On Mon, Jun 11, 2018 at 6:44 AM Adam Ford <aford173@xxxxxxxxx> wrote: > > I am running some experiments on the DM3730 trying to read the > abb_mpu_iva regulator voltage. The min and max voltage readings of > abb_mpu_iva seem to match the table from the device tree for the > omap36xx, however when I read the microvolts, it returns and invalid > number. > Adding in Viresh Kumar... > # cat min_microvolts > 1012500 > # cat max_microvolts > 1375000 > # cat microvolts > -22 > > While booting the omap3 also returns are messages: > [ 3.105560] omap2_set_init_voltage: unable to find boot up OPP for > vdd_mpu_iva > [ 3.113037] omap2_set_init_voltage: unable to set vdd_mpu_iva > > I don't know if their related. > > I am trying to push the DM3730 to OPP1G which means enabling the ABB. > According to the device tree and the TRM, this register should be > 483072f0 and when I read this register with the DM3730 pushed to 1000 > MHz, the returned value is 0x00, meaning ABB is still off. > > Since the twl4030 PMIC is sourcing the power to the processor from > I am associating the cpu supply to to vcc with cpu0-supply = <&vcc>; > > I have modified the operating table for 1000MHz, and from what I can > tell the vcc is the correct operating voltage when operating at 1GHz, > but it looks like we somehow need to associate the abb_mpu_iva > regulator it properly sets up the operating points so ABB is enabled. > > I looked through how other boards do this. The dra7 references the > abb regulator by vbb-supply = <&abb_mpu> but I am not seeing anything > like that for omap3/4/5. > > So I have two questions: > > How does the abb driver know what the current state of the processor > is so can set the values properly? > > Does it seem reasonable that other processors should have a vbb-supply > reference to use the ti-abb-regulator driver to keep it in sync with > the operating voltage and frequency of the processor? It appears as if there is a newer way to define the operating points using "operating-points-v2" If we were to migrate the omap3630 to this, would it be able to associate the abb_mpu_iva in addition the the existing regulator to get it to switch on the ABB and correctly identify the operating voltage of abb_mpu_iva? adam > > adam -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
