Re: [PATCH 8/8] phy: add driver for TI TUSB1210 ULPI PHY

[David Cohen <david.a.cohen@linux.intel.com>]

Hi Heikki,

On Mon, Jan 26, 2015 at 02:55:03PM +0200, Heikki Krogerus wrote:
> Hi David,
> 
> On Sat, Jan 24, 2015 at 03:58:11PM -0800, David Cohen wrote:
> > > +static int tusb1210_power_on(struct phy *phy)
> > > +{
> > > +	struct tusb1210 *tusb = phy_get_drvdata(phy);
> > > +
> > > +	gpiod_set_value_cansleep(tusb->gpio_reset, 1);
> > > +	gpiod_set_value_cansleep(tusb->gpio_cs, 1);
> > > +
> > > +	/* Restore eye diagram optimisation value */
> > > +	ulpi_write(tusb->ulpi, TUSB1210_VENDOR_SPECIFIC2,
> > > +		   tusb->eye_diagram_tuning);
> > 
> > After you power on phy, ulpi bus may not be available right away. In
> > intel case, phy power on happens during dwc3 power on. ULPI bus is not
> > available until OTG controller and phy are in sync.
> > 
> > In resume, you can't restore eye diagram from here.
> 
> I'm sorry but I don't think I understand? Where do we power on the phy
> before dwc3 is powered on? Or is this a Baytrail-CR specific problem?
> I can't see any problems with the hardware I have.

You can't see in single accesses. But you may see when running stability
tests overnight.

Anyway, look for dwc3_core_soft_reset() function:
- dwc3 goes to reset state
- phy is initialized (or at least gets ready to sync clocks)
- dwc3 goes out or reset state

During tusb1210 phy init from dwc3, you shouldn't access ulpi bus.

> 
> In any case, this sounds like purely dwc3 issue and not tusb1210
> issue.

That's neither purely dwc3's not tusb1210's, that's your problem :)
Since it's a potential bug introduced by your patch set here.

> 
> > > +static int tusb1210_probe(struct ulpi *ulpi)
> > > +{
> > > +	struct gpio_desc *gpio;
> > > +	struct tusb1210 *tusb;
> > > +	int ret;
> > > +
> > > +	tusb = devm_kzalloc(&ulpi->dev, sizeof(*tusb), GFP_KERNEL);
> > > +	if (!tusb)
> > > +		return -ENOMEM;
> > > +
> > > +	gpio = devm_gpiod_get(&ulpi->dev, "reset");
> > > +	if (!IS_ERR(gpio)) {
> > > +		ret = gpiod_direction_output(gpio, 0);
> > > +		if (ret)
> > > +			return ret;
> > > +		tusb->gpio_reset = gpio;
> > > +	}
> > 
> > You cannot proceed with probe if gpio reset is not available. Different
> > from CS, it's a mandatory pin to toggle in order to power on/off phy and
> > get it in sync with OTG controller.
> 
> Well, let's check -ENOENT and -ENODEV return values separately. The
> reset pin is not used on all platforms so getting this gpio is
> optional. This is the case even with some Intel's platforms using
> TUSB1210.

I doublechecked the tusb1210 datasheet. Despite the power on sequence
mentions RESET toggling as required, it has a comment later on that it
can be tied to VDDIO.
So as you mentioned, it'd be better to ignore -ENOENT and -ENODEV and
raise error otherwise.

> 
> > > +
> > > +	gpio = devm_gpiod_get(&ulpi->dev, "cs");
> > > +	if (!IS_ERR(gpio)) {
> > > +		ret = gpiod_direction_output(gpio, 0);
> > > +		if (ret)
> > > +			return ret;
> > > +		tusb->gpio_cs = gpio;
> > > +	}
> > > +
> > > +	/* Store initial eye diagram optimisation value */
> > > +	ret = ulpi_read(ulpi, TUSB1210_VENDOR_SPECIFIC2);
> > 
> > It's unclear if ulpi is accessible at this point. You can't read it at
> > this point.
> 
> We wouldn't have reached this point if ulpi wasn't accessible.
> Registering the ulpi interface would have already failed so no driver
> would have been probed.

You have a chicken/egg problem here:
- dwc3 needs phy to complete soft reset during probe
- tusb1210 needs dwc3 soft reset completed to be accessible via ULPI

Can you share how tusb1210 is connected on the platform you're using as
test for this patch? I don't think this driver would work reliably with
this device:
http://liliputing.com/2014/11/trekstor-launches-first-android-tablet-based-intels-irda-reference-design.html

Br, David

> 
> 
> Thanks!
> 
> -- 
> heikki