Re: linux-next: build failure after merge of the vhost tree

[Xuan Zhuo <xuanzhuo@linux.alibaba.com>]

On Thu, 29 Feb 2024 12:01:44 +0800, Xuan Zhuo <xuanzhuo@linux.alibaba.com> wrote:
> On Thu, 29 Feb 2024 14:55:31 +1100, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> > Hi all,
> >
> > After merging the vhost tree, today's linux-next build (powerpc
> > ppc64_defconfig) failed like this:
> >
> > drivers/virtio/virtio_pci_modern.c: In function 'vp_modern_create_avq':
> > drivers/virtio/virtio_pci_modern.c:755:34: warning: passing argument 5 of 'vp_dev->setup_vq' makes integer from pointer without a cast [-Wint-conversion]
> >   755 |                               avq->name, NULL, VIRTIO_MSI_NO_VECTOR);
> >       |                               ~~~^~~~~~
> >       |                                  |
> >       |                                  char *
> > drivers/virtio/virtio_pci_modern.c:755:34: note: expected 'u16' {aka 'short unsigned int'} but argument is of type 'char *'
> > drivers/virtio/virtio_pci_modern.c:754:14: error: too many arguments to function 'vp_dev->setup_vq'
> >   754 |         vq = vp_dev->setup_vq(vp_dev, &vp_dev->admin_vq.info, avq->vq_index, NULL,
> >       |              ^~~~~~
> >
> > Caused by commit
> >
> >   4cceb2591a87 ("virtio: find_vqs: pass struct instead of multi parameters")
> >
> > I have used the vhost tree from next-20240228 for today.
>
> I will post new version for powerpc soon.


This is not a bug on the powerpc. This is due to a conflict with the admin
queue commit. Because my patch set contains files for many other platforms. So I
have to check each file manually. Maybe when I do rebase I missed
checking the code again and my automated script didn't work because some files
can not compile on x86.

But the good news is that this patch set contains the ability to add new virtio
features without impacting files on other platforms.

The new version is sent.

	http://lore.kernel.org/all/20240229072044.77388-1-xuanzhuo@linux.alibaba.com

Thanks.


>
> Thanks.
>
>
> >
> > --
> > Cheers,
> > Stephen Rothwell
> >
>