On Sat, Oct 25, 2014 at 3:10 AM, Christian Ege <[email protected]> wrote: > > Am 24.10.2014 22:34 schrieb "Tim Tisdall" <[email protected]>: > > On Fri, Oct 24, 2014 at 3:51 PM, Christian Ege <[email protected]> > wrote: > >> > >> Am 23.10.2014 20:17 schrieb "Tim Tisdall" <[email protected]>: > >> > > >> > Does anyone have a device with an ICplus IP101A PHYceiver that's > properly working? I've had it working intermittently but not > consistently. I'm at a loss as to what the issue could be. > >> > > >> I do not own one. But maybe this is an issue with clock skew. Some > Phy's allow the setting of a clock skew. We had stability issues with a > micrel phy and fixed them by tuning clock skew. > > > > How does one do that? Did you have to modify the kernel driver source > code to make those changes? > > > We did this for a different platform I.mx6 based. As a first approach we > patched u-boot and kernel a second one was using device tree. As I remember > there had been patches to tweak this from user space as well. > > The values itself have been calculated be some h/w eng. We used the dived > and concur approach which resulted in the same values. > I guess I forgot to mention this is on an A20 device running the latest 3.4 kernel in the linux-sunxi repo. I've actually tried putting some printk statements in the PHY driver and I don't see any of them. I'm not sure if those (should) show up in dmesg or just on the console, so I need to hook up a serial cable to confirm.
-- You received this message because you are subscribed to the Google Groups "linux-sunxi" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
