Knowing that I am neither in the group, nor qualified enough to give any statements that go beyond my opinion on this issue; Is it possible to change the strategy on the fly by writing to a kernel pseudo-device at /proc ? This would keep the fallback solution available without recompiling the kernel.
Sorry for the interference Regards Ullrich -----Original Message----- From: David Brownell [mailto:[EMAIL PROTECTED] Sent: Freitag, 25. Juni 2004 18:09 To: Alan Stern Cc: Greg KH; Ullrich Sigwanz; USB development list Subject: Re: PATCH: (as332) Don't ask for string descriptor lengths Alan Stern wrote: > > usb_string() is moved up in the file to right after usb_get_string(), > putting it together with all the other descriptor-reading utilities. > More importantly, it no longer tries to read a string's length before > transferring the string. In this respect it's more like it used to be > back in 2.4. > > .... > > I realize that we have discussed making this sort of change before, > although I don't remember coming to any definite conclusion. Does anyone > feel very strongly about not applying this? No, though I wish I knew more about why it was changed. The only thing that sticks in my memory is that some devices had bugs that prevented the "read all bytes at once" sort of strategy from working properly. Probably the most compatible approach would be to keep one strategy as a fallback. I know that until I modified "lsusb" to know about both strategies, there were devices that wouldn't deliver strings to it. - Dave ------------------------------------------------------- This SF.Net email sponsored by Black Hat Briefings & Training. Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital self defense, top technical experts, no vendor pitches, unmatched networking opportunities. Visit www.blackhat.com _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
