Alan Stern wrote:
On Sat, 27 Dec 2003, David Brownell wrote:


Yes, a bug!  I fixed it in the updated patch which I've attached;
it applies cleanly to Greg's current usb-devel-2.6 BK.

[ Split out patches will coming out too, in dribs and drabs. ]


Okay. I won't have time to look at the updated patch for a little while.

In fact, I would like to see this stuff get into one of Greg's trees before I begin adding the "device-morphed" parts.

I'll send in some of those patches in the next few days, but I don't know when Greg will be integrating any of them.


Maybe by the time we finish, usb_set_configuration() will rely on callers
to grab the lock ... easy to do through sysfs, and usbfs needs changes
in any case.


That would make Duncan's life easier :-)

Likely someone should change that call, and all its callers, as a separate patch ... just to unblock other work.


> Nowhere does the core call disable_endpoint for ep0 at address 0.  I
> suppose the host drivers are supposed to take care of that themselves when
> they exit.

Actually that's one of the few remaining bits of automagic in the
interface to an HCD:  they're expected to morph that address's ep0
into a descriptor for the "real" address.  It's a UHCI-ism that
OHCI (and now EHCI) have had to cope with forever.

Removing that would be an excuse to clean up some minor cruft in
OHCI and EHCI.


Now might be a good time to take care of it.

Someone else would have to contribute that work, and test it...


- Dave




Alan Stern








------------------------------------------------------- This SF.net email is sponsored by: IBM Linux Tutorials. Become an expert in LINUX or just sharpen your skills. Sign up for IBM's Free Linux Tutorials. Learn everything from the bash shell to sys admin. Click now! http://ads.osdn.com/?ad_id=1278&alloc_id=3371&op=click _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to