dm-crypt: disable block encryption with arc4
Subsystems (7)
Mikulas Patocka
maintainerPATCH 1/1 dm-crypt: disable block encryption with arc4
Hi
When using arc4 to encrypt a block device, the resulting device is unreliable. It reads garbage. That's because arc4 is a stream cipher, if you write something, it advances its state and if you attempt to decrypt the same sector, it uses new state that is different.
This patch disables the use of arc4 on block devices.
A question to crypto maintainers: Is there some general method how to determine that the cipher is a stream cipher, changes its state as it progresses and thus is unusable for block devices? I haven't found any flag for that.
Mikulas
---
Disable arc4 for encrypting block device
Arc4 is a stream cipher, it's once initialized with a key, it outputs a stream of bytes (that are xored with the data to be encrypted) and changes it's internal state.
Because the cipher changes it's internal state, it is not useable for encrypting block devices --- once someone encrypts a sector of data, the internal state changes --- and further attempts to decrypt the same block of data use the new internal state. Thus, the encrypted device returns garbage.
This patch disables the use of arc4 for dm-crypt.
If we wanted to use arc4, we would have to setup the key before encrypting each sector. That is slow. Because arc4 works by xoring the bitstream with the data, it is not suitable for encrypting block devices anyway: if the attacker obtains two images of the same block device at two different times, he can xor them with each other, eliminating the cipher and getting two xored plaintexts.
--- drivers/md/dm-crypt.c | 5 +++++ 1 file changed, 5 insertions(+)
drivers/md/dm-crypt.c +5 -0
@@ -1035,6 +1035,11 @@ static int crypt_ctr(struct dm_target *t goto bad_cipher; } +if (!strcmp (cc->cipher,"arc4" )) { + ti->error ="Stream cipher arc4 not supported" ; + goto bad_cipher; + } +if (snprintf (cc->cipher,CRYPTO_MAX_ALG_NAME ,"%s(%s)" , chainmode, cipher) >=CRYPTO_MAX_ALG_NAME ) { ti->error ="Chain mode + cipher name is too long" ;
Quoted 8 lines · Mikulas Patocka
Hi
When using arc4 to encrypt a block device, the resulting device is unreliable. It reads garbage. That's because arc4 is a stream cipher, if you write something, it advances its state and if you attempt to decrypt the same sector, it uses new state that is different.
This patch disables the use of arc4 on block devices.
arc4 again. it is simply not a block cipher:-)
This should be solved inside cryptoAPI and not blacklist it in dm-crypt, see that thread
http://article.gmane.org/gmane.linux.kernel.cryptoapi/3441
Milan
Quoted 8 lines · * Milan Broz | 2010-01-25 19:39:11 [+0100]:
Quoted 7 lines · Mikulas Patocka
Hi
When using arc4 to encrypt a block device, the resulting device is unreliable. It reads garbage. That's because arc4 is a stream cipher, if you write something, it advances its state and if you attempt to decrypt the same sector, it uses new state that is different.
Quoted 7 lines
Quoted 1 line
This patch disables the use of arc4 on block devices.
arc4 again. it is simply not a block cipher:-)
This should be solved inside cryptoAPI and not blacklist it in dm-crypt, see that thread
http://article.gmane.org/gmane.linux.kernel.cryptoapi/3441
I some how remember Herbert saying to test for block size > 1. Wouldn't this be acceptable to block all stream cipher in one go?
Quoted 1 line
Milan
Sebastian
Quoted 9 lines · Sebastian Andrzej Siewior
Quoted 8 lines · * Milan Broz | 2010-01-25 19:39:11 [+0100]:
Quoted 6 lines · Mikulas Patocka
When using arc4 to encrypt a block device, the resulting device is unreliable. It reads garbage. That's because arc4 is a stream cipher, if you write something, it advances its state and if you attempt to decrypt the same sector, it uses new state that is different.
This patch disables the use of arc4 on block devices.
Quoted 8 lines
Quoted 5 lines
arc4 again. it is simply not a block cipher:-)
This should be solved inside cryptoAPI and not blacklist it in dm-crypt, see that thread
http://article.gmane.org/gmane.linux.kernel.cryptoapi/3441I some how remember Herbert saying to test for block size > 1. Wouldn't this be acceptable to block all stream cipher in one go?
yes, I think it is better.
(...and I just forgot to add that test to dm-crypt after that suggestion.)
Milan
Mikulas Patocka
maintainerQuoted 8 lines
Quoted 8 lines
Quoted 7 lines
Quoted 1 line
This patch disables the use of arc4 on block devices.
arc4 again. it is simply not a block cipher:-)
This should be solved inside cryptoAPI and not blacklist it in dm-crypt, see that thread
http://article.gmane.org/gmane.linux.kernel.cryptoapi/3441
Quoted 7 lines
Quoted 2 lines
I some how remember Herbert saying to test for block size > 1. Wouldn't this be acceptable to block all stream cipher in one go?
yes, I think it is better.
(...and I just forgot to add that test to dm-crypt after that suggestion.)Milan
Hmm, there is salsa20 that has block size 1, larger initialization vectors, and can be used to encrypt disks (although salsa20 doesn't currently work with dm-crypt, because it doesn't accept "ecb(), cbc(), etc." chaining modes --- but if you remove the chaining mode manually, it works).
You should rather add a flag CRYPTO_ALG_CHANGES_STATE to determine that a cipher can't be used to encrypt disks.
Mikulas
Quoted 5 lines · * Mikulas Patocka | 2010-01-26 07:27:18 [-0500]:
Quoted 4 lines
yes, I think it is better.
(...and I just forgot to add that test to dm-crypt after that suggestion.)Milan
Quoted 8 lines
Hmm, there is salsa20 that has block size 1, larger initialization vectors, and can be used to encrypt disks (although salsa20 doesn't currently work with dm-crypt, because it doesn't accept "
ecb(),cbc(), etc." chaining modes --- but if you remove the chaining mode manually, it works).You should rather add a flag
CRYPTO_ALG_CHANGES_STATEto determine that a cipher can't be used to encrypt disks.
Just because it will work does not make it a good idea.
SALSA20 is a stream cipher not a block cipher.
Block ciphers are used to encrypt data.
Stream ciphers are used to create one time pads, a set of encryption keys, ...
There are block modes like CTR which can turn a block cipher into a stream cipher. Those should not be used for disk encryption as well.
Quoted 2 lines
Mikulas
Sebastian
Mikulas Patocka
maintainerQuoted 16 lines · Sebastian Andrzej Siewior
Quoted 5 lines · * Mikulas Patocka | 2010-01-26 07:27:18 [-0500]:
Quoted 4 lines
yes, I think it is better.
(...and I just forgot to add that test to dm-crypt after that suggestion.)Milan
Quoted 8 lines
Hmm, there is salsa20 that has block size 1, larger initialization vectors, and can be used to encrypt disks (although salsa20 doesn't currently work with dm-crypt, because it doesn't accept "
ecb(),cbc(), etc." chaining modes --- but if you remove the chaining mode manually, it works).You should rather add a flag
CRYPTO_ALG_CHANGES_STATEto determine that a cipher can't be used to encrypt disks.
Quoted 8 lines
Just because it will work does not make it a good idea.
SALSA20 is a stream cipher not a block cipher.
Block ciphers are used to encrypt data.
Stream ciphers are used to create one time pads, a set of encryption keys, ...
There are block modes like CTR which can turn a block cipher into a stream cipher. Those should not be used for disk encryption as well.
Salsa20 is unsuitable for disk encryption in most cases. It would be suitable if we knew that the attacker can obtain the image of encrypted device at most once --- it is OK to protect against laptop theft (it happens just once), but it is not OK to protect against support technician spying on your data (he can read them multiple times, if you have multiple support requests).
Anyway, what I wanted to say, is that block_size <= 1 test is no less hacky than !strcmp(cipher, "arc4") test.
Mikulas
Herbert Xu
maintainerQuoted 3 lines · Mikulas Patocka
You should rather add a flag
CRYPTO_ALG_CHANGES_STATEto determine that a cipher can't be used to encrypt disks.
No, please see my reply in the previous thread. What we should do is fix arc4. I just haven't got around to doing it yet.
As to blacklisting algorithms not suitable for disk encryption, that is up to the dm-crypt maintainers to decide.
Cheers,
Signature · 4 lines
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txtMikulas Patocka
maintainerQuoted 7 lines · Herbert Xu
Quoted 3 lines · Mikulas Patocka
You should rather add a flag
CRYPTO_ALG_CHANGES_STATEto determine that a cipher can't be used to encrypt disks.No, please see my reply in the previous thread. What we should do is fix arc4. I just haven't got around to doing it yet.
What is the fix for arc4? Copy the internal state after a key schedule and restore it with every encryption?
Quoted 4 lines
As to blacklisting algorithms not suitable for disk encryption, that is up to the dm-crypt maintainers to decide.
Cheers,
I think blacklisting "arc4" is better, because it provides a fix now. Otherwise, people will just keep on arguing what is the "clean" solution and nothing gets done.
Mikulas
Herbert Xu
maintainerQuoted 3 lines · Mikulas Patocka
What is the fix for arc4? Copy the internal state after a key schedule and restore it with every encryption?
arc4 should be a blkcipher, not a cipher. Then it can have an IV which is where thie should be stored.
Quoted 3 lines
I think blacklisting "arc4" is better, because it provides a fix now. Otherwise, people will just keep on arguing what is the "clean" solution and nothing gets done.
The crypto layer makes no guarantee that every algorithm that is available is suitable for a particular application such as disk encryption. FWIW we also export an algorithm called null!
People should not be making uninformed choices on crypto algorithms.
Cheers,
Signature · 4 lines
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txtQuoted 10 lines · * Herbert Xu | 2010-02-09 18:37:18 [+1100]:
Quoted 3 lines · Mikulas Patocka
You should rather add a flag
CRYPTO_ALG_CHANGES_STATEto determine that a cipher can't be used to encrypt disks.No, please see my reply in the previous thread. What we should do is fix arc4. I just haven't got around to doing it yet.
As to blacklisting algorithms not suitable for disk encryption, that is up to the dm-crypt maintainers to decide.
Herbert, what happend to the "check for streamcipher" idea you had? Is it gone? On the other hand it wouldn't be probably that bad to have a seprate interface to grab a block cipher _or_ a stream cipher. So something like this wouldn't happen. This is basically the "check for stream cipher" without encrypt = decrypt plus with a struct the cra_u union.
I can't imaging how you want to fix arc4 that it will work in dm-crypt. The algorithm relies more or less on the fact that it envolves itself during processing of data.
Salsa works with dm-crypt because the internal state is taken from the IV and is never written back. dm-crypt always encrypts/decrypts a 512 block in one go. Splitting it in two and requesting two 256 block encryptions would work with _every_ other block cipher but break salsa.
Quoted 2 lines
Cheers,
Sebastian
Herbert Xu
maintainerQuoted 6 lines · Sebastian Andrzej Siewior
Quoted 3 lines
As to blacklisting algorithms not suitable for disk encryption, that is up to the dm-crypt maintainers to decide.
Herbert, what happend to the "check for streamcipher" idea you had? Is it gone? On the other hand it wouldn't be probably that bad to have a
Well again whether that should be done is up to the dm-crypt maintainers.
Quoted 1 line
seprate interface to grab a block cipher _or_ a stream cipher. So
Just because something isn't a stream cipher doesn't mean that it is safe for disk encryption. People simply shouldn't be using random algorithms for disk encryption.
Quoted 1 line
I can't imaging how you want to fix arc4 that it will work in dm-crypt.
I thought I've explained this before. Just turn it into a blkcipher and add IV.
Quoted 2 lines
The algorithm relies more or less on the fact that it envolves itself during processing of data.
This is no different to any stream cipher.
Cheers,
Signature · 4 lines
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txtQuoted 5 lines · * Herbert Xu | 2010-02-10 07:45:19 [+1100]:
Quoted 2 lines
Herbert, what happend to the "check for streamcipher" idea you had? Is it gone? On the other hand it wouldn't be probably that bad to have a
Well again whether that should be done is up to the dm-crypt maintainers.
Milan liked that afaik.
Quoted 10 lines
Quoted 1 line
seprate interface to grab a block cipher _or_ a stream cipher. So
Just because something isn't a stream cipher doesn't mean that it is safe for disk encryption. People simply shouldn't be using random algorithms for disk encryption.
Quoted 1 line
I can't imaging how you want to fix arc4 that it will work in dm-crypt.
I thought I've explained this before. Just turn it into a blkcipher and add IV.
I beg your pardon. I probably mixed things up.
Quoted 4 lines
Quoted 2 lines
The algorithm relies more or less on the fact that it envolves itself during processing of data.
This is no different to any stream cipher.
Sure. So we fix arc4 and don't play mother . Okay I will into this :)
Quoted 2 lines
Cheers,
Sebastian
Herbert Xu
maintainerQuoted 3 lines · Sebastian Andrzej Siewior
Quoted 2 lines
This is no different to any stream cipher.
Sure. So we fix arc4 and don't play mother . Okay I will into this :)
That would be awesome. Thanks!
Signature · 4 lines
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txtPATCH 1/1 crypto/arc4: convert this stream cipher into a block cipher
the state has been moved from ctx into iv. That way encrypt()/decrypt() can deliver the same result for a given IV. This patch makes the cipher work with dm-crypt not that it is a good thing. However, the performance may have improved :)
The name is still ecb(aes) but since this is provided by the blkcipher itself, I removed the select statement.
---
I had it run with wireless and dm-crypt. No problems so far. Not sure if it makes sense to rename it to arc4 and strip the ecb prefix. It would make it consistent with salsa but would require another patch.
crypto/Kconfig | 2 +- crypto/arc4.c | 120 ++++++++++++++++++++++++---------- crypto/testmgr.h | 3 +- drivers/net/Kconfig | 1 - drivers/net/wireless/hostap/Kconfig | 2 - drivers/net/wireless/ipw2x00/Kconfig | 2 - net/mac80211/Kconfig | 1 - 7 files changed, 87 insertions(+), 44 deletions(-)
crypto/Kconfig +1 -1
@@ -561,7 +561,7 @@ config CRYPTO_ANUBIS config CRYPTO_ARC4 tristate "ARC4 cipher algorithm" - select CRYPTO_ALGAPI + select CRYPTO_BLKCIPHER help ARC4 cipher algorithm.
crypto/arc4.c +84 -36
164 lines not on this page · raw message
crypto/testmgr.h +2 -1
@@ -24,7 +24,8 @@#define MAX_TAP 8#define MAX_KEYLEN 56 -#define MAX_IVLEN 32 +/* sizeof arc4_iv */ +#define MAX_IVLEN 260struct hash_testvec {/* only used with keyed hash algorithms */
drivers/net/Kconfig +0 -1
@@ -3076,7 +3076,6 @@ config PPP_MPPE select CRYPTO select CRYPTO_SHA1 select CRYPTO_ARC4 - select CRYPTO_ECB ---help--- Support for the MPPE Encryption protocol, as employed by the Microsoft Point-to-Point Tunneling Protocol.
drivers/net/wireless/hostap/Kconfig +0 -2
@@ -5,10 +5,8 @@ config HOSTAP select WEXT_PRIV select CRYPTO select CRYPTO_ARC4 - select CRYPTO_ECB select CRYPTO_AES select CRYPTO_MICHAEL_MIC - select CRYPTO_ECB select CRC32 select LIB80211 select LIB80211_CRYPT_WEP
drivers/net/wireless/ipw2x00/Kconfig +0 -2
@@ -159,10 +159,8 @@ config LIBIPW select WEXT_SPY select CRYPTO select CRYPTO_ARC4 - select CRYPTO_ECB select CRYPTO_AES select CRYPTO_MICHAEL_MIC - select CRYPTO_ECB select CRC32 select LIB80211 select LIB80211_CRYPT_WEP
net/mac80211/Kconfig +0 -1
@@ -2,7 +2,6 @@ config MAC80211 tristate "Generic IEEE 802.11 Networking Stack (mac80211)" depends on CFG80211 select CRYPTO - select CRYPTO_ECB select CRYPTO_ARC4 select CRYPTO_AES select CRC32
Signature · 1 line
1.6.6.1
Re: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
Hi,
Quoted 5 lines · Sebastian Andrzej Siewior
the state has been moved from ctx into iv. That way encrypt()/decrypt() can deliver the same result for a given IV. This patch makes the cipher work with dm-crypt not that it is a good thing. However, the performance may have improved :)
The name is still ecb(aes) but since this is provided by the blkcipher itself,
Just to avoid any confusion you meant ecb(arc4) not ecb(aes) here right?
-Adrian
Re: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
Quoted 1 line · * Adrian-Ken Rueegsegger | 2010-02-12 10:34:27 [+0100]:
Hi,
Hi,
Quoted 3 lines
Quoted 1 line · Sebastian Andrzej Siewior
The name is still ecb(aes) but since this is provided by the blkcipher itself,
Just to avoid any confusion you meant ecb(arc4) not ecb(aes) here right?
Yes, I do. Not sure how I got aes in there. I did not mix it up in the patch :)
Quoted 2 lines
-Adrian
Sebastian
Re: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
In reply to Sebastian Andrzej Siewiorcrypto/arc4.c line 59
Quoted 4 lines · * Sebastian Andrzej Siewior | 2010-02-12 09:42:28 [+0100]:
+static void
arc4_ivsetup(struct arc4_ctx *ctx, u8 *iv){crypto/arc4.c +1
-1-struct arc4_ctx *ctx =crypto_tfm_ctx (tfm); +if (unlikely (!ctx->new_key))
That should be likely(). Do you want me resend the whole thing? Haven't noticed anything else :)
crypto/arc4.c line 63
Quoted 4 lines
crypto/arc4.c +4
-0+return ; +memcpy (iv, &ctx->iv,sizeof (ctx->iv)); + ctx->new_key =0 ; +}
Sebastian
Herbert Xu
maintainerRe: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
Quoted 9 lines · Sebastian Andrzej Siewior
crypto/arc4.c line 59
Quoted 4 lines · * Sebastian Andrzej Siewior | 2010-02-12 09:42:28 [+0100]:
+static void
arc4_ivsetup(struct arc4_ctx *ctx, u8 *iv){crypto/arc4.c +1
-1-struct arc4_ctx *ctx =crypto_tfm_ctx (tfm); +if (unlikely (!ctx->new_key))That should be
likely(). Do you want me resend the whole thing? Haven't noticed anything else :)
How about we just remove it? It's not on a hot path anyway.
I can do this when integrating the patch so you don't have to resend.
Thanks,
Signature · 4 lines
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txtRe: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
Quoted 1 line · * Herbert Xu | 2010-02-15 08:10:08 [+0800]:
How about we just remove it? It's not on a hot path anyway.
Sure.
Quoted 2 lines
I can do this when integrating the patch so you don't have to resend.
Okay, thanks.
Quoted 1 line
Thanks,
Sebastian
Herbert Xu
maintainerRe: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
In reply to Sebastian Andrzej Siewiorcrypto/arc4.c line 62
Quoted 9 lines · Sebastian Andrzej Siewior
crypto/arc4.c +5
-2-static void arc4_crypt (struct crypto_tfm *tfm,u8 *out,const u8 *in) +static void arc4_ivsetup (struct arc4_ctx *ctx,u8 *iv) { -struct arc4_ctx *ctx =crypto_tfm_ctx (tfm); +if (unlikely (!ctx->new_key)) +return ; +memcpy (iv, &ctx->iv,sizeof (ctx->iv)); + ctx->new_key =0 ;
Sorry, but this doesn't work.
A ctx is supposed to be reentrant. That is, while one thread is working away with a given ctx I should be able to use that same ctx in a different thread without them clobbering each other.
So that means (in general) you must not modify the ctx in any function other than setkey.
This also brings up the bigger question of how we transition to this new arc4. I don't think we need to maintain exactly the same behaviour as the existing ecb(arc4).
So what we could do is simply add a new blkcipher arc4, alongside the existing cipher arc4. Then we can convert the existing users across, and finally remove the old arc4.
Cheers,
Signature · 4 lines
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txtRe: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
Quoted 13 lines · * Herbert Xu | 2010-02-16 20:51:25 [+0800]:
crypto/arc4.c line 62
Quoted 9 lines · Sebastian Andrzej Siewior
crypto/arc4.c +5
-2-static void arc4_crypt (struct crypto_tfm *tfm,u8 *out,const u8 *in) +static void arc4_ivsetup (struct arc4_ctx *ctx,u8 *iv) { -struct arc4_ctx *ctx =crypto_tfm_ctx (tfm); +if (unlikely (!ctx->new_key)) +return ; +memcpy (iv, &ctx->iv,sizeof (ctx->iv)); + ctx->new_key =0 ;Sorry, but this doesn't work.
Quoted 4 lines
A ctx is supposed to be reentrant. That is, while one thread is working away with a given ctx I should be able to use that same ctx in a different thread without them clobbering each other.
I also destroy the user supplied IV. You don't care about that? :) So I have to know that someone called setkey() on this ctx but I can't leave hints.
salsa also does not stick to plan here. ctx->input[6-9] is initialized in encrypt() path. So two threads sharing a ctx are going to clobber their state.
What about a new api for the stream cipher? We would merge the ctx part and the iv into one handle. So the user would call setup_iv() instead of setkey(). The difference would be that I can access the iv from within setkey(). And the algorithm can fully express himself since he is no longer trapped in the wrong body :)
Quoted 2 lines
So that means (in general) you must not modify the ctx in any function other than setkey.
That is hard because I have a new state after encryption which I am only allowed to save in the iv. And the new state may be reset in setkey() where I can't touch the iv.
salsa does not keep/update its state. So the input[6-9] problem could be fixed. Who/where is it used anyway? I can't see any user besides the possible once (i.e. dm-crypt/ipsec).
Quoted 7 lines
This also brings up the bigger question of how we transition to this new arc4. I don't think we need to maintain exactly the same behaviour as the existing ecb(arc4).
So what we could do is simply add a new blkcipher arc4, alongside the existing cipher arc4. Then we can convert the existing users across, and finally remove the old arc4.
This has worked out before, lets stick to this :)
Quoted 1 line
Cheers,
Sebastian
Herbert Xu
maintainerRe: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
Quoted 4 lines · Sebastian Andrzej Siewior
I also destroy the user supplied IV. You don't care about that? :) So I have to know that someone called
setkey()on this ctx but I can't leave hints.
How about this? You extend the IV by one more byte, and use that byte as a boolean flag to indicate whether the IV is valid. All users that cannot supply their own IVs can then set the IV to zero.
When you see the zero flag in the IV, you reinitialise the IV per the key.
Quoted 3 lines
salsa also does not stick to plan here. ctx->input[6-9] is initialized in
encrypt()path. So two threads sharing a ctx are going to clobber their state.
Salsa should also be fixed.
Quoted 5 lines
What about a new api for the stream cipher? We would merge the ctx part and the iv into one handle. So the user would call
setup_iv()instead ofsetkey(). The difference would be that I can access the iv from withinsetkey(). And the algorithm can fully express himself since he is no longer trapped in the wrong body :)
There is some merit in that, but as the current API can be made to do the same thing (see above) I'm not convinced that this is worth the cost for the moment.
Cheers,
Signature · 4 lines
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txtHerbert Xu
maintainerRe: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
Quoted 7 lines · Herbert Xu
How about this? You extend the IV by one more byte, and use that byte as a boolean flag to indicate whether the IV is valid. All users that cannot supply their own IVs can then set the IV to zero.
When you see the zero flag in the IV, you reinitialise the IV per the key.
In fact for arc4 we could just drop the key altogether since it plays no part after setting the initial state.
Quoted 5 lines
Quoted 3 lines
salsa also does not stick to plan here. ctx->input[6-9] is initialized in
encrypt()path. So two threads sharing a ctx are going to clobber their state.Salsa should also be fixed.
For Salsa on the other hand the key is rather useful since all we need is a two-byte IV that's just a sequence number.
Cheers,
Signature · 4 lines
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txtRe: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
Quoted 4 lines · * Herbert Xu | 2010-02-22 08:52:17 [+0800]:
Quoted 3 lines · Herbert Xu
How about this? You extend the IV by one more byte, and use that byte as a boolean flag to indicate whether the IV is valid. All
So I trick the crypto api to allocate more bytes than ->ivsize says.
Quoted 1 line
Quoted 1 line
users that cannot supply their own IVs can then set the IV to zero.
which works fine with kzalloc()
Quoted 2 lines
Quoted 2 lines
When you see the zero flag in the IV, you reinitialise the IV per the key.
Okay. When we have to re-key and the user calls setkey() without re-allocating thr cipher then I would not notice this. So I need a counter. And all this will make it work but I still think it is fishy. Plus we waste 258bytes.
Quoted 2 lines
In fact for arc4 we could just drop the key altogether since it plays no part after setting the initial state.
Since I'm not allowed to kfree() the ctx in encrypt() are you proposing tfm->setup_iv(iv, key)?
Quoted 5 lines
Quoted 5 lines
Quoted 3 lines
salsa also does not stick to plan here. ctx->input[6-9] is initialized in
encrypt()path. So two threads sharing a ctx are going to clobber their state.Salsa should also be fixed.
I saw that comming. And I complaind back then that the assembly code was not pretty enough... and removing the assembly is probably not option :)
Quoted 2 lines
For Salsa on the other hand the key is rather useful since all we need is a two-byte IV that's just a sequence number.
No it's 8 bytes. Berstein's U8TO32_LITTLE() is actually a cpu_to_be32(). Not sure if he knows it :)
However I'm not sure where you going with this. salsa is fine besides the clobber thing, isn't it?
Quoted 1 line
Cheers,
Sebastian
Herbert Xu
maintainerRe: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
Quoted 7 lines · Sebastian Andrzej Siewior
Quoted 4 lines · * Herbert Xu | 2010-02-22 08:52:17 [+0800]:
Quoted 3 lines · Herbert Xu
How about this? You extend the IV by one more byte, and use that byte as a boolean flag to indicate whether the IV is valid. All
So I trick the crypto api to allocate more bytes than ->ivsize says.
No tricks needed, just add the flag to the struct.
struct arc4_iv {
u8 S[256];
u8 x, y;
u8 valid;
};Quoted 4 lines
Okay. When we have to re-key and the user calls
setkey()without re-allocating thr cipher then I would not notice this. So I need a counter. And all this will make it work but I still think it is fishy. Plus we waste 258bytes.
No you don't need to refresh the IV when the key changes. The key should only be consulted when the valid flag in the IV is zero.
You need the 258 + flag bytes because that's just the amount of state carried between the encrypt/decrypt operation. So it isn't really wasted.
If you can find a way that allows arc4 to be used by multiple threads at the same time while storing less than 258 bytes in each thread, please let me know :)
Quoted 4 lines
Quoted 2 lines
In fact for arc4 we could just drop the key altogether since it plays no part after setting the initial state.
Since I'm not allowed to
kfree()the ctx inencrypt()are you proposingtfm->setup_iv(iv, key)?
No, what you could do is structure the IV differently based on the flag:
struct arc4_iv {
union {
struct key {
u8 key[256];
u16 keylen;
};
struct iv {
u8 S[256];
u8 x, y;
};
};
u8 type;
};This relies on the fact that we never use more than 256 bytes in the key so limiting its length is OK.
Quoted 3 lines
Quoted 1 line
Quoted 1 line
Salsa should also be fixed.
I saw that comming. And I complaind back then that the assembly code was not pretty enough... and removing the assembly is probably not option :)
Well if nobody steps in to fix the assembly then removing it is the only option.
Quoted 4 lines
Quoted 2 lines
For Salsa on the other hand the key is rather useful since all we need is a two-byte IV that's just a sequence number.
No it's 8 bytes. Berstein's
U8TO32_LITTLE()is actually acpu_to_be32(). Not sure if he knows it :)
Right.
Quoted 2 lines
However I'm not sure where you going with this. salsa is fine besides the clobber thing, isn't it?
I don't know of any other problems.
Basically salsa should look pretty much like CTR from the outside when it's fixed.
Cheers,
Signature · 4 lines
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txtRe: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
Quoted 3 lines · * Herbert Xu | 2010-02-23 08:32:39 [+0800]:
If you can find a way that allows arc4 to be used by multiple threads at the same time while storing less than 258 bytes in each thread, please let me know :)
:)
Quoted 17 lines
No, what you could do is structure the IV differently based on the flag:
struct arc4_iv { union { struct key { u8 key[256]; u16 keylen; }; struct iv { u8 S[256]; u8 x, y; }; }; u8 type; };
Quoted 2 lines
This relies on the fact that we never use more than 256 bytes in the key so limiting its length is OK.
Okay. So so are we talking about something like that below then? This is untested and I break other users bexcept lib80211_crypt_tkip.
the state has been moved from ctx into iv. That way encrypt()/decrypt() can deliver the same result for a given IV. If the IV is supplied as a plain key then it wil be converted into a different internal state. The name is now arc4.
--- crypto/Kconfig | 2 +- crypto/arc4.c | 131 +++++++++++++++++++++++----------- crypto/testmgr.h | 3 +- drivers/net/Kconfig | 1 - drivers/net/wireless/hostap/Kconfig | 2 - drivers/net/wireless/ipw2x00/Kconfig | 2 - include/crypto/arc4.h | 26 +++++++ net/mac80211/Kconfig | 1 - net/wireless/lib80211_crypt_tkip.c | 10 ++- 9 files changed, 123 insertions(+), 55 deletions(-) create mode 100644 include/crypto/arc4.h
crypto/Kconfig +1 -1
@@ -561,7 +561,7 @@ config CRYPTO_ANUBIS config CRYPTO_ARC4 tristate "ARC4 cipher algorithm" - select CRYPTO_ALGAPI + select CRYPTO_BLKCIPHER help ARC4 cipher algorithm.
crypto/arc4.c +88 -43
175 lines not on this page · raw message
crypto/testmgr.h +2 -1
@@ -24,7 +24,8 @@#define MAX_TAP 8#define MAX_KEYLEN 56 -#define MAX_IVLEN 32 +/* sizeof arc4_iv */ +#define MAX_IVLEN 260struct hash_testvec {/* only used with keyed hash algorithms */
drivers/net/Kconfig +0 -1
@@ -3076,7 +3076,6 @@ config PPP_MPPE select CRYPTO select CRYPTO_SHA1 select CRYPTO_ARC4 - select CRYPTO_ECB ---help--- Support for the MPPE Encryption protocol, as employed by the Microsoft Point-to-Point Tunneling Protocol.
drivers/net/wireless/hostap/Kconfig +0 -2
@@ -5,10 +5,8 @@ config HOSTAP select WEXT_PRIV select CRYPTO select CRYPTO_ARC4 - select CRYPTO_ECB select CRYPTO_AES select CRYPTO_MICHAEL_MIC - select CRYPTO_ECB select CRC32 select LIB80211 select LIB80211_CRYPT_WEP
drivers/net/wireless/ipw2x00/Kconfig +0 -2
@@ -159,10 +159,8 @@ config LIBIPW select WEXT_SPY select CRYPTO select CRYPTO_ARC4 - select CRYPTO_ECB select CRYPTO_AES select CRYPTO_MICHAEL_MIC - select CRYPTO_ECB select CRC32 select LIB80211 select LIB80211_CRYPT_WEP
include/crypto/arc4.h new file +26 -0
@@ -0,0 +1,26 @@ +#ifndef __CRYPTO_ARC4_H__ +#define __CRYPTO_ARC4_H__ + +struct arc4_iv { +union { +struct arc4_key { +u8 key[256 ]; +u16 key_len; + } key; +struct arc4_riv { +u8 S[256 ]; +u8 x, y; + } iv; + }; +#define ARC4_TYPE_KEY 0 +#define ARC4_TYPE_IV 1 +u8 type; +}; + +static inline void arc4_setup_iv (struct arc4_iv *iv,const u8 *key,u32 len) +{ +memcpy (iv->key.key, key, len); + iv->key.key_len = len; + iv->type =ARC4_TYPE_KEY ; +} +#endif
net/mac80211/Kconfig +0 -1
@@ -2,7 +2,6 @@ config MAC80211 tristate "Generic IEEE 802.11 Networking Stack (mac80211)" depends on CFG80211 select CRYPTO - select CRYPTO_ECB select CRYPTO_ARC4 select CRYPTO_AES select CRC32
net/wireless/lib80211_crypt_tkip.c +6 -4
@@ -94,7 +94,7 @@ static void *lib80211_tkip_init(int key_idx) priv->key_idx = key_idx; - priv->tx_tfm_arc4 =crypto_alloc_blkcipher ("ecb(arc4)" ,0 , + priv->tx_tfm_arc4 =crypto_alloc_blkcipher ("arc4" ,0 ,CRYPTO_ALG_ASYNC );if (IS_ERR (priv->tx_tfm_arc4)) { printk(KERN_DEBUG "lib80211_crypt_tkip: could not allocate "
@@ -112,7 +112,7 @@ static void *lib80211_tkip_init(int key_idx) goto fail; } - priv->rx_tfm_arc4 =crypto_alloc_blkcipher ("ecb(arc4)" ,0 , + priv->rx_tfm_arc4 =crypto_alloc_blkcipher ("arc4" ,0 ,CRYPTO_ALG_ASYNC );if (IS_ERR (priv->rx_tfm_arc4)) { printk(KERN_DEBUG "lib80211_crypt_tkip: could not allocate "
@@ -360,6 +360,7 @@ static int lib80211_tkip_encrypt(struct sk_buff *skb, int hdr_len, void *priv) {struct lib80211_tkip_data *tkey = priv;struct blkcipher_desc desc = { .tfm = tkey->tx_tfm_arc4 }; +struct arc4_iv *iv =crypto_blkcipher_crt (tkey->tx_tfm_arc4)->iv;int len;u8 rc4key[16 ], *pos, *icv;u32 crc;
@@ -392,7 +393,7 @@ static int lib80211_tkip_encrypt(struct sk_buff *skb, int hdr_len, void *priv) icv[2 ] = crc >>16 ; icv[3 ] = crc >>24 ; -crypto_blkcipher_setkey (tkey->tx_tfm_arc4, rc4key,16 ); +arc4_setup_iv (iv, rc4key,16 );sg_init_one (&sg, pos, len +4 );return crypto_blkcipher_encrypt (&desc, &sg, &sg, len +4 ); }
@@ -414,6 +415,7 @@ static int lib80211_tkip_decrypt(struct sk_buff *skb, int hdr_len, void *priv) {struct lib80211_tkip_data *tkey = priv;struct blkcipher_desc desc = { .tfm = tkey->rx_tfm_arc4 }; +struct arc4_iv *iv =crypto_blkcipher_crt (tkey->rx_tfm_arc4)->iv;u8 rc4key[16 ];u8 keyidx, *pos;u32 iv32;
@@ -485,7 +487,7 @@ static int lib80211_tkip_decrypt(struct sk_buff *skb, int hdr_len, void *priv) plen = skb->len - hdr_len -12 ; -crypto_blkcipher_setkey (tkey->rx_tfm_arc4, rc4key,16 ); +arc4_setup_iv (iv, rc4key,16 );sg_init_one (&sg, pos, plen +4 );if (crypto_blkcipher_decrypt (&desc, &sg, &sg, plen +4 )) {if (net_ratelimit ()) {
Signature · 1 line
1.6.6
Herbert Xu
maintainerRe: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
Quoted 2 lines · Sebastian Andrzej Siewior
Okay. So so are we talking about something like that below then? This is
Pretty much.
Quoted 1 line
untested and I break other users bexcept
lib80211_crypt_tkip.
For the sake of compatibility please do this as a 3-step dance. First add a new arc4 blkcipher, then convert the users, and finally delete the old arc4.
crypto/arc4.c line 31
Quoted 10 lines
static int arc4_set_key(struct
crypto_tfm*tfm, const u8*in_key,crypto/arc4.c +6
-1unsigned int key_len) { -struct arc4_ctx *ctx =crypto_tfm_ctx (tfm); +/* +* ARC4 is special: The user should supply an IV as struct arc4_iv and +* fill either the key or the iv. +*/ +return -EOPNOTSUPP ; +}
You should return 0 here. You should also set the min/max key size to zero. The new arc4 no longer has a key as far as the API is concerned.
Cheers,
Signature · 4 lines
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txtA message that is not in the archive (sent off-list or not yet received): [email protected]
Herbert Xu
maintainerRe: [PATCH] crypto/arc4: convert this stream cipher into a block cipher
Quoted 10 lines · Mikulas Patocka
Quoted 4 lines
So what we could do is simply add a new blkcipher arc4, alongside the existing cipher arc4. Then we can convert the existing users across, and finally remove the old arc4.
arc4 can't be used as a block cipher --- see this paper
http://www.wisdom.weizmann.ac.il/~itsik/RC4/Papers/Rc4_ksa.ps , it says that initialization vectors on RC4 are unreliable, if you use (unknown key concatenated with known IV) or (known IV concatenated with unknown key) as a RC4 key, the RC4 state can be exposed and the cipher is broken.
What we call a blkcipher is not really a block cipher. In fact, what we call "cipher" is really a block cipher. So we're actually changing arc4 so that it doesn't get used as a block cipher, i.e., you will no longer be able to say "cbc(arc4)" or some such.
I know it's confusing and perhaps one day we will rename blkcipher to skcipher and cipher to blkcipher.
Cheers,
Signature · 4 lines
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txtAlasdair G Kergon
maintainerQuoted 2 lines · Milan Broz
This should be solved inside cryptoAPI and not blacklist it in dm-crypt, see that thread
Agreed. I'm not going to apply a dm patch that maintains a hard-coded "broken" list.
Alasdair
Mikulas Patocka
maintainerBTW. I created a script that tests all possible ciphers, keysizes, chaining modes and iv modes for dm-crypt. arc4 is the only one that fails. You can add it your regression testsuite if you want.
Mikulas
#!/bin/sh
dmsetup remove cr0
set -e
cipher="$@"
cryptsetup -d key $cipher create cr0 /dev/ram0
mkfs.ext3 -b 1024 /dev/mapper/cr0
fsck.ext2 -fn /dev/mapper/cr0
dmsetup remove cr0
cryptsetup -d key $cipher create cr0 /dev/ram0
fsck.ext2 -fn /dev/mapper/cr0
dmsetup remove cr0
#!/bin/sh
set -e
Quoted 1 line
log
# arc4 - doesn't work for cipher in aes anubis blowfish camellia cast5 cast6 des des3_ede fcrypt khazad seed serpent tnepres tea xtea xeta; do if [ -n "$1" -a "$cipher" != "$1" ]; then continue fi if [ $cipher = aes ]; then keysizes="128 192 256" blocksize=128 elif [ $cipher = anubis ]; then keysizes="`seq 128 32 320`" blocksize=128 elif [ $cipher = arc4 ]; then keysizes="`seq 8 256 2048` 2048" blocksize=1 elif [ $cipher = blowfish ]; then keysizes="`seq 32 8 448`" blocksize=64 elif [ $cipher = camellia ]; then keysizes="128 192 256" blocksize=128 elif [ $cipher = cast5 ]; then keysizes="`seq 40 8 128`" blocksize=64 elif [ $cipher = cast6 ]; then keysizes="`seq 128 32 256`" blocksize=128 elif [ $cipher = des ]; then keysizes=64 blocksize=64 elif [ $cipher = des3_ede ]; then keysizes=192 blocksize=64 elif [ $cipher = fcrypt ]; then keysizes=64 blocksize=64 elif [ $cipher = khazad ]; then keysizes=128 blocksize=64 elif [ $cipher = seed ]; then keysizes=128 blocksize=128 elif [ $cipher = serpent -o $cipher = tnepres ]; then keysizes="`seq 0 8 256`" blocksize=128 elif [ $cipher = tea -o $cipher = xtea -o $cipher = xeta ]; then keysizes=128 blocksize=64 elif [ $cipher = twofish ]; then keysizes="128 192 256" blocksize=128 else echo UNKNOWN CIPHER exit 1 fi for keysize in $keysizes; do # if ! echo $keysizes | grep -qw $keysize$; then continue; fi for chaining in cbc ctr ecb lrw pcbc xts; do chaining_keysize=$keysize if [ $chaining = xts ]; then chaining_keysize=`expr $chaining_keysize \* 2 | cat`; if [ $blocksize != 128 ]; then continue; fi fi if [ $chaining = lrw ]; then chaining_keysize=`expr $chaining_keysize + $blocksize | cat` if [ $blocksize != 128 ]; then continue; fi fi for iv in null plain benbi essiv:md4 essiv:md5 essiv:michael_mic essiv:rmd128 essiv:rmd160 essiv:rmd256 essiv:rmd320 essiv:sha1 essiv:sha224 essiv:sha256 essiv:sha384 essiv:sha512 essiv:tgr128 essiv:tgr160 essiv:tgr192 essiv:wp256 essiv:wp384 essiv:wp512; do if [ $chaining = ecb ] && echo "$iv" | grep -q ^essiv; then continue; fi if [ $iv = essiv:md4 ] && ! echo $keysizes | grep -qw 128; then continue; fi if [ $iv = essiv:md5 ] && ! echo $keysizes | grep -qw 128; then continue; fi if [ $iv = essiv:michael_mic ] && ! echo $keysizes | grep -qw 64; then continue; fi if [ $iv = essiv:rmd128 ] && ! echo $keysizes | grep -qw 128; then continue; fi if [ $iv = essiv:rmd160 ] && ! echo $keysizes | grep -qw 160; then continue; fi if [ $iv = essiv:rmd256 ] && ! echo $keysizes | grep -qw 256; then continue; fi if [ $iv = essiv:rmd320 ] && ! echo $keysizes | grep -qw 320; then continue; fi if [ $iv = essiv:sha1 ] && ! echo $keysizes | grep -qw 160; then continue; fi if [ $iv = essiv:sha224 ] && ! echo $keysizes | grep -qw 224; then continue; fi if [ $iv = essiv:sha256 ] && ! echo $keysizes | grep -qw 256; then continue; fi if [ $iv = essiv:sha384 ] && ! echo $keysizes | grep -qw 384; then continue; fi if [ $iv = essiv:sha512 ] && ! echo $keysizes | grep -qw 512; then continue; fi if [ $iv = essiv:tgr128 ] && ! echo $keysizes | grep -qw 128; then continue; fi if [ $iv = essiv:tgr160 ] && ! echo $keysizes | grep -qw 160; then continue; fi if [ $iv = essiv:tgr192 ] && ! echo $keysizes | grep -qw 192; then continue; fi if [ $iv = essiv:wp256 ] && ! echo $keysizes | grep -qw 256; then continue; fi if [ $iv = essiv:wp384 ] && ! echo $keysizes | grep -qw 384; then continue; fi if [ $iv = essiv:wp512 ] && ! echo $keysizes | grep -qw 512; then continue; fi echo TESTING: CIPHER $cipher, KEYSIZE $chaining_keysize, CHAINING $chaining, IV $iv | tee -a log ./test-cipher -s $chaining_keysize -c $cipher-$chaining-$iv done done done done echo PASSED