> The main problem with the data memory is simply that some locations are not
> guaranteed to map continuosly to PCI space, so that accidently crossing a
> page boundary may fail.
That sounds like an issue that pci_map_single() and siblings should
be addressing ... but the last time I checked, on x86 they were just
wrappers around virt_to_bus(), and couldn't report errors. (Some
other platforms have more complex implementations...) The API
docs didn't mention any such issues either.
Is there a complete list of those locations ("not guaranteed to map
continuously to PCI space")?
- Dave
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
http://lists.sourceforge.net/lists/listinfo/linux-usb-devel
- Re: [linux-usb-devel] Bulk URB submission. Brad Hards
- Re: [linux-usb-devel] Bulk URB submission. xavier
- Re: [linux-usb-devel] Bulk URB submission. Tony Hoyle
- Re: [linux-usb-devel] Bulk URB submission. Brad Hards
- Re: [linux-usb-devel] Bulk URB submission. Martin Diehl
- Re: [linux-usb-devel] Bulk URB submission. David Brownell
- Re: [linux-usb-devel] Bulk URB submission. Martin Diehl
- Re: [linux-usb-devel] Bulk URB submission. David Brownell
- Re: [linux-usb-devel] Bulk URB submission. Martin Diehl
- Re: [linux-usb-devel] Bulk URB submission. Georg Acher
- Re: [linux-usb-devel] Bulk URB submission. David Brownell
- Re: [linux-usb-devel] Bulk URB submission. David Brownell
- Re: [linux-usb-devel] Bulk URB submission. Brad Hards
- Re: [linux-usb-devel] Bulk URB submission. David Brownell
