Hi, On Thu, Nov 12, 2020 at 2:34 AM Chuanhong Guo <gch981213@xxxxxxxxx> wrote: > > On Thu, Nov 12, 2020 at 9:26 AM Chuanhong Guo <gch981213@xxxxxxxxx> wrote: > > > > I've already said in previous threads that clock assignment in > > current linux kernel is not trustworthy. > > I've got the clock plan for mt7621 now. (Can't share it, sorry.) > > Most of your clock assumptions above are incorrect. > > I've made a clock driver with gate support a few months ago.[0] > > but I don't have much time to really finish it. > > Maybe you could rework your clock gate driver based on it. > > > > [0] https://github.com/981213/linux/commit/2eca1f045e4c3db18c941135464c0d7422ad8133 > > hsdma/eth/pio clocks are still missing in mediatek doc and > I just made them up in the driver. Correct clock frequency for > them aren't really important for them to work though. > And another part I didn't finish is checking clock support for > every drivers mt7621 used. Many drivers don't explicitly > enable the clock and may be problematic when kernel > gates unused clocks. > Well, I think they are not important either. Also, by default gate register has all the gate bits enabled. When a gate driver is added, the kernel by default will try to disable those clocks that haven't been requested. To avoid weird behaviour because of some drivers are not using properly clocks we have the CLK_IGNORED_UNUSED, which as you can see is currently being used in my code. Using that all seems to work as expected as it is now. > -- > Regards, > Chuanhong Guo Best regards, Sergio Paracuellos
