Hi Frank,
> From: Frank Li, Sent: Tuesday, March 26, 2024 11:09 PM
>
> On Tue, Mar 26, 2024 at 05:47:23AM +0000, Yoshihiro Shimoda wrote:
> > Hi Frank,
> >
> > > From: Frank Li, Sent: Tuesday, March 26, 2024 12:21 PM
> >
> > > On Tue, Mar 26, 2024 at 11:45:40AM +0900, Yoshihiro Shimoda wrote:
> > > > Add Renesas R8A779G0 in pci_device_id table so that pci-epf-test
> > > > can be used for testing PCIe EP on R-Car V4H.
> > > >
> > > > Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@xxxxxxxxxxx>
> > > > Reviewed-by: Geert Uytterhoeven <geert+renesas@xxxxxxxxx>
> > > > ---
> > > > drivers/misc/pci_endpoint_test.c | 4 ++++
> > > > 1 file changed, 4 insertions(+)
> > > >
> > > > diff --git a/drivers/misc/pci_endpoint_test.c b/drivers/misc/pci_endpoint_test.c
> > > > index c38a6083f0a7..2fa3c6473c7d 100644
> > > > --- a/drivers/misc/pci_endpoint_test.c
> > > > +++ b/drivers/misc/pci_endpoint_test.c
> > > > @@ -83,6 +83,7 @@
> > > > #define PCI_DEVICE_ID_RENESAS_R8A774C0 0x002d
> > > > #define PCI_DEVICE_ID_RENESAS_R8A774E1 0x0025
> > > > #define PCI_DEVICE_ID_RENESAS_R8A779F0 0x0031
> > > > +#define PCI_DEVICE_ID_RENESAS_R8A779G0 0x0030
> > > >
> > > > static DEFINE_IDA(pci_endpoint_test_ida);
> > > >
> > > > @@ -1005,6 +1006,9 @@ static const struct pci_device_id pci_endpoint_test_tbl[] = {
> > > > { PCI_DEVICE(PCI_VENDOR_ID_RENESAS, PCI_DEVICE_ID_RENESAS_R8A779F0),
> > > > .driver_data = (kernel_ulong_t)&default_data,
> > > > },
> > > > + { PCI_DEVICE(PCI_VENDOR_ID_RENESAS, PCI_DEVICE_ID_RENESAS_R8A779G0),
> > > > + .driver_data = (kernel_ulong_t)&default_data,
> > > > + },
> > >
> > > You use default_data, why need new device_id? I think you can use 0x0031
> > > to do test.
> >
> > I thought we can add a new device_id freely like other devices.
> > Since the PCIe controller's endpoint mode can configure the device id,
> > I can use 0x0031 to do test though.
> >
> > If such a reusable entry exists, is adding a new device id into the driver prohibited?
>
> I just think it is not necessary. This list will become longer and longer.
> And difference device id can't help us at all.
I agreed. To record it, I'll make a patch to add such description in the pci_endpoint_test.c.
> We should use difference production as difference functions, or difference
> configuration. Such as usb gadget product id, we use 0x4545 for all mass
> storage.
I see.
> Using difference devices id for difference function, such as 0x31 for
> ep_test 0x30 for virtual net, 0x29 for virtual console ...
>
> Or using difference devices id indicate some features. For example, use
> 0x30 means support write to EP MSI ITS to trigger irq.
>
> Donate a device_id to more valuable things.
I think so.
Best regards,
Yoshihiro Shimoda
> Frank
>
> >
> > Best regards,
> > Yoshihiro Shimoda
> >
> > > Frank
> > >
> > > > { PCI_DEVICE(PCI_VENDOR_ID_TI, PCI_DEVICE_ID_TI_J721E),
> > > > .driver_data = (kernel_ulong_t)&j721e_data,
> > > > },
> > > > --
> > > > 2.25.1
> > > >
