Re: [PATCH 2/2] vfio: Introduce strict PFNMAP mappings

[Jason Gunthorpe]

On Thu, May 14, 2020 at 10:52:09AM -0600, Alex Williamson wrote:
>  	vfio_unregister_iommu_driver(&vfio_noiommu_ops);
> diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c
> index 62ba6bd8a486..8d6286d89230 100644
> +++ b/drivers/vfio/vfio_iommu_type1.c
> @@ -61,6 +61,11 @@ module_param_named(dma_entry_limit, dma_entry_limit, uint, 0644);
>  MODULE_PARM_DESC(dma_entry_limit,
>  		 "Maximum number of user DMA mappings per container (65535).");
>  
> +static bool strict_mmio_maps = true;
> +module_param_named(strict_mmio_maps, strict_mmio_maps, bool, 0644);
> +MODULE_PARM_DESC(strict_mmio_maps,
> +		 "Restrict DMA mappings of MMIO to those provided by vfio bus drivers supporting invalidation (true).");
> +

This should probably explain that 'false' allows some kind of security
issue and maybe taint the kernel?

Do you think there is a reason to have this anyhow?

Jason