diff mbox series

[v2] ext4: avoid utf8_strncasecmp() with unstable name

Message ID 20200601200543.59417-1-ebiggers@kernel.org
State Awaiting Upstream
Headers show
Series [v2] ext4: avoid utf8_strncasecmp() with unstable name | expand

Commit Message

Eric Biggers June 1, 2020, 8:05 p.m. UTC
From: Eric Biggers <ebiggers@google.com>

If the dentry name passed to ->d_compare() fits in dentry::d_iname, then
it may be concurrently modified by a rename.  This can cause undefined
behavior (possibly out-of-bounds memory accesses or crashes) in
utf8_strncasecmp(), since fs/unicode/ isn't written to handle strings
that may be concurrently modified.

Fix this by first copying the filename to a stack buffer if needed.
This way we get a stable snapshot of the filename.

Fixes: b886ee3e778e ("ext4: Support case-insensitive file name lookups")
Cc: <stable@vger.kernel.org> # v5.2+
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Daniel Rosenberg <drosen@google.com>
Cc: Gabriel Krisman Bertazi <krisman@collabora.co.uk>
Signed-off-by: Eric Biggers <ebiggers@google.com>
---

v2: use memcpy() + barrier() instead of a byte-by-byte copy.

 fs/ext4/dir.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

Comments

Andreas Dilger June 1, 2020, 8:49 p.m. UTC | #1
On Jun 1, 2020, at 2:05 PM, Eric Biggers <ebiggers@kernel.org> wrote:
> 
> From: Eric Biggers <ebiggers@google.com>
> 
> If the dentry name passed to ->d_compare() fits in dentry::d_iname, then
> it may be concurrently modified by a rename.  This can cause undefined
> behavior (possibly out-of-bounds memory accesses or crashes) in
> utf8_strncasecmp(), since fs/unicode/ isn't written to handle strings
> that may be concurrently modified.
> 
> Fix this by first copying the filename to a stack buffer if needed.
> This way we get a stable snapshot of the filename.
> 
> Fixes: b886ee3e778e ("ext4: Support case-insensitive file name lookups")
> Cc: <stable@vger.kernel.org> # v5.2+
> Cc: Al Viro <viro@zeniv.linux.org.uk>
> Cc: Daniel Rosenberg <drosen@google.com>
> Cc: Gabriel Krisman Bertazi <krisman@collabora.co.uk>
> Signed-off-by: Eric Biggers <ebiggers@google.com>

LGTM.

Reviewed-by: Andreas Dilger <adilger@dilger.ca>

> ---
> 
> v2: use memcpy() + barrier() instead of a byte-by-byte copy.
> 
> fs/ext4/dir.c | 16 ++++++++++++++++
> 1 file changed, 16 insertions(+)
> 
> diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
> index c654205f648dd..1d82336b1cd45 100644
> --- a/fs/ext4/dir.c
> +++ b/fs/ext4/dir.c
> @@ -675,6 +675,7 @@ static int ext4_d_compare(const struct dentry *dentry, unsigned int len,
> 	struct qstr qstr = {.name = str, .len = len };
> 	const struct dentry *parent = READ_ONCE(dentry->d_parent);
> 	const struct inode *inode = READ_ONCE(parent->d_inode);
> +	char strbuf[DNAME_INLINE_LEN];
> 
> 	if (!inode || !IS_CASEFOLDED(inode) ||
> 	    !EXT4_SB(inode->i_sb)->s_encoding) {
> @@ -683,6 +684,21 @@ static int ext4_d_compare(const struct dentry *dentry, unsigned int len,
> 		return memcmp(str, name->name, len);
> 	}
> 
> +	/*
> +	 * If the dentry name is stored in-line, then it may be concurrently
> +	 * modified by a rename.  If this happens, the VFS will eventually retry
> +	 * the lookup, so it doesn't matter what ->d_compare() returns.
> +	 * However, it's unsafe to call utf8_strncasecmp() with an unstable
> +	 * string.  Therefore, we have to copy the name into a temporary buffer.
> +	 */
> +	if (len <= DNAME_INLINE_LEN - 1) {
> +		memcpy(strbuf, str, len);
> +		strbuf[len] = 0;
> +		qstr.name = strbuf;
> +		/* prevent compiler from optimizing out the temporary buffer */
> +		barrier();
> +	}
> +
> 	return ext4_ci_compare(inode, name, &qstr, false);
> }
> 
> --
> 2.26.2
> 


Cheers, Andreas
Eric Biggers June 9, 2020, 8:28 p.m. UTC | #2
On Mon, Jun 01, 2020 at 01:05:43PM -0700, Eric Biggers wrote:
> From: Eric Biggers <ebiggers@google.com>
> 
> If the dentry name passed to ->d_compare() fits in dentry::d_iname, then
> it may be concurrently modified by a rename.  This can cause undefined
> behavior (possibly out-of-bounds memory accesses or crashes) in
> utf8_strncasecmp(), since fs/unicode/ isn't written to handle strings
> that may be concurrently modified.
> 
> Fix this by first copying the filename to a stack buffer if needed.
> This way we get a stable snapshot of the filename.
> 
> Fixes: b886ee3e778e ("ext4: Support case-insensitive file name lookups")
> Cc: <stable@vger.kernel.org> # v5.2+
> Cc: Al Viro <viro@zeniv.linux.org.uk>
> Cc: Daniel Rosenberg <drosen@google.com>
> Cc: Gabriel Krisman Bertazi <krisman@collabora.co.uk>
> Signed-off-by: Eric Biggers <ebiggers@google.com>
> ---
> 
> v2: use memcpy() + barrier() instead of a byte-by-byte copy.
> 
>  fs/ext4/dir.c | 16 ++++++++++++++++
>  1 file changed, 16 insertions(+)

Ted, could you take this through the ext4 tree as a fix for 5.8?
The f2fs patch has been merged already.

- Eric
Theodore Ts'o June 11, 2020, 3:01 p.m. UTC | #3
On Mon, Jun 01, 2020 at 02:49:51PM -0600, Andreas Dilger wrote:
> On Jun 1, 2020, at 2:05 PM, Eric Biggers <ebiggers@kernel.org> wrote:
> > 
> > From: Eric Biggers <ebiggers@google.com>
> > 
> > If the dentry name passed to ->d_compare() fits in dentry::d_iname, then
> > it may be concurrently modified by a rename.  This can cause undefined
> > behavior (possibly out-of-bounds memory accesses or crashes) in
> > utf8_strncasecmp(), since fs/unicode/ isn't written to handle strings
> > that may be concurrently modified.
> > 
> > Fix this by first copying the filename to a stack buffer if needed.
> > This way we get a stable snapshot of the filename.
> > 
> > Fixes: b886ee3e778e ("ext4: Support case-insensitive file name lookups")
> > Cc: <stable@vger.kernel.org> # v5.2+
> > Cc: Al Viro <viro@zeniv.linux.org.uk>
> > Cc: Daniel Rosenberg <drosen@google.com>
> > Cc: Gabriel Krisman Bertazi <krisman@collabora.co.uk>
> > Signed-off-by: Eric Biggers <ebiggers@google.com>
> 
> LGTM.
> 
> Reviewed-by: Andreas Dilger <adilger@dilger.ca>

Thanks, applied.

					- Ted
diff mbox series

Patch

diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
index c654205f648dd..1d82336b1cd45 100644
--- a/fs/ext4/dir.c
+++ b/fs/ext4/dir.c
@@ -675,6 +675,7 @@  static int ext4_d_compare(const struct dentry *dentry, unsigned int len,
 	struct qstr qstr = {.name = str, .len = len };
 	const struct dentry *parent = READ_ONCE(dentry->d_parent);
 	const struct inode *inode = READ_ONCE(parent->d_inode);
+	char strbuf[DNAME_INLINE_LEN];
 
 	if (!inode || !IS_CASEFOLDED(inode) ||
 	    !EXT4_SB(inode->i_sb)->s_encoding) {
@@ -683,6 +684,21 @@  static int ext4_d_compare(const struct dentry *dentry, unsigned int len,
 		return memcmp(str, name->name, len);
 	}
 
+	/*
+	 * If the dentry name is stored in-line, then it may be concurrently
+	 * modified by a rename.  If this happens, the VFS will eventually retry
+	 * the lookup, so it doesn't matter what ->d_compare() returns.
+	 * However, it's unsafe to call utf8_strncasecmp() with an unstable
+	 * string.  Therefore, we have to copy the name into a temporary buffer.
+	 */
+	if (len <= DNAME_INLINE_LEN - 1) {
+		memcpy(strbuf, str, len);
+		strbuf[len] = 0;
+		qstr.name = strbuf;
+		/* prevent compiler from optimizing out the temporary buffer */
+		barrier();
+	}
+
 	return ext4_ci_compare(inode, name, &qstr, false);
 }