Linux filesystem development
 help / color / mirror / Atom feed
From: "Ritesh Harjani (IBM)" <ritesh.list@gmail.com>
To: linux-xfs@vger.kernel.org
Cc: "Darrick J . Wong" <djwong@kernel.org>,
	Jens Axboe <axboe@kernel.dk>,
	linux-fsdevel@vger.kernel.org,
	"Ritesh Harjani (IBM)" <ritesh.list@gmail.com>
Subject: [PATCH v3 3/3] xfs_io: Add RWF_DONTCACHE support to preadv2
Date: Sat, 15 Mar 2025 13:50:13 +0530	[thread overview]
Message-ID: <af58062f6e4e0f2acc72884f2a398cd8dcfcae2c.1741340857.git.ritesh.list@gmail.com> (raw)
In-Reply-To: <cover.1741340857.git.ritesh.list@gmail.com>

Add per-io RWF_DONTCACHE support flag to preadv2().
This enables xfs_io to perform uncached buffered-io reads.

	e.g. xfs_io -c "pread -U -V 1 0 16K" /mnt/f1

Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@gmail.com>
---
 io/pread.c        | 17 +++++++++++++++--
 man/man8/xfs_io.8 |  8 +++++++-
 2 files changed, 22 insertions(+), 3 deletions(-)

diff --git a/io/pread.c b/io/pread.c
index b314fbc7..606bfe36 100644
--- a/io/pread.c
+++ b/io/pread.c
@@ -38,6 +38,9 @@ pread_help(void)
 " -Z N -- zeed the random number generator (used when reading randomly)\n"
 "         (heh, zorry, the -s/-S arguments were already in use in pwrite)\n"
 " -V N -- use vectored IO with N iovecs of blocksize each (preadv)\n"
+#ifdef HAVE_PREADV2
+" -U   -- Perform the preadv2() with RWF_DONTCACHE\n"
+#endif
 "\n"
 " When in \"random\" mode, the number of read operations will equal the\n"
 " number required to do a complete forward/backward scan of the range.\n"
@@ -388,7 +391,7 @@ pread_f(
 	init_cvtnum(&fsblocksize, &fssectsize);
 	bsize = fsblocksize;

-	while ((c = getopt(argc, argv, "b:BCFRquvV:Z:")) != EOF) {
+	while ((c = getopt(argc, argv, "b:BCFRquUvV:Z:")) != EOF) {
 		switch (c) {
 		case 'b':
 			tmp = cvtnum(fsblocksize, fssectsize, optarg);
@@ -417,6 +420,11 @@ pread_f(
 		case 'u':
 			uflag = 1;
 			break;
+#ifdef HAVE_PREADV2
+		case 'U':
+			preadv2_flags |= RWF_DONTCACHE;
+			break;
+#endif
 		case 'v':
 			vflag = 1;
 			break;
@@ -446,6 +454,11 @@ pread_f(
 		exitcode = 1;
 		return command_usage(&pread_cmd);
 	}
+	if (preadv2_flags != 0 && vectors == 0) {
+		printf(_("preadv2 flags require vectored I/O (-V)\n"));
+		exitcode = 1;
+		return command_usage(&pread_cmd);
+	}

 	offset = cvtnum(fsblocksize, fssectsize, argv[optind]);
 	if (offset < 0 && (direction & (IO_RANDOM|IO_BACKWARD))) {
@@ -514,7 +527,7 @@ pread_init(void)
 	pread_cmd.argmin = 2;
 	pread_cmd.argmax = -1;
 	pread_cmd.flags = CMD_NOMAP_OK | CMD_FOREIGN_OK;
-	pread_cmd.args = _("[-b bs] [-qv] [-i N] [-FBR [-Z N]] off len");
+	pread_cmd.args = _("[-b bs] [-qUv] [-i N] [-FBR [-Z N]] off len");
 	pread_cmd.oneline = _("reads a number of bytes at a specified offset");
 	pread_cmd.help = pread_help;

diff --git a/man/man8/xfs_io.8 b/man/man8/xfs_io.8
index 47af5232..df508054 100644
--- a/man/man8/xfs_io.8
+++ b/man/man8/xfs_io.8
@@ -200,7 +200,7 @@ option will set the file permissions to read-write (0644). This allows xfs_io to
 set up mismatches between the file permissions and the open file descriptor
 read/write mode to exercise permission checks inside various syscalls.
 .TP
-.BI "pread [ \-b " bsize " ] [ \-qv ] [ \-FBR [ \-Z " seed " ] ] [ \-V " vectors " ] " "offset length"
+.BI "pread [ \-b " bsize " ] [ \-qUv ] [ \-FBR [ \-Z " seed " ] ] [ \-V " vectors " ] " "offset length"
 Reads a range of bytes in a specified blocksize from the given
 .IR offset .
 .RS 1.0i
@@ -214,6 +214,12 @@ requests will be split. The default blocksize is 4096 bytes.
 .B \-q
 quiet mode, do not write anything to standard output.
 .TP
+.B \-U
+Perform the
+.BR preadv2 (2)
+call with
+.IR RWF_DONTCACHE .
+.TP
 .B \-v
 dump the contents of the buffer after reading,
 by default only the count of bytes actually read is dumped.
--
2.48.1


  parent reply	other threads:[~2025-03-15  8:20 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-03-15  8:20 [PATCH v3 0/3] xfsprogs: Add support for preadv2() and RWF_DONTCACHE Ritesh Harjani (IBM)
2025-03-15  8:20 ` [PATCH v3 1/3] xfs_io: Add support for preadv2 Ritesh Harjani (IBM)
2025-03-15  8:20 ` [PATCH v3 2/3] xfs_io: Add RWF_DONTCACHE support to pwritev2 Ritesh Harjani (IBM)
2025-03-15  8:20 ` Ritesh Harjani (IBM) [this message]
2025-03-15 12:49 ` [PATCH v3 0/3] xfsprogs: Add support for preadv2() and RWF_DONTCACHE Jens Axboe

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=af58062f6e4e0f2acc72884f2a398cd8dcfcae2c.1741340857.git.ritesh.list@gmail.com \
    --to=ritesh.list@gmail.com \
    --cc=axboe@kernel.dk \
    --cc=djwong@kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-xfs@vger.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