Hi,
On Thu, 2025-10-23 at 15:59 +0000, Chun-Kuang Hu wrote:
> Hi, Louis:
>
> Louis-Alexis Eyraud <[email protected]> 於 2025年10月23日
> 週四 上午10:32寫道：
> >
> > From: AngeloGioacchino Del Regno
> > <[email protected]>
> >
> > In preparation for adding a new driver for HDMIv2, for which CEC
> > is not strictly required, change the of_get_compatible_child()
> > failure error to -ENOTSUPP to be able to differentiate between
> > error conditions in mtk_hdmi_dt_parse_pdata().
> >
> > In that case, if -ENOTSUPP is returned, this driver will print
> > an informative message saying that CEC support is unavailable,
> > as the devicetree node for that was not found, but after that,
> > function mtk_hdmi_dt_parse_pdata() will not return error to
> > the caller.
> >
> > This will not change functionality of the mtk_hdmi (v1) driver
> > as that is still checking whether CEC is present and, if not,
> > will fail probing with an error saying that CEC is required
> > by HDMIv1.
> >
> > Reviewed-by: CK Hu <[email protected]>
> > Signed-off-by: AngeloGioacchino Del Regno
> > <[email protected]>
> > Signed-off-by: Louis-Alexis Eyraud
> > <[email protected]>
> > ---
> > drivers/gpu/drm/mediatek/mtk_hdmi_common.c | 9 +++++----
> > 1 file changed, 5 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/mediatek/mtk_hdmi_common.c
> > b/drivers/gpu/drm/mediatek/mtk_hdmi_common.c
> > index
> > f28e63716d7c2e2c6bdf5b28d24789bc926d2beb..989479fdfe40952762a1b6379
> > 7bf602b520dcac5 100644
> > --- a/drivers/gpu/drm/mediatek/mtk_hdmi_common.c
> > +++ b/drivers/gpu/drm/mediatek/mtk_hdmi_common.c
> > @@ -253,12 +253,11 @@ static int mtk_hdmi_get_cec_dev(struct
> > mtk_hdmi *hdmi, struct device *dev, struc
> > /* The CEC module handles HDMI hotplug detection */
> > cec_np = of_get_compatible_child(np->parent,
> > "mediatek,mt8173-cec");
> > if (!cec_np)
> > - return dev_err_probe(dev, -EINVAL, "Failed to find
> > CEC node\n");
> > + return dev_err_probe(dev, -ENOTSUPP, "Failed to
> > find CEC node\n");
>
> The checkpatch show the warning:
>
> WARNING: ENOTSUPP is not a SUSV4 error code, prefer EOPNOTSUPP
> #40: FILE: drivers/gpu/drm/mediatek/mtk_hdmi_common.c:256:
> + return dev_err_probe(dev, -ENOTSUPP, "Failed to find CEC
> node\n");
>
> I find many driver use ENOTSUPP, so I'm not sure which one is better.
> I really don't understand the description about these two error code
> [1].
> If you choose ENOTSUPP, please give me some explain.
>
> [1] https://man7.org/linux/man-pages/man3/errno.3.html
>
> Regards,
> Chun-Kuang.
>
>
I think ENOTSUPP can be replaced by EOPNOTSUPP in this patch safely.
There is only a single call of the mtk_hdmi_get_cec_dev function in the
driver and the code that checks this particular return code value in
mtk_hdmi_dt_parse_pdata function is also added by this patch.
I also don't see no other occurrence of ENOTSUPP code in mediatek-drm
and since checkpatch recommends to prefer EOPNOTSUPP over ENOTSUPP,
let's replace it in this patch (both code and commit message).
Do you want me to send a new version of the series for this change?
Regards,
Louis-Alexis
> > cec_pdev = of_find_device_by_node(cec_np);
> > if (!cec_pdev) {
> > - dev_err(hdmi->dev, "Waiting for CEC device %pOF\n",
> > - cec_np);
> > + dev_err(hdmi->dev, "Waiting for CEC device %pOF\n",
> > cec_np);
> > of_node_put(cec_np);
> > return -EPROBE_DEFER;
> > }
> > @@ -319,7 +318,9 @@ static int mtk_hdmi_dt_parse_pdata(struct
> > mtk_hdmi *hdmi, struct platform_device
> > return dev_err_probe(dev, -EINVAL, "Failed to get
> > ddc i2c adapter by node\n");
> >
> > ret = mtk_hdmi_get_cec_dev(hdmi, dev, np);
> > - if (ret)
> > + if (ret == -ENOTSUPP)
> > + dev_info(dev, "CEC support unavailable: node not
> > found\n");
> > + else if (ret)
> > return ret;
> >
> > return 0;
> >
> > --
> > 2.51.0
> >
