From: John Garry <john.g.garry@oracle.com>
To: chandan.babu@oracle.com, djwong@kernel.org, dchinner@redhat.com,
hch@lst.de
Cc: viro@zeniv.linux.org.uk, brauner@kernel.org, jack@suse.cz,
linux-xfs@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-fsdevel@vger.kernel.org, catherine.hoang@oracle.com,
martin.petersen@oracle.com
Subject: Re: [PATCH v3 02/14] xfs: always tail align maxlen allocations
Date: Tue, 13 Aug 2024 16:01:22 +0100 [thread overview]
Message-ID: <8f499ed7-b220-496d-a008-2320266d394f@oracle.com> (raw)
In-Reply-To: <20240801163057.3981192-3-john.g.garry@oracle.com>
On 01/08/2024 17:30, John Garry wrote:
> From: Dave Chinner <dchinner@redhat.com>
>
> When we do a large allocation, the core free space allocation code
> assumes that args->maxlen is aligned to args->prod/args->mod. hence
> if we get a maximum sized extent allocated, it does not do tail
> alignment of the extent.
>
> However, this assumes that nothing modifies args->maxlen between the
> original allocation context setup and trimming the selected free
> space extent to size. This assumption has recently been found to be
> invalid - xfs_alloc_space_available() modifies args->maxlen in low
> space situations - and there may be more situations we haven't yet
> found like this.
>
> Force aligned allocation introduces the requirement that extents are
> correctly tail aligned, resulting in this occasional latent
> alignment failure to be reclassified from an unimportant curiousity
> to a must-fix bug.
>
> Removing the assumption about args->maxlen allocations always being
> tail aligned is trivial, and should not impact anything because
> args->maxlen for inodes with extent size hints configured are
> already aligned. Hence all this change does it avoid weird corner
> cases that would have resulted in unaligned extent sizes by always
> trimming the extent down to an aligned size.
>
> Signed-off-by: Dave Chinner <dchinner@redhat.com>
> Reviewed-by: "Darrick J. Wong" <djwong@kernel.org> [provisional on v1 series comment]
Note that I marked this as provisional, as not all questions were
answered from v1:
https://lore.kernel.org/linux-xfs/20240621195058.GS3058325@frogsfrogsfrogs/
I plan on sending v4 today like this, and hope to clear up any
unanswered questions then.
> Signed-off-by: John Garry <john.g.garry@oracle.com>
> ---
> fs/xfs/libxfs/xfs_alloc.c | 12 +++++-------
> 1 file changed, 5 insertions(+), 7 deletions(-)
>
> diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
> index d559d992c6ef..bf08b9e9d9ac 100644
> --- a/fs/xfs/libxfs/xfs_alloc.c
> +++ b/fs/xfs/libxfs/xfs_alloc.c
> @@ -433,20 +433,18 @@ xfs_alloc_compute_diff(
> * Fix up the length, based on mod and prod.
> * len should be k * prod + mod for some k.
> * If len is too small it is returned unchanged.
> - * If len hits maxlen it is left alone.
> */
> -STATIC void
> +static void
> xfs_alloc_fix_len(
> - xfs_alloc_arg_t *args) /* allocation argument structure */
> + struct xfs_alloc_arg *args)
> {
> - xfs_extlen_t k;
> - xfs_extlen_t rlen;
> + xfs_extlen_t k;
> + xfs_extlen_t rlen = args->len;
>
> ASSERT(args->mod < args->prod);
> - rlen = args->len;
> ASSERT(rlen >= args->minlen);
> ASSERT(rlen <= args->maxlen);
> - if (args->prod <= 1 || rlen < args->mod || rlen == args->maxlen ||
> + if (args->prod <= 1 || rlen < args->mod ||
> (args->mod == 0 && rlen < args->prod))
> return;
> k = rlen % args->prod;
next prev parent reply other threads:[~2024-08-13 15:01 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-08-01 16:30 [PATCH v3 00/14] forcealign for xfs John Garry
2024-08-01 16:30 ` [PATCH v3 01/14] xfs: only allow minlen allocations when near ENOSPC John Garry
2024-08-06 18:51 ` Darrick J. Wong
2024-08-07 0:26 ` Dave Chinner
2024-08-01 16:30 ` [PATCH v3 02/14] xfs: always tail align maxlen allocations John Garry
2024-08-13 15:01 ` John Garry [this message]
2024-08-01 16:30 ` [PATCH v3 03/14] xfs: simplify extent allocation alignment John Garry
2024-08-06 18:56 ` Darrick J. Wong
2024-08-06 23:52 ` Dave Chinner
2024-08-07 0:23 ` Darrick J. Wong
2024-08-07 0:34 ` Dave Chinner
2024-08-01 16:30 ` [PATCH v3 04/14] xfs: make EOF allocation simpler John Garry
2024-08-06 18:58 ` Darrick J. Wong
2024-08-07 0:00 ` Dave Chinner
2024-08-07 0:24 ` Darrick J. Wong
2024-08-01 16:30 ` [PATCH v3 05/14] xfs: introduce forced allocation alignment John Garry
2024-08-01 16:30 ` [PATCH v3 06/14] xfs: align args->minlen for " John Garry
2024-08-01 16:30 ` [PATCH v3 07/14] xfs: Introduce FORCEALIGN inode flag John Garry
2024-08-06 19:02 ` Darrick J. Wong
2024-08-07 11:42 ` John Garry
2024-08-07 14:40 ` Darrick J. Wong
2024-08-01 16:30 ` [PATCH v3 08/14] xfs: Update xfs_inode_alloc_unitsize() for forcealign John Garry
2024-08-06 19:02 ` Darrick J. Wong
2024-08-01 16:30 ` [PATCH v3 09/14] xfs: Update xfs_setattr_size() " John Garry
2024-08-06 19:03 ` Darrick J. Wong
2024-08-01 16:30 ` [PATCH v3 10/14] xfs: Do not free EOF blocks " John Garry
2024-08-06 19:24 ` Darrick J. Wong
2024-08-07 12:33 ` John Garry
2024-08-07 15:12 ` Darrick J. Wong
2024-08-01 16:30 ` [PATCH v3 11/14] xfs: Only free full extents " John Garry
2024-08-06 19:27 ` Darrick J. Wong
2024-08-07 0:08 ` Dave Chinner
2024-08-07 13:06 ` John Garry
2024-08-01 16:30 ` [PATCH v3 12/14] xfs: Unmap blocks according to forcealign John Garry
2024-08-06 20:14 ` Darrick J. Wong
2024-08-07 13:40 ` John Garry
2024-08-07 16:19 ` Darrick J. Wong
2024-08-01 16:30 ` [PATCH v3 13/14] xfs: Don't revert allocated offset for forcealign John Garry
2024-08-01 16:30 ` [PATCH v3 14/14] xfs: Enable file data forcealign feature John Garry
2024-08-06 19:43 ` Darrick J. Wong
2024-08-07 13:50 ` John Garry
2024-08-07 15:17 ` Darrick J. Wong
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=8f499ed7-b220-496d-a008-2320266d394f@oracle.com \
--to=john.g.garry@oracle.com \
--cc=brauner@kernel.org \
--cc=catherine.hoang@oracle.com \
--cc=chandan.babu@oracle.com \
--cc=dchinner@redhat.com \
--cc=djwong@kernel.org \
--cc=hch@lst.de \
--cc=jack@suse.cz \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-xfs@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=viro@zeniv.linux.org.uk \
/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).