On Wed, Jul 19, 2023 at 11:36:46AM +0300, Alexey Minnekhanov wrote: > On 19.07.2023 10:39, Krzysztof Kozlowski wrote: > > It does not look like you tested the DTS against bindings. Please run > > `make dtbs_check` > > > > If DT schema for interconnect requires bus clocks to be specified, I don't > even know what to put there now. Can we change schema? > I think I mentioned the DT schema updates during the review of Konrad's interconnect changes and he mentioned he would like to clean those up after getting the series in. (Which would be sometime soon now I guess) For now, having the &rpmcc "bus"/"bus_a"/"ipa" clocks specified on the interconnect@... nodes is still valid. At runtime they will just be ignored. Feel free to just keep them there for this initial fix. For the other two usages (iommu@, usb@) these votes with minimal frequency look a bit related to the "keep_alive" stuff Konrad added. [1] Maybe that could be used here instead of bypassing interconnect with the clock votes? Thanks, Stephan [1]: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?id=b979049c38e170286158e97290c892957c836903
