fscrypt: derive dirhash key for casefolded directories

When we allow indexed directories to use both encryption and
casefolding, for the dirhash we can't just hash the ciphertext filenames
that are stored on-disk (as is done currently) because the dirhash must
be case insensitive, but the stored names are case-preserving.  Nor can
we hash the plaintext names with an unkeyed hash (or a hash keyed with a
value stored on-disk like ext4's s_hash_seed), since that would leak
information about the names that encryption is meant to protect.

Instead, if we can accept a dirhash that's only computable when the
fscrypt key is available, we can hash the plaintext names with a keyed
hash using a secret key derived from the directory's fscrypt master key.
We'll use SipHash-2-4 for this purpose.

Prepare for this by deriving a SipHash key for each casefolded encrypted
directory.  Make sure to handle deriving the key not only when setting
up the directory's fscrypt_info, but also in the case where the casefold
flag is enabled after the fscrypt_info was already set up.  (We could
just always derive the key regardless of casefolding, but that would
introduce unnecessary overhead for people not using casefolding.)

Signed-off-by: Daniel Rosenberg <drosen@google.com>
[EB: improved commit message, updated fscrypt.rst, squashed with change
 that avoids unnecessarily deriving the key, and many other cleanups]
Link: https://lore.kernel.org/r/20200120223201.241390-3-ebiggers@kernel.org
Signed-off-by: Eric Biggers <ebiggers@google.com>

Documentation/filesystems/fscrypt.rst

@@ -302,6 +302,16 @@ For master keys used for v2 encryption policies, a unique 16-byte "key
 identifier" is also derived using the KDF.  This value is stored in
 the clear, since it is needed to reliably identify the key itself.
 
+Dirhash keys
+------------
+
+For directories that are indexed using a secret-keyed dirhash over the
+plaintext filenames, the KDF is also used to derive a 128-bit
+SipHash-2-4 key per directory in order to hash filenames.  This works
+just like deriving a per-file encryption key, except that a different
+KDF context is used.  Currently, only casefolded ("case-insensitive")
+encrypted directories use this style of hashing.
+
 Encryption modes and usage
 ==========================
 

fs/crypto/fname.c

@@ -402,6 +402,27 @@ errout:
 }
 EXPORT_SYMBOL(fscrypt_setup_filename);
 
+/**
+ * fscrypt_fname_siphash() - calculate the SipHash of a filename
+ * @dir: the parent directory
+ * @name: the filename to calculate the SipHash of
+ *
+ * Given a plaintext filename @name and a directory @dir which uses SipHash as
+ * its dirhash method and has had its fscrypt key set up, this function
+ * calculates the SipHash of that name using the directory's secret dirhash key.
+ *
+ * Return: the SipHash of @name using the hash key of @dir
+ */
+u64 fscrypt_fname_siphash(const struct inode *dir, const struct qstr *name)
+{
+	const struct fscrypt_info *ci = dir->i_crypt_info;
+
+	WARN_ON(!ci->ci_dirhash_key_initialized);
+
+	return siphash(name->name, name->len, &ci->ci_dirhash_key);
+}
+EXPORT_SYMBOL_GPL(fscrypt_fname_siphash);
+
 /*
  * Validate dentries in encrypted directories to make sure we aren't potentially
  * caching stale dentries after a key has been added.

fs/crypto/fscrypt_private.h

@@ -12,6 +12,7 @@
 #define _FSCRYPT_PRIVATE_H
 
 #include <linux/fscrypt.h>
+#include <linux/siphash.h>
 #include <crypto/hash.h>
 
 #define CONST_STRLEN(str)	(sizeof(str) - 1)
@@ -188,6 +189,14 @@ struct fscrypt_info {
 	 */
 	struct fscrypt_direct_key *ci_direct_key;
 
+	/*
+	 * This inode's hash key for filenames.  This is a 128-bit SipHash-2-4
+	 * key.  This is only set for directories that use a keyed dirhash over
+	 * the plaintext filenames -- currently just casefolded directories.
+	 */
+	siphash_key_t ci_dirhash_key;
+	bool ci_dirhash_key_initialized;
+
 	/* The encryption policy used by this inode */
 	union fscrypt_policy ci_policy;
 
@@ -263,6 +272,7 @@ extern int fscrypt_init_hkdf(struct fscrypt_hkdf *hkdf, const u8 *master_key,
 #define HKDF_CONTEXT_PER_FILE_KEY	2
 #define HKDF_CONTEXT_DIRECT_KEY		3
 #define HKDF_CONTEXT_IV_INO_LBLK_64_KEY	4
+#define HKDF_CONTEXT_DIRHASH_KEY	5
 
 extern int fscrypt_hkdf_expand(const struct fscrypt_hkdf *hkdf, u8 context,
 			       const u8 *info, unsigned int infolen,
@@ -434,6 +444,9 @@ fscrypt_allocate_skcipher(struct fscrypt_mode *mode, const u8 *raw_key,
 extern int fscrypt_set_derived_key(struct fscrypt_info *ci,
 				   const u8 *derived_key);
 
+extern int fscrypt_derive_dirhash_key(struct fscrypt_info *ci,
+				      const struct fscrypt_master_key *mk);
+
 /* keysetup_v1.c */
 
 extern void fscrypt_put_direct_key(struct fscrypt_direct_key *dk);

fs/crypto/hooks.c

@@ -5,6 +5,8 @@
  * Encryption hooks for higher-level filesystem operations.
  */
 
+#include <linux/key.h>
+
 #include "fscrypt_private.h"
 
 /**
@@ -137,8 +139,14 @@ int fscrypt_prepare_setflags(struct inode *inode,
 			     unsigned int oldflags, unsigned int flags)
 {
 	struct fscrypt_info *ci;
+	struct fscrypt_master_key *mk;
 	int err;
 
+	/*
+	 * When the CASEFOLD flag is set on an encrypted directory, we must
+	 * derive the secret key needed for the dirhash.  This is only possible
+	 * if the directory uses a v2 encryption policy.
+	 */
 	if (IS_ENCRYPTED(inode) && (flags & ~oldflags & FS_CASEFOLD_FL)) {
 		err = fscrypt_require_key(inode);
 		if (err)
@@ -146,6 +154,14 @@ int fscrypt_prepare_setflags(struct inode *inode,
 		ci = inode->i_crypt_info;
 		if (ci->ci_policy.version != FSCRYPT_POLICY_V2)
 			return -EINVAL;
+		mk = ci->ci_master_key->payload.data[0];
+		down_read(&mk->mk_secret_sem);
+		if (is_master_key_secret_present(&mk->mk_secret))
+			err = fscrypt_derive_dirhash_key(ci, mk);
+		else
+			err = -ENOKEY;
+		up_read(&mk->mk_secret_sem);
+		return err;
 	}
 	return 0;
 }

fs/crypto/keysetup.c

@@ -174,10 +174,24 @@ done:
 	return 0;
 }
 
+int fscrypt_derive_dirhash_key(struct fscrypt_info *ci,
+			       const struct fscrypt_master_key *mk)
+{
+	int err;
+
+	err = fscrypt_hkdf_expand(&mk->mk_secret.hkdf, HKDF_CONTEXT_DIRHASH_KEY,
+				  ci->ci_nonce, FS_KEY_DERIVATION_NONCE_SIZE,
+				  (u8 *)&ci->ci_dirhash_key,
+				  sizeof(ci->ci_dirhash_key));
+	if (err)
+		return err;
+	ci->ci_dirhash_key_initialized = true;
+	return 0;
+}
+
 static int fscrypt_setup_v2_file_key(struct fscrypt_info *ci,
 				     struct fscrypt_master_key *mk)
 {
-	u8 derived_key[FSCRYPT_MAX_KEY_SIZE];
 	int err;
 
 	if (ci->ci_policy.v2.flags & FSCRYPT_POLICY_FLAG_DIRECT_KEY) {
@@ -189,8 +203,8 @@ static int fscrypt_setup_v2_file_key(struct fscrypt_info *ci,
 		 * This ensures that the master key is consistently used only
 		 * for HKDF, avoiding key reuse issues.
 		 */
-		return setup_per_mode_key(ci, mk, mk->mk_direct_tfms,
-					  HKDF_CONTEXT_DIRECT_KEY, false);
+		err = setup_per_mode_key(ci, mk, mk->mk_direct_tfms,
+					 HKDF_CONTEXT_DIRECT_KEY, false);
 	} else if (ci->ci_policy.v2.flags &
 		   FSCRYPT_POLICY_FLAG_IV_INO_LBLK_64) {
 		/*
@@ -199,21 +213,33 @@ static int fscrypt_setup_v2_file_key(struct fscrypt_info *ci,
 		 * the IVs.  This format is optimized for use with inline
 		 * encryption hardware compliant with the UFS or eMMC standards.
 		 */
-		return setup_per_mode_key(ci, mk, mk->mk_iv_ino_lblk_64_tfms,
-					  HKDF_CONTEXT_IV_INO_LBLK_64_KEY,
-					  true);
-	}
+		err = setup_per_mode_key(ci, mk, mk->mk_iv_ino_lblk_64_tfms,
+					 HKDF_CONTEXT_IV_INO_LBLK_64_KEY, true);
+	} else {
+		u8 derived_key[FSCRYPT_MAX_KEY_SIZE];
 
-	err = fscrypt_hkdf_expand(&mk->mk_secret.hkdf,
-				  HKDF_CONTEXT_PER_FILE_KEY,
-				  ci->ci_nonce, FS_KEY_DERIVATION_NONCE_SIZE,
-				  derived_key, ci->ci_mode->keysize);
+		err = fscrypt_hkdf_expand(&mk->mk_secret.hkdf,
+					  HKDF_CONTEXT_PER_FILE_KEY,
+					  ci->ci_nonce,
+					  FS_KEY_DERIVATION_NONCE_SIZE,
+					  derived_key, ci->ci_mode->keysize);
+		if (err)
+			return err;
+
+		err = fscrypt_set_derived_key(ci, derived_key);
+		memzero_explicit(derived_key, ci->ci_mode->keysize);
+	}
 	if (err)
 		return err;
 
-	err = fscrypt_set_derived_key(ci, derived_key);
-	memzero_explicit(derived_key, ci->ci_mode->keysize);
-	return err;
+	/* Derive a secret dirhash key for directories that need it. */
+	if (S_ISDIR(ci->ci_inode->i_mode) && IS_CASEFOLDED(ci->ci_inode)) {
+		err = fscrypt_derive_dirhash_key(ci, mk);
+		if (err)
+			return err;
+	}
+
+	return 0;
 }
 
 /*

include/linux/fscrypt.h

@@ -247,6 +247,9 @@ static inline bool fscrypt_match_name(const struct fscrypt_name *fname,
 	return !memcmp(de_name, fname->disk_name.name, fname->disk_name.len);
 }
 
+extern u64 fscrypt_fname_siphash(const struct inode *dir,
+				 const struct qstr *name);
+
 /* bio.c */
 extern void fscrypt_decrypt_bio(struct bio *);
 extern int fscrypt_zeroout_range(const struct inode *, pgoff_t, sector_t,
@@ -479,6 +482,13 @@ static inline bool fscrypt_match_name(const struct fscrypt_name *fname,
 	return !memcmp(de_name, fname->disk_name.name, fname->disk_name.len);
 }
 
+static inline u64 fscrypt_fname_siphash(const struct inode *dir,
+					const struct qstr *name)
+{
+	WARN_ON_ONCE(1);
+	return 0;
+}
+
 /* bio.c */
 static inline void fscrypt_decrypt_bio(struct bio *bio)
 {