Linux FSCRYPT development
 help / color / mirror / Atom feed
From: Eric Biggers <ebiggers@kernel.org>
To: linux-ext4@vger.kernel.org
Cc: linux-fscrypt@vger.kernel.org,
	Harshad Shirwadkar <harshadshirwadkar@gmail.com>,
	stable@vger.kernel.org
Subject: [PATCH 4/7] ext4: add missing validation of fast-commit record lengths
Date: Sun,  6 Nov 2022 14:48:38 -0800	[thread overview]
Message-ID: <20221106224841.279231-5-ebiggers@kernel.org> (raw)
In-Reply-To: <20221106224841.279231-1-ebiggers@kernel.org>

From: Eric Biggers <ebiggers@google.com>

Validate the inode and filename lengths in fast-commit journal records
so that a malicious fast-commit journal cannot cause a crash by having
invalid values for these.  Also validate EXT4_FC_TAG_DEL_RANGE.

Fixes: aa75f4d3daae ("ext4: main fast-commit commit path")
Cc: <stable@vger.kernel.org> # v5.10+
Signed-off-by: Eric Biggers <ebiggers@google.com>
---
 fs/ext4/fast_commit.c | 38 +++++++++++++++++++-------------------
 fs/ext4/fast_commit.h |  2 +-
 2 files changed, 20 insertions(+), 20 deletions(-)

diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c
index 1e8be05542396..d5ad4b2b235d6 100644
--- a/fs/ext4/fast_commit.c
+++ b/fs/ext4/fast_commit.c
@@ -1991,32 +1991,31 @@ void ext4_fc_replay_cleanup(struct super_block *sb)
 	kfree(sbi->s_fc_replay_state.fc_modified_inodes);
 }
 
-static inline bool ext4_fc_tag_len_isvalid(struct ext4_fc_tl *tl,
-					   u8 *val, u8 *end)
+static bool ext4_fc_value_len_isvalid(struct ext4_sb_info *sbi,
+				      int tag, int len)
 {
-	if (val + tl->fc_len > end)
-		return false;
-
-	/* Here only check ADD_RANGE/TAIL/HEAD which will read data when do
-	 * journal rescan before do CRC check. Other tags length check will
-	 * rely on CRC check.
-	 */
-	switch (tl->fc_tag) {
+	switch (tag) {
 	case EXT4_FC_TAG_ADD_RANGE:
-		return (sizeof(struct ext4_fc_add_range) == tl->fc_len);
-	case EXT4_FC_TAG_TAIL:
-		return (sizeof(struct ext4_fc_tail) <= tl->fc_len);
-	case EXT4_FC_TAG_HEAD:
-		return (sizeof(struct ext4_fc_head) == tl->fc_len);
+		return len == sizeof(struct ext4_fc_add_range);
 	case EXT4_FC_TAG_DEL_RANGE:
+		return len == sizeof(struct ext4_fc_del_range);
+	case EXT4_FC_TAG_CREAT:
 	case EXT4_FC_TAG_LINK:
 	case EXT4_FC_TAG_UNLINK:
-	case EXT4_FC_TAG_CREAT:
+		len -= sizeof(struct ext4_fc_dentry_info);
+		return len >= 1 && len <= EXT4_NAME_LEN;
 	case EXT4_FC_TAG_INODE:
+		len -= sizeof(struct ext4_fc_inode);
+		return len >= EXT4_GOOD_OLD_INODE_SIZE &&
+			len <= sbi->s_inode_size;
 	case EXT4_FC_TAG_PAD:
-	default:
-		return true;
+		return true; /* padding can have any length */
+	case EXT4_FC_TAG_TAIL:
+		return len >= sizeof(struct ext4_fc_tail);
+	case EXT4_FC_TAG_HEAD:
+		return len == sizeof(struct ext4_fc_head);
 	}
+	return false;
 }
 
 /*
@@ -2079,7 +2078,8 @@ static int ext4_fc_replay_scan(journal_t *journal,
 	     cur = cur + EXT4_FC_TAG_BASE_LEN + tl.fc_len) {
 		ext4_fc_get_tl(&tl, cur);
 		val = cur + EXT4_FC_TAG_BASE_LEN;
-		if (!ext4_fc_tag_len_isvalid(&tl, val, end)) {
+		if (tl.fc_len > end - val ||
+		    !ext4_fc_value_len_isvalid(sbi, tl.fc_tag, tl.fc_len)) {
 			ret = state->fc_replay_num_tags ?
 				JBD2_FC_REPLAY_STOP : -ECANCELED;
 			goto out_err;
diff --git a/fs/ext4/fast_commit.h b/fs/ext4/fast_commit.h
index 256f2ad27204a..2fadb2c4780c8 100644
--- a/fs/ext4/fast_commit.h
+++ b/fs/ext4/fast_commit.h
@@ -58,7 +58,7 @@ struct ext4_fc_dentry_info {
 	__u8 fc_dname[];
 };
 
-/* Value structure for EXT4_FC_TAG_INODE and EXT4_FC_TAG_INODE_PARTIAL. */
+/* Value structure for EXT4_FC_TAG_INODE. */
 struct ext4_fc_inode {
 	__le32 fc_ino;
 	__u8 fc_raw_inode[];
-- 
2.38.1


  parent reply	other threads:[~2022-11-06 22:52 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-06 22:48 [PATCH 0/7] ext4 fast-commit fixes Eric Biggers
2022-11-06 22:48 ` [PATCH 1/7] ext4: disable fast-commit of encrypted dir operations Eric Biggers
2022-11-06 22:48 ` [PATCH 2/7] ext4: don't set up encryption key during jbd2 transaction Eric Biggers
2022-11-06 22:48 ` [PATCH 3/7] ext4: fix leaking uninitialized memory in fast-commit journal Eric Biggers
2022-11-06 22:48 ` Eric Biggers [this message]
2022-11-06 22:48 ` [PATCH 5/7] ext4: fix unaligned memory access in ext4_fc_reserve_space() Eric Biggers
2022-11-06 22:48 ` [PATCH 6/7] ext4: fix off-by-one errors in fast-commit block filling Eric Biggers
2022-11-06 22:48 ` [PATCH 7/7] ext4: simplify fast-commit CRC calculation Eric Biggers
2022-11-16  1:18 ` [PATCH 0/7] ext4 fast-commit fixes Eric Biggers
2022-11-28 19:03   ` Eric Biggers
2022-12-06 21:04 ` Theodore Ts'o
2022-12-09 16:12   ` harshad shirwadkar

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=20221106224841.279231-5-ebiggers@kernel.org \
    --to=ebiggers@kernel.org \
    --cc=harshadshirwadkar@gmail.com \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-fscrypt@vger.kernel.org \
    --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