Messages in this thread | | | From | Moritz Fischer <> | Date | Tue, 4 Sep 2018 20:27:43 -0700 | Subject | Re: [RFC/PATCH] net: nixge: Add PHYLINK support |
| |
Hi Andrew,
On Tue, Sep 4, 2018 at 6:01 PM, Andrew Lunn <andrew@lunn.ch> wrote: >> 3) I'm again not sure about the 'select PHYLINK', wouldn't >> wanna break the build again... > > Hi Moritz > > I think it is safe. PHYLINK has no stated dependencies on OF. But i > suspect it currently is pretty useless without OF.
Ok, great. Thanks!
>> @@ -1286,7 +1329,13 @@ static int nixge_probe(struct platform_device *pdev) >> priv->coalesce_count_rx = XAXIDMA_DFT_RX_THRESHOLD; >> priv->coalesce_count_tx = XAXIDMA_DFT_TX_THRESHOLD; >> >> - err = nixge_mdio_setup(priv, pdev->dev.of_node); >> + mn = of_get_child_by_name(pdev->dev.of_node, "mdio"); >> + if (!mn) { >> + dev_warn(&pdev->dev, "No \"mdio\" subnode found, defaulting to legacy\n"); >> + mn = pdev->dev.of_node; >> + } >> + >> + err = nixge_mdio_setup(priv, mn); > > I would suggest making this a patch of its own.
Yeah, will do. > Also, do you need the legacy behaviour? If there are no boards out in > the wild which this will break, just make the change.
Well all users basically use devicetree overlays that we distribute with the FPGA images as part of a filesystem update. So I suppose it wouldn't break anyone, yet... It would certainly make the code easier to read.
> Please also update the device tree binding documentation.
Yeah will do.
Thanks for the feedback,
Moritz
| |