linux-kernel - [PATCH v2] 9p: fix i_size update race in getattr with writeback caching

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <20251227180137.730385-1-pierre@barre.sh>
Date: Sat, 27 Dec 2025 19:01:37 +0100
From: Pierre Barre <pierre@...re.sh>
To: ericvh@...nel.org,
	lucho@...kov.net,
	asmadeus@...ewreck.org
Cc: linux_oss@...debyte.com,
	v9fs@...ts.linux.dev,
	linux-kernel@...r.kernel.org,
	Pierre Barre <pierre@...re.sh>
Subject: [PATCH v2] 9p: fix i_size update race in getattr with writeback caching

With writeback caching (cache=mmap), v9fs_stat2inode() and
v9fs_stat2inode_dotl() unconditionally overwrite i_size from the server
response, even when dirty pages may exist locally. This causes processes
using lseek(SEEK_END) to see incorrect file sizes, leading to data
corruption when extending files while concurrent stat() calls occur.

Fix by passing V9FS_STAT2INODE_KEEP_ISIZE when CACHE_WRITEBACK is
enabled to preserve the client's authoritative i_size.

Signed-off-by: Pierre Barre <pierre@...re.sh>
---
 fs/9p/vfs_inode.c      | 7 ++++++-
 fs/9p/vfs_inode_dotl.c | 7 ++++++-
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index 97abe65bf7c1..bfba21f5d8a9 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -993,7 +993,12 @@ v9fs_vfs_getattr(struct mnt_idmap *idmap, const struct path *path,
 	if (IS_ERR(st))
 		return PTR_ERR(st);
 
-	v9fs_stat2inode(st, d_inode(dentry), dentry->d_sb, 0);
+	/*
+	 * With writeback caching, the client is authoritative for i_size.
+	 * Don't let the server overwrite it with a potentially stale value.
+	 */
+	v9fs_stat2inode(st, d_inode(dentry), dentry->d_sb,
+		(v9ses->cache & CACHE_WRITEBACK) ? V9FS_STAT2INODE_KEEP_ISIZE : 0);
 	generic_fillattr(&nop_mnt_idmap, request_mask, d_inode(dentry), stat);
 
 	p9stat_free(st);
diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c
index 643e759eacb2..67a0ded2e223 100644
--- a/fs/9p/vfs_inode_dotl.c
+++ b/fs/9p/vfs_inode_dotl.c
@@ -453,7 +453,12 @@ v9fs_vfs_getattr_dotl(struct mnt_idmap *idmap,
 	if (IS_ERR(st))
 		return PTR_ERR(st);
 
-	v9fs_stat2inode_dotl(st, d_inode(dentry), 0);
+	/*
+	 * With writeback caching, the client is authoritative for i_size.
+	 * Don't let the server overwrite it with a potentially stale value.
+	 */
+	v9fs_stat2inode_dotl(st, d_inode(dentry),
+		(v9ses->cache & CACHE_WRITEBACK) ? V9FS_STAT2INODE_KEEP_ISIZE : 0);
 	generic_fillattr(&nop_mnt_idmap, request_mask, d_inode(dentry), stat);
 	/* Change block size to what the server returned */
 	stat->blksize = st->st_blksize;
-- 
2.43.0