Em Fri, 29 Jan 2021 13:29:10 +0000
Lee Jones <lee.jones@xxxxxxxxxx> escreveu:
> On Fri, 29 Jan 2021, Mauro Carvalho Chehab wrote:
>
> > Hi Lee,
> >
> > Em Wed, 27 Jan 2021 11:05:37 +0000
> > Lee Jones <lee.jones@xxxxxxxxxx> escreveu:
>
> > > > +static const struct mfd_cell hi6421v600_devs[] = {
> > > > + { .name = "hi6421v600-regulator", },
> > > > +};
> > >
> > > Where are the reset of the devices?
> >
> > Not sure what you mean here.
> >
> > This MFD device provides:
> >
> > - an IRQ handler;
> > - several LDO lines used by a regulator driver.
> >
> > The IRQ handler is properly initialized here, while the
> > regulators are initialized by the regulator driver. The initial
> > state of this device is set up by u-boot.
> >
> > So, AFAIKT, there's no need to have any reset line
> > attached here.
>
> Oh wow! Sorry about that. It's a misspelling.
>
> "Where are the *rest* of the devices?"
>
> Registering one device does not qualify this as an MFD.
This is a MFD device. The current support has two different functions:
- it is an IRQ controller
- it regulator
>From the Hikey 970 schematics (pags. 12-14):
https://www.96boards.org/documentation/consumer/hikey/hikey970/hardware-docs/files/hikey970-schematics.pdf
This chip also has some A/D converts, a battery voltage sensor
and some clock generators. Those are unused on Hikey 970.
Btw, the initial version of this chipset (without SPMI bus) is
also under drivers/mfd, as hi6421-pmic-core.
> >
> > > > +static void hi6421_spmi_irq_mask(struct irq_data *d)
> > > > +{
> > > > + struct hi6421_spmi_pmic *pmic = irq_data_get_irq_chip_data(d);
> > > > + unsigned long flags;
> > > > + unsigned int data;
> > > > + u32 offset;
> > > > +
> > > > + offset = (irqd_to_hwirq(d) >> 3);
> > >
> > > Why 3?
> >
> > No idea. I don't have any datasheets from this device.
> >
> > > Probably better to define these shifts/masks rather than use
> > > magic numbers with no comments.
> >
> > I'll change the above to:
> >
> > #define HISI_IRQ_MASK GENMASK(1, 0)
> >
> > offset = (irqd_to_hwirq(d) >> HISI_IRQ_MASK);
>
> This is a shift surely? Marks are usually &'ed.
>
> > > > + regmap_read(pmic->map, offset, &data);
> > > > + data |= (1 << (irqd_to_hwirq(d) & 0x07));
> > >
> > > What are you doing here?
> > >
> > > Maybe improved defines will be enough. If not, please supply a
> > > suitable comment.
> >
> > Again, no idea. The only documentation I had access to this chip is
> > at:
> >
> > https://www.96boards.org/documentation/consumer/hikey/hikey970/
> >
> > With doesn't mention any register details.
> >
> > The driver itself came from the Linaro's 96boards git tree, with also
> > doesn't contain any register mapping.
>
> Well that's awkward.
>
> I'm not keen on merging code that no one knows what it does.
>
> Maybe I can do some digging.
Hmm... after re-reading the code:
enum hi6421_spmi_pmic_irq_list {
OTMP = 0,
VBUS_CONNECT,
VBUS_DISCONNECT,
ALARMON_R,
HOLD_6S,
HOLD_1S,
POWERKEY_UP,
POWERKEY_DOWN,
OCP_SCP_R,
COUL_R,
SIM0_HPD_R,
SIM0_HPD_F,
SIM1_HPD_R,
SIM1_HPD_F,
PMIC_IRQ_LIST_MAX,
};
...
#define SOC_PMIC_IRQ_MASK_0_ADDR 0x0202
#define SOC_PMIC_IRQ0_ADDR 0x0212
...
static void hi6421_spmi_irq_mask(struct irq_data *d)
{
struct hi6421_spmi_pmic *ddata = irq_data_get_irq_chip_data(d);
unsigned long flags;
unsigned int data;
u32 offset;
offset = (irqd_to_hwirq(d) >> 3);
offset += SOC_PMIC_IRQ_MASK_0_ADDR;
spin_lock_irqsave(&ddata->lock, flags);
regmap_read(ddata->regmap, offset, &data);
data |= (1 << (irqd_to_hwirq(d) & 0x07));
regmap_write(ddata->regmap, offset, data);
spin_unlock_irqrestore(&ddata->lock, flags);
}
Actually does this mapping:
====================== ============= =====
IRQ MASK REGISTER BIT
====================== ============= =====
OTMP 0x0202 bit 0
VBUS_CONNECT 0x0202 bit 1
VBUS_DISCONNECT 0x0202 bit 2
ALARMON_R 0x0202 bit 3
HOLD_6S 0x0202 bit 4
HOLD_1S 0x0202 bit 5
POWERKEY_UP 0x0202 bit 6
POWERKEY_DOWN 0x0202 bit 7
OCP_SCP_R 0x0203 bit 0
COUL_R 0x0203 bit 1
SIM0_HPD_R 0x0203 bit 2
SIM0_HPD_F 0x0203 bit 3
SIM1_HPD_R 0x0203 bit 4
SIM1_HPD_F 0x0203 bit 5
====================== ============= =====
The IRQ register itself is the mask register + 0x10:
static irqreturn_t hi6421_spmi_irq_handler(int irq, void *priv)
{
struct hi6421_spmi_pmic *ddata = (struct hi6421_spmi_pmic *)priv;
unsigned long pending;
unsigned int in;
int i, offset;
for (i = 0; i < HISI_IRQ_ARRAY; i++) {
regmap_read(ddata->regmap, SOC_PMIC_IRQ0_ADDR + i, &in);
pending = HISI_MASK & in;
regmap_write(ddata->regmap, SOC_PMIC_IRQ0_ADDR + i, pending);
...
}
return IRQ_HANDLED;
}
Anyway, it sounds a good idea to have a table like the above on some
comment inside the driver's code.
> > > > + pr_debug("PMU IRQ address value:irq[0x%x] = 0x%x\n",
> > > > + SOC_PMIC_IRQ0_ADDR + i, pending);
> > >
> > > Again, is this actually useful to anyone now that the driver is nearly
> > > 10 years old. Particularly anyone who can't add a quick printk()
> > > during a debug session?
> >
> > With regards to the debug stuff, I'm dropping everything.
>
> Great.
>
> > On a side comment, I doubt that the driver has 10 years old ;-)
> >
> > See, Hikey 970 uses Kirin 970 SoC, which it was launched in Sept, 2017.
>
> The header has a copyright from 2011.
>
> // Copyright (c) 2013 Linaro Ltd.
> // Copyright (c) 2011 Hisilicon.
>
> > The original version of this driver publicly debuted on this tree:
> >
> > https://github.com/96boards-hikey/linux/blob/hikey970-v4.9/drivers/mfd/hisi_pmic_spmi.c
> >
> > On a commit made on Feb, 2018.
> >
> > Ok, Hi6421v600 is a separate silicon, probably derivative from
> > Hi6421 (used on Hikey 960). Its copyright mentions 2011, but
> > that's probably because the code itself came from older generations
> > of the regulator chipset.
>
> So we've inherited a copyright from another driver?
>
> Sounds suspect.
Not really. The non-SPMI version of this chipset
(at drivers/mfd/hi6421-pmic-core.c) has its copyright starting in
2011:
/*
* Device driver for Hi6421 PMIC
*
* Copyright (c) <2011-2014> HiSilicon Technologies Co., Ltd.
* http://www.hisilicon.com
* Copyright (c) <2013-2017> Linaro Ltd.
* https://www.linaro.org
> > Please see the enclosed patch for the new code after fixing the issues
> > you pointed. I'll re-submit it as a series once you're ok with the
> > code.
>
> Would you mind just resubmitting please? We can go from there.
Ok. Will do it on a next spin.
Thanks,
Mauro
_______________________________________________
devel mailing list
devel@xxxxxxxxxxxxxxxxxxxxxx
http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
