On Mon, Jun 17, 2019 at 3:58 PM Paul Walmsley <paul.walmsley@xxxxxxxxxx> wrote: > > On Mon, 17 Jun 2019, Yash Shah wrote: > > > On Mon, Jun 17, 2019 at 3:28 PM Paul Walmsley <paul.walmsley@xxxxxxxxxx> wrote: > > > > > On Mon, 17 Jun 2019, Andreas Schwab wrote: > > > > > > > On Jun 17 2019, Yash Shah <yash.shah@xxxxxxxxxx> wrote: > > > > > > > > > - Add "MACB_SIFIVE_FU540" in Kconfig to support SiFive FU540 in macb > > > > > driver. This is needed because on FU540, the macb driver depends on > > > > > SiFive GPIO driver. > > > > > > > > This of course requires that the GPIO driver is upstreamed first. > > > > > > What's the impact of enabling CONFIG_MACB_SIFIVE_FU540 when the GPIO > > > driver isn't present? (After modifying the Kconfig "depends" line > > > appropriately.) > > > > > > Looks to me that it shouldn't have an impact unless the DT string is > > > present, and even then, the impact might simply be that the MACB driver > > > may not work? > > > > Yes, there won't be an impact other than MACB driver not working. > > OK. In that case, there doesn't seem much point to adding the Kconfig > option. Could you please post a new version without it? Sure, will do that. > > > In any case, without GPIO driver, PHY won't get reset and the network > > interface won't come up. > > Naturally, in the medium term, we want Linux to handle the reset. But if > there's no GPIO driver present, and the bootloader handles the PHY reset > before the kernel starts, would the network driver work in that case? Yes, if bootloader handles the PHY reset then the network driver will work in that case. I will post a new version without the GPIO driver dependency. > > > - Paul
