On Thu, 2020-05-28 at 15:39 +0200, Matthias Brugger wrote:
>
> On 28/05/2020 11:57, EastL wrote:
> > This patch fixes mediatek-cqdma compatible to common.
> >
> > Signed-off-by: EastL <EastL.Lee@xxxxxxxxxxxx>
> > ---
> > drivers/dma/mediatek/mtk-cqdma.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/dma/mediatek/mtk-cqdma.c b/drivers/dma/mediatek/mtk-cqdma.c
> > index 905bbcb..bca7118 100644
> > --- a/drivers/dma/mediatek/mtk-cqdma.c
> > +++ b/drivers/dma/mediatek/mtk-cqdma.c
> > @@ -544,7 +544,7 @@ static void mtk_cqdma_hw_deinit(struct mtk_cqdma_device *cqdma)
> > }
> >
> > static const struct of_device_id mtk_cqdma_match[] = {
> > - { .compatible = "mediatek,mt6765-cqdma" },
> > + { .compatible = "mediatek,cqdma" },
>
> We can't just delete and old compatible. If other cqdma IP blocks are the same
> as mt6795, we should instead add entries in the binding description with
> fallback compatible. For example for mt6779 the DTS would look like this:
> compatible = "mediatek,mt6779-cqdma", "mediatek,mt6765-cqdma";
>
> This way we the kernel will take care to bind the device against the driver with
> mt7665-cqdma, but leaves us the posibillity to add any changes to the driver in
> the future if we find some bugs/features for mt6779 that are not present in mt6765.
>
> Regards,
> Matthias
>
> > { /* sentinel */ }
> > };
> > MODULE_DEVICE_TABLE(of, mtk_cqdma_match);
> >
OK, Can I add a common compatible? Like this
static const struct of_device_id mtk_cqdma_match[] = {
{ .compatible = "mediatek,mt6765-cqdma" },
{ .compatible = "mediatek,common-cqdma" },
Regards,
EastL
