All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
To: linux-crypto@vger.kernel.org
Cc: herbert@gondor.apana.org.au, Ard Biesheuvel <ard.biesheuvel@linaro.org>
Subject: [PATCH v5 01/30] crypto: des/3des_ede - add new helpers to verify keys
Date: Thu, 15 Aug 2019 12:00:43 +0300	[thread overview]
Message-ID: <20190815090112.9377-2-ard.biesheuvel@linaro.org> (raw)
In-Reply-To: <20190815090112.9377-1-ard.biesheuvel@linaro.org>

The recently added helper routine to perform key strength validation
of triple DES keys is slightly inadequate, since it comes in two versions,
neither of which are highly useful for anything other than skciphers (and
many drivers still use the older blkcipher interfaces).

So let's add a new helper and, considering that this is a helper function
that is only intended to be used by crypto code itself, put it in a new
des.h header under crypto/internal.

While at it, implement a similar helper for single DES, so that we can
start replacing the pattern of calling des_ekey() into a temp buffer
that occurs in many drivers in drivers/crypto.

Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
---
 crypto/des_generic.c          |  13 --
 include/crypto/internal/des.h | 141 ++++++++++++++++++++
 2 files changed, 141 insertions(+), 13 deletions(-)

diff --git a/crypto/des_generic.c b/crypto/des_generic.c
index dc085514408a..c4d8ecda4ddf 100644
--- a/crypto/des_generic.c
+++ b/crypto/des_generic.c
@@ -841,19 +841,6 @@ static void des_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src)
 	d[1] = cpu_to_le32(L);
 }
 
-/*
- * RFC2451:
- *
- *   For DES-EDE3, there is no known need to reject weak or
- *   complementation keys.  Any weakness is obviated by the use of
- *   multiple keys.
- *
- *   However, if the first two or last two independent 64-bit keys are
- *   equal (k1 == k2 or k2 == k3), then the DES3 operation is simply the
- *   same as DES.  Implementers MUST reject keys that exhibit this
- *   property.
- *
- */
 int __des3_ede_setkey(u32 *expkey, u32 *flags, const u8 *key,
 		      unsigned int keylen)
 {
diff --git a/include/crypto/internal/des.h b/include/crypto/internal/des.h
new file mode 100644
index 000000000000..f5d2e696522e
--- /dev/null
+++ b/include/crypto/internal/des.h
@@ -0,0 +1,141 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * DES & Triple DES EDE key verification helpers
+ */
+
+#ifndef __CRYPTO_INTERNAL_DES_H
+#define __CRYPTO_INTERNAL_DES_H
+
+#include <linux/crypto.h>
+#include <linux/fips.h>
+#include <crypto/des.h>
+#include <crypto/aead.h>
+#include <crypto/skcipher.h>
+
+/**
+ * crypto_des_verify_key - Check whether a DES key is weak
+ * @tfm: the crypto algo
+ * @key: the key buffer
+ *
+ * Returns -EINVAL if the key is weak and the crypto TFM does not permit weak
+ * keys. Otherwise, 0 is returned.
+ *
+ * It is the job of the caller to ensure that the size of the key equals
+ * DES_KEY_SIZE.
+ */
+static inline int crypto_des_verify_key(struct crypto_tfm *tfm, const u8 *key)
+{
+	u32 tmp[DES_EXPKEY_WORDS];
+	int err = 0;
+
+	if (!(crypto_tfm_get_flags(tfm) & CRYPTO_TFM_REQ_FORBID_WEAK_KEYS))
+		return 0;
+
+	if (!des_ekey(tmp, key)) {
+		crypto_tfm_set_flags(tfm, CRYPTO_TFM_RES_WEAK_KEY);
+		err = -EINVAL;
+	}
+
+	memzero_explicit(tmp, sizeof(tmp));
+	return err;
+}
+
+/*
+ * RFC2451:
+ *
+ *   For DES-EDE3, there is no known need to reject weak or
+ *   complementation keys.  Any weakness is obviated by the use of
+ *   multiple keys.
+ *
+ *   However, if the first two or last two independent 64-bit keys are
+ *   equal (k1 == k2 or k2 == k3), then the DES3 operation is simply the
+ *   same as DES.  Implementers MUST reject keys that exhibit this
+ *   property.
+ *
+ */
+
+/**
+ * crypto_des3_ede_verify_key - Check whether a DES3-EDE key is weak
+ * @tfm: the crypto algo
+ * @key: the key buffer
+ *
+ * Returns -EINVAL if the key is weak and the crypto TFM does not permit weak
+ * keys or when running in FIPS mode. Otherwise, 0 is returned. Note that some
+ * keys are rejected in FIPS mode even if weak keys are permitted by the TFM
+ * flags.
+ *
+ * It is the job of the caller to ensure that the size of the key equals
+ * DES3_EDE_KEY_SIZE.
+ */
+static inline int crypto_des3_ede_verify_key(struct crypto_tfm *tfm,
+					     const u8 *key)
+{
+	int err = -EINVAL;
+	u32 K[6];
+
+	memcpy(K, key, DES3_EDE_KEY_SIZE);
+
+	if ((!((K[0] ^ K[2]) | (K[1] ^ K[3])) ||
+	     !((K[2] ^ K[4]) | (K[3] ^ K[5]))) &&
+	    (fips_enabled || (crypto_tfm_get_flags(tfm) &
+		              CRYPTO_TFM_REQ_FORBID_WEAK_KEYS)))
+		goto bad;
+
+	if ((!((K[0] ^ K[4]) | (K[1] ^ K[5]))) && fips_enabled)
+		goto bad;
+
+	err = 0;
+out:
+	memzero_explicit(K, DES3_EDE_KEY_SIZE);
+	return err;
+
+bad:
+	crypto_tfm_set_flags(tfm, CRYPTO_TFM_RES_WEAK_KEY);
+	goto out;
+}
+
+static inline int verify_skcipher_des_key(struct crypto_skcipher *tfm,
+					  const u8 *key)
+{
+	return crypto_des_verify_key(crypto_skcipher_tfm(tfm), key);
+}
+
+static inline int verify_skcipher_des3_key(struct crypto_skcipher *tfm,
+					   const u8 *key)
+{
+	return crypto_des3_ede_verify_key(crypto_skcipher_tfm(tfm), key);
+}
+
+static inline int verify_ablkcipher_des_key(struct crypto_ablkcipher *tfm,
+					    const u8 *key)
+{
+	return crypto_des_verify_key(crypto_ablkcipher_tfm(tfm), key);
+}
+
+static inline int verify_ablkcipher_des3_key(struct crypto_ablkcipher *tfm,
+					     const u8 *key)
+{
+	return crypto_des3_ede_verify_key(crypto_ablkcipher_tfm(tfm), key);
+}
+
+static inline int verify_aead_des_key(struct crypto_aead *tfm, const u8 *key,
+				      int keylen)
+{
+	if (keylen != DES_KEY_SIZE) {
+		crypto_aead_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
+		return -EINVAL;
+	}
+	return crypto_des_verify_key(crypto_aead_tfm(tfm), key);
+}
+
+static inline int verify_aead_des3_key(struct crypto_aead *tfm, const u8 *key,
+				       int keylen)
+{
+	if (keylen != DES3_EDE_KEY_SIZE) {
+		crypto_aead_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
+		return -EINVAL;
+	}
+	return crypto_des3_ede_verify_key(crypto_aead_tfm(tfm), key);
+}
+
+#endif /* __CRYPTO_INTERNAL_DES_H */
-- 
2.17.1


  reply	other threads:[~2019-08-15  9:01 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-15  9:00 [PATCH v5 00/30] crypto: DES/3DES cleanup Ard Biesheuvel
2019-08-15  9:00 ` Ard Biesheuvel [this message]
2019-08-15  9:00 ` [PATCH v5 02/30] crypto: s390/des - switch to new verification routines Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 03/30] crypto: sparc/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 04/30] crypto: atmel/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 05/30] crypto: bcm/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 06/30] crypto: caam/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 07/30] crypto: cpt/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 08/30] crypto: nitrox/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 09/30] crypto: ccp/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 10/30] crypto: ccree/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 11/30] crypto: hifn/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 12/30] crypto: hisilicon/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 13/30] crypto: safexcel/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 14/30] crypto: ixp4xx/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 15/30] crypto: cesa/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 16/30] crypto: n2/des " Ard Biesheuvel
2019-08-15  9:00 ` [PATCH v5 17/30] crypto: omap/des " Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 18/30] crypto: picoxcell/des " Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 19/30] crypto: qce/des " Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 20/30] crypto: rk3288/des " Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 21/30] crypto: stm32/des " Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 22/30] crypto: sun4i/des " Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 23/30] crypto: talitos/des " Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 24/30] crypto: ux500/des " Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 25/30] crypto: 3des - move verification out of exported routine Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 26/30] crypto: des - remove unused function Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 27/30] crypto: des - split off DES library from generic DES cipher driver Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 28/30] crypto: x86/des - switch to library interface Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 29/30] crypto: des - remove now unused __des3_ede_setkey() Ard Biesheuvel
2019-08-15  9:01 ` [PATCH v5 30/30] fs: cifs: move from the crypto cipher API to the new DES library interface Ard Biesheuvel
2019-08-22  5:56 ` [PATCH v5 00/30] crypto: DES/3DES cleanup Herbert Xu

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=20190815090112.9377-2-ard.biesheuvel@linaro.org \
    --to=ard.biesheuvel@linaro.org \
    --cc=herbert@gondor.apana.org.au \
    --cc=linux-crypto@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.