Re: [PATCH 16/21] KVM: guest_memfd: pass error up from filemap_grab_folio

[Xu Yilun]

On Tue, Feb 27, 2024 at 06:20:55PM -0500, Paolo Bonzini wrote:
> Some SNP ioctls will require the page not to be in the pagecache, and as such they
> will want to return EEXIST to userspace.  Start by passing the error up from
> filemap_grab_folio.
> 
> Signed-off-by: Paolo Bonzini <pbonzini@xxxxxxxxxx>
> ---
>  virt/kvm/guest_memfd.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/virt/kvm/guest_memfd.c b/virt/kvm/guest_memfd.c
> index 0f4e0cf4f158..de0d5a5c210c 100644
> --- a/virt/kvm/guest_memfd.c
> +++ b/virt/kvm/guest_memfd.c
> @@ -20,7 +20,7 @@ static struct folio *kvm_gmem_get_folio(struct inode *inode, pgoff_t index)
>  	/* TODO: Support huge pages. */
>  	folio = filemap_grab_folio(inode->i_mapping, index);
>  	if (IS_ERR_OR_NULL(folio))
> -		return NULL;
> +		return folio;

I think it impacts kvm_gmem_get_pfn()

{
	...

	folio = kvm_gmem_get_folio(file_inode(file), index);
	if (!folio) {
		r = -ENOMEM;
		goto out_fput;
	}

	...
}

Thanks,
Yilun

>  
>  	/*
>  	 * Use the up-to-date flag to track whether or not the memory has been
> @@ -146,8 +146,8 @@ static long kvm_gmem_allocate(struct inode *inode, loff_t offset, loff_t len)
>  		}
>  
>  		folio = kvm_gmem_get_folio(inode, index);
> -		if (!folio) {
> -			r = -ENOMEM;
> +		if (IS_ERR_OR_NULL(folio)) {
> +			r = folio ? PTR_ERR(folio) : -ENOMEM;
>  			break;
>  		}
>  
> -- 
> 2.39.0
> 
> 
>