USB support for device tree

USB support for device tree

[Pavan Kondeti]

Hi

I am working on adding USB device tree support for MSM platform.  One of
our chip set has 2 hsusb cores. The first core is configured as otg and
the other core is configured in host only mode (EHCI compliant). Are the
below device node names Okay? Please suggest.

hsusb0-otg: usb@0xa6000000 {
	compatible = "qcom,hsusb-otg";
        ---
};

hsusb0-device: usb@0xa6000000 {
	compatible = "qcom,hsusb-device";
        ---
};

hsusb0-host: usb@0xa6000000 {
	compatible = "qcom,hsusb-host", "usb-ehci";
        ---
};

hsusb1-host: usb@0xa6000000 {
	compatible = "qcom,hsusb-host", "usb-ehci";
        ---
};

/* super speed support

ssusb0-device: usb@0xa6000000 {
	compatible = "qcom,ssusb-device";
};

*/

Thanks,
Pavan

-- 
Sent by a consultant of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.

Re: USB support for device tree

[Greg KH]

On Fri, Nov 04, 2011 at 01:55:09PM +0530, Pavan Kondeti wrote:
> Hi
> 
> I am working on adding USB device tree support for MSM platform.  One of
> our chip set has 2 hsusb cores. The first core is configured as otg and
> the other core is configured in host only mode (EHCI compliant). Are the
> below device node names Okay? Please suggest.

Why do you need to name the speed of the device and the host?  That
doesn't really matter if you plug a low-speed device into that super
speed host controller, right?

I'm not familiar with device tree naming conventions, but is this
something you do for network devices?  If not, I'd not recommend it for
USB controllers either.

thanks,

greg k-h
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: USB support for device tree

[Grant Likely]

On Fri, Nov 4, 2011 at 11:43 AM, Greg KH <greg@kroah.com> wrote:
> On Fri, Nov 04, 2011 at 01:55:09PM +0530, Pavan Kondeti wrote:
>> Hi
>>
>> I am working on adding USB device tree support for MSM platform.  One of
>> our chip set has 2 hsusb cores. The first core is configured as otg and
>> the other core is configured in host only mode (EHCI compliant). Are the
>> below device node names Okay? Please suggest.
>
> Why do you need to name the speed of the device and the host?  That
> doesn't really matter if you plug a low-speed device into that super
> speed host controller, right?
>
> I'm not familiar with device tree naming conventions, but is this
> something you do for network devices?  If not, I'd not recommend it for
> USB controllers either.

I'd like to comment, but I wasn't cc'd on the original email so I
don't know what is being proposed.  Pavan, can you please resend your
question and cc devicetree-discuss and me?

thanks,
g.

Re: USB support for device tree

[Grant Likely]

On Fri, Nov 4, 2011 at 12:08 PM, Grant Likely <grant.likely@secretlab.ca> wrote:
> On Fri, Nov 4, 2011 at 11:43 AM, Greg KH <greg@kroah.com> wrote:
>> On Fri, Nov 04, 2011 at 01:55:09PM +0530, Pavan Kondeti wrote:
>>> Hi
>>>
>>> I am working on adding USB device tree support for MSM platform.  One of
>>> our chip set has 2 hsusb cores. The first core is configured as otg and
>>> the other core is configured in host only mode (EHCI compliant). Are the
>>> below device node names Okay? Please suggest.
>>
>> Why do you need to name the speed of the device and the host?  That
>> doesn't really matter if you plug a low-speed device into that super
>> speed host controller, right?
>>
>> I'm not familiar with device tree naming conventions, but is this
>> something you do for network devices?  If not, I'd not recommend it for
>> USB controllers either.
>
> I'd like to comment, but I wasn't cc'd on the original email so I
> don't know what is being proposed.  Pavan, can you please resend your
> question and cc devicetree-discuss and me?

Oops, never mind.  I see that it got hung up in the mailing list.
I'll clear it now.

g.

Re: USB support for device tree

[Grant Likely]

On Fri, Nov 4, 2011 at 4:25 AM, Pavan Kondeti <pkondeti-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org> wrote:
> Hi
>
> I am working on adding USB device tree support for MSM platform.  One of
> our chip set has 2 hsusb cores. The first core is configured as otg and
> the other core is configured in host only mode (EHCI compliant). Are the
> below device node names Okay? Please suggest.
>
> hsusb0-otg: usb@0xa6000000 {
>        compatible = "qcom,hsusb-otg";
>        ---
> };
>
> hsusb0-device: usb@0xa6000000 {
>        compatible = "qcom,hsusb-device";
>        ---
> };
>
> hsusb0-host: usb@0xa6000000 {
>        compatible = "qcom,hsusb-host", "usb-ehci";
>        ---
> };
>
> hsusb1-host: usb@0xa6000000 {
>        compatible = "qcom,hsusb-host", "usb-ehci";
>        ---
> };
>
> /* super speed support
>
> ssusb0-device: usb@0xa6000000 {
>        compatible = "qcom,ssusb-device";
> };


The host controller node names as "usb@<adddr>" as you have here is
exactly right.  The driver doesn't care and will only look at the
compatible list.  OTG controllers can also use "usb@" as the prefix.

Controllers that are only in device mode should probably be called
something like "usb-gadget@<addr>" or similar, because "usb@" is used
for host controllers.

The label names (hsusb*-host, hsusb*-device) are completely irrelevant
since Linux never sees them.  Use whatever you want for the label
names.

Also, the form of the node name is: "usb@a6000000" (without '0x' in
the address).

g.
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: USB support for device tree

[Pavan Kondeti]

Hi

On 11/4/2011 10:15 PM, Grant Likely wrote:
> On Fri, Nov 4, 2011 at 4:25 AM, Pavan Kondeti <pkondeti@codeaurora.org> wrote:
>> Hi
>>
>> I am working on adding USB device tree support for MSM platform.  One of
>> our chip set has 2 hsusb cores. The first core is configured as otg and
>> the other core is configured in host only mode (EHCI compliant). Are the
>> below device node names Okay? Please suggest.
>>
>> hsusb0-otg: usb@0xa6000000 {
>>        compatible = "qcom,hsusb-otg";
>>        ---
>> };
>>
>> hsusb0-device: usb@0xa6000000 {
>>        compatible = "qcom,hsusb-device";
>>        ---
>> };
>>
>> hsusb0-host: usb@0xa6000000 {
>>        compatible = "qcom,hsusb-host", "usb-ehci";
>>        ---
>> };
>>
>> hsusb1-host: usb@0xa6000000 {
>>        compatible = "qcom,hsusb-host", "usb-ehci";
>>        ---
>> };
>>
>> /* super speed support
>>
>> ssusb0-device: usb@0xa6000000 {
>>        compatible = "qcom,ssusb-device";
>> };
> 
> 
> The host controller node names as "usb@<adddr>" as you have here is
> exactly right.  The driver doesn't care and will only look at the
> compatible list.  OTG controllers can also use "usb@" as the prefix.
> 
> Controllers that are only in device mode should probably be called
> something like "usb-gadget@<addr>" or similar, because "usb@" is used
> for host controllers.
> 

In our case OTG, gadget and host controllers are part of a single core
i.e they all share the same register address space.

If I use usb@addr for OTG controller, then host and OTG device node
names becomes same. Is that okay? Can two devices have the same device
node name in device tree source file?

> The label names (hsusb*-host, hsusb*-device) are completely irrelevant
> since Linux never sees them.  Use whatever you want for the label
> names.
> 
Is it a good practice to give label names? I thought, it improves a bit
readability of device tree source file.

> Also, the form of the node name is: "usb@a6000000" (without '0x' in
> the address).
> 
Thanks for the correction.

Thanks,
Pavan


-- 
Sent by a consultant of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.

Re: USB support for device tree

[Pavan Kondeti]

Hi Greg,

On 11/4/2011 9:13 PM, Greg KH wrote:
> On Fri, Nov 04, 2011 at 01:55:09PM +0530, Pavan Kondeti wrote:
>> Hi
>>
>> I am working on adding USB device tree support for MSM platform.  One of
>> our chip set has 2 hsusb cores. The first core is configured as otg and
>> the other core is configured in host only mode (EHCI compliant). Are the
>> below device node names Okay? Please suggest.
> 
> Why do you need to name the speed of the device and the host?  That
> doesn't really matter if you plug a low-speed device into that super
> speed host controller, right?
> 

Right :-)

> I'm not familiar with device tree naming conventions, but is this
> something you do for network devices?  If not, I'd not recommend it for
> USB controllers either.
> 

I am also not sure about the label names. The reason I would like to
have them in device tree source file is just for readability. One can
easily correlates the device node properties with the label. Say if I
want to change the IRQ number or some property value of 1st instance
HSUSB device controller (while porting a different board), I look for
hsusb0-device.

Thanks,
Pavan

-- 
Sent by a consultant of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.

Re: USB support for device tree

[Grant Likely]

On Fri, Nov 4, 2011 at 1:46 PM, Pavan Kondeti <pkondeti-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org> wrote:
> On 11/4/2011 10:15 PM, Grant Likely wrote:
>> The host controller node names as "usb@<adddr>" as you have here is
>> exactly right.  The driver doesn't care and will only look at the
>> compatible list.  OTG controllers can also use "usb@" as the prefix.
>>
>> Controllers that are only in device mode should probably be called
>> something like "usb-gadget@<addr>" or similar, because "usb@" is used
>> for host controllers.
>>
>
> In our case OTG, gadget and host controllers are part of a single core
> i.e they all share the same register address space.
>
> If I use usb@addr for OTG controller, then host and OTG device node
> names becomes same. Is that okay? Can two devices have the same device
> node name in device tree source file?

It is not legal for two device nodes to have overlapping 'reg' regions
(unless one is a child of the other), so by extension it is not okay
for two nodes to have the same 'name@addr'.  However, it is perfectly
acceptable and encouraged for two nodes at different addresses to
start with the same value for 'name@'.  This is called the generic
names recommended practice, and it can also be found in the ePAPR
documentation on node names.

If you want to have both host and device drivers bound to a single
device for OTG mode, then you should use a wrapper driver in Linux
that binds to the single node and instantiates each of the interfaces
as a child device.  For an example take a look at
drivers/usb/host/fsl-mph-dr-of.c.

>
>> The label names (hsusb*-host, hsusb*-device) are completely irrelevant
>> since Linux never sees them.  Use whatever you want for the label
>> names.
>>
> Is it a good practice to give label names? I thought, it improves a bit
> readability of device tree source file.

Yes, label names are important; the kernel just never sees them so I
don't care much.

g.
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: USB support for device tree

[Pavan Kondeti]

Hi

On 11/4/2011 11:42 PM, Grant Likely wrote:
> It is not legal for two device nodes to have overlapping 'reg' regions
> (unless one is a child of the other), so by extension it is not okay
> for two nodes to have the same 'name@addr'.  However, it is perfectly
> acceptable and encouraged for two nodes at different addresses to
> start with the same value for 'name@'.  This is called the generic
> names recommended practice, and it can also be found in the ePAPR
> documentation on node names.
> 
> If you want to have both host and device drivers bound to a single
> device for OTG mode, then you should use a wrapper driver in Linux
> that binds to the single node and instantiates each of the interfaces
> as a child device.  For an example take a look at
> drivers/usb/host/fsl-mph-dr-of.c.

Currently we have two platform devices one for OTG and one for host,
corresponding drivers for them. If I would like to keep it this way, the
device tree becomes something like below

hsusb0-otg: usb-otg@0xa6000000 {
	compatible = "qcom,hsusb-otg";
        ---
};

hsusb0-device: usb-gadget@0xa6000000 {
	compatible = "qcom,hsusb-device";
        ---
};

hsusb0-host: usb@0xa6000000 {
	compatible = "qcom,hsusb-host", "usb-ehci";
        ---
};

Are you okay with above naming convention?

-- 
Sent by a consultant of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: USB support for device tree

[Grant Likely]

[-- Attachment #1.1: Type: text/plain, Size: 1979 bytes --]

On Nov 4, 2011 11:08 PM, "Pavan Kondeti" <pkondeti-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org> wrote:
>
> Hi
>
> On 11/4/2011 11:42 PM, Grant Likely wrote:
> > It is not legal for two device nodes to have overlapping 'reg' regions
> > (unless one is a child of the other), so by extension it is not okay
> > for two nodes to have the same 'name@addr'.  However, it is perfectly
> > acceptable and encouraged for two nodes at different addresses to
> > start with the same value for 'name@'.  This is called the generic
> > names recommended practice, and it can also be found in the ePAPR
> > documentation on node names.
> >
> > If you want to have both host and device drivers bound to a single
> > device for OTG mode, then you should use a wrapper driver in Linux
> > that binds to the single node and instantiates each of the interfaces
> > as a child device.  For an example take a look at
> > drivers/usb/host/fsl-mph-dr-of.c.
>
> Currently we have two platform devices one for OTG and one for host,
> corresponding drivers for them. If I would like to keep it this way, the
> device tree becomes something like below
>
> hsusb0-otg: usb-otg@0xa6000000 {
>        compatible = "qcom,hsusb-otg";
>        ---
> };
>
> hsusb0-device: usb-gadget@0xa6000000 {
>        compatible = "qcom,hsusb-device";
>        ---
> };
>
> hsusb0-host: usb@0xa6000000 {
>        compatible = "qcom,hsusb-host", "usb-ehci";
>        ---
> };

No, you don't need three nodes. Only one node for the whole thing since
from the hardware perspective it is still a single device. The driver for
that node should create the child otg and gadget platform_devices so that
you can preserve the existing driver structure.

There does not need to be a device tree node for every struct device in the
kernel.

g.

>
> Are you okay with above naming convention?
>
> --
> Sent by a consultant of the Qualcomm Innovation Center, Inc.
> The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.

[-- Attachment #1.2: Type: text/html, Size: 2553 bytes --]

[-- Attachment #2: Type: text/plain, Size: 192 bytes --]

_______________________________________________
devicetree-discuss mailing list
devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org
https://lists.ozlabs.org/listinfo/devicetree-discuss

Re: USB support for device tree

[Pavan Kondeti]

Hi

On 11/5/2011 9:22 AM, Grant Likely wrote:
> On Nov 4, 2011 11:08 PM, "Pavan Kondeti" <pkondeti@codeaurora.org> wrote:
>>
>> Hi
>>
>> On 11/4/2011 11:42 PM, Grant Likely wrote:
>>> It is not legal for two device nodes to have overlapping 'reg' regions
>>> (unless one is a child of the other), so by extension it is not okay
>>> for two nodes to have the same 'name@addr'.  However, it is perfectly
>>> acceptable and encouraged for two nodes at different addresses to
>>> start with the same value for 'name@'.  This is called the generic
>>> names recommended practice, and it can also be found in the ePAPR
>>> documentation on node names.
>>>
>>> If you want to have both host and device drivers bound to a single
>>> device for OTG mode, then you should use a wrapper driver in Linux
>>> that binds to the single node and instantiates each of the interfaces
>>> as a child device.  For an example take a look at
>>> drivers/usb/host/fsl-mph-dr-of.c.
>>
>> Currently we have two platform devices one for OTG and one for host,
>> corresponding drivers for them. If I would like to keep it this way, the
>> device tree becomes something like below
>>
>> hsusb0-otg: usb-otg@0xa6000000 {
>>        compatible = "qcom,hsusb-otg";
>>        ---
>> };
>>
>> hsusb0-device: usb-gadget@0xa6000000 {
>>        compatible = "qcom,hsusb-device";
>>        ---
>> };
>>
>> hsusb0-host: usb@0xa6000000 {
>>        compatible = "qcom,hsusb-host", "usb-ehci";
>>        ---
>> };
> 
> No, you don't need three nodes. Only one node for the whole thing since
> from the hardware perspective it is still a single device. The driver for
> that node should create the child otg and gadget platform_devices so that
> you can preserve the existing driver structure.
>

Okay. I got it. Thanks for pointing me to
drivers/usb/host/fsl-mph-dr-of.c. I can have one device node for otg and
create host and/or gadget based on the operational mode.

> There does not need to be a device tree node for every struct device in the
> kernel.
> 

Agreed.

Thanks,
Pavan
-- 
Sent by a consultant of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.