On Tuesday 18 January 2005 3:07 pm, Pete Zaitcev wrote:
> On Tue, 18 Jan 2005 13:51:22 -0800, David Brownell <[EMAIL PROTECTED]> wrote:
> 
> > Thereby breaking usb on various systems ... by overwriting the root hub
> > status message data as part of bounce buffer handling.   NOT a good idea!
> > Those patches were added for important reasons.  (Or did you add some
> > other solution to the issue described in that comment?)
> 
> What was the nature of the breakage exactly? Like I mentioned, everything
> is peachy here, but I don't have weirdo controllers like sl81xx or what is
> it called.

That driver doesn't do DMA, it wouldn't have that problem.  I think
the problem was reported by Russell King with some hardware using the
SA-1111 OHCI controller on an SA-1100 board ... one of several types
of system where DMA bounce buffers get much used.  (No, you wouldn't
usually see that with a PC.  I think there should be a kernel debug
option to use bounce buffers for all DMA...)  Check the list archives
and BK.  The current resolution is the cleanest patch we found for
that articular problem.


> > There are a lot of non-usbmon patches mixed into this ...
> 
> Not a lot, but yes, I'm sorry about that.

Mind you, I'd like to see better handling of _that particular_ error!

I think some folk have been mucking around with early boot logic, since
that message started to appear a lot recently.  The fix has been to
arrange to make the root filesystem have the three /dev nodes that
the initramfs is supposed to have ... it's as if initramfs isn't
doing the job, or is vanishing too early, or something equally rude.

- Dave


> 
> -- Pete
> 


-------------------------------------------------------
The SF.Net email is sponsored by: Beat the post-holiday blues
Get a FREE limited edition SourceForge.net t-shirt from ThinkGeek.
It's fun and FREE -- well, almost....http://www.thinkgeek.com/sfshirt
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to