Re: HEADSUP usb2 (usb4bsd) to become default in 2 weeks.

----- Original Message ----- 
From: "Hans Petter Selasky" <hselasky_at_c2i.net>
To: <remko_at_elvandar.org>
Cc: <freebsd-current_at_freebsd.org>; "Pegasus Mc Cleaft" <ken_at_mthelicon.com>
Sent: Tuesday, December 23, 2008 10:30 AM
Subject: Re: HEADSUP usb2 (usb4bsd) to become default in 2 weeks.


> On Tuesday 23 December 2008, Remko Lodder wrote:
>> On Tue, December 23, 2008 11:03 am, Hans Petter Selasky wrote:
>> > On Tuesday 23 December 2008, Pegasus Mc Cleaft wrote:
>> >> I believe that the Prolific USB serial adapter is broken on AMD64.  I
>> >> tried
>> >> one of the adapters on my machine and no matter what serial speed I 
>> >> was
>> >> selecting with cu it would come back with an error stating the speed 
>> >> was
>> >> unsupported. I moved the adapter to a 8-current machine (i386) with 
>> >> the
>> >> usb1 stack and everything worked fine.
>> >>
>> >> Peg
>> >
>> > Are you sure that this is not an mpsafetty problem?
>>
>> Given that he states that it works under the original USB stack, I would
>> guess that it's not an MPsafeTTY thingy else it wouldn't work under the
>> original stack either.
>
> Try to turn on USB debugging for:
>
> sysctl hw.usb2.ucom.debug=15
> sysctl hw.usb2.uftdi.debug=15
>
> Have a look in "/usr/src/sys/dev/usb2/serial/uftdi2.c" and the function:
>
> uftdi_8u232am_getrate().
>
> Extra checks have been added to detect invalid line speeds between USB1 
> and
> USB2.
>
> What rate are you trying to set?
>
> Minimum and maximum rate is defined like this:
>
>        if ((speed < 178) || (speed > ((3000000 * 100) / 97)))
>                return (1);             /* prevent numerical overflow */
>
> --HPS
>
> uftdi_8u232am_getrate

Hi Hans,

    I was trying something like:  cu -l /dev/cuaU0 -s 9600

    I did also try the same with /dev/ttyU0, both with various speeds from 
300 to 115200.

    Interestingly, I also tried doing a cat /dev/random > /dev/cuaU0 and 
that didnt throw an error (Dont know if it actually sent anything as I didnt 
have a scope hooked to the line). Like I said in another email, I will try 
moving the machines around between the USB1 and 2 stack tongiht and see if I 
can pin it down to a specific combination (MP vrs non-MP, AMD64 vrs i386)

Peg