[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <e55136cf-db2b-1305-2c21-3d003a0218df@nvidia.com>
Date: Fri, 27 Dec 2019 12:36:05 +0530
From: Nagarjuna Kristam <nkristam@...dia.com>
To: Thierry Reding <thierry.reding@...il.com>
CC: <balbi@...nel.org>, <gregkh@...uxfoundation.org>,
<jonathanh@...dia.com>, <mark.rutland@....com>,
<robh+dt@...nel.org>, <kishon@...com>,
<devicetree@...r.kernel.org>, <linux-tegra@...r.kernel.org>,
<linux-usb@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [Patch V2 04/18] phy: tegra: xusb: Add usb-phy support
On 19-12-2019 19:07, Thierry Reding wrote:
> On Wed, Dec 18, 2019 at 02:46:17PM +0530, Nagarjuna Kristam wrote:
>> For USB 2 ports that has usb-role-switch enabled, add usb-phy for
>> corresponding USB 2 phy. USB role changes from role switch are then
>> updated to corresponding host and device mode drivers via usb-phy notifier
>> block.
>>
>> Signed-off-by: Nagarjuna Kristam<nkristam@...dia.com>
>> ---
>> V2:
>> - Added dev_set_drvdata for port->dev.
>> ---
>> drivers/phy/tegra/xusb.c | 71 ++++++++++++++++++++++++++++++++++++++++++++++++
>> drivers/phy/tegra/xusb.h | 2 ++
>> 2 files changed, 73 insertions(+)
>>
>> diff --git a/drivers/phy/tegra/xusb.c b/drivers/phy/tegra/xusb.c
>> index dc00b42..5bde8f1 100644
>> --- a/drivers/phy/tegra/xusb.c
>> +++ b/drivers/phy/tegra/xusb.c
>> @@ -533,6 +533,8 @@ static int tegra_xusb_port_init(struct tegra_xusb_port *port,
>> if (err < 0)
>> goto unregister;
>>
>> + dev_set_drvdata(&port->dev, port);
>> +
>> return 0;
>>
>> unregister:
>> @@ -545,6 +547,8 @@ static void tegra_xusb_port_unregister(struct tegra_xusb_port *port)
>> if (!IS_ERR_OR_NULL(port->usb_role_sw)) {
>> of_platform_depopulate(&port->dev);
>> usb_role_switch_unregister(port->usb_role_sw);
>> + cancel_work_sync(&port->usb_phy_work);
>> + usb_remove_phy(&port->usb_phy);
>> }
>> device_unregister(&port->dev);
>> }
>> @@ -556,16 +560,59 @@ static const char *const modes[] = {
>> [USB_DR_MODE_OTG] = "otg",
>> };
>>
>> +static void tegra_xusb_usb_phy_work(struct work_struct *work)
>> +{
>> + struct tegra_xusb_port *port = container_of(work,
>> + struct tegra_xusb_port, usb_phy_work);
> Perhaps add a static inline function to cast this? Might not be worth it
> since we only need to cast once. In that case, perhaps make this look a
> little prettier by aligning arguments on subsequent lines with "work" on
> the first line?
>
Will align arguments with "work".
>> + enum usb_role role = usb_role_switch_get_role(port->usb_role_sw);
>> +
>> + dev_dbg(&port->dev, "%s calling notifier for role %d\n", __func__,
>> + role);
>> +
>> + atomic_notifier_call_chain(&port->usb_phy.notifier, role,
>> + &port->usb_phy);
> I'm curious: you use an atomic notifier call chain here but then you
> schedule work to call it. Typically you only need to schedule work if
> you get notified in atomic context and you need to process the event
> outside of the atomic context.
>
> Since these are atomic notifiers, do we really need the work? Or the
> other way around: why not use regular notifiers if we're processing them
> from non-atomic contexts only anyway?
>
notifier used by usb-phy are atomic notifiers and hence need to call
atomic_notifier_call_chain only in this context., regular notifiers
cannot be used.
>> +}
>> +
>> static int tegra_xusb_role_sw_set(struct device *dev, enum usb_role role)
>> {
>> + struct tegra_xusb_port *port = dev_get_drvdata(dev);
>> +
>> dev_dbg(dev, "%s calling notifier for role is %d\n", __func__, role);
>>
>> + schedule_work(&port->usb_phy_work);
>> +
>> + return 0;
>> +}
>> +
>> +static int tegra_xusb_set_peripheral(struct usb_otg *otg,
>> + struct usb_gadget *gadget)
>> +{
>> + struct tegra_xusb_port *port = container_of(otg->usb_phy,
>> + struct tegra_xusb_port, usb_phy);
>> +
>> + if (gadget != NULL)
>> + schedule_work(&port->usb_phy_work);
>> +
>> + return 0;
>> +}
>> +
>> +static int tegra_xusb_set_host(struct usb_otg *otg, struct usb_bus *host)
>> +{
>> + struct tegra_xusb_port *port = container_of(otg->usb_phy,
>> + struct tegra_xusb_port, usb_phy);
>> +
>> + if (host != NULL)
>> + schedule_work(&port->usb_phy_work);
>> +
>> return 0;
>> }
>>
>> +
>> static int tegra_xusb_setup_usb_role_switch(struct tegra_xusb_port *port)
>> {
>> int err = 0;
>> + struct tegra_xusb_lane *lane = tegra_xusb_find_lane(port->padctl,
>> + "usb2", port->index);
> You're not properly aligning the arguments here.
>
> Thierry
>
Will align here and other places accordingly.
-Nagarjuna
>> struct usb_role_switch_desc role_sx_desc = {
>> .set = tegra_xusb_role_sw_set,
>> .fwnode = dev_fwnode(&port->dev),
>> @@ -578,6 +625,30 @@ static int tegra_xusb_setup_usb_role_switch(struct tegra_xusb_port *port)
>> if (err != EPROBE_DEFER)
>> dev_err(&port->dev, "Failed to register USB role SW: %d",
>> err);
>> + return err;
>> + }
>> +
>> + INIT_WORK(&port->usb_phy_work, tegra_xusb_usb_phy_work);
>> +
>> + port->usb_phy.otg = devm_kzalloc(&port->dev,
>> + sizeof(struct usb_otg), GFP_KERNEL);
>> + if (!port->usb_phy.otg)
>> + return -ENOMEM;
>> +
>> + /*
>> + * Assign phy dev to usb-phy dev. Host/device drivers can use phy
>> + * reference to retrieve usb-phy details.
>> + */
>> + port->usb_phy.dev = &lane->pad->lanes[port->index]->dev;
>> + port->usb_phy.dev->driver = port->padctl->dev->driver;
>> + port->usb_phy.otg->usb_phy = &port->usb_phy;
>> + port->usb_phy.otg->set_peripheral = tegra_xusb_set_peripheral;
>> + port->usb_phy.otg->set_host = tegra_xusb_set_host;
>> +
>> + err = usb_add_phy_dev(&port->usb_phy);
>> + if (err < 0) {
>> + dev_err(&port->dev, "Failed to add usbphy: %d\n", err);
>> + return err;
>> }
>>
>> /* populate connector entry */
>> diff --git a/drivers/phy/tegra/xusb.h b/drivers/phy/tegra/xusb.h
>> index 9f27899..2345657 100644
>> --- a/drivers/phy/tegra/xusb.h
>> +++ b/drivers/phy/tegra/xusb.h
>> @@ -268,6 +268,8 @@ struct tegra_xusb_port {
>> struct device dev;
>>
>> struct usb_role_switch *usb_role_sw;
>> + struct work_struct usb_phy_work;
>> + struct usb_phy usb_phy;
>>
>> const struct tegra_xusb_port_ops *ops;
>> };
>> --
>> 2.7.4
Powered by blists - more mailing lists