Hi Fabio, On Tue, Jan 05, 2021 at 11:32:29AM -0300, Fabio Estevam wrote: > On Tue, Jan 5, 2021 at 10:45 AM Fabio Estevam <festevam@xxxxxxxxx> wrote: > > > Just tested against next-20210105 and the original warning happens and > > csi is no longer probed. > > > > I am using the same dtb that worked on 5.4.84. > > > > It looks like we have a regression. > > And here is a fix that allows csi to probe: > https://pastebin.com/raw/g6ijDf2N > > Makes sense? yup. > > There is another error though: I do not see the message below as seen > on 5.4 kernel: > [ 10.690711] imx-media: ov5640 1-003c:0 -> csi:0 > > And the same pipeline that worked with 5.4 does not work with linux-next: > > # media-ctl -l "'ov5640 1-003c':0 -> 'csi':0[1]" > media-ctl -l "'csi':1 -> 'csi capture':0[1]" > media-ctl -v -V "'ov5640 1-003c':0 [fmt:UYVY8_2X8/320x240 field:none]" > > # gst-launch-1.0 -v v4l2src device=/dev/video1 ! v4l2convert ! fbdevsink > Setting pipeline to PAUSED ... > Pipeline is live and does not need PREROLL ... > Pipeline is PREROLLED ... > Setting pipeline to PLAYING ... > New clock: GstSystemClock > /GstPipeline:pipeline0/GstV4l2Src:v4l2src0.GstPad:src: caps = > video/x-raw, format=(string)UYVY, width=(int)320, height=(int)240, > framerate=(fraction)30000/1001, interlace-mode=(string)progressive, > colorim > etry=(string)1:4:7:1 > /GstPipeline:pipeline0/v4l2convert:v4l2convert0.GstPad:src: caps = > video/x-raw, format=(string)BGRx, width=(int)320, height=(int)240, > framerate=(fraction)30000/1001, interlace-mode=(string)progressive > /GstPipeline:pipeline0/GstFBDEVSink:fbdevsink0.GstPad:sink: caps = > video/[ 421.495561] alloc_contig_range: [9c480, 9c4a6) PFNs busy > x-raw, format=(string)BGRx, width=(int)320, height=(int)240, fra[ > 421.504399] alloc_contig_range: [9c480, 9c4a6) PFNs busy > merate=(fraction)30000/1001, interlace-mode=(string)progressive > /GstPipeline:pipeline0/v4l2convert:v4l2convert0.GstPad:sink: c[ > 421.520989] alloc_contig_range: [9c480, 9c4a6) PFNs busy > aps = video/x-raw, format=(string)UYVY, width=(int)320, height=([ > 421.533523] csi: pipeline start failed with -19 > int)240, framerate=(fraction)30000/1001, > interlace-mode=(string)progressive, colorimetry=(string)1:4:7:1 > ERROR: from element /GstPipeline:pipeline0/GstV4l2Src:v4l2src0: Failed > to allocate required memory. > Additional debug info: > ../sys/v4l2/gstv4l2src.c(659): gst_v4l2src_decide_allocation (): > /GstPipeline:pipeline0/GstV4l2Src:v4l2src0: > Buffer pool activation failed > Execution ended after 0:00:00.106613417 > Setting pipeline to NULL ... > Freeing pipeline ... > # > > Any ideas? can you see if the following patch make it work again? 8<---------------------------------------------------- diff --git a/drivers/staging/media/imx/imx7-media-csi.c b/drivers/staging/media/imx/imx7-media-csi.c index a3f3df901704..fa8db9f1cfc8 100644 --- a/drivers/staging/media/imx/imx7-media-csi.c +++ b/drivers/staging/media/imx/imx7-media-csi.c @@ -499,6 +499,7 @@ static int imx7_csi_pad_link_validate(struct v4l2_subdev *sd, struct v4l2_subdev_format *sink_fmt) { struct imx7_csi *csi = v4l2_get_subdevdata(sd); + struct media_entity *src; struct media_pad *pad; int ret; @@ -509,11 +510,21 @@ static int imx7_csi_pad_link_validate(struct v4l2_subdev *sd, if (!csi->src_sd) return -EPIPE; + src = &csi->src_sd->entity; + + /* + * if the source is neither a mux or csi2 get the one directly upstream + * from this csi + */ + if (src->function != MEDIA_ENT_F_VID_IF_BRIDGE && + src->function != MEDIA_ENT_F_VID_MUX) + src = &csi->sd.entity; + /* * find the entity that is selected by the CSI mux. This is needed * to distinguish between a parallel or CSI-2 pipeline. */ - pad = imx_media_pipeline_pad(&csi->src_sd->entity, 0, 0, true); + pad = imx_media_pipeline_pad(src, 0, 0, true); if (!pad) return -ENODEV; > > Thanks
