[PATCH v2 1/4] dt-bindings: imx6q-pcie: Add reg-name "dbi2" and "atu" for i.MX8M PCIe Endpoint

[Hongxing Zhu]

> -----Original Message-----
> From: Conor Dooley <conor at kernel.org>
> Sent: 2024年7月25日 0:24
> To: Krzysztof Kozlowski <krzk at kernel.org>
> Cc: Hongxing Zhu <hongxing.zhu at nxp.com>; robh at kernel.org;
> krzk+dt at kernel.org; conor+dt at kernel.org; shawnguo at kernel.org;
> l.stach at pengutronix.de; devicetree at vger.kernel.org; linux-pci at vger.kernel.org;
> linux-arm-kernel at lists.infradead.org; linux-kernel at vger.kernel.org;
> kernel at pengutronix.de; imx at lists.linux.dev
> Subject: Re: [PATCH v2 1/4] dt-bindings: imx6q-pcie: Add reg-name "dbi2" and
> "atu" for i.MX8M PCIe Endpoint
> 
> On Wed, Jul 24, 2024 at 08:32:34AM +0200, Krzysztof Kozlowski wrote:
> > On 24/07/2024 08:26, Hongxing Zhu wrote:
> > >> -----Original Message-----
> > >> From: Krzysztof Kozlowski <krzk at kernel.org>
> > >> Sent: 2024年7月24日 14:07
> > >> To: Hongxing Zhu <hongxing.zhu at nxp.com>; robh at kernel.org;
> > >> krzk+dt at kernel.org; conor+dt at kernel.org; shawnguo at kernel.org;
> > >> l.stach at pengutronix.de
> > >> Cc: devicetree at vger.kernel.org; linux-pci at vger.kernel.org;
> > >> linux-arm-kernel at lists.infradead.org; linux-kernel at vger.kernel.org;
> > >> kernel at pengutronix.de; imx at lists.linux.dev
> > >> Subject: Re: [PATCH v2 1/4] dt-bindings: imx6q-pcie: Add reg-name
> > >> "dbi2" and "atu" for i.MX8M PCIe Endpoint
> > >>
> > >> On 24/07/2024 05:03, Richard Zhu wrote:
> > >>> Add reg-name: "dbi2", "atu" for i.MX8M PCIe Endpoint.
> > >>
> > >> This we see in the diff. What I do not see is why? Hardware changed? How
> come?
> > >>
> > > For i.MX8M PCIe EP, the dbi2 and atu address are pre-defined in the driver.
> > > This method is not good.
> > > In commit b7d67c6130ee ("PCI: imx6: Add iMX95 Endpoint (EP)
> > > support"), Frank suggests to fetch the dbi2 and atu from DT directly.
> > > This series is preparation to do that for i.MX8M PCIe EP.
> >
> > This all must be explained in commit msg.
> >
> > Anyway, this will be an ABI break, so explain exactly why it is OK to
> > break the ABI.
> 
> And the driver needs to be written in such a way that if only two reg properties are
> provided, it falls back to the old method of acquiring the two new reg regions. I
> didn't see a driver patch on v1, so I missed that this is what was actually the plan - I
> thought that this was just adding two missing regions.
Hi Corner:
Thanks for your comments.
Yes, the current driver just uses this way to handle dbi2 and atu base
 address assignments.
If only two reg properties are provided, driver would falls back to the
 old method.

Best Regards
Richard Zhu