From: "Darrick J. Wong" <djwong@kernel.org>
To: Christoph Hellwig <hch@lst.de>
Cc: Zorro Lang <zlang@kernel.org>,
fstests@vger.kernel.org, linux-xfs@vger.kernel.org
Subject: Re: [PATCH 2/5] add a new min_dio_alignment helper
Date: Tue, 13 Aug 2024 07:44:39 -0700 [thread overview]
Message-ID: <20240813144439.GG6047@frogsfrogsfrogs> (raw)
In-Reply-To: <20240813144004.GD6047@frogsfrogsfrogs>
On Tue, Aug 13, 2024 at 07:40:04AM -0700, Darrick J. Wong wrote:
> On Tue, Aug 13, 2024 at 09:35:01AM +0200, Christoph Hellwig wrote:
> > Add a new C program to find the minimum direct I/O alignment. This
> > uses the statx stx_dio_offset_align field if provided, then falls
> > back to the BLKSSZGET ioctl for block backed file systems and finally
> > the page size. It is intended as a more capable replacement for the
> > _min_dio_alignment bash helper.
> >
> > Signed-off-by: Christoph Hellwig <hch@lst.de>
> > ---
> > src/Makefile | 2 +-
> > src/min_dio_alignment.c | 66 +++++++++++++++++++++++++++++++++++++++++
> > 2 files changed, 67 insertions(+), 1 deletion(-)
> > create mode 100644 src/min_dio_alignment.c
> >
> > diff --git a/src/Makefile b/src/Makefile
> > index 559209be9..b3da59a0e 100644
> > --- a/src/Makefile
> > +++ b/src/Makefile
> > @@ -34,7 +34,7 @@ LINUX_TARGETS = xfsctl bstat t_mtab getdevicesize preallo_rw_pattern_reader \
> > attr_replace_test swapon mkswap t_attr_corruption t_open_tmpfiles \
> > fscrypt-crypt-util bulkstat_null_ocount splice-test chprojid_fail \
> > detached_mounts_propagation ext4_resize t_readdir_3 splice2pipe \
> > - uuid_ioctl t_snapshot_deleted_subvolume fiemap-fault
> > + uuid_ioctl t_snapshot_deleted_subvolume fiemap-fault min_dio_alignment
Also this program ought to be listed in gitignore.
--D
> >
> > EXTRA_EXECS = dmerror fill2attr fill2fs fill2fs_check scaleread.sh \
> > btrfs_crc32c_forged_name.py popdir.pl popattr.py \
> > diff --git a/src/min_dio_alignment.c b/src/min_dio_alignment.c
> > new file mode 100644
> > index 000000000..c3345bfb2
> > --- /dev/null
> > +++ b/src/min_dio_alignment.c
> > @@ -0,0 +1,66 @@
> > +// SPDX-License-Identifier: GPL-2.0
> > +/*
> > + * Copyright (c) 2024 Christoph Hellwig
> > + */
> > +#include <fcntl.h>
> > +#include <stdio.h>
> > +#include <stdlib.h>
> > +#include <unistd.h>
> > +#include <sys/mount.h>
> > +#include <sys/ioctl.h>
> > +#include <sys/stat.h>
> > +#include "statx.h"
> > +
> > +static int min_dio_alignmenent(const char *mntpnt, const char *devname)
>
> min_dio_alignment
>
> > +{
> > + struct statx stx = { };
> > + struct stat st;
> > + int fd;
> > +
> > + /*
> > + * If the file system supports STATX_DIOALIGN, use the dio_offset_align
> > + * member, as that reports exactly the information that we are asking
> > + * for.
> > + *
> > + * STATX_DIOALIGN is only reported on regular files, so use O_TMPFILE
> > + * to create one without leaving a trace.
> > + */
> > + fd = open(mntpnt, O_TMPFILE | O_RDWR | O_EXCL, 0600);
> > + if (fd >= 0 &&
> > + xfstests_statx(fd, "", AT_EMPTY_PATH, STATX_DIOALIGN, &stx) == 0 &&
> > + (stx.stx_mask & STATX_DIOALIGN))
> > + return stx.stx_dio_offset_align;
> > +
> > + /*
> > + * If we are on a block device and no explicit aligned is reported, use
> > + * the logical block size as a guestimate.
> > + */
> > + if (stat(devname, &st) == 0 && S_ISBLK(st.st_mode)) {
> > + int dev_fd = open(devname, O_RDONLY);
> > + int logical_block_size;
> > +
> > + if (dev_fd > 0 &&
> > + fstat(dev_fd, &st) == 0 &&
> > + S_ISBLK(st.st_mode) &&
> > + ioctl(dev_fd, BLKSSZGET, &logical_block_size)) {
> > + return logical_block_size;
> > + }
> > + }
> > +
> > + /*
> > + * No support for STATX_DIOALIGN and not a block device:
> > + * default to PAGE_SIZE.
>
> Should we try DIOINFO here as a second to last gasp?
>
> --D
>
> > + */
> > + return getpagesize();
> > +}
> > +
> > +int main(int argc, char **argv)
> > +{
> > + if (argc != 3) {
> > + fprintf(stderr, "usage: %s mountpoint devicename\n", argv[0]);
> > + exit(1);
> > + }
> > +
> > + printf("%d\n", min_dio_alignmenent(argv[1], argv[2]));
> > + exit(0);
> > +}
> > --
> > 2.43.0
> >
> >
>
next prev parent reply other threads:[~2024-08-13 14:44 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-08-13 7:34 improve minalign handling Christoph Hellwig
2024-08-13 7:35 ` [PATCH 1/5] statx.h: update to latest kernel UAPI Christoph Hellwig
2024-08-13 14:37 ` Darrick J. Wong
2024-08-13 14:54 ` Christoph Hellwig
2024-08-13 7:35 ` [PATCH 2/5] add a new min_dio_alignment helper Christoph Hellwig
2024-08-13 14:40 ` Darrick J. Wong
2024-08-13 14:44 ` Darrick J. Wong [this message]
2024-08-13 14:55 ` Christoph Hellwig
2024-08-14 3:43 ` Christoph Hellwig
2024-08-13 7:35 ` [PATCH 3/5] xfs/424: don't use _min_dio_alignment Christoph Hellwig
2024-08-13 14:40 ` Darrick J. Wong
2024-08-13 7:35 ` [PATCH 4/5] generic: don't use _min_dio_alignment without a device argument Christoph Hellwig
2024-08-13 14:43 ` Darrick J. Wong
2024-08-13 14:56 ` Christoph Hellwig
2024-08-13 7:35 ` [PATCH 5/5] replace _min_dio_alignment with calls to src/min_dio_alignment Christoph Hellwig
2024-08-13 14:45 ` Darrick J. Wong
-- strict thread matches above, loose matches on Subject: below --
2024-08-14 4:52 improve minalign handling v2 Christoph Hellwig
2024-08-14 4:52 ` [PATCH 2/5] add a new min_dio_alignment helper Christoph Hellwig
2024-08-14 5:12 ` Darrick J. Wong
2024-09-10 17:09 ` Eric Biggers
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=20240813144439.GG6047@frogsfrogsfrogs \
--to=djwong@kernel.org \
--cc=fstests@vger.kernel.org \
--cc=hch@lst.de \
--cc=linux-xfs@vger.kernel.org \
--cc=zlang@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