Re: [PATCH 0/2] "pcie_aux" clock for i.MX8MQ

From: Lucas Stach
Date: Tue Feb 19 2019 - 07:46:40 EST


Hi Lorenzo,

Am Dienstag, den 19.02.2019, 12:42 +0000 schrieb Lorenzo Pieralisi:
> On Mon, Feb 11, 2019 at 05:51:06PM -0800, Andrey Smirnov wrote:
> > Lorenzo:
> >
> > This small series adds code to control "pcie_aux" clock. This is an
> > oversight from original submission [pcie-imx8mq-v7], which was only
> > discovered once I submitted an RFC for corresponding DT changes going
> > via i.MX tree [imx-dt-rfc].
> >
> > Thanks,
> > Andrey Smirnov
> >
> > [imx-dt-rfc] https://lore.kernel.org/lkml/20190131204333.31846-1-andrew.smirnov@xxxxxxxxx
> > [pcie-imx8mq-v7] https://lore.kernel.org/lkml/20190202001523.12517-1-andrew.smirnov@xxxxxxxxx
> >
> > Andrey Smirnov (2):
> > dt-bindings: imx6q-pcie: Add "pcie_aux" clock for imx8mq
> > PCI: imx6: Add code to request/control "pcie_aux" clock for i.MX8MQ
> >
> > .../devicetree/bindings/pci/fsl,imx6q-pcie.txt | 4 ++++
> > drivers/pci/controller/dwc/pci-imx6.c | 16 ++++++++++++++++
> > 2 files changed, 20 insertions(+)
>
> Hi Andrey,
>
> I have applied it to pci/dwc for v5.1, however it looks like it would
> break the driver with an old dts - I assume that's expected but let
> me know if there is a better way to handle this.

There is no upstream DT using the imx8mq binding, yet. We've actually
noticed the issue due to the DT patches adding the PCIe nodes and those
2 patches fix the driver _before_ we introduce any DT using it. So I
think it's okay not to worry about backwards compatibility here.

Regards,
Lucas