Re: [PATCH RFC 05/12] remoteproc: always report VIRTIO_F_VERSION_1 feature

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] remoteproc: host-firmware protocol mismatch by forcing VIRTIO_F_VERSION_1
--

commit 425d88ad507f29f8586fd7c8c447cf1bcdba6678
Author: Francesco Valla <francesco@valla.it>

remoteproc: always report VIRTIO_F_VERSION_1 feature

There is currently no way to report the VIRTIO_F_VERSION_1 using the
resource table alone, as the per-vdev feature array is limited to 32
bits.

> diff --git a/drivers/remoteproc/remoteproc_virtio.c b/drivers/remoteproc/remoteproc_virtio.c
> index 74e0da970f1d8..cfd66d9d1c9e5 100644
> --- a/drivers/remoteproc/remoteproc_virtio.c
> +++ b/drivers/remoteproc/remoteproc_virtio.c
> @@ -249,7 +249,7 @@ static u64 rproc_virtio_get_features(struct virtio_device *vdev)
>  
>  	rsc = (void *)rvdev->rproc->table_ptr + rvdev->rsc_offset;
>  
> -	return rsc->dfeatures;
> +	return rsc->dfeatures | (1ULL << VIRTIO_F_VERSION_1);

[Severity: High]
Will this forcefully advertise Virtio 1.0 support to the host virtio
core, misleading it into Modern mode?

By injecting VIRTIO_F_VERSION_1 into rsc->dfeatures, the host virtio core
will preserve the feature since it is a transport feature. This configures
virtio_ring as strict Little-Endian and assumes Modern struct layouts.

[ ... ]

> @@ -275,14 +275,16 @@ static int rproc_virtio_finalize_features(struct virtio_device *vdev)
>  	/* Give virtio_rproc a chance to accept features. */
>  	rproc_transport_features(vdev);
>  
> -	/* Make sure we don't have any features > 32 bits! */
> -	BUG_ON((u32)vdev->features != vdev->features);
> +	/* Make sure we don't have any features > 32 bits except VIRTIO_F_VERSION_1 */
> +	if (WARN_ON_ONCE((u32)vdev->features !=
> +			 (vdev->features & ~(1ULL << VIRTIO_F_VERSION_1))))
> +		return -1;
>  
>  	/*
>  	 * Remember the finalized features of our vdev, and provide it
>  	 * to the remote processor once it is powered on.
>  	 */
> -	rsc->gfeatures = vdev->features;
> +	rsc->gfeatures = vdev->features & ~(1ULL << VIRTIO_F_VERSION_1);

[Severity: High]
Does hiding the VIRTIO_F_VERSION_1 feature from the remote firmware
guarantee a feature mismatch between the host and device?

Since rproc_virtio_finalize_features() masks the bit out before passing
the negotiated features back to the firmware, the firmware remains unaware
and operates in Legacy mode using guest-native endianness and Legacy struct
sizes.

Could this lead to dropped messages or crashes due to differing endianness
assumptions, particularly on Big-Endian hosts where vring formatting will
break?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260916-remoteproc_virtio_map-v1-0-dac8c5eb4aa9@valla.it?part=5