public inbox for linux-ext4@vger.kernel.org
 help / color / mirror / Atom feed
From: John Garry <john.g.garry@oracle.com>
To: axboe@kernel.dk, tytso@mit.edu, dchinner@redhat.com,
	viro@zeniv.linux.org.uk, brauner@kernel.org, djwong@kernel.org,
	jack@suse.com, chandan.babu@oracle.com, hch@lst.de
Cc: linux-block@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-btrfs@vger.kernel.org, linux-erofs@lists.ozlabs.org,
	linux-ext4@vger.kernel.org,
	linux-f2fs-devel@lists.sourceforge.net,
	linux-fsdevel@vger.kernel.org, gfs2@lists.linux.dev,
	linux-xfs@vger.kernel.org, catherine.hoang@oracle.com,
	ritesh.list@gmail.com, mcgrof@kernel.org,
	mikulas@artax.karlin.mff.cuni.cz, agruenba@redhat.com,
	miklos@szeredi.hu, martin.petersen@oracle.com,
	John Garry <john.g.garry@oracle.com>
Subject: [PATCH v4 04/22] xfs: only allow minlen allocations when near ENOSPC
Date: Fri,  7 Jun 2024 14:39:01 +0000	[thread overview]
Message-ID: <20240607143919.2622319-5-john.g.garry@oracle.com> (raw)
In-Reply-To: <20240607143919.2622319-1-john.g.garry@oracle.com>

From: Dave Chinner <dchinner@redhat.com>

When we are near ENOSPC and don't have enough free
space for an args->maxlen allocation, xfs_alloc_space_available()
will trim args->maxlen to equal the available space. However, this
function has only checked that there is enough contiguous free space
for an aligned args->minlen allocation to succeed. Hence there is no
guarantee that an args->maxlen allocation will succeed, nor that the
available space will allow for correct alignment of an args->maxlen
allocation.

Further, by trimming args->maxlen arbitrarily, it breaks an
assumption made in xfs_alloc_fix_len() that if the caller wants
aligned allocation, then args->maxlen will be set to an aligned
value. It then skips the tail alignment and so we end up with
extents that aren't aligned to extent size hint boundaries as we
approach ENOSPC.

To avoid this problem, don't reduce args->maxlen by some random,
arbitrary amount. If args->maxlen is too large for the available
space, reduce the allocation to a minlen allocation as we know we
have contiguous free space available for this to succeed and always
be correctly aligned.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: John Garry <john.g.garry@oracle.com>
---
 fs/xfs/libxfs/xfs_alloc.c | 19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)

diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
index 6c55a6e88eba..5855a21d4864 100644
--- a/fs/xfs/libxfs/xfs_alloc.c
+++ b/fs/xfs/libxfs/xfs_alloc.c
@@ -2409,14 +2409,23 @@ xfs_alloc_space_available(
 	if (available < (int)max(args->total, alloc_len))
 		return false;
 
+	if (flags & XFS_ALLOC_FLAG_CHECK)
+		return true;
+
 	/*
-	 * Clamp maxlen to the amount of free space available for the actual
-	 * extent allocation.
+	 * If we can't do a maxlen allocation, then we must reduce the size of
+	 * the allocation to match the available free space. We know how big
+	 * the largest contiguous free space we can allocate is, so that's our
+	 * upper bound. However, we don't exaclty know what alignment/size
+	 * constraints have been placed on the allocation, so we can't
+	 * arbitrarily select some new max size. Hence make this a minlen
+	 * allocation as we know that will definitely succeed and match the
+	 * callers alignment constraints.
 	 */
-	if (available < (int)args->maxlen && !(flags & XFS_ALLOC_FLAG_CHECK)) {
-		args->maxlen = available;
+	alloc_len = args->maxlen + (args->alignment - 1) + args->minalignslop;
+	if (longest < alloc_len) {
+		args->maxlen = args->minlen;
 		ASSERT(args->maxlen > 0);
-		ASSERT(args->maxlen >= args->minlen);
 	}
 
 	return true;
-- 
2.31.1


  parent reply	other threads:[~2024-06-07 14:40 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-06-07 14:38 [PATCH v4 00/22] block atomic writes for xfs John Garry
2024-06-07 14:38 ` [PATCH v4 01/22] fs: Add generic_atomic_write_valid_size() John Garry
2024-06-12 21:10   ` Darrick J. Wong
2024-06-13  7:35     ` John Garry
2024-06-20 21:24       ` Darrick J. Wong
2024-06-07 14:38 ` [PATCH v4 02/22] iomap: Allow filesystems set IO block zeroing size John Garry
2024-06-12 21:32   ` Darrick J. Wong
2024-06-13 10:31     ` John Garry
2024-06-21 21:18       ` Darrick J. Wong
2024-06-24 13:58         ` John Garry
2024-06-07 14:39 ` [PATCH v4 03/22] xfs: Use extent size granularity for iomap->io_block_size John Garry
2024-06-12 21:47   ` Darrick J. Wong
2024-06-13 11:13     ` John Garry
2024-06-07 14:39 ` John Garry [this message]
2024-06-07 14:39 ` [PATCH v4 05/22] xfs: always tail align maxlen allocations John Garry
2024-06-07 14:39 ` [PATCH v4 06/22] xfs: simplify extent allocation alignment John Garry
2024-06-07 14:39 ` [PATCH v4 07/22] xfs: make EOF allocation simpler John Garry
2024-06-07 14:39 ` [PATCH v4 08/22] xfs: introduce forced allocation alignment John Garry
2024-06-07 14:39 ` [PATCH v4 09/22] xfs: align args->minlen for " John Garry
2024-06-07 14:39 ` [PATCH v4 10/22] xfs: Introduce FORCEALIGN inode flag John Garry
2024-06-07 14:39 ` [PATCH v4 11/22] xfs: Do not free EOF blocks for forcealign John Garry
2024-06-07 14:39 ` [PATCH v4 12/22] xfs: Update xfs_inode_alloc_unitsize_fsb() " John Garry
2024-06-07 14:39 ` [PATCH v4 13/22] xfs: Unmap blocks according to forcealign John Garry
2024-06-11 10:08   ` John Garry
2024-06-07 14:39 ` [PATCH v4 14/22] xfs: Only free full extents for forcealign John Garry
2024-06-07 14:39 ` [PATCH v4 15/22] xfs: Don't revert allocated offset " John Garry
2024-06-07 14:39 ` [PATCH v4 16/22] xfs: Enable file data forcealign feature John Garry
2024-06-07 14:39 ` [PATCH v4 17/22] fs: Add FS_XFLAG_ATOMICWRITES flag John Garry
2024-06-07 14:39 ` [PATCH v4 18/22] iomap: Atomic write support John Garry
2024-06-07 14:39 ` [PATCH v4 19/22] xfs: Support FS_XFLAG_ATOMICWRITES for forcealign John Garry
2024-06-07 14:39 ` [PATCH v4 20/22] xfs: Support atomic write for statx John Garry
2024-06-07 14:39 ` [PATCH v4 21/22] xfs: Validate atomic writes John Garry
2024-06-07 14:39 ` [PATCH v4 22/22] xfs: Support setting FMODE_CAN_ATOMIC_WRITE John Garry

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=20240607143919.2622319-5-john.g.garry@oracle.com \
    --to=john.g.garry@oracle.com \
    --cc=agruenba@redhat.com \
    --cc=axboe@kernel.dk \
    --cc=brauner@kernel.org \
    --cc=catherine.hoang@oracle.com \
    --cc=chandan.babu@oracle.com \
    --cc=dchinner@redhat.com \
    --cc=djwong@kernel.org \
    --cc=gfs2@lists.linux.dev \
    --cc=hch@lst.de \
    --cc=jack@suse.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=linux-erofs@lists.ozlabs.org \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-f2fs-devel@lists.sourceforge.net \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-xfs@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=mcgrof@kernel.org \
    --cc=miklos@szeredi.hu \
    --cc=mikulas@artax.karlin.mff.cuni.cz \
    --cc=ritesh.list@gmail.com \
    --cc=tytso@mit.edu \
    --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