diff mbox series

[v6] KEYS: encrypted: fix key instantiation with user-provided data

Message ID 20221019164526.B70DF1C59@mail.steuer-voss.de (mailing list archive)
State Handled Elsewhere
Headers show
Series [v6] KEYS: encrypted: fix key instantiation with user-provided data | expand

Commit Message

Nikolaus Voss Oct. 19, 2022, 4:38 p.m. UTC
Commit cd3bc044af48 ("KEYS: encrypted: Instantiate key with
user-provided decrypted data") added key instantiation with user
provided decrypted data.  The user data is hex-ascii-encoded but was
just memcpy'ed to the binary buffer. Fix this to use hex2bin instead.

Old keys created from user provided decrypted data saved with "keyctl
pipe" are still valid, however if the key is recreated from decrypted
data the old key must be converted to the correct format. This can be
done with a small shell script, e.g.:

BROKENKEY=abcdefABCDEF1234567890aaaaaaaaaa
NEWKEY=$(echo -ne $BROKENKEY | xxd -p -c32)
keyctl add user masterkey "$(cat masterkey.bin)" @u
keyctl add encrypted testkey "new user:masterkey 32 $NEWKEY" @u

However, NEWKEY is still broken: If for BROKENKEY 32 bytes were
specified, a brute force attacker knowing the key properties would only
need to try at most 2^(16*8) keys, as if the key was only 16 bytes long.

The security issue is a result of the combination of limiting the input
range to hex-ascii and using memcpy() instead of hex2bin(). It could
have been fixed either by allowing binary input or using hex2bin() (and
doubling the ascii input key length). This patch implements the latter.

The corresponding test for the Linux Test Project ltp has also been
fixed (see link below).

Fixes: cd3bc044af48 ("KEYS: encrypted: Instantiate key with user-provided decrypted data")
Cc: stable@kernel.org
Link: https://lore.kernel.org/ltp/20221006081709.92303897@mail.steuer-voss.de/
Reviewed-by: Mimi Zohar <zohar@linux.ibm.com>
Signed-off-by: Nikolaus Voss <nikolaus.voss@haag-streit.com>
---
Changes
=======
v6: - fix checkpatch warnings
v5: - explain security issue in commit message, add Reviewd-by
v4: - change "Link:" address
v3: - use generated random key in example, reformat commit message
v2: - clarify commit message, add example to recover old/broken keys
    - improve example in Documentation/security/keys/trusted-encrypted.rst
    - add link to ltp patch

 Documentation/security/keys/trusted-encrypted.rst | 3 ++-
 security/keys/encrypted-keys/encrypted.c          | 6 +++---
 2 files changed, 5 insertions(+), 4 deletions(-)

Comments

Mimi Zohar Oct. 19, 2022, 9:28 p.m. UTC | #1
On Wed, 2022-10-19 at 18:38 +0200, Nikolaus Voss wrote:
> Commit cd3bc044af48 ("KEYS: encrypted: Instantiate key with
> user-provided decrypted data") added key instantiation with user
> provided decrypted data.  The user data is hex-ascii-encoded but was
> just memcpy'ed to the binary buffer. Fix this to use hex2bin instead.
> 
> Old keys created from user provided decrypted data saved with "keyctl
> pipe" are still valid, however if the key is recreated from decrypted
> data the old key must be converted to the correct format. This can be
> done with a small shell script, e.g.:
> 
> BROKENKEY=abcdefABCDEF1234567890aaaaaaaaaa
> NEWKEY=$(echo -ne $BROKENKEY | xxd -p -c32)
> keyctl add user masterkey "$(cat masterkey.bin)" @u
> keyctl add encrypted testkey "new user:masterkey 32 $NEWKEY" @u
> 
> However, NEWKEY is still broken: If for BROKENKEY 32 bytes were
> specified, a brute force attacker knowing the key properties would only
> need to try at most 2^(16*8) keys, as if the key was only 16 bytes long.
> 
> The security issue is a result of the combination of limiting the input
> range to hex-ascii and using memcpy() instead of hex2bin(). It could
> have been fixed either by allowing binary input or using hex2bin() (and
> doubling the ascii input key length). This patch implements the latter.
> 
> The corresponding test for the Linux Test Project ltp has also been
> fixed (see link below).
> 
> Fixes: cd3bc044af48 ("KEYS: encrypted: Instantiate key with user-provided decrypted data")
> Cc: stable@kernel.org
> Link: https://lore.kernel.org/ltp/20221006081709.92303897@mail.steuer-voss.de/
> Reviewed-by: Mimi Zohar <zohar@linux.ibm.com>
> Signed-off-by: Nikolaus Voss <nikolaus.voss@haag-streit.com>

Thanks!  This patch is now queued in next-integrity/next-integrity-
testing.
Jarkko Sakkinen Oct. 23, 2022, 8:40 p.m. UTC | #2
On Wed, Oct 19, 2022 at 06:38:20PM +0200, Nikolaus Voss wrote:
> Commit cd3bc044af48 ("KEYS: encrypted: Instantiate key with
> user-provided decrypted data") added key instantiation with user
> provided decrypted data.  The user data is hex-ascii-encoded but was
> just memcpy'ed to the binary buffer. Fix this to use hex2bin instead.
> 
> Old keys created from user provided decrypted data saved with "keyctl
> pipe" are still valid, however if the key is recreated from decrypted
> data the old key must be converted to the correct format. This can be
> done with a small shell script, e.g.:
> 
> BROKENKEY=abcdefABCDEF1234567890aaaaaaaaaa
> NEWKEY=$(echo -ne $BROKENKEY | xxd -p -c32)
> keyctl add user masterkey "$(cat masterkey.bin)" @u
> keyctl add encrypted testkey "new user:masterkey 32 $NEWKEY" @u
> 
> However, NEWKEY is still broken: If for BROKENKEY 32 bytes were
> specified, a brute force attacker knowing the key properties would only
> need to try at most 2^(16*8) keys, as if the key was only 16 bytes long.
> 
> The security issue is a result of the combination of limiting the input
> range to hex-ascii and using memcpy() instead of hex2bin(). It could
> have been fixed either by allowing binary input or using hex2bin() (and
> doubling the ascii input key length). This patch implements the latter.
> 
> The corresponding test for the Linux Test Project ltp has also been
> fixed (see link below).
> 
> Fixes: cd3bc044af48 ("KEYS: encrypted: Instantiate key with user-provided decrypted data")
> Cc: stable@kernel.org
> Link: https://lore.kernel.org/ltp/20221006081709.92303897@mail.steuer-voss.de/
> Reviewed-by: Mimi Zohar <zohar@linux.ibm.com>
> Signed-off-by: Nikolaus Voss <nikolaus.voss@haag-streit.com>
> ---
> Changes
> =======
> v6: - fix checkpatch warnings
> v5: - explain security issue in commit message, add Reviewd-by
> v4: - change "Link:" address
> v3: - use generated random key in example, reformat commit message
> v2: - clarify commit message, add example to recover old/broken keys
>     - improve example in Documentation/security/keys/trusted-encrypted.rst
>     - add link to ltp patch
> 
>  Documentation/security/keys/trusted-encrypted.rst | 3 ++-
>  security/keys/encrypted-keys/encrypted.c          | 6 +++---
>  2 files changed, 5 insertions(+), 4 deletions(-)
> 
> diff --git a/Documentation/security/keys/trusted-encrypted.rst b/Documentation/security/keys/trusted-encrypted.rst
> index 0bfb4c339748..9bc9db8ec651 100644
> --- a/Documentation/security/keys/trusted-encrypted.rst
> +++ b/Documentation/security/keys/trusted-encrypted.rst
> @@ -350,7 +350,8 @@ Load an encrypted key "evm" from saved blob::
>  
>  Instantiate an encrypted key "evm" using user-provided decrypted data::
>  
> -    $ keyctl add encrypted evm "new default user:kmk 32 `cat evm_decrypted_data.blob`" @u
> +    $ evmkey=$(dd if=/dev/urandom bs=1 count=32 | xxd -c32 -p)
> +    $ keyctl add encrypted evm "new default user:kmk 32 $evmkey" @u
>      794890253
>  
>      $ keyctl print 794890253
> diff --git a/security/keys/encrypted-keys/encrypted.c b/security/keys/encrypted-keys/encrypted.c
> index e05cfc2e49ae..1e313982af02 100644
> --- a/security/keys/encrypted-keys/encrypted.c
> +++ b/security/keys/encrypted-keys/encrypted.c
> @@ -627,7 +627,7 @@ static struct encrypted_key_payload *encrypted_key_alloc(struct key *key,
>  			pr_err("encrypted key: instantiation of keys using provided decrypted data is disabled since CONFIG_USER_DECRYPTED_DATA is set to false\n");
>  			return ERR_PTR(-EINVAL);
>  		}
> -		if (strlen(decrypted_data) != decrypted_datalen) {
> +		if (strlen(decrypted_data) != decrypted_datalen * 2) {
>  			pr_err("encrypted key: decrypted data provided does not match decrypted data length provided\n");
>  			return ERR_PTR(-EINVAL);
>  		}
> @@ -791,8 +791,8 @@ static int encrypted_init(struct encrypted_key_payload *epayload,
>  		ret = encrypted_key_decrypt(epayload, format, hex_encoded_iv);
>  	} else if (decrypted_data) {
>  		get_random_bytes(epayload->iv, ivsize);
> -		memcpy(epayload->decrypted_data, decrypted_data,
> -				   epayload->decrypted_datalen);
> +		ret = hex2bin(epayload->decrypted_data, decrypted_data,
> +			      epayload->decrypted_datalen);
>  	} else {
>  		get_random_bytes(epayload->iv, ivsize);
>  		get_random_bytes(epayload->decrypted_data, epayload->decrypted_datalen);
> -- 
> 2.34.1
> 

Did you check that this passes "checkpatch.pl --strict"?

Your earlier version did not.

BR, Jarkko
Jarkko Sakkinen Oct. 23, 2022, 9:24 p.m. UTC | #3
On Wed, Oct 19, 2022 at 05:28:23PM -0400, Mimi Zohar wrote:
> On Wed, 2022-10-19 at 18:38 +0200, Nikolaus Voss wrote:
> > Commit cd3bc044af48 ("KEYS: encrypted: Instantiate key with
> > user-provided decrypted data") added key instantiation with user
> > provided decrypted data.  The user data is hex-ascii-encoded but was
> > just memcpy'ed to the binary buffer. Fix this to use hex2bin instead.
> > 
> > Old keys created from user provided decrypted data saved with "keyctl
> > pipe" are still valid, however if the key is recreated from decrypted
> > data the old key must be converted to the correct format. This can be
> > done with a small shell script, e.g.:
> > 
> > BROKENKEY=abcdefABCDEF1234567890aaaaaaaaaa
> > NEWKEY=$(echo -ne $BROKENKEY | xxd -p -c32)
> > keyctl add user masterkey "$(cat masterkey.bin)" @u
> > keyctl add encrypted testkey "new user:masterkey 32 $NEWKEY" @u
> > 
> > However, NEWKEY is still broken: If for BROKENKEY 32 bytes were
> > specified, a brute force attacker knowing the key properties would only
> > need to try at most 2^(16*8) keys, as if the key was only 16 bytes long.
> > 
> > The security issue is a result of the combination of limiting the input
> > range to hex-ascii and using memcpy() instead of hex2bin(). It could
> > have been fixed either by allowing binary input or using hex2bin() (and
> > doubling the ascii input key length). This patch implements the latter.
> > 
> > The corresponding test for the Linux Test Project ltp has also been
> > fixed (see link below).
> > 
> > Fixes: cd3bc044af48 ("KEYS: encrypted: Instantiate key with user-provided decrypted data")
> > Cc: stable@kernel.org
> > Link: https://lore.kernel.org/ltp/20221006081709.92303897@mail.steuer-voss.de/
> > Reviewed-by: Mimi Zohar <zohar@linux.ibm.com>
> > Signed-off-by: Nikolaus Voss <nikolaus.voss@haag-streit.com>
> 
> Thanks!  This patch is now queued in next-integrity/next-integrity-
> testing.

Did you check the checkpatch.pl because earlier versions did not pass.

BR, Jarkko
diff mbox series

Patch

diff --git a/Documentation/security/keys/trusted-encrypted.rst b/Documentation/security/keys/trusted-encrypted.rst
index 0bfb4c339748..9bc9db8ec651 100644
--- a/Documentation/security/keys/trusted-encrypted.rst
+++ b/Documentation/security/keys/trusted-encrypted.rst
@@ -350,7 +350,8 @@  Load an encrypted key "evm" from saved blob::
 
 Instantiate an encrypted key "evm" using user-provided decrypted data::
 
-    $ keyctl add encrypted evm "new default user:kmk 32 `cat evm_decrypted_data.blob`" @u
+    $ evmkey=$(dd if=/dev/urandom bs=1 count=32 | xxd -c32 -p)
+    $ keyctl add encrypted evm "new default user:kmk 32 $evmkey" @u
     794890253
 
     $ keyctl print 794890253
diff --git a/security/keys/encrypted-keys/encrypted.c b/security/keys/encrypted-keys/encrypted.c
index e05cfc2e49ae..1e313982af02 100644
--- a/security/keys/encrypted-keys/encrypted.c
+++ b/security/keys/encrypted-keys/encrypted.c
@@ -627,7 +627,7 @@  static struct encrypted_key_payload *encrypted_key_alloc(struct key *key,
 			pr_err("encrypted key: instantiation of keys using provided decrypted data is disabled since CONFIG_USER_DECRYPTED_DATA is set to false\n");
 			return ERR_PTR(-EINVAL);
 		}
-		if (strlen(decrypted_data) != decrypted_datalen) {
+		if (strlen(decrypted_data) != decrypted_datalen * 2) {
 			pr_err("encrypted key: decrypted data provided does not match decrypted data length provided\n");
 			return ERR_PTR(-EINVAL);
 		}
@@ -791,8 +791,8 @@  static int encrypted_init(struct encrypted_key_payload *epayload,
 		ret = encrypted_key_decrypt(epayload, format, hex_encoded_iv);
 	} else if (decrypted_data) {
 		get_random_bytes(epayload->iv, ivsize);
-		memcpy(epayload->decrypted_data, decrypted_data,
-				   epayload->decrypted_datalen);
+		ret = hex2bin(epayload->decrypted_data, decrypted_data,
+			      epayload->decrypted_datalen);
 	} else {
 		get_random_bytes(epayload->iv, ivsize);
 		get_random_bytes(epayload->decrypted_data, epayload->decrypted_datalen);