[PATCH RFC] fs/ntfs3: handle partial allocation in attr_data_get_block_locked()

[syzbot]

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+7be6ad...@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=7be6ad5cb228286af3e3
Link: https://syzkaller.appspot.com/ai_job?id=3c9dee6e-001d-4972-8bf6-f6c960708246
To: "Konstantin Komarov" <almaz.ale...@paragon-software.com>
To: <nt...@lists.linux.dev>
Cc: <linux-...@vger.kernel.org>

---
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
--
This is an AI-generated patch subject to moderation.
Reply with '#syz upstream' to Sign-off the patch as a human author
and send it to the upstream kernel mailing lists.
Reply with '#syz reject' to reject it ('#syz unreject' to undo).

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 syzk...@googlegroups.com.

[Slawomir Stepien]

#syz upstream

Slawomir Stepien

[syzbot]

From: Slawomir Stepien <s...@poczta.fm>

Signed-off-by: Slawomir Stepien <s...@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
--