mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "syzbot" <syzbot@kernel.org>
To: syzkaller-bugs@googlegroups.com, Slawomir Stepien <sst@poczta.fm>,
	"Konstantin Komarov" <almaz.alexandrovich@paragon-software.com>,
	<ntfs3@lists.linux.dev>
Cc: linux-kernel@vger.kernel.org, syzbot@lists.linux.dev,
	syzkaller-upstream-moderation@googlegroups.com
Subject: [PATCH] fs/ntfs3: handle partial allocation in attr_data_get_block_locked()
Date: Wed, 15 Jul 2026 12:29:34 +0000 (UTC)	[thread overview]
Message-ID: <8fc44517-e231-4539-a353-3dd47045a203@mail.kernel.org> (raw)

From: Slawomir Stepien <sst@poczta.fm>

When allocating clusters for a sparse or compressed attribute in
attr_data_get_block_locked(), the requested virtual cluster number (vcn0)
is aligned down to a frame boundary (vcn). The function then attempts to
allocate a full frame of clusters.

If the disk is nearly full and highly fragmented, attr_allocate_clusters()
might only manage to allocate a small fragment and return success with a
partial allocation length. If this partial allocation is so small that it
does not even reach the originally requested vcn0 (i.e., end <= vcn0), vcn0
remains unallocated (a sparse hole).

When the code subsequently calls run_lookup_entry() for vcn0, it finds the
original SPARSE_LCN entry, which triggers a WARN_ON(1).

Fix this by adding a check to verify if the partial allocation reached
vcn0. If it did not (vcn0 >= end), treat this as an -ENOSPC failure and
jump to the undo1 label. This safely rolls back the partial allocation,
restores the original size, and repacks the runs, leaving the inode in a
consistent state before returning the error to the caller.

WARNING: fs/ntfs3/attrib.c:1236 at attr_data_get_block_locked+0x1980/0x25d0
fs/ntfs3/attrib.c:1236
RIP: 0010:attr_data_get_block_locked+0x1980/0x25d0 fs/ntfs3/attrib.c:1236
Call Trace:
 <TASK>
 attr_data_get_block+0x1f1/0x2a0 fs/ntfs3/attrib.c:979
 ntfs_fallocate+0xd3d/0xf70 fs/ntfs3/file.c:670
 vfs_fallocate+0x663/0x7f0 fs/open.c:338
 ksys_fallocate fs/open.c:362 [inline]
 __do_sys_fallocate fs/open.c:367 [inline]
 __se_sys_fallocate fs/open.c:365 [inline]
 __x64_sys_fallocate+0xbf/0x110 fs/open.c:365
 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline]
 do_syscall_64+0x174/0x580 arch/x86/entry/syscall_64.c:94
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
 </TASK>

Fixes: 099ef9ab9203 ("fs/ntfs3: implement iomap-based file operations")
Assisted-by: Gemini:gemini-3.1-pro-preview Gemini:gemini-3-flash-preview syzbot
Reported-by: syzbot+7be6ad5cb228286af3e3@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=7be6ad5cb228286af3e3
Link: https://syzkaller.appspot.com/ai_job?id=3c9dee6e-001d-4972-8bf6-f6c960708246
Signed-off-by: Slawomir Stepien <sst@poczta.fm>

---
diff --git a/fs/ntfs3/attrib.c b/fs/ntfs3/attrib.c
index c621a4c58..786fd0ec0 100644
--- a/fs/ntfs3/attrib.c
+++ b/fs/ntfs3/attrib.c
@@ -1227,6 +1227,10 @@ int attr_data_get_block_locked(struct ntfs_inode *ni, CLST vcn, CLST clen,
 	total_size = total_size0 + ((u64)*len << cluster_bits);
 
 	if (vcn != vcn0) {
+		if (vcn0 >= end) {
+			err = -ENOSPC;
+			goto undo1;
+		}
 		if (!run_lookup_entry(run, vcn0, lcn, len, NULL)) {
 			err = -EINVAL;
 			goto out;


base-commit: 8cdeaa50eae8dad34885515f62559ee83e7e8dda
-- 
See https://goo.gle/syzbot-ai-patches for information about AI-generated patches.
You can comment on the patch as usual, syzbot will try to address
the comments and send a new version of the patch if necessary.
syzbot engineers can be reached at syzkaller@googlegroups.com.

                 reply	other threads:[~2026-07-15 12:29 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=8fc44517-e231-4539-a353-3dd47045a203@mail.kernel.org \
    --to=syzbot@kernel.org \
    --cc=almaz.alexandrovich@paragon-software.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ntfs3@lists.linux.dev \
    --cc=sst@poczta.fm \
    --cc=syzbot@lists.linux.dev \
    --cc=syzkaller-bugs@googlegroups.com \
    --cc=syzkaller-upstream-moderation@googlegroups.com \
    /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

Powered by JetHome