Re: [PATCH v2 2/2] vfio/cdx: update driver to build without CONFIG_GENERIC_MSI_IRQ

[Alex Williamson <alex.williamson@redhat.com>]

On Thu, 21 Aug 2025 12:43:08 +0530
Nipun Gupta <nipun.gupta@amd.com> wrote:

> Define dummy MSI related APIs in VFIO CDX driver to build the
> driver without enabling CONFIG_GENERIC_MSI_IRQ flag.
> 
> Reported-by: kernel test robot <lkp@intel.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202508070308.opy5dIFX-lkp@intel.com/
> Reviewed-by: Nikhil Agarwal <nikhil.agarwal@amd.com>
> Signed-off-by: Nipun Gupta <nipun.gupta@amd.com>
> ---
> Changes v1->v2:
> - fix linking intr.c file in Makefile
> 
>  drivers/vfio/cdx/Makefile  |  6 +++++-
>  drivers/vfio/cdx/private.h | 14 ++++++++++++++
>  2 files changed, 19 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/vfio/cdx/Makefile b/drivers/vfio/cdx/Makefile
> index df92b320122a..dadbef2419ea 100644
> --- a/drivers/vfio/cdx/Makefile
> +++ b/drivers/vfio/cdx/Makefile
> @@ -5,4 +5,8 @@
>  
>  obj-$(CONFIG_VFIO_CDX) += vfio-cdx.o
>  
> -vfio-cdx-objs := main.o intr.o
> +vfio-cdx-objs := main.o
> +
> +ifdef CONFIG_GENERIC_MSI_IRQ
> +vfio-cdx-objs += intr.o
> +endif
> diff --git a/drivers/vfio/cdx/private.h b/drivers/vfio/cdx/private.h
> index dc56729b3114..0482a1516439 100644
> --- a/drivers/vfio/cdx/private.h
> +++ b/drivers/vfio/cdx/private.h
> @@ -38,11 +38,25 @@ struct vfio_cdx_device {
>  	u8			config_msi;
>  };
>  
> +#ifdef CONFIG_GENERIC_MSI_IRQ
>  int vfio_cdx_set_irqs_ioctl(struct vfio_cdx_device *vdev,
>  			    u32 flags, unsigned int index,
>  			    unsigned int start, unsigned int count,
>  			    void *data);
>  
>  void vfio_cdx_irqs_cleanup(struct vfio_cdx_device *vdev);
> +#else
> +static int vfio_cdx_set_irqs_ioctl(struct vfio_cdx_device *vdev,
> +				   u32 flags, unsigned int index,
> +				   unsigned int start, unsigned int count,
> +				   void *data)
> +{
> +	return 0;
> +}
> +
> +static void vfio_cdx_irqs_cleanup(struct vfio_cdx_device *vdev)
> +{
> +}
> +#endif
>  
>  #endif /* VFIO_CDX_PRIVATE_H */

Why are we returning success if we don't implement the function?  This
provides the wrong return value to the user.

Also, aren't we also missing in patch 1/ that cdx_dev->num_msi is
initialized and therefore vfio_cdx_ioctl_get_info() might return that
an interrupt is supported when it is not?  Thanks,

Alex