Re: OHCI hangs after failing to free resources

On Fri, 27 Apr 2007, Mike Nuss wrote:

> > Look at ohci-hcd.c:ohci_endpoint_disable().  There's a
> "switch(ed->state)"
> > line; see what it does in the ED_UNLINK case.  It retries for up to
> 1000
> > ms, waiting for ed->state to change to ED_IDLE.  That's where the "IRQ
> > INTR_SF lossage" message comes from, if ed->state never changes.
> 
> I have indeed looked in there; it was originally where I thought the
> problem was coming from. It's interesting that you say it should retry
> for up to 1000 ms. schedule_timeout_uninterruptible(1) sleeps for one
> kernel tick, correct? And the limit is decreased by one each time
> through the loop. So for smaller values of HZ, it would retry for
> longer. Was this the intended behavior?

You're right; for some reason I was thinking that the argument to 
schedule_timeout_interruptible() was in milliseconds, not jiffies.

I don't know the intended behavior because I didn't write the code.  In 
principle one would think that retrying for 10 ms or even less would be 
sufficient, since the SOF interrupt is supposed to occur every 
millisecond.


> Of course. Which reminds me - there's a comment in ohci-pci.c:
> 
> /* Check for Compaq's ZFMicro chipset, which needs short
>  * delays before control or bulk queues get re-activated
>  * in finish_unlinks()
>  */
> This is the chipset I'm using (and the quirk is enabled). Is it possible
> there's also an issue with interrupt queues handling? I don't know what
> the issue was that led to the creation of the quirk, but it seems like
> it's in the same ballpark.

Certainly it's possible.  I don't know what led to the creation of the 
ZFMicro quirk either, but Dave does.

> I think it's pretty clear there's an underlying hardware issue, but
> maybe there is a workaround. I added the additional debug code you
> suggested to get a better idea of what finish_unlinks() is doing (or not
> doing).

And the results are... ?

Alan Stern


-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel