Hi > Gesendet: Samstag, 26. Februar 2022 um 19:15 Uhr > Von: "Peter Geis" <pgwipeout@xxxxxxxxx> > On Sat, Feb 26, 2022 at 1:08 PM Heiko Stübner <heiko@xxxxxxxxx> wrote: > > > > Hi Frank, > > > > Am Samstag, 26. Februar 2022, 14:57:24 CET schrieb Frank Wunderlich: > > > From: Frank Wunderlich <frank-w@xxxxxxxxxxxxxxx> > > > > > > RK356x supports up to 3 sata controllers which were compatible with the > > > existing snps,dwc-ahci binding. > > > > > > My board has only sata2 connected to combphy2 so only add this one. > > > > how far does the added node diverge from the vendor kernel? > > > > If it's pretty much similar between both, we can assume the other nodes > > should work pretty well as well and therefore should all of them at once > > and hope for the best? > > There's essentially zero divergence (minus the change due to combophy > changing), and likely won't be until the ahci-platform.txt is > converted to yaml. > > I have tested both SATA1 and SATA2 successfully on the rk3566. > I don't have any rk3568 boards that are operational yet to test SATA0. Like Peter says, the sata-node(s) is basicly same as downstream, only combphy-label has changed. Sata1 from Peters tree [1] seems to have only this change. I'll send an followup containing all 3, but had only tested sata2. already sent a bindings-patch for the ahci-platform.txt [2] regards Frank [1] https://gitlab.com/pgwipeout/linux-next/-/blob/main/arch/arm64/boot/dts/rockchip/rk356x.dtsi#L238 [2] https://lore.kernel.org/lkml/20220226204404.109867-1-linux@xxxxxxxxx/T/
