[9fans] usb disk error

[9fans] usb disk error

[erik quanstrom]

i don't recall getting these before.  this seems
like new behavior.  i got 4 of these when downloading
4gb of pictures from a sd card.  never in
the same place twice.  sb600 ohci controller.

disk... reset: device is detached

it seems that umsrecover -> umsreset -> usbcmd and

	if(usbcmd(ums->dev, r, Umsreset, 0, 0, nil, 0) < 0){
>		fprint(2, "disk: reset: %r\n");

i don't see exactly where it is, but could this be
a timeout issue?  that's the main difference i see
between now and the previous time.  turning on
debugging might not help.  there's just too much
of it.  sometimes it's many gb of transfer before
an error.

- erik

Re: [9fans] usb disk error

[W B Hacker]

erik quanstrom wrote:
> i don't recall getting these before.  this seems
> like new behavior.  i got 4 of these when downloading
> 4gb of pictures from a sd card.  never in
> the same place twice.  sb600 ohci controller.
>
> disk... reset: device is detached
>
> it seems that umsrecover -> umsreset -> usbcmd and
>
> 	if(usbcmd(ums->dev, r, Umsreset, 0, 0, nil, 0) < 0){
>> 		fprint(2, "disk: reset: %r\n");
>
> i don't see exactly where it is, but could this be
> a timeout issue?  that's the main difference i see
> between now and the previous time.  turning on
> debugging might not help.  there's just too much
> of it.  sometimes it's many gb of transfer before
> an error.
>
> - erik
>
>

If a new problem, it 'should' be a hardware issue. Bits, as you know as well as
anyone, do not really 'rot'.

Wots the ambient temperature, device temp, planar temp and surface
cleanliness/humidity at the area of the driving silicon and on any
sockets/conectors?

And if clean, cool, and well-connected, do the PS rail and signals look clean on
a 'scope?

Bill

Re: [9fans] usb disk error

[Francisco J Ballesteros]

Can you get the error with the old version of the code?

Nothing really changed, the disk should be asking for timeouts IIRC.


On Tue, Oct 20, 2009 at 9:33 PM, erik quanstrom <quanstro@quanstro.net> wrote:
> i don't recall getting these before.  this seems
> like new behavior.  i got 4 of these when downloading
> 4gb of pictures from a sd card.  never in
> the same place twice.  sb600 ohci controller.
>
> disk... reset: device is detached
>
> it seems that umsrecover -> umsreset -> usbcmd and
>
>        if(usbcmd(ums->dev, r, Umsreset, 0, 0, nil, 0) < 0){
>>               fprint(2, "disk: reset: %r\n");
>
> i don't see exactly where it is, but could this be
> a timeout issue?  that's the main difference i see
> between now and the previous time.  turning on
> debugging might not help.  there's just too much
> of it.  sometimes it's many gb of transfer before
> an error.
>
> - erik
>
>

Re: [9fans] usb disk error

[erik quanstrom]

> Can you get the error with the old version of the code?

nope.

- erik