Re: usb suspend backtraces.
David Brownell <[email protected]>
| Newsgroups | gmane.linux.usb.devel |
|---|---|
| Message-ID | <[email protected]> |
On Sunday 15 July 2007, Dave Jones wrote: > Before I dig deeper, are there other side-effects of changing > this default ? Other than spending more power than necessary? :) I don't think there should be, but Alan's most on top of these particular issues. The main thing I'd worry about is root hubs, which may be performing undesirable periodic DMAs (1/msec or more often) if they aren't allowed to suspend their downstream links. (Called, misleadingly, "global" suspend even though it's limited to a single USB bus.) As I recall, both OHCI and EHCI hardware is smart enough that such DMA polling can be turned off ... leaving the main cost of leaving some clocks/PLLs running that could otherwise be turned off. UHCI had some issues there. I think we've been finding out a consequence of Microsoft's rarely using the USB suspend mode: devices don't get tested with it, so many of them don't support it correctly. - Dave ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ [email protected] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel