NTFS3 file system kernel mode driver
 help / color / mirror / Atom feed
From: Weiming Shi <bestswngs@gmail.com>
To: ntfs3@lists.linux.dev
Cc: Konstantin Komarov <almaz.alexandrovich@paragon-software.com>,
	Xiang Mei <xmei5@asu.edu>,
	linux-kernel@vger.kernel.org, Weiming Shi <bestswngs@gmail.com>
Subject: [PATCH] fs/ntfs3: take run_lock when freeing offs_folio in ni_decompress_file
Date: Fri, 26 Jun 2026 03:23:43 -0700	[thread overview]
Message-ID: <20260626102341.1909289-3-bestswngs@gmail.com> (raw)

attr_wof_frame_info() allocates, caches and dereferences
ni->file.offs_folio under down_write(&ni->file.run_lock).
ni_decompress_file() frees the same folio with folio_put() without
taking run_lock. When a WOF externally-compressed file is opened for
write, ni_decompress_file() can drop the last reference while a
concurrent O_RDONLY reader is dereferencing the folio in
attr_wof_frame_info(), leading to a use-after-free:

 BUG: KASAN: use-after-free in attr_wof_frame_info (fs/ntfs3/attrib.c:1619)
 Read of size 4 at addr ffff8880120c9000 by task exploit
 Call Trace:
  attr_wof_frame_info (fs/ntfs3/attrib.c:1619)
  ni_read_frame (fs/ntfs3/frecord.c:2421)
  ni_read_folio_cmpr (fs/ntfs3/frecord.c:1916)
  ntfs_read_folio (fs/ntfs3/inode.c:648)
  read_pages (mm/readahead.c:181)
  page_cache_ra_unbounded (mm/readahead.c:292)
  force_page_cache_ra (mm/readahead.c:364)
  page_cache_sync_ra (mm/readahead.c:573)
  filemap_get_pages (mm/filemap.c:2688)
  filemap_read (mm/filemap.c:2806)
  generic_file_read_iter (mm/filemap.c:2994)
  ntfs_file_read_iter (fs/ntfs3/file.c:842)
  vfs_read (fs/read_write.c:574)
  __x64_sys_pread64 (fs/read_write.c:773)

Take run_lock around the folio_put(), as is done for every other
access to ni->file.offs_folio.

Fixes: 4342306f0f0d ("fs/ntfs3: Add file operations and implementation")
Reported-by: Xiang Mei <xmei5@asu.edu>
Assisted-by: Claude:claude-opus-4-8
Signed-off-by: Weiming Shi <bestswngs@gmail.com>
---
 fs/ntfs3/frecord.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/fs/ntfs3/frecord.c b/fs/ntfs3/frecord.c
index 2b49bc077558..b11961a1f8d8 100644
--- a/fs/ntfs3/frecord.c
+++ b/fs/ntfs3/frecord.c
@@ -2148,10 +2148,13 @@ int ni_decompress_file(struct ntfs_inode *ni)
 
 	/* Clear cached flag. */
 	ni->ni_flags &= ~NI_FLAG_COMPRESSED_MASK;
+	/* offs_folio is accessed under run_lock in attr_wof_frame_info(). */
+	down_write(&ni->file.run_lock);
 	if (ni->file.offs_folio) {
 		folio_put(ni->file.offs_folio);
 		ni->file.offs_folio = NULL;
 	}
+	up_write(&ni->file.run_lock);
 	mapping->a_ops = &ntfs_aops;
 
 out:
-- 
2.43.0


                 reply	other threads:[~2026-06-26 10:24 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=20260626102341.1909289-3-bestswngs@gmail.com \
    --to=bestswngs@gmail.com \
    --cc=almaz.alexandrovich@paragon-software.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=ntfs3@lists.linux.dev \
    --cc=xmei5@asu.edu \
    /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