From: Jan Kara <jack@suse.cz>
To: <linux-fsdevel@vger.kernel.org>
Cc: Yunpeng Tian <shionthanatos@gmail.com>,
Gongming Wang <gmwgg05@gmail.com>, Mingda Zhang <npczmd@qq.com>,
Qinrun Dai <jupmouse@gmail.com>,
stable@vger.kernel.org, Jan Kara <jack@suse.cz>
Subject: [PATCH 1/2] udf: validate the device specification EA before using it
Date: Tue, 1 Sep 2026 11:12:25 +0200 [thread overview]
Message-ID: <20260901091256.1226297-3-jack@suse.cz> (raw)
In-Reply-To: <20260901091037.30587-1-jack@suse.cz>
From: Yunpeng Tian <shionthanatos@gmail.com>
udf_get_extendedattr() bounds an extended attribute only against
struct genericFormat, which is 12 bytes. A type-12 device
specification is struct deviceSpec, which is 24 bytes plus a flexible
impUse array. Thus a corrupted filesystem can contain extended attribute
that is accepted by udf_get_extendedattr() but udf_write_inode() or
udf_read_inode() may access beyond the end of provided buffer.
Provide a udf_device_spec_valid() check to validate correctness of
extended attribute that is a deviceSpec. While here, check the result of
udf_add_extendedattr() before dereferencing it as it can return NULL in
case of error.
[JK: Removed duplicated validity checks]
Reported-by: Yunpeng Tian <shionthanatos@gmail.com>
Reported-by: Gongming Wang <gmwgg05@gmail.com>
Reported-by: Mingda Zhang <npczmd@qq.com>
Reported-by: Qinrun Dai <jupmouse@gmail.com>
Cc: stable@vger.kernel.org
Signed-off-by: Yunpeng Tian <shionthanatos@gmail.com>
Signed-off-by: Jan Kara <jack@suse.cz>
---
fs/udf/inode.c | 43 ++++++++++++++++++++++++++++++++++++++-----
1 file changed, 38 insertions(+), 5 deletions(-)
diff --git a/fs/udf/inode.c b/fs/udf/inode.c
index e45e546a739a..baf9845ac06b 100644
--- a/fs/udf/inode.c
+++ b/fs/udf/inode.c
@@ -1336,6 +1336,26 @@ int udf_setsize(struct inode *inode, loff_t newsize)
return err;
}
+/*
+ * Verify validity of struct deviceSpec on disk. udf_get_extendedattr() has
+ * already verified the generic header and made sure attribute fits in the
+ * inode so we just have to make sure attribute space is large enough for
+ * deviceSpec struct and required impUse information.
+ */
+static bool udf_device_spec_valid(struct deviceSpec *dsea)
+{
+ u32 attr_length, imp_use_length;
+
+ attr_length = le32_to_cpu(dsea->attrLength);
+ imp_use_length = le32_to_cpu(dsea->impUseLength);
+ if (attr_length < sizeof(struct deviceSpec) ||
+ imp_use_length < sizeof(struct regid) ||
+ imp_use_length > attr_length - sizeof(struct deviceSpec))
+ return false;
+
+ return true;
+}
+
/*
* Maximum length of linked list formed by ICB hierarchy. The chosen number is
* arbitrary - just that we hopefully don't limit any real use of rewritten
@@ -1654,13 +1674,12 @@ static int udf_read_inode(struct inode *inode, bool hidden_inode)
if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) {
struct deviceSpec *dsea =
(struct deviceSpec *)udf_get_extendedattr(inode, 12, 1);
- if (dsea) {
- init_special_inode(inode, inode->i_mode,
+
+ if (!dsea || !udf_device_spec_valid(dsea))
+ goto out;
+ init_special_inode(inode, inode->i_mode,
MKDEV(le32_to_cpu(dsea->majorDeviceIdent),
le32_to_cpu(dsea->minorDeviceIdent)));
- /* Developer ID ??? */
- } else
- goto out;
}
ret = 0;
out:
@@ -1757,6 +1776,7 @@ int udf_write_inode(struct inode *inode, struct writeback_control *wbc)
struct udf_sb_info *sbi = UDF_SB(inode->i_sb);
unsigned char blocksize_bits = inode->i_sb->s_blocksize_bits;
struct udf_inode_info *iinfo = UDF_I(inode);
+ int err;
bh = sb_getblk(inode->i_sb,
udf_get_lb_pblock(inode->i_sb, &iinfo->i_location, 0));
@@ -1821,6 +1841,10 @@ int udf_write_inode(struct inode *inode, struct writeback_control *wbc)
udf_add_extendedattr(inode,
sizeof(struct deviceSpec) +
sizeof(struct regid), 12, 0x3);
+ if (!dsea) {
+ err = -ENOSPC;
+ goto out_unlock;
+ }
dsea->attrType = cpu_to_le32(12);
dsea->attrSubtype = 1;
dsea->attrLength = cpu_to_le32(
@@ -1828,6 +1852,10 @@ int udf_write_inode(struct inode *inode, struct writeback_control *wbc)
sizeof(struct regid));
dsea->impUseLength = cpu_to_le32(sizeof(struct regid));
}
+ if (!udf_device_spec_valid(dsea)) {
+ err = -EFSCORRUPTED;
+ goto out_unlock;
+ }
eid = (struct regid *)dsea->impUse;
memset(eid, 0, sizeof(*eid));
strcpy(eid->ident, UDF_ID_DEVELOPER);
@@ -1962,6 +1990,11 @@ int udf_write_inode(struct inode *inode, struct writeback_control *wbc)
set_inode_metadata_writeback(inode);
return 0;
+
+out_unlock:
+ unlock_buffer(bh);
+ brelse(bh);
+ return err;
}
struct inode *__udf_iget(struct super_block *sb, struct kernel_lb_addr *ino,
--
2.51.0
next prev parent reply other threads:[~2026-09-01 9:13 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-01 9:12 [PATCH 0/2] udf: Fix handling of device spec attrs Jan Kara
2026-09-01 9:12 ` Jan Kara [this message]
2026-09-01 9:12 ` [PATCH 2/2] udf: Avoid confusion of missing and corrupted extended attributes Jan Kara
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=20260901091256.1226297-3-jack@suse.cz \
--to=jack@suse.cz \
--cc=gmwgg05@gmail.com \
--cc=jupmouse@gmail.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=npczmd@qq.com \
--cc=shionthanatos@gmail.com \
--cc=stable@vger.kernel.org \
/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