From: Mike Snitzer <snitzer@kernel.org>
To: Chuck Lever <chuck.lever@oracle.com>,
Jeff Layton <jlayton@kernel.org>, NeilBrown <neil@brown.name>,
Olga Kornievskaia <okorniev@redhat.com>,
Dai Ngo <Dai.Ngo@oracle.com>, Tom Talpey <tom@talpey.com>,
Trond Myklebust <trondmy@kernel.org>,
Anna Schumaker <anna@kernel.org>
Cc: linux-nfs@vger.kernel.org, linus-fsdevel@vger.kernel.org,
linux-mm@kvack.org, snitzer@kernel.org
Subject: [RFC PATCH v2 7/8] NFSD: add io_cache_write controls to debugfs interface
Date: Tue, 8 Jul 2025 12:06:18 -0400 [thread overview]
Message-ID: <20250708160619.64800-8-snitzer@kernel.org> (raw)
In-Reply-To: <20250708160619.64800-1-snitzer@kernel.org>
Add 'io_cache_write' to NFSD's debugfs interface so that: Any data
written by NFSD will either be:
- cached using page cache (NFSD_IO_BUFFERED=0)
- cached but removed from the page cache upon completion
(NFSD_IO_DONTCACHE=1).
- not cached (NFSD_IO_DIRECT=2)
io_cache_write is 0 by default. It may be set by writing to:
/sys/kernel/debug/nfsd/io_cache_write
If NFSD_IO_DONTCACHE is specified using 1, FOP_DONTCACHE must be
advertised as supported by the underlying filesystem (e.g. XFS),
otherwise all IO flagged with RWF_DONTCACHE will fail with
-EOPNOTSUPP.
If NFSD_IO_DIRECT is specified using 2, the IO must be aligned
relative to the underlying block device's logical_block_size. Also the
memory buffer used to store the WRITE payload must be aligned relative
to the underlying block device's dma_alignment.
Signed-off-by: Mike Snitzer <snitzer@kernel.org>
---
fs/nfsd/debugfs.c | 41 +++++++++++++++++++++++++++++++++++++++++
fs/nfsd/nfsd.h | 1 +
fs/nfsd/vfs.c | 15 +++++++++++++++
3 files changed, 57 insertions(+)
diff --git a/fs/nfsd/debugfs.c b/fs/nfsd/debugfs.c
index 709646af797a..fadf1d88f640 100644
--- a/fs/nfsd/debugfs.c
+++ b/fs/nfsd/debugfs.c
@@ -82,6 +82,44 @@ static int nfsd_io_cache_read_set(void *data, u64 val)
DEFINE_DEBUGFS_ATTRIBUTE(nfsd_io_cache_read_fops, nfsd_io_cache_read_get,
nfsd_io_cache_read_set, "%llu\n");
+/*
+ * /sys/kernel/debug/nfsd/io_cache_write
+ *
+ * Contents:
+ * %0: NFS WRITE will use buffered IO (default)
+ * %1: NFS WRITE will use dontcache (buffered IO w/ dropbehind)
+ * %2: NFS WRITE will use direct IO
+ *
+ * The default value of this setting is zero (buffered IO is
+ * used). This setting takes immediate effect for all NFS
+ * versions, all exports, and in all NFSD net namespaces.
+ */
+
+static int nfsd_io_cache_write_get(void *data, u64 *val)
+{
+ *val = nfsd_io_cache_write;
+ return 0;
+}
+
+static int nfsd_io_cache_write_set(void *data, u64 val)
+{
+ switch (val) {
+ case NFSD_IO_BUFFERED:
+ case NFSD_IO_DONTCACHE:
+ case NFSD_IO_DIRECT:
+ nfsd_io_cache_write = val;
+ break;
+ default:
+ nfsd_io_cache_write = NFSD_IO_BUFFERED;
+ break;
+ }
+
+ return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(nfsd_io_cache_write_fops, nfsd_io_cache_write_get,
+ nfsd_io_cache_write_set, "%llu\n");
+
void nfsd_debugfs_exit(void)
{
debugfs_remove_recursive(nfsd_top_dir);
@@ -97,4 +135,7 @@ void nfsd_debugfs_init(void)
debugfs_create_file("io_cache_read", S_IWUSR | S_IRUGO,
nfsd_top_dir, NULL, &nfsd_io_cache_read_fops);
+
+ debugfs_create_file("io_cache_write", S_IWUSR | S_IRUGO,
+ nfsd_top_dir, NULL, &nfsd_io_cache_write_fops);
}
diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h
index 4740567f4e7e..1ae38c5557c4 100644
--- a/fs/nfsd/nfsd.h
+++ b/fs/nfsd/nfsd.h
@@ -160,6 +160,7 @@ enum {
};
extern u64 nfsd_io_cache_read __read_mostly;
+extern u64 nfsd_io_cache_write __read_mostly;
extern int nfsd_max_blksize;
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 632ce417f4ef..05a7ba383334 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -50,6 +50,7 @@
bool nfsd_disable_splice_read __read_mostly;
u64 nfsd_io_cache_read __read_mostly;
+u64 nfsd_io_cache_write __read_mostly;
/**
* nfserrno - Map Linux errnos to NFS errnos
@@ -1228,6 +1229,20 @@ 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);
+
+ switch (nfsd_io_cache_write) {
+ case NFSD_IO_DIRECT:
+ if (iov_iter_is_aligned(&iter, nf->nf_dio_mem_align - 1,
+ nf->nf_dio_offset_align - 1))
+ kiocb.ki_flags = IOCB_DIRECT;
+ break;
+ case NFSD_IO_DONTCACHE:
+ kiocb.ki_flags = IOCB_DONTCACHE;
+ break;
+ case NFSD_IO_BUFFERED:
+ break;
+ }
+
since = READ_ONCE(file->f_wb_err);
if (verf)
nfsd_copy_write_verifier(verf, nn);
--
2.44.0
next prev parent reply other threads:[~2025-07-08 16:06 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-07-08 16:06 [RFC PATCH v2 0/8] NFSD: support DIO Mike Snitzer
2025-07-08 16:06 ` [RFC PATCH v2 1/8] NFSD: Relocate the fh_want_write() and fh_drop_write() helpers Mike Snitzer
2025-07-10 13:59 ` Jeff Layton
2025-07-08 16:06 ` [RFC PATCH v2 2/8] NFSD: Move the fh_getattr() helper Mike Snitzer
2025-07-10 13:59 ` Jeff Layton
2025-07-08 16:06 ` [RFC PATCH v2 3/8] NFSD: filecache: add STATX_DIOALIGN and STATX_DIO_READ_ALIGN support Mike Snitzer
2025-07-10 7:45 ` Christoph Hellwig
2025-07-14 17:46 ` Mike Snitzer
2025-07-08 16:06 ` [RFC PATCH v2 4/8] lib/iov_iter: remove piecewise bvec length checking in iov_iter_aligned_bvec Mike Snitzer
2025-07-10 7:24 ` Christoph Hellwig
2025-07-10 7:32 ` Mike Snitzer
2025-07-10 7:44 ` Christoph Hellwig
2025-07-10 13:52 ` Jeff Layton
2025-07-10 14:48 ` Keith Busch
2025-07-10 16:12 ` Mike Snitzer
2025-07-10 16:29 ` Keith Busch
2025-07-10 17:22 ` Mike Snitzer
2025-07-10 19:51 ` Keith Busch
2025-07-10 19:57 ` Keith Busch
2025-08-01 15:23 ` Keith Busch
2025-08-01 16:10 ` Mike Snitzer
2025-07-08 16:06 ` [RFC PATCH v2 5/8] NFSD: pass nfsd_file to nfsd_iter_read() Mike Snitzer
2025-07-08 16:06 ` [RFC PATCH v2 6/8] NFSD: add io_cache_read controls to debugfs interface Mike Snitzer
2025-07-10 7:47 ` Christoph Hellwig
2025-07-14 17:33 ` Mike Snitzer
2025-07-10 14:06 ` Jeff Layton
2025-07-10 22:46 ` Chuck Lever
2025-07-14 16:47 ` Mike Snitzer
2025-07-15 11:57 ` Jeff Layton
2025-07-08 16:06 ` Mike Snitzer [this message]
2025-07-08 16:06 ` [RFC PATCH v2 8/8] NFSD: issue READs using O_DIRECT even if IO is misaligned Mike Snitzer
2025-07-08 21:22 ` Mike Snitzer
2025-07-10 7:51 ` Christoph Hellwig
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=20250708160619.64800-8-snitzer@kernel.org \
--to=snitzer@kernel.org \
--cc=Dai.Ngo@oracle.com \
--cc=anna@kernel.org \
--cc=chuck.lever@oracle.com \
--cc=jlayton@kernel.org \
--cc=linus-fsdevel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-nfs@vger.kernel.org \
--cc=neil@brown.name \
--cc=okorniev@redhat.com \
--cc=tom@talpey.com \
--cc=trondmy@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.