Re: [PATCH RFC] NFSD: Reduce svc_rqst::rq_pages churn during READDIR operations

[bfields@fieldses.org (J. Bruce Fields)]

Makes sense to me.--b.

On Fri, Jan 22, 2021 at 03:21:10PM -0500, Chuck Lever wrote:
> During NFSv2 and NFSv3 READDIR/PLUS operations, NFSD advances
> rq_next_page to the full size of the client-requested buffer, then
> releases all those pages at the end of the request. The next request
> to use that nfsd thread has to refill the pages.
> 
> NFSD does this even when the dirlist in the reply is small. With
> NFSv3 clients that send READDIR operations with large buffer sizes,
> that can be 256 put_page/alloc_page pairs per READDIR request, even
> though those pages often remain unused.
> 
> We can save some work by not releasing dirlist buffer pages that
> were not used to form the READDIR Reply. I've left the NFSv2 code
> alone since there are never more than three pages involved in an
> NFSv2 READDIR operation.
> 
> Eventually we should nail down why these pages need to be released
> at all in order to avoid allocating and releasing pages
> unnecessarily.
> 
> Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
> ---
>  fs/nfsd/nfs3proc.c |    2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c
> index 8675851199f8..569bfff314f0 100644
> --- a/fs/nfsd/nfs3proc.c
> +++ b/fs/nfsd/nfs3proc.c
> @@ -492,6 +492,7 @@ nfsd3_proc_readdir(struct svc_rqst *rqstp)
>  		}
>  		count += PAGE_SIZE;
>  	}
> +	rqstp->rq_next_page = p;
>  	resp->count = count >> 2;
>  	if (resp->offset) {
>  		if (unlikely(resp->offset1)) {
> @@ -557,6 +558,7 @@ nfsd3_proc_readdirplus(struct svc_rqst *rqstp)
>  		}
>  		count += PAGE_SIZE;
>  	}
> +	rqstp->rq_next_page = p;
>  	resp->count = count >> 2;
>  	if (resp->offset) {
>  		if (unlikely(resp->offset1)) {
>