[Xen-devel] [acooks@xxxxxxxxx: Re: [RFC PATCH v2 1/2] pci: Create PCIe requester ID interface]

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 24.07.13 at 18:26, Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx> wrote:
> Does it make sense to integrate these PCI ids as part of the
> phantom device? Right now we have a parameter where one
> can specify them, but this is a nice complete list of the
> actual devices.

I don't mind adding such a table, but I'm also not eager to do
so myself...

Jan

> Here's a table of affected devices with links to the
> bug reports. In each case both functions 0 and 1 are used.
> 
> static const struct pci_dev_dma_multi_source_map {
>        u16 vendor;
>        u16 device;
> } pci_dev_dma_multi_source_map[] = {
>         /* Reported by Patrick Bregman
>          * https://bugzilla.redhat.com/show_bug.cgi?id=863653 */
>        { PCI_VENDOR_ID_MARVELL_EXT, 0x9120},
> 
>        /* Reported by  PaweÅ Åak, Korneliusz JarzÄbski, Daniel Mayer
>         * https://bugzilla.kernel.org/show_bug.cgi?id=42679 and by
>         * Justin Piszcz  https://lkml.org/lkml/2012/11/24/94 */
>        { PCI_VENDOR_ID_MARVELL_EXT, 0x9123},
> 
>        /* Reported by Robert Cicconetti
>         * https://bugzilla.kernel.org/show_bug.cgi?id=42679 and by
>         * Fernando https://bugzilla.redhat.com/show_bug.cgi?id=757166 */
>        { PCI_VENDOR_ID_MARVELL_EXT, 0x9128},
> 
>        /* Reported by Stijn Tintel
>         * https://bugzilla.kernel.org/show_bug.cgi?id=42679 */
>        { PCI_VENDOR_ID_MARVELL_EXT, 0x9130},
> 
>        /* Reported by Gaudenz Steinlin
>         * https://lkml.org/lkml/2013/3/5/288 */
>        { PCI_VENDOR_ID_MARVELL_EXT, 0x9143},
> 
>        /* Reported by Andrew Cooks
>         * https://bugzilla.kernel.org/show_bug.cgi?id=42679 */
>        { PCI_VENDOR_ID_MARVELL_EXT, 0x9172}
> };


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel