by Anand Moon

[permalink] [raw]

Subject: Re: [RFCv1 0/8] RK3399 clean shutdown issue

Hi Robin,

On Mon, 9 Dec 2019 at 18:59, Robin Murphy <[email protected]> wrote:
>
> On 06/12/2019 6:45 pm, Anand Moon wrote:
> > Most of the RK3399 SBC boards do not perform clean
> > shutdown and clean reboot.
>
> FWIW reboot problems on RK3399 have been tracked down to issues in
> upstream ATF, and are unrelated to the PMIC.

Yes I am aware of this changes.
But, I have tired to study *RK808 datasheet V1.4* [0] below section
*5.2.3 Power Channel Control/Monitor Registers*
for clean reboot I was going to try disable some bit in below
into reboot handle in the future patch.

DCDC_EN_REG
SLEEP_SET_OFF_REG1
SLEEP_SET_OFF_REG2
DCDC_UV_STS_REG

I was going see if this helps to do clean reboot.
further more use this in suspend/resume operation.

[0] http://rockchip.fr/RK808%20datasheet%20V1.4.pdf

But I feed that their is some more issue with related to mmc or PCIe
not able to cleanly release the resources while reboot which caused
then to disable after reboot.

>
> > These patches try to help resolve the issue with proper
> > shutdown by turning off the PMIC.
>
> As mentioned elsewhere[1], although this is what the BSP kernel seems to
> do, and in practice it's unlikely to matter for the majority of devboard
> users like you and me, I still feel a bit uncomfortable with this
> solution for systems using ATF as in principle the secure world might
> want to know about orderly shutdowns, and this effectively makes every
> shutdown an unexpected power loss from secure software's point of view.
>
> Robin.
>
> [1]
> http://lists.infradead.org/pipermail/linux-rockchip/2019-December/028183.html
>

Yes I have follow the mailing list and I read this thread.
I am not aware of ATF complete architecture.

-Anand

2019-12-09 15:40:34

by Anand Moon

[permalink] [raw]

Subject: Re: [RFCv1 2/8] mfd: rk808: use syscore for RK805 PMIC shutdown

Hi Robin,

On Mon, 9 Dec 2019 at 19:04, Robin Murphy <[email protected]> wrote:
>
> On 06/12/2019 6:45 pm, Anand Moon wrote:
> > Use common syscore_shutdown for RK805 PMIC to do
> > clean I2C shutdown, drop the unused pm_pwroff_prep_fn
> > and pm_pwroff_fn function pointers.
>
> Coincidentally, I've also been looking at RK805 for the sake of trying
> to get suspend to behave on my RK3328 box, and I've ended up with some
> slightly different cleanup patches - I'll tidy them up and post them for
> comparison as soon as I can.

No issue if their is better clean approach, I will definitely test that series.
>
> > Cc: Heiko Stuebner <[email protected]>
> > Signed-off-by: Anand Moon <[email protected]>
> > ---
> > drivers/mfd/rk808.c | 33 +++++++++++++++++----------------
> > 1 file changed, 17 insertions(+), 16 deletions(-)
> >
> > diff --git a/drivers/mfd/rk808.c b/drivers/mfd/rk808.c
> > index e637f5bcc8bb..713d989064ba 100644
> > --- a/drivers/mfd/rk808.c
> > +++ b/drivers/mfd/rk808.c
> > @@ -467,16 +467,6 @@ static void rk808_update_bits(unsigned int reg, unsigned int mask,
> > "can't write to register 0x%x: %x!\n", reg, ret);
> > }
> >
> > -static void rk805_device_shutdown(void)
> > -{
> > - rk808_update_bits(RK805_DEV_CTRL_REG, DEV_OFF, DEV_OFF);
> > -}
> > -
> > -static void rk805_device_shutdown_prepare(void)
> > -{
> > - rk808_update_bits(RK805_GPIO_IO_POL_REG, SLP_SD_MSK, SHUTDOWN_FUN);
> > -}
> > -
> > static void rk808_device_shutdown(void)
> > {
> > rk808_update_bits(RK808_DEVCTRL_REG, DEV_OFF_RST, DEV_OFF_RST);
> > @@ -491,10 +481,23 @@ static void rk8xx_syscore_shutdown(void)
> > {
> > struct rk808 *rk808 = i2c_get_clientdata(rk808_i2c_client);
> >
> > - if (system_state == SYSTEM_POWER_OFF &&
> > - (rk808->variant == RK809_ID || rk808->variant == RK817_ID)) {
> > - rk808_update_bits(RK817_SYS_CFG(3), RK817_SLPPIN_FUNC_MSK,
> > - SLPPIN_DN_FUN);
> > + if (system_state == SYSTEM_POWER_OFF) {
> > + dev_info(&rk808_i2c_client->dev, "System Shutdown Event\n");
> > +
> > + switch (rk808->variant) {
> > + case RK805_ID:
> > + rk808_update_bits(RK805_GPIO_IO_POL_REG,
> > + SLP_SD_MSK, SHUTDOWN_FUN);
> > + rk808_update_bits(RK805_DEV_CTRL_REG, DEV_OFF, DEV_OFF);
>
> Why this change? Shutdown via the SLEEP pin is working just fine on my
> box :/
>
> Robin.

As per RK-805 datasheet [0] below.
For clean poweroff we need to set in DEV_CTRL_REG reg
Bit 0 DEV_OFF: write “1” to turn down the PMU.

[0] http://files.pine64.org/doc/rock64/Rockchip_RK805_Datasheet_V1.1%C2%A020160921.pdf

-Anand