Hi Łukasz,
On Mon, Feb 19, 2024 at 01:41:39PM +0100, Łukasz Stelmach wrote:
> The first time virtio_init() gets called from board_init() PCI isn't
> ready. Thus any virtio-over-PCI (e.g. network interfaces) devices can't
> be detected and used without additional `virtio scan` scan in the shell
> or a script.
>
> Signed-off-by: Łukasz Stelmach
> ---
> The patch works for my but:
>
> a) maybe virtio_init() should called only from board_init() the same
>way as on ARM?
> b) can a repeated virtio_init() break already detected/initialized
>devices?
>
A repeated virtio_init() call would not break the already probed devices.
But we should prevent this redundancy by moving the virtio_init() call to
board_late_init() like what this commit[1] does for ARM.
Could you send a v2 to move the virtio_init() into board_late_init() ?
[1]
https://patchwork.ozlabs.org/project/uboot/patch/20201230135712.5289-3-sughosh.g...@linaro.org/
e1ee06dde7 ("qemu: arm: Initialise virtio devices in board_late_init")
Best regards,
Leo
> board/emulation/qemu-riscv/qemu-riscv.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/board/emulation/qemu-riscv/qemu-riscv.c
> b/board/emulation/qemu-riscv/qemu-riscv.c
> index 181abbbf97d..567b9dc6170 100644
> --- a/board/emulation/qemu-riscv/qemu-riscv.c
> +++ b/board/emulation/qemu-riscv/qemu-riscv.c
> @@ -46,6 +46,9 @@ int board_late_init(void)
> if (CONFIG_IS_ENABLED(USB_KEYBOARD))
> usb_init();
>
> + /* Repeat virtio scan to detect PCI attached virtio devices. */
> + virtio_init();
> +
> return 0;
> }
>
