[PATCH] media: s5p-mfc fix vidioc_g_crop() to return crop info.
From: Hans Verkuil <hidden>
Date: 2016-06-30 06:41:26
Also in:
linux-media, lkml
On 06/30/2016 01:33 AM, Shuah Khan wrote:
quoted hunk ↗ jump to hunk
Fix vidioc_g_crop() to report crop information irrepective of ctx state. g_crop is expected to return crop information as long as the passed in v4l2_crop type field is vV4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE. Signed-off-by: Shuah Khan <redacted> --- drivers/media/platform/s5p-mfc/s5p_mfc_dec.c | 9 +++------ 1 file changed, 3 insertions(+), 6 deletions(-)diff --git a/drivers/media/platform/s5p-mfc/s5p_mfc_dec.c b/drivers/media/platform/s5p-mfc/s5p_mfc_dec.c index a01a373..4ace9e1 100644 --- a/drivers/media/platform/s5p-mfc/s5p_mfc_dec.c +++ b/drivers/media/platform/s5p-mfc/s5p_mfc_dec.c@@ -774,12 +774,9 @@ static int vidioc_g_crop(struct file *file, void *priv, struct s5p_mfc_dev *dev = ctx->dev; u32 left, right, top, bottom; - if (ctx->state != MFCINST_HEAD_PARSED && - ctx->state != MFCINST_RUNNING && ctx->state != MFCINST_FINISHING - && ctx->state != MFCINST_FINISHED) { - mfc_err("Cannont set crop\n"); - return -EINVAL; - }
Well, without this information all you can do is to return a fake rectangle. The decoder needs to have parsed the header to know the size.
+ if (cr->type != V4L2_BUF_TYPE_VIDEO_OUTPUT_MPLANE)
Actually, this should be V4L2_BUF_TYPE_VIDEO_OUTPUT. See the spec. I actually wonder why there is a g_crop at all since without a s_crop it doesn't provide any useful information. And if it is needed, then g_selection is the right callback since g_crop for output provides the compose rectangle, not the crop rectangle (one of the reasons we introduced g/s_selection). Regards, Hans
+ return -EINVAL;
+
if (ctx->src_fmt->fourcc == V4L2_PIX_FMT_H264) {
left = s5p_mfc_hw_call(dev->mfc_ops, get_crop_info_h, ctx);
right = left >> S5P_FIMV_SHARED_CROP_RIGHT_SHIFT;