On 8/16/2022 7:14 PM, Zhu, Lingshan
wrote:
Can we conclude we can drop the misleading comment?Yes it is a little messy, and we can not check _F_VERSION_1 because of transitional devices, so maybe this is the best we can do for nowWhile the fix is fine, the comment is misleading in giving readers false hope. This is in vdpa_dev_net_config_fill() the vdpa tool query path, instead of calls from the VMM dealing with vhost/virtio plumbing specifics. I think what's missing today in vdpa core is the detection of guest type (legacy, transitional, or modern) regarding endianness through F_VERSION_1 and legacy interface access, the latter of which would need some assistance from VMM for sure. However, the presence of information via the vdpa tool query is totally orthogonal. I don't get a good reason for why it has to couple with endianness. How vdpa tool users space is supposed to tweak it? I don't get it...
Thanks,
-Siwei
_______________________________________________ Virtualization mailing list Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linuxfoundation.org/mailman/listinfo/virtualization