Hi Hans, On Wed, Jul 19, 2023 at 12:26:37PM +0200, Hans Verkuil wrote: > On 11/07/2023 14:32, Alain Volmat wrote: > > Cascade the s_stream call to the source subdev whenever the bridge > > streaming is enable / disabled. > > > > Signed-off-by: Alain Volmat <alain.volmat@xxxxxxxxxxx> > > --- > > drivers/media/i2c/st-mipid02.c | 11 +++++++++++ > > 1 file changed, 11 insertions(+) > > > > diff --git a/drivers/media/i2c/st-mipid02.c b/drivers/media/i2c/st-mipid02.c > > index 906553a28676..703d2f06f552 100644 > > --- a/drivers/media/i2c/st-mipid02.c > > +++ b/drivers/media/i2c/st-mipid02.c > > @@ -547,6 +547,13 @@ static int mipid02_stream_disable(struct mipid02_dev *bridge) > > struct i2c_client *client = bridge->i2c_client; > > int ret; > > > > + if (!bridge->s_subdev) > > + goto error; > > I'm getting this compiler warning: > > media-git/drivers/media/i2c/st-mipid02.c: In function 'mipid02_stream_disable': > media-git/drivers/media/i2c/st-mipid02.c:568:12: warning: 'ret' may be used uninitialized [-Wmaybe-uninitialized] > 568 | if (ret) > | ^ > media-git/drivers/media/i2c/st-mipid02.c:548:13: note: 'ret' was declared here > 548 | int ret; > | ^~~ > > I'm dropping this series, waiting for a v2. Indeed, this was a real issue. I added KCFLAGS=-Wmaybe-uninitialized in my build command line to spot thoses issues from now on. v2 posted with the fix. > > Regards, > > Hans Regards, Alain > > > + > > + ret = v4l2_subdev_call(bridge->s_subdev, video, s_stream, 0); > > + if (ret) > > + goto error; > > + > > /* Disable all lanes */ > > ret = mipid02_write_reg(bridge, MIPID02_CLK_LANE_REG1, 0); > > if (ret) > > @@ -633,6 +640,10 @@ static int mipid02_stream_enable(struct mipid02_dev *bridge) > > if (ret) > > goto error; > > > > + ret = v4l2_subdev_call(bridge->s_subdev, video, s_stream, 1); > > + if (ret) > > + goto error; > > + > > return 0; > > > > error: >
