp->colorspace = V4L2_COLORSPACE_REC709;
-	else /* SD */
-		pix_mp->colorspace = V4L2_COLORSPACE_SMPTE170M;
 	pix_mp->num_planes	= frame->fmt->num_planes;
 
+	if (pix_mp->colorspace != V4L2_COLORSPACE_REC709 &&
+		pix_mp->colorspace != V4L2_COLORSPACE_SMPTE170M &&
+		pix_mp->colorspace != V4L2_COLORSPACE_DEFAULT) {
+		if (pix_mp->width > 720 && pix_mp->height > 576) /* HD */
+		  pix_mp->colorspace = V4L2_COLORSPACE_REC709;
+		else /* SD */
+		  pix_mp->colorspace = V4L2_COLORSPACE_SMPTE170M;
+	  }
+
 	for (i = 0; i < pix_mp->num_planes; ++i) {
 		pix_mp->plane_fmt[i].bytesperline = (frame->f_width *
 			frame->fmt->depth[i]) / 8;
-- 
2.11.1

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web