On 04/04/2023 12:06:14+0200, Krzysztof Kozlowski wrote: > On 04/04/2023 11:56, Alexandre Belloni wrote: > > On 04/04/2023 10:14:33+0200, Krzysztof Kozlowski wrote: > >> On 04/04/2023 09:44, Alexandre Belloni wrote: > >>>> > >>>>> charging will always > >>>>> enable a diode, select a resistor and then have or not an extra diode. > >>>>> Figure2 of the MAX31329 datasheet is great. > >>>> > >>>> So the diode is in the max313xx? Then why enabling it is a property of > >>>> DT? Either this should be inferred from compatible or is even a policy, > >>>> not a DT property. Just because device has a register for something, is > >>>> not an argument that "something" should be in DT. > >>> > >>> Well, it depends on the battery that is installed on the board so it > >>> makes sense to have it in DT. > >> > >> OK, that would be a good reason, but I wonder why? Why choosing diode or > >> not depends on the battery? Wouldn't you always want to have the diode? > >> > > > > It limits the maximum current used to charge the battery or supercap to > > not exceed what is allowed. > > and I still wonder why someone would like to allow exceeding what is > allowed. IOW, what is the use case to disable the diode? > The battery or supercap is the part defining the current limit. why would you want to limit the current if you can charge faster? -- Alexandre Belloni, co-owner and COO, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
