On Wed, 13 Feb 2002, Johannes Erdfelt wrote: > Yes, you *must* double buffer if you want Isochronous to be seamless. > Isochronous transfers occur at the beginning of the frame and if you're > scheduling for the next frame at the end of the previous, you'be already > lost.
> Have you tried more than 2 frames in your double buffer. Yes, at one time we did. There was no noticeable improvement over what I'm already getting. (Clear with OHCI, kind of poppy with UHCI) > > I wouldn't trust a 2 frame double buffer. > I suspect it may have something to do with the fact that it's a VIA controller(1). I just read an email from a guy that's noticed similar troubles. I have not yet (IIRC) tried on a non-VIA UHCI controller. (1) From a contact we have at a USB-client chipset manufacturer, VIA USB chipsets suck pretty badly. He told me that Microsoft had to release a special driver for the VIA USB chipset because it was so bad, so I presume there are some quirks about it that may not be accounted for. > > More data: IIRC, on the vanilla uhci.c driver, ISOC sucks. Sucks really > > badly. Kind of funny though, 'cause I was helping somebody with writing a > > driver for something that used BULK transfers, and the usb-uhci driver > > kept on Oops!'ing his system (dereferencing a NULL pointer in the host > > controller code) whenever he tried to initiate his transfer. Switching to > > the uhci.c driver corrected this. Go Figure. So both drivers are > > good/bad at different things, though the usb-uhci driver seems to > > perform/behave better for most cases I've seen. > > Define "sucks badly". I'll have to do some testing again to verify, but when using the same code that I was running just fine on OHCI, and not terribly badly on the usb-uhci driver, it is REALLY crappy on the uhci driver(2). Maybe it's the way it does ISOC transfers, so adding another buffer frame would improve it. I will try this. (2) I think the sound was not even barely understandable. Really choppy, and IIRC I got of bus ISOC scheduling printks all over my logs :-) Could be a bad thing. > > Anyway, hope this helps. By the time I get done with emails like this, I > > always seem to forget something important I intended to put in, so if this > > seems incomplete, or you have any questions, feel free to reply :-) > > JE Thanks for replying. I hadn't submitted a bug report or anything like that partly because I've been pretty busy, but mostly because I wasn't sure what could be done about the VIA chipset stuff, so I kept quiet. :-) Matthew Fredrickson _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
