Status of isp116x driver... (on mpc8xx)?

Status of isp116x driver... (on mpc8xx)?

[David Jander]

Hi,

I connected a Philips ISP1160 usb host controller to a MPC852T-based custom 
board. I am using CVS kernel from DENX. 
I added some stuff to hcd-1161.h and hcd-1161.c to support our hardware, and 
it seems to work at first.
The controller is recognized, initialized and everything seems fine, but when 
I connect a device to one of the two ports, every time the function 
fnvIsp1161AtlRead() is called, it reads in a 72 byte-long block of zeroes.
Then I get "not enough configurations..." (error=-22).
If I printk-out the contents of the buffer status register (REG_BUFF_STS) at 
that point, it is zero (=no data read).
Nevertheless, putting a scope on the physical USB wires, I can see quite some 
amount of non-zero data flying by.

Is the isp1160 chip really supported by the driver? tested?
Are access timings very critical? I see some ugly iodelay_hcd() calls here and 
there, are they necessary? why?

What about the other (older??) hc_isp116x.c driver floating around somewhere 
on the net, from Roman Weissgaerber? I found version 0.9.6 (apparently for 
kernel 2.4.22) somewhere, made some adaptations to get it loaded in our 
kernel but it either oopses after initializing the hc, or simply freezes.
Whats the relationship between this and the driver hacked by Wolfgang Denk?

Is someone using a isp116x on a mpc8xx and can tell his experience on getting 
it working?

Greetings,

-- 
David Jander
Protonic Holland.

Re: Status of isp116x driver... (on mpc8xx)?

[Wolfgang Denk]

Dear David,

in message <200505111240.35425.david.jander@protonic.nl> you wrote:
> 
> I connected a Philips ISP1160 usb host controller to a MPC852T-based custom 
> board. I am using CVS kernel from DENX. 
> I added some stuff to hcd-1161.h and hcd-1161.c to support our hardware, and 
> it seems to work at first.
...
> Is the isp1160 chip really supported by the driver? tested?

We only use (and tested) this driver with a ISP1361 ...

> Is someone using a isp116x on a mpc8xx and can tell his experience on getting 
> it working?

Sorry, we don't have any experience with a 1160. The  driver  in  our
tree  works fine with the 1362, both as host driver and as device (as
a gadget driver).

Best regards,

Wolfgang Denk

-- 
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de
No man knows what true happiness is until he gets married.  By  then,
of course, its too late.

Re: Status of isp116x driver... (on mpc8xx)?

[David Jander]

On Wednesday 11 May 2005 15:28, Wolfgang Denk wrote:
>[...]
> > Is the isp1160 chip really supported by the driver? tested?
>
> We only use (and tested) this driver with a ISP1361 ...

Ok. The sources of your driver and the one from Roman Weissgaerber look very 
different, so I can't really tell: did you start from scratch and picked some 
things from RW's driver, or did you modify (beyond recognition) and extend 
his driver?
I ask, because now I have RW's driver (version 0.9.6) working almost 100% 
alright. The issues we still have are most probably related to bus-timing 
problems (the ISP1160 is connected via a CPLD on an 8-bit wide bus controlled 
by UPMA!) and occur sporadically. Your driver OTOH still doesn't manage to 
receive Atl packets that are not entirely filled with zeroes ;-)

> > Is someone using a isp116x on a mpc8xx and can tell his experience on
> > getting it working?
>
> Sorry, we don't have any experience with a 1160. The  driver  in  our
> tree  works fine with the 1362, both as host driver and as device (as
> a gadget driver).

I'll work on the problem, and if I get your driver working, I'll send a patch 
that hopefully fixes 116x support again ;-)

Greetings,

-- 
David Jander
Protonic Holland.

Re: Status of isp116x driver... (on mpc8xx)?

[Wolfgang Denk]

Dear David,

in message <200505121455.15864.david.jander@protonic.nl> you wrote:
> 
> Ok. The sources of your driver and the one from Roman Weissgaerber look very 
> different, so I can't really tell: did you start from scratch and picked some 
> things from RW's driver, or did you modify (beyond recognition) and extend 
> his driver?

We used and extended a driver based on RW code. If you want to  folow
the history of changes in our CVS then please note thatthe driver was
renamed from drivers/usb/hcd-1161.* into drivers/usb/host/hcd-1161.*,
so for the first changes look for drivers/usb/hcd-1161.*.

> I'll work on the problem, and if I get your driver working, I'll send a patch 
> that hopefully fixes 116x support again ;-)

Thanks in advance, and good luck!

Best regards,

Wolfgang Denk

-- 
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de
The evolution of the human race will not be accomplished in  the  ten
thousand  years  of  tame  animals,  but in the million years of wild
animals, because man is and will always be a wild animal.
                                              - Charles Galton Darwin