Hi,
I have sent approximately two months ago mails about the problems concerning resetting a usb device and I was pointed to usb_physical_reset_device. Then I saw that this function was never exported as others like usb_reset_device or declared at <linux/usb.h>.
I noticed that too. Also: that elsewhere Linux uses the "__" convention to distinguish safe from dangerous versions of a function (like with the __list_* variants of list_* calls). And in this case I'm starting to think that calling threads (khubd, usb-storage EH, usbfs) should be made to grab dev->serialize themselves, anyway.
The result of this was a warning of the compiler for implicit declaration of this function ( in plain text the function was never called ). Is there a patch to fix this for
linux-2.6.0.
There's some work afoot to clean up the reset processing so it'll work more reasonably for DFU-style devices ... hmm, like certain Atmel WLAN adapters! I posted an updated patch on the "re-factor enumeration/reset paths" thread just yesterday; the original one described the changes, including a relevant rename+export.
Eventually it should be possible for a driver probe() to download firmware then reset the device (the name of that call may change). DFU-ish devices then "fail", so the device probe will finish with an error since the original device is gone ... to be replaced by a new device, with new descriptors, which gets then probed in turn.
- Dave
------------------------------------------------------- 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
