On 11/01/2017 11:32 AM, Michael S. Tsirkin wrote:
On Wed, Nov 01, 2017 at 02:48:17PM +0000, Nawrocki, Michael wrote:
On 10/31/17, 13:50, "Dr. David Alan Gilbert" <dgilb...@redhat.com> wrote:
* Mike Nawrocki (michael.nawro...@gtri.gatech.edu) wrote:
> Signed-off-by: Mike Nawrocki <michael.nawro...@gtri.gatech.edu>
> ---
> hw/net/eepro100.c | 2 +-
> hw/pci/pci.c | 2 ++
> include/hw/pci/pci.h | 1 +
> qemu-options.hx | 2 +-
> 4 files changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/hw/net/eepro100.c b/hw/net/eepro100.c
> index 91dd058010..2c7c5cdeb9 100644
> --- a/hw/net/eepro100.c
> +++ b/hw/net/eepro100.c
> @@ -1973,7 +1973,7 @@ static E100PCIDeviceInfo e100_devices[] = {
> .name = "i82559a",
> .desc = "Intel i82559A Ethernet",
> .device = i82559A,
> - .device_id = PCI_DEVICE_ID_INTEL_82557,
> + .device_id = PCI_DEVICE_ID_INTEL_82559,
Doesn't that break migration/compatibility - if one side suddenly
sees a different device_id ?
Dave
I’m not sure. I searched for code references to the i82559a interface and
couldn’t find any outside of the driver itself; it wasn’t exported as a
pci_nic_model before my patch, so I don’t think the i82559a device was being
used, programmatically or via the command line, before this point. Perhaps
there’s a different way to use the device that I’m not aware of?
the modern way to create devices is with -devices.
Gotcha. I've updated my usage to reflect this.
I think the ID I’m using (0x8086 0x1030 – Intel Corporation 82559 InBusiness
10/100) is more accurate to the chipset, and a driver I’ve got explicitly
searches for this ID. Most drivers seem to search for a number of PCI IDs in
the eepro100 family, at any rate, so I think it should be okay. Let me know if
another approach is preferable.
Thanks,
Mike
You need a property and set it to a compat value for
old machine types.
I've added a "use-alt-device-id" boolean property, an "alt_device_id"
field in the info struct, and reverted the device_id to the 82557 one.
I'll push the v2 patches shortly.
> .revision = 0x06,
> .stats_size = 80,
> .has_extended_tcb_support = true,
> diff --git a/hw/pci/pci.c b/hw/pci/pci.c
> index 5ed3c8dca4..1126ad1eb3 100644
> --- a/hw/pci/pci.c
> +++ b/hw/pci/pci.c
> @@ -1826,6 +1826,7 @@ static const char * const pci_nic_models[] = {
> "ne2k_pci",
> "i82551",
> "i82557b",
> + "i82559a",
> "i82559er",
> "rtl8139",
> "e1000",
> @@ -1839,6 +1840,7 @@ static const char * const pci_nic_names[] = {
> "ne2k_pci",
> "i82551",
> "i82557b",
> + "i82559a",
> "i82559er",
> "rtl8139",
> "e1000",
There's no need to add more nic names here. Just create with -device.
Gotcha. I've removed these in the upcoming patches.
Thanks,
Mike
> diff --git a/include/hw/pci/pci.h b/include/hw/pci/pci.h
> index 8d02a0a383..f30e2cfb72 100644
> --- a/include/hw/pci/pci.h
> +++ b/include/hw/pci/pci.h
> @@ -70,6 +70,7 @@ extern bool pci_available;
> /* Intel (0x8086) */
> #define PCI_DEVICE_ID_INTEL_82551IT 0x1209
> #define PCI_DEVICE_ID_INTEL_82557 0x1229
> +#define PCI_DEVICE_ID_INTEL_82559 0x1030
> #define PCI_DEVICE_ID_INTEL_82801IR 0x2922
>
> /* Red Hat / Qumranet (for QEMU) -- see pci-ids.txt */
> diff --git a/qemu-options.hx b/qemu-options.hx
> index 3728e9b4dd..a39c7e44b3 100644
> --- a/qemu-options.hx
> +++ b/qemu-options.hx
> @@ -2047,7 +2047,7 @@ that the card should have; this option currently
only affects virtio cards; set
> @var{v} = 0 to disable MSI-X. If no @option{-net} option is specified,
a single
> NIC is created. QEMU can emulate several different models of network
card.
> Valid values for @var{type} are
> -@code{virtio}, @code{i82551}, @code{i82557b}, @code{i82559er},
> +@code{virtio}, @code{i82551}, @code{i82557b}, @code{i82559a},
@code{i82559er},
> @code{ne2k_pci}, @code{ne2k_isa}, @code{pcnet}, @code{rtl8139},
> @code{e1000}, @code{smc91c111}, @code{lance} and @code{mcf_fec}.
> Not all devices are supported on all targets. Use @code{-net
nic,model=help}
> --
> 2.14.2
>
>
--
Dr. David Alan Gilbert / dgilb...@redhat.com / Manchester, UK
