linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Chuck Lever <chuck.lever@oracle.com>
To: Mike Snitzer <snitzer@kernel.org>, Jeff Layton <jlayton@kernel.org>
Cc: linux-nfs@vger.kernel.org, linux-fsdevel@vger.kernel.org,
	Jens Axboe <axboe@kernel.dk>
Subject: Re: [PATCH 1/6] NFSD: add the ability to enable use of RWF_DONTCACHE for all IO
Date: Wed, 11 Jun 2025 10:31:20 -0400	[thread overview]
Message-ID: <4b858fb1-25f6-457f-8908-67339e20318e@oracle.com> (raw)
In-Reply-To: <20250610205737.63343-2-snitzer@kernel.org>

On 6/10/25 4:57 PM, Mike Snitzer wrote:
> Add 'enable-dontcache' to NFSD's debugfs interface so that: Any data
> read or written by NFSD will either not be cached (thanks to O_DIRECT)
> or will be removed from the page cache upon completion (DONTCACHE).

I thought we were going to do two switches: One for reads and one for
writes? I could be misremembering.

After all, you are describing two different facilities here: a form of
direct I/O for READs, and RWF_DONTCACHE for WRITEs (I think?).


> enable-dontcache is 0 by default.  It may be enabled with:
>   echo 1 > /sys/kernel/debug/nfsd/enable-dontcache
> 
> FOP_DONTCACHE must be advertised as supported by the underlying
> filesystem (e.g. XFS), otherwise if/when 'enable-dontcache' is 1
> all IO flagged with RWF_DONTCACHE will fail with -EOPNOTSUPP.



> Signed-off-by: Mike Snitzer <snitzer@kernel.org>
> ---
>  fs/nfsd/debugfs.c | 39 +++++++++++++++++++++++++++++++++++++++
>  fs/nfsd/nfsd.h    |  1 +
>  fs/nfsd/vfs.c     | 12 +++++++++++-
>  3 files changed, 51 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/nfsd/debugfs.c b/fs/nfsd/debugfs.c
> index 84b0c8b559dc..8decdec60a8e 100644
> --- a/fs/nfsd/debugfs.c
> +++ b/fs/nfsd/debugfs.c
> @@ -32,6 +32,42 @@ static int nfsd_dsr_set(void *data, u64 val)
>  
>  DEFINE_DEBUGFS_ATTRIBUTE(nfsd_dsr_fops, nfsd_dsr_get, nfsd_dsr_set, "%llu\n");
>  
> +/*
> + * /sys/kernel/debug/nfsd/enable-dontcache
> + *
> + * Contents:
> + *   %0: NFS READ and WRITE are not allowed to use dontcache
> + *   %1: NFS READ and WRITE are allowed to use dontcache
> + *
> + * NFSD's dontcache support reserves the right to use O_DIRECT
> + * if it chooses (instead of dontcache's usual pagecache-based
> + * dropbehind semantics).
> + *
> + * The default value of this setting is zero (dontcache is
> + * disabled). This setting takes immediate effect for all NFS
> + * versions, all exports, and in all NFSD net namespaces.
> + */
> +
> +static int nfsd_dontcache_get(void *data, u64 *val)
> +{
> +	*val = nfsd_enable_dontcache ? 1 : 0;
> +	return 0;
> +}
> +
> +static int nfsd_dontcache_set(void *data, u64 val)
> +{
> +	if (val > 0) {
> +		/* Must first also disable-splice-read */
> +		nfsd_disable_splice_read = true;
> +		nfsd_enable_dontcache = true;
> +	} else
> +		nfsd_enable_dontcache = false;
> +	return 0;
> +}
> +
> +DEFINE_DEBUGFS_ATTRIBUTE(nfsd_dontcache_fops, nfsd_dontcache_get,
> +			 nfsd_dontcache_set, "%llu\n");
> +
>  void nfsd_debugfs_exit(void)
>  {
>  	debugfs_remove_recursive(nfsd_top_dir);
> @@ -44,4 +80,7 @@ void nfsd_debugfs_init(void)
>  
>  	debugfs_create_file("disable-splice-read", S_IWUSR | S_IRUGO,
>  			    nfsd_top_dir, NULL, &nfsd_dsr_fops);
> +
> +	debugfs_create_file("enable-dontcache", S_IWUSR | S_IRUGO,
> +			    nfsd_top_dir, NULL, &nfsd_dontcache_fops);
>  }
> diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h
> index 1bfd0b4e9af7..00546547eae6 100644
> --- a/fs/nfsd/nfsd.h
> +++ b/fs/nfsd/nfsd.h
> @@ -155,6 +155,7 @@ static inline void nfsd_debugfs_exit(void) {}
>  #endif
>  
>  extern bool nfsd_disable_splice_read __read_mostly;
> +extern bool nfsd_enable_dontcache __read_mostly;
>  
>  extern int nfsd_max_blksize;
>  
> diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
> index 7d94fae1dee8..bba3e6f4f56b 100644
> --- a/fs/nfsd/vfs.c
> +++ b/fs/nfsd/vfs.c
> @@ -49,6 +49,7 @@
>  #define NFSDDBG_FACILITY		NFSDDBG_FILEOP
>  
>  bool nfsd_disable_splice_read __read_mostly;
> +bool nfsd_enable_dontcache __read_mostly;
>  
>  /**
>   * nfserrno - Map Linux errnos to NFS errnos
> @@ -1086,6 +1087,7 @@ __be32 nfsd_iter_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
>  	unsigned long v, total;
>  	struct iov_iter iter;
>  	loff_t ppos = offset;
> +	rwf_t flags = 0;
>  	ssize_t host_err;
>  	size_t len;
>  
> @@ -1103,7 +1105,11 @@ __be32 nfsd_iter_read(struct svc_rqst *rqstp, struct svc_fh *fhp,
>  
>  	trace_nfsd_read_vector(rqstp, fhp, offset, *count);
>  	iov_iter_bvec(&iter, ITER_DEST, rqstp->rq_bvec, v, *count);
> -	host_err = vfs_iter_read(file, &iter, &ppos, 0);
> +
> +	if (nfsd_enable_dontcache)
> +		flags |= RWF_DONTCACHE;

Two things:

- Maybe NFSD should record whether the file system is DONTCACHE-enabled
in @fhp or in the export it is associated with, and then check that
setting here before asserting RWF_DONTCACHE

- I thought we were going with O_DIRECT for READs.


> +
> +	host_err = vfs_iter_read(file, &iter, &ppos, flags);
>  	return nfsd_finish_read(rqstp, fhp, file, offset, count, eof, host_err);
>  }
>  
> @@ -1209,6 +1215,10 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp,
>  
>  	nvecs = xdr_buf_to_bvec(rqstp->rq_bvec, rqstp->rq_maxpages, payload);
>  	iov_iter_bvec(&iter, ITER_SOURCE, rqstp->rq_bvec, nvecs, *cnt);
> +
> +	if (nfsd_enable_dontcache)
> +		flags |= RWF_DONTCACHE;
> +
>  	since = READ_ONCE(file->f_wb_err);
>  	if (verf)
>  		nfsd_copy_write_verifier(verf, nn);


-- 
Chuck Lever

  parent reply	other threads:[~2025-06-11 14:31 UTC|newest]

Thread overview: 75+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-10 20:57 [PATCH 0/6] NFSD: add enable-dontcache and initially use it to add DIO support Mike Snitzer
2025-06-10 20:57 ` [PATCH 1/6] NFSD: add the ability to enable use of RWF_DONTCACHE for all IO Mike Snitzer
2025-06-11  6:57   ` Christoph Hellwig
2025-06-11 10:44     ` Mike Snitzer
2025-06-11 13:04       ` Jeff Layton
2025-06-11 13:56     ` Chuck Lever
2025-06-11 14:31   ` Chuck Lever [this message]
2025-06-11 19:18     ` Mike Snitzer
2025-06-11 20:29       ` Jeff Layton
2025-06-11 21:36         ` need SUNRPC TCP to receive into aligned pages [was: Re: [PATCH 1/6] NFSD: add the ability to enable use of RWF_DONTCACHE for all IO] Mike Snitzer
2025-06-12 10:28           ` Jeff Layton
2025-06-12 11:28             ` Jeff Layton
2025-06-12 13:28             ` Chuck Lever
2025-06-12 14:17               ` Benjamin Coddington
2025-06-12 15:56                 ` Mike Snitzer
2025-06-12 15:58                   ` Chuck Lever
2025-06-12 16:12                     ` Mike Snitzer
2025-06-12 16:32                       ` Chuck Lever
2025-06-13  5:39                     ` Christoph Hellwig
2025-06-12 16:22               ` Jeff Layton
2025-06-13  5:46                 ` Christoph Hellwig
2025-06-13  9:23                   ` Mike Snitzer
2025-06-13 13:02                     ` Jeff Layton
2025-06-16 12:35                       ` Christoph Hellwig
2025-06-16 12:29                     ` Christoph Hellwig
2025-06-16 16:07                       ` Mike Snitzer
2025-06-17  4:37                         ` Christoph Hellwig
2025-06-17 20:26                           ` Mike Snitzer
2025-06-17 22:23                             ` [RFC PATCH] lib/iov_iter: remove piecewise bvec length checking in iov_iter_aligned_bvec [was: Re: need SUNRPC TCP to receive into aligned pages] Mike Snitzer
2025-07-03  0:12             ` need SUNRPC TCP to receive into aligned pages [was: Re: [PATCH 1/6] NFSD: add the ability to enable use of RWF_DONTCACHE for all IO] NeilBrown
2025-06-12  7:13         ` [PATCH 1/6] NFSD: add the ability to enable use of RWF_DONTCACHE for all IO Christoph Hellwig
2025-06-12 13:15           ` Chuck Lever
2025-06-12 13:21       ` Chuck Lever
2025-06-12 16:00         ` Mike Snitzer
2025-06-16 13:32           ` Chuck Lever
2025-06-16 16:10             ` Mike Snitzer
2025-06-17 17:22               ` Mike Snitzer
2025-06-17 17:31                 ` Chuck Lever
2025-06-19 20:19                   ` Mike Snitzer
2025-06-30 14:50                     ` Chuck Lever
2025-07-04 19:46                       ` Mike Snitzer
2025-07-04 19:49                         ` Chuck Lever
2025-06-10 20:57 ` [PATCH 2/6] NFSD: filecache: add STATX_DIOALIGN and STATX_DIO_READ_ALIGN support Mike Snitzer
2025-06-10 20:57 ` [PATCH 3/6] NFSD: pass nfsd_file to nfsd_iter_read() Mike Snitzer
2025-06-10 20:57 ` [PATCH 4/6] fs: introduce RWF_DIRECT to allow using O_DIRECT on a per-IO basis Mike Snitzer
2025-06-11  6:58   ` Christoph Hellwig
2025-06-11 10:51     ` Mike Snitzer
2025-06-11 14:17     ` Chuck Lever
2025-06-12  7:15       ` Christoph Hellwig
2025-06-10 20:57 ` [PATCH 5/6] NFSD: leverage DIO alignment to selectively issue O_DIRECT reads and writes Mike Snitzer
2025-06-11  7:00   ` Christoph Hellwig
2025-06-11 12:23     ` Mike Snitzer
2025-06-11 13:30       ` Jeff Layton
2025-06-12  7:22         ` Christoph Hellwig
2025-06-12  7:23       ` Christoph Hellwig
2025-06-11 14:42   ` Chuck Lever
2025-06-11 15:07     ` Jeff Layton
2025-06-11 15:11       ` Chuck Lever
2025-06-11 15:44         ` Jeff Layton
2025-06-11 20:51           ` Mike Snitzer
2025-06-12  7:32           ` Christoph Hellwig
2025-06-12  7:28         ` Christoph Hellwig
2025-06-12  7:25       ` Christoph Hellwig
2025-06-10 20:57 ` [PATCH 6/6] NFSD: issue READs using O_DIRECT even if IO is misaligned Mike Snitzer
2025-06-11 12:55 ` [PATCH 0/6] NFSD: add enable-dontcache and initially use it to add DIO support Jeff Layton
2025-06-12  7:39   ` Christoph Hellwig
2025-06-12 20:37     ` Mike Snitzer
2025-06-13  5:31       ` Christoph Hellwig
2025-06-11 14:16 ` Chuck Lever
2025-06-11 18:02   ` Mike Snitzer
2025-06-11 19:06     ` Chuck Lever
2025-06-11 19:58       ` Mike Snitzer
2025-06-12 13:46 ` Chuck Lever
2025-06-12 19:08   ` Mike Snitzer
2025-06-12 20:17     ` Chuck Lever

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4b858fb1-25f6-457f-8908-67339e20318e@oracle.com \
    --to=chuck.lever@oracle.com \
    --cc=axboe@kernel.dk \
    --cc=jlayton@kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=snitzer@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).