Re: [PATCH 01/16] power: supply: axp288_charger: Do not stop + restart charging at boot
From: Hans de Goede <hidden>
Date: 2017-12-26 09:26:53
Hi, On 26-12-17 08:29, Chen-Yu Tsai wrote:
On Fri, Dec 22, 2017 at 8:37 PM, Hans de Goede [off-list ref] wrote:quoted
Before this commit we were relying solely on the extcon interface for cable detection, including to determine if a cable providing vbus is connected at all. This caused us to turn off charging at boot, because when we run the initial state processing the axp288-extcon driver is still running charger-type detection most of the time, so all charger cable types read as disconnected when we run the initial state processing. This commit reworks the axp288_charger_extcon_evt_worker flow to use the VBUS_VALID bit from the PWR_INPUT_STATUS register to determine if we should turn charging on/off. Note this is the same bit as we use for the online property. If VBUS_VALID is set, but the extcon code has not completed the charger type detection yet, we now simply bail leaving things as configured by the BIOS (we will get a notifier call when the extcon code is done and reschedule the axp288_charger_extcon_evt_worker). The extcon code is the only one to trigger the worker (outside of the initial run) and we can rely on it to only call us if things have changed, so while we are completely refactoring axp288_charger_extcon_evt_worker, also remove the code to check if the state has changed. Signed-off-by: Hans de Goede <redacted>Other than the two patches I replied to, this looks good overall. For the remaining patches, Reviewed-by: Chen-Yu Tsai <redacted> Hans, I don't have any devices with this PMIC, so I'm unlikely to spend much time on the axp288 specific drivers. Any chance you want to maintain them?
Yes that probably is a good idea, I will add a patch to v2 of the series adding a MAINTAINERS entry for this. Regards, Hans