Hi Andrew, On mar., févr. 27 2018, Andrew Lunn <and...@lunn.ch> wrote:
> On Tue, Feb 27, 2018 at 11:24:02AM +0100, Gregory CLEMENT wrote: >> Hi Andrew, >> >> On jeu., févr. 22 2018, Andrew Lunn <and...@lunn.ch> wrote: >> >> > Not all boards using the mv88e6xxx switches have the interrupt output >> > connected to a GPIO. On these boards phylib has to poll the PHYs, >> > rather than use interrupts. Have the driver poll the interrupt status >> > register, which is more efficient than having phylib do it. And it >> > enables other switch interrupts to be services. >> > >> > The Armada 370RD is such a board without a interrupt GPIO. Now that >> > interrupts work, wire up the PHYs to make use if them. >> > >> > Gregory: Are you O.K. for the second patch to go through netdev? >> >> Why do you need that the second patch to go through netdev. Is there any >> dependency between the 2 patches? >> >> If it is the case does it means that an new kernel won't work with an >> old device tree? > > Hi Gregory > > There is a runtime dependency between the two. A new device tree blob > will not run on an old kernel. So if you take the second patch alone > via mvebu, the PHYs will stop working, no link up reported. > > But an old blob will run on a new kernel. Backwards compatibility is > maintained. Ok so you wanted to keep the kernel bisctable. So I'm fine with having this patch in netdev (and I think it is alrdeay the case). Gregory > > Andrew -- Gregory Clement, Bootlin (formerly Free Electrons) Embedded Linux and Kernel engineering http://bootlin.com
