On 9/14/26 3:34 PM, Zhuoying Cai wrote:
> Enable network booting via virtio-net-pci by implementing PCI transport
> support for virtio-net.
>
> This patch also extends endianness handling for virtio PCI ring operations in
> little-endian to ensure correct behavior on s390x.
>
> Signed-off-by: Zhuoying Cai <[email protected]>
> Reviewed-by: Jared Rossi <[email protected]>
> ---
> pc-bios/s390-ccw/main.c | 1 +
> pc-bios/s390-ccw/netmain.c | 9 +++++++-
> pc-bios/s390-ccw/virtio-net.c | 41 ++++++++++++++++++++++++++++-------
> pc-bios/s390-ccw/virtio-pci.c | 20 +++++++++++++++++
> 4 files changed, 62 insertions(+), 9 deletions(-)
>
> diff --git a/pc-bios/s390-ccw/main.c b/pc-bios/s390-ccw/main.c
> index 8fc1cbf653..28cab3b411 100644
> --- a/pc-bios/s390-ccw/main.c
> +++ b/pc-bios/s390-ccw/main.c
> @@ -332,6 +332,7 @@ static void ipl_pci_device(void)
> }
>
> switch (vdev->dev_type) {
> + case VIRTIO_ID_NET:
block, scsi (when you rebase on top of Jared's series) and now net will
all use the exact same code here by falling through.
How about a comment block above the switch statement explaining they all
intentionally fall through / use the same shared setup
> case VIRTIO_ID_BLOCK:
> if (virtio_setup() == 0) {
> zipl_load(); /* only return on error */
> diff --git a/pc-bios/s390-ccw/netmain.c b/pc-bios/s390-ccw/netmain.c
> index bcb62bf6ca..e42e049063 100644
> --- a/pc-bios/s390-ccw/netmain.c
> +++ b/pc-bios/s390-ccw/netmain.c
> @@ -498,7 +498,14 @@ int netmain(void)
> sclp_setup();
> puts("Network boot starting...");
>
> - if (!virtio_ccw_net_setup()) {
> + /*
> + * CCW devices require subchannel enumeration here.
> + * PCI devices don't require specific setup: the transport is already
> + * initialized by ipl_pci_device(), and virtio_is_supported() is checked
> + * later in virtio_net_init().
> + */
This comment seems not quite accurate?
AFAICT ipl_pci_device() is just going to enable the PCI device, the
transport isn't setup until
virtio_net_init()->virtio_pci_setup()
With those 2 minor issues addressed:
Reviewed-by: Matthew Rosato <[email protected]>
