> On 03/04/2024 18:20, Lorenzo Bianconi wrote:
> > Introduce EN7581 clock support to clk-en7523 driver.
> >
> > Tested-by: Zhengping Zhang <zhengping.zhang@xxxxxxxxxx>
> > Signed-off-by: Lorenzo Bianconi <lorenzo@xxxxxxxxxx>
>
>
> > + return 0;
> > +}
> > +
> > static int en7523_clk_probe(struct platform_device *pdev)
> > {
> > struct device_node *node = pdev->dev.of_node;
> > @@ -306,6 +413,12 @@ static int en7523_clk_probe(struct platform_device *pdev)
> > if (IS_ERR(np_base))
> > return PTR_ERR(np_base);
> >
> > + if (of_device_is_compatible(node, "airoha,en7581-scu")) {
>
> Having matching and compatible comparisons inside various code is
> discouraged. Does not scale. Use driver/match data to store some sort of
> flags and check for the flag or some other parameter. The best if
> compatible appears once and only once: in of_device_id.
ack, I will fix it.
Regards,
Lorenzo
>
> > + r = en7581_clk_hw_init(pdev, base, np_base);
> > + if (r)
> > + return r;
> > + }
> > +
> > clk_data = devm_kzalloc(&pdev->dev,
> > struct_size(clk_data, hws, EN7523_NUM_CLOCKS),
> > GFP_KERNEL);
> > @@ -329,8 +442,15 @@ static const struct clk_ops en7523_pcie_ops = {
> > .unprepare = en7523_pci_unprepare,
> > };
> >
> > +static const struct clk_ops en7581_pcie_ops = {
> > + .is_enabled = en7581_pci_is_enabled,
> > + .prepare = en7581_pci_prepare,
> > + .unprepare = en7581_pci_unprepare,
> > +};
> > +
> > static const struct of_device_id of_match_clk_en7523[] = {
> > { .compatible = "airoha,en7523-scu", .data = &en7523_pcie_ops },
> > + { .compatible = "airoha,en7581-scu", .data = &en7581_pcie_ops },
> > { /* sentinel */ }
> > };
> >
>
> Best regards,
> Krzysztof
>
Attachment:
signature.asc
Description: PGP signature
