On Sun, Oct 13, 2002, David Brownell <[EMAIL PROTECTED]> wrote: > > >> > However, I don't think BW can be calculated effectively by common code, it > >> > is dependent on the HCD. > >> > >>That's how it behaves already, with the different HCDs putting more > >>or less work into their scheduling policy. > > > > That's how it is done now, but since the original code was the same > > across all HCD's, I don't see why it should differ between HCD's. > > Because the HCDs do scheduling differently. UHCI can significantly > undercommit periodic bandwidth, and does no balancing. OHCI does > balancing, so it does a lot less undercommitting. EHCI doesn't do > much periodic scheduling yet; drivers haven't needed it, and anyway > its problem is about ten times harder because of split transactions.
Yeah, I think interrupt URB's will vary significantly regardless. Too many limitation withs UHCI. Which reminds me, I need to add balancing to my TODO list for UHCI. > > USB 1.1 is the same across all HCD's. The only thing I can think of that > > would differ is what frames interrupt URB's get scheduled into. > > Example, the balancing that OHCI does to even out loads. (Which > applies equally to iso transfers -- they don't need to run every > frame, though they often seem to.) Careful with Isochronous. By definition, it shouldn't be delayed and should work identically across all HCD's. If they don't, we should fix that. JE ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
