Re: [PATCH 3/5] coda: use iterate_dir() in coda_readdir()

[Jeff Layton <jlayton@kernel.org>]

On Mon, 2025-06-09 at 09:09 +1000, NeilBrown wrote:
> The code in coda_readdir() is nearly identical to iterate_dir().
> Differences are:
>  - iterate_dir() is killable
>  - iterate_dir() adds permission checking and accessing notifications
> 
> I believe these are not harmful for coda so it is best to use
> iterate_dir() directly.  This will allow locking changes without
> touching the code in coda.
> 
> Signed-off-by: NeilBrown <neil@brown.name>
> ---
>  fs/coda/dir.c | 12 ++----------
>  1 file changed, 2 insertions(+), 10 deletions(-)
> 
> diff --git a/fs/coda/dir.c b/fs/coda/dir.c
> index ab69d8f0cec2..ca9990017265 100644
> --- a/fs/coda/dir.c
> +++ b/fs/coda/dir.c
> @@ -429,17 +429,9 @@ static int coda_readdir(struct file *coda_file, struct dir_context *ctx)
>  	cfi = coda_ftoc(coda_file);
>  	host_file = cfi->cfi_container;
>  
> -	if (host_file->f_op->iterate_shared) {
> -		struct inode *host_inode = file_inode(host_file);
> -		ret = -ENOENT;
> -		if (!IS_DEADDIR(host_inode)) {
> -			inode_lock_shared(host_inode);
> -			ret = host_file->f_op->iterate_shared(host_file, ctx);
> -			file_accessed(host_file);
> -			inode_unlock_shared(host_inode);
> -		}
> +	ret = iterate_dir(host_file, ctx);
> +	if (ret != -ENOTDIR)
>  		return ret;
> -	}
>  	/* Venus: we must read Venus dirents from a file */
>  	return coda_venus_readdir(coda_file, ctx);
>  }


Is it already time for my annual ask of "Who the heck is using coda
these days?" Anyway, this patch looks fine to me.

Reviewed-by: Jeff Layton <jlayton@kernel.org>