Re: [PATCH v2 10/20] power: supply: bq25890: Add bq25890_set_otg_cfg() helper

From: Hans de Goede
Date: Sun Nov 28 2021 - 10:04:31 EST


Hi,

On 11/16/21 12:07, Yauhen Kharuzhy wrote:
>
>
> аў, 16 ліс 2021, 12:33 карыстальнік Hans de Goede <hdegoede@xxxxxxxxxx <mailto:hdegoede@xxxxxxxxxx>> напісаў:
>
> Hi Yauhen,
>
> On 11/15/21 11:11, Yauhen Kharuzhy wrote:
> > On Sun, Nov 14, 2021 at 06:03:25PM +0100, Hans de Goede wrote:
> >> Add a bq25890_set_otg_cfg() helper function, this is a preparation
> >> patch for adding regulator support.
> >>
> >> Signed-off-by: Hans de Goede <hdegoede@xxxxxxxxxx <mailto:hdegoede@xxxxxxxxxx>>
> >> ---
> >>  drivers/power/supply/bq25890_charger.c | 28 ++++++++++++++------------
> >>  1 file changed, 15 insertions(+), 13 deletions(-)
> >>
> >> diff --git a/drivers/power/supply/bq25890_charger.c b/drivers/power/supply/bq25890_charger.c
> >> index 2bdfb58cda75..3c41fe86b3d3 100644
> >> --- a/drivers/power/supply/bq25890_charger.c
> >> +++ b/drivers/power/supply/bq25890_charger.c
> >> @@ -801,6 +801,17 @@ static int bq25890_power_supply_init(struct bq25890_device *bq)
> >>      return PTR_ERR_OR_ZERO(bq->charger);
> >>  }
> >> 
> >> +static int bq25890_set_otg_cfg(struct bq25890_device *bq, u8 val)
> >> +{
> >> +    int ret;
> >> +
> >> +    ret = bq25890_field_write(bq, F_OTG_CFG, val);
> >> +    if (ret < 0)
> >> +            dev_err(bq->dev, "Error switching to boost/charger mode: %d\n", ret);
> >
> > Just a note: if a connected USB device has relative big capacitor
> > at power wires inside, then a starting current pulse may be enough to
> > overload the boost reguator and VBUS will not be powered. I met this
> > at Yoga Book: the firmware set boost current limit to 1.4 A (default
> > value for bq25892) but when USB hub connected, the BOOST_FAULT event
> > appeared.
> >
> > To avoid this, Lenovo uses following trick in its kernel: set a boost
> > current limit to big value (2.1 A), wait some time (500 ms) and set
> > the current limit to right value (1.4A). This provides enough current to
> > charge capacitors in the connected device but saves desired long-time limit
> > to prevent overloading if the device consumes too much power itself.
>
> Right I saw this in your git repo, but I cannot reproduce the issue (1)
> I was hoping that since you can reproduce this, that you can rebase
> your fix on top of my patch-set ?
>
> Also I'm wondering if this behavior should be the default, I believe
> that the max. boost current may also be dependent on some external
> factors, so maybe we should make this behavior conditional on a
> new device-property ?
>
> Yes, defining of max VBUS current may be a good idea. Another possible approach may be to use some empirical multiplier, like 150% of max 'long time' current limit setting. I almost sure that all hardware will work with short impulse of such current, its usual condition at device connection.
>
>
> Regards,
>
> Hans
>
>
>
> 1) I must admit I did not try really hard, I guess I could try an
> USB powered hdd enclosure with a spinning disk
>
> What device are you seeing this with?
>
> I cannot remember exactly device but this was a USB hub, possible with keyboard, mouse receiver and USB dongle inserted. I can recheck this issue but one week after, when will return home.

So as I mentioned before I've just tried to reproduce this problem, but
I cannot reproduce it with an 2.5" USB disk enclosure with a spinning
disk, which typically will cause a nice current-peak when spinning up.

I think this might also require an almost empty battery to reproduce ?

Regards,

Hans