[9fans] usb vid/did

[9fans] usb vid/did

[erik quanstrom]

it doesn't seem to be available in anyone's control files.

usb(3) says

        This may result from the read of an endpoint control file:

               (the first line is wrapped to make it fit here)
               enabled control rw speed full maxpkt 64 pollival 0
                    samplesz 0 hz 0 hub 1 port 3 busy
               storage csp 0x500608 vid 0x951 did 0x1613 Kingston 'DT 101 II'

yet my endpoints never have a vid nor a did.

kw# grep . */ctl
ep1.0/ctl:enabled control rw speed high maxpkt 64 pollival 0 samplesz 0 hz 0 hub 0 port 0 busy
ep1.0/ctl:roothub csp 0x000009 ports 1 ehci
ep2.0/ctl:enabled control rw speed high maxpkt 64 pollival 0 samplesz 0 hz 0 hub 1 port 1 busy
ep2.0/ctl:hub csp 0x020009 ports 7 none <nil> ehci
ep3.0/ctl:enabled control rw speed full maxpkt 8 pollival 0 samplesz 0 hz 0 hub 2 port 2 busy
ep3.0/ctl:hub csp 0x000009 ports 3 'Mitsumi Electric' 'Hub in Apple Extended USB Keyboard' ehci

a little grep-o-matic indicates this is just the way it goes.
is this an oversite, or does usb(3) just not have that information?

if not, usbd(4) does have that information.  are there any problems
with adding it and correcting the man page?

- erik

Re: [9fans] usb vid/did

[Nemo]

i think usbd added that. 
iirc it should do it. I'll take a look anyway. maybe the code was gone
in some change...

On Jan 7, 2011, at 2:12 PM, erik quanstrom <quanstro@quanstro.net> wrote:

> it doesn't seem to be available in anyone's control files.
> 
> usb(3) says
> 
>        This may result from the read of an endpoint control file:
> 
>               (the first line is wrapped to make it fit here)
>               enabled control rw speed full maxpkt 64 pollival 0
>                    samplesz 0 hz 0 hub 1 port 3 busy
>               storage csp 0x500608 vid 0x951 did 0x1613 Kingston 'DT 101 II'
> 
> yet my endpoints never have a vid nor a did.
> 
> kw# grep . */ctl
> ep1.0/ctl:enabled control rw speed high maxpkt 64 pollival 0 samplesz 0 hz 0 hub 0 port 0 busy
> ep1.0/ctl:roothub csp 0x000009 ports 1 ehci
> ep2.0/ctl:enabled control rw speed high maxpkt 64 pollival 0 samplesz 0 hz 0 hub 1 port 1 busy
> ep2.0/ctl:hub csp 0x020009 ports 7 none <nil> ehci
> ep3.0/ctl:enabled control rw speed full maxpkt 8 pollival 0 samplesz 0 hz 0 hub 2 port 2 busy
> ep3.0/ctl:hub csp 0x000009 ports 3 'Mitsumi Electric' 'Hub in Apple Extended USB Keyboard' ehci
> 
> a little grep-o-matic indicates this is just the way it goes.
> is this an oversite, or does usb(3) just not have that information?
> 
> if not, usbd(4) does have that information.  are there any problems
> with adding it and correcting the man page?
> 
> - erik
> 

Re: [9fans] usb vid/did

[Gorka Guardiola]

I do see vids/dids just not for hubs.

-
Curiosity sKilled the cat

G.

On Jan 7, 2011, at 3:28 PM, Nemo <nemo.mbox@gmail.com> wrote:

> i think usbd added that. 
> iirc it should do it. I'll take a look anyway. maybe the code was gone
> in some change...
> 
> On Jan 7, 2011, at 2:12 PM, erik quanstrom <quanstro@quanstro.net> wrote:
> 
>> it doesn't seem to be available in anyone's control files.
>> 
>> usb(3) says
>> 
>>       This may result from the read of an endpoint control file:
>> 
>>              (the first line is wrapped to make it fit here)
>>              enabled control rw speed full maxpkt 64 pollival 0
>>                   samplesz 0 hz 0 hub 1 port 3 busy
>>              storage csp 0x500608 vid 0x951 did 0x1613 Kingston 'DT 101 II'
>> 
>> yet my endpoints never have a vid nor a did.
>> 
>> kw# grep . */ctl
>> ep1.0/ctl:enabled control rw speed high maxpkt 64 pollival 0 samplesz 0 hz 0 hub 0 port 0 busy
>> ep1.0/ctl:roothub csp 0x000009 ports 1 ehci
>> ep2.0/ctl:enabled control rw speed high maxpkt 64 pollival 0 samplesz 0 hz 0 hub 1 port 1 busy
>> ep2.0/ctl:hub csp 0x020009 ports 7 none <nil> ehci
>> ep3.0/ctl:enabled control rw speed full maxpkt 8 pollival 0 samplesz 0 hz 0 hub 2 port 2 busy
>> ep3.0/ctl:hub csp 0x000009 ports 3 'Mitsumi Electric' 'Hub in Apple Extended USB Keyboard' ehci
>> 
>> a little grep-o-matic indicates this is just the way it goes.
>> is this an oversite, or does usb(3) just not have that information?
>> 
>> if not, usbd(4) does have that information.  are there any problems
>> with adding it and correcting the man page?
>> 
>> - erik
>> 
> 

Re: [9fans] usb vid/did

[erik quanstrom]

>> ep3.0/ctl:hub csp 0x000009 ports 3 'Mitsumi Electric' 'Hub in Apple Extended USB Keyboard' ehci

>I do see vids/dids just not for hubs.

perhaps i misunderstand what is ment by a "root hub".
i thought that ment that it was a hub that usb(3) invented.
a hub in a keyboard can't be a root hub, can it?

- erik

Re: [9fans] usb vid/did

[Francisco J Ballesteros]

Ah.
A root hub is an invention of the usb std.
There's no such thing. It's the controller.
It can be that usbd does not write vid/did for root hubs, perhaps.

I've not seen a root hub in a keyboard. If it's a usb keyboard, it can't be
a root hub, because you have to plug the kbd into a port, which
belongs to a hub,
root or not.

On Fri, Jan 7, 2011 at 4:59 PM, erik quanstrom <quanstro@labs.coraid.com> wrote:
>>> ep3.0/ctl:hub csp 0x000009 ports 3 'Mitsumi Electric' 'Hub in Apple Extended USB Keyboard' ehci
>
>>I do see vids/dids just not for hubs.
>
> perhaps i misunderstand what is ment by a "root hub".
> i thought that ment that it was a hub that usb(3) invented.
> a hub in a keyboard can't be a root hub, can it?
>
> - erik
>
>

Re: [9fans] usb vid/did

[erik quanstrom]

On Fri Jan  7 11:51:23 EST 2011, nemo@lsub.org wrote:
> Ah.
> A root hub is an invention of the usb std.
> There's no such thing. It's the controller.
> It can be that usbd does not write vid/did for root hubs, perhaps.
>
> I've not seen a root hub in a keyboard. If it's a usb keyboard, it can't be
> a root hub, because you have to plug the kbd into a port, which
> belongs to a hub,
> root or not.

okay, so there's some sort of bug.  what kind of output
would you like.  this is for the kw ehci.

- erik

Re: [9fans] usb vid/did

[Gorka Guardiola]

On Fri, Jan 7, 2011 at 5:52 PM, erik quanstrom <quanstro@labs.coraid.com> wrote:
> okay, so there's some sort of bug.  what kind of output
> would you like.  this is for the kw ehci.
>

Hmm, if it is a bug it is on purpose I think. You are right
it could probably change to show vid/did.

All is started by startdev in
/sys/src/cmd/usb/usbd/usbd.c:224

Devices have their info written by
writeinfo()
/sys/src/cmd/usb/usbd/dev.c:217

Hubs (root or not) have their info written by usbd in newhub()
/sys/src/cmd/usb/usbd/usbd.c:177

Hubs all have the same format. As root hubs are invented,
they don't have vid/did, so plain hubs don't either to have the same format.
The vid/did can be added  and to make the string
similar maybe the vid/did for roohubs can be set to 0xffff/0xffff or something.

You only have to change the info strings of newhub to be more like
the ones in writeinfo and see if anything breaks (I don't think it
should but...).

HTH
G.