[PATCH] Documentation: dts: fsl-usb: Document USB node compatible string for IP version

[Scott Wood]

On Wed, 2014-08-27 at 00:46 -0500, Mehresh Ramneek-B31383 wrote:
> 
> > -----Original Message-----
> > From: Wood Scott-B07421
> > Sent: Wednesday, August 27, 2014 4:23 AM
> > To: Mehresh Ramneek-B31383
> > Cc: Badola Nikhil-B46172; linuxppc-dev at lists.ozlabs.org;
> > devicetree at vger.kernel.org
> > Subject: Re: [PATCH] Documentation: dts: fsl-usb: Document USB node
> > compatible string for IP version
> > 
> > On Fri, 2014-08-22 at 00:05 -0500, Mehresh Ramneek-B31383 wrote:
> > >
> > > -----Original Message-----
> > > From: Badola Nikhil-B46172
> > > Sent: Friday, August 22, 2014 10:18 AM
> > > To: Wood Scott-B07421
> > > Cc: linuxppc-dev at lists.ozlabs.org; devicetree at vger.kernel.org; Mehresh
> > > Ramneek-B31383
> > > Subject: RE: [PATCH] Documentation: dts: fsl-usb: Document USB node
> > > compatible string for IP version
> > >
> > > Adding Ramneek
> > >
> > > > -----Original Message-----
> > > > From: Wood Scott-B07421
> > > > Sent: Friday, August 22, 2014 3:53 AM
> > > > To: Badola Nikhil-B46172
> > > > Cc: linuxppc-dev at lists.ozlabs.org; devicetree at vger.kernel.org
> > > > Subject: Re: [PATCH] Documentation: dts: fsl-usb: Document USB node
> > > > compatible string for IP version
> > > >
> > > > On Thu, 2014-08-21 at 14:48 +0530, Nikhil Badola wrote:
> > > > > Document compatible string containing IP version in USB device
> > > > > tree node
> > > > >
> > > > > Signed-off-by: Nikhil Badola <nikhil.badola at freescale.com>
> > > > > ---
> > > > >  Documentation/devicetree/bindings/usb/fsl-usb.txt | 13
> > > > > ++++++++-----
> > > > >  1 file changed, 8 insertions(+), 5 deletions(-)
> > > >
> > > > Please CC devicetree at vger.kernel.org on all device tree patches (in
> > > > addition to linuxppc-dev).
> > > >
> > > > >
> > > > > diff --git a/Documentation/devicetree/bindings/usb/fsl-usb.txt
> > > > > b/Documentation/devicetree/bindings/usb/fsl-usb.txt
> > > > > index 4779c02..5a3a0a8 100644
> > > > > --- a/Documentation/devicetree/bindings/usb/fsl-usb.txt
> > > > > +++ b/Documentation/devicetree/bindings/usb/fsl-usb.txt
> > > > > @@ -10,7 +10,10 @@ Required properties :
> > > > >     controllers, or "fsl-usb2-dr" for dual role USB controllers
> > > > >     or "fsl,mpc5121-usb2-dr" for dual role USB controllers of MPC5121.
> > > > >     Wherever applicable, the IP version of the USB controller should
> > > > > -   also be mentioned (for eg. fsl-usb2-dr-v2.2 for bsc9132).
> > > > > +   also be mentioned in another string.
> > > > > +   For multi port host USB controller with IP version <IP_Ver>, it should
> > be
> > > > > +   "fsl-usb2-mph-<IP_Ver>". For dual role USB controller with IP version
> > > > > +   <IP_Ver>, it should be "fsl-usb2-dr-<IP_Ver>".
> > > >
> > > > It was documented before -- this is just making it more explicit, right?
> > > >
> > > > FWIW, the version number can be read out of a USB register, so I'd
> > > > rather remove the suggestion to specify the version number and
> > > > replace it with a reference to the ID register.
> > > we have following two issues -
> > > (a) our USBIP version register doesn't have consistent "version field
> > > size" over multiple version(s). This is why we couldn't use it for
> > > reading version info across various IP versions
> > > (b) this register is not exposed in all SoC RMs (probably because of
> > > above reason)
> > 
> > :-(
> > 
> > If this is just a problem with older chips, we could have a new compatible name
> > that designates the family of USB block versions with a sane version register.
> > 
> we could have done...but we have a requirement to write version specific code...
> for instance, usb controller init sequence has changes from version 2.5 onwards...
> then there are version specific errata fixe(s) also. Hence we decided to go for
> compatible string containing hw ip version (major no.) so that our workaround/code is
> consistent with hw ip version(s) published in errata(s)

I wasn't saying ignore version differences.  I was suggesting that we
make a special case out of older versions where the version register is
different/broken.

-Scott