linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Artem Bityutskiy <dedekind1@gmail.com>
To: Joel Reardon <joel@clambassador.com>
Cc: linux-mtd@lists.infradead.org, linux-kernel@vger.kernel.org,
	linux-fsdevel@vger.kernel.org
Subject: Re: [patch] UBIFS: Add cryptographic functionality when a key is passed to the compress / decompress functions
Date: Mon, 02 Apr 2012 17:36:23 +0300	[thread overview]
Message-ID: <1333377383.22146.14.camel@sauron.fi.intel.com> (raw)
In-Reply-To: <alpine.DEB.2.00.1203291639200.912@eristoteles.iwoars.net>

[-- Attachment #1: Type: text/plain, Size: 3361 bytes --]

On Thu, 2012-03-29 at 16:39 +0200, Joel Reardon wrote:
>  /* Fake description object for the "none" compressor */
>  static struct ubifs_compressor none_compr = {
>  	.compr_type = UBIFS_COMPR_NONE,
> @@ -75,6 +78,55 @@ static struct ubifs_compressor zlib_compr = {
>  struct ubifs_compressor *ubifs_compressors[UBIFS_COMPR_TYPES_CNT];
> 
>  /**
> + * ubifs_aes_crypt - encrypt / decrypt data.
> + * @str: data to crypt
> + * @len: length of the data
> + * @crypto_key: the cryptographic key to use to crypt the data
> + * @crypto_key_len: the length of the crypto_key
> + * @iv: the initialization vector to use
> + * @ivlen: the length of the initialization vector
> + *
> + * This function applies aes encryption to the data. It is done in counter
> + * mode, which means that encryption and decryption are the same operation,
> + * i.e., it XORs the same generated bitstream, so it can be used both for
> + * encryption / decryption. The operation is done in-place, so str mutates.
> + */
> +int ubifs_aes_crypt(void *str, int len, u8 *crypto_key,
> +		    int crypto_key_len, u8 *iv, int ivlen)

You support only one length - please, kill ivlen parameter.

Also, should ubifs_aes_crypt be static? I do not see any users outside
of compress.c. In this case remove the "ubifs_" prefix. But a
non-written convention, in UBIFS we _tend_ to prefix only non-static
functions with "ubifs_" and avoid having it for static functions.

> +{
> +	struct crypto_blkcipher *tfm;
> +	struct blkcipher_desc desc;
> +	struct scatterlist sg;
> +	int err = 0;
> +
> +	tfm = crypto_alloc_blkcipher(UBIFS_CRYPTO_ALGORITHM, 0, 0);
> +

Unnecessary empty line.

> +	if (IS_ERR(tfm)) {
> +		ubifs_err("failed to load transform for aes: %ld",
> +			  PTR_ERR(tfm));
> +		return err;
> +	}
> +
> +	err = crypto_blkcipher_setkey(tfm, crypto_key, crypto_key_len);
> +	desc.tfm = tfm;
> +	desc.flags = 0;
> +	if (err) {
> +		ubifs_err("crypto_blkcipher_setkey() failed  flags=%#x",
> +			  crypto_blkcipher_get_flags(tfm));
> +		return err;
> +	}
> +	memset(&sg, 0, sizeof(struct scatterlist));
> +

Empty lines mean grouping, and I think this memeset should be grouped
with sg_set_buf instead.


>  no_compr:
>  	memcpy(out_buf, in_buf, in_len);
>  	*out_len = in_len;
>  	*compr_type = UBIFS_COMPR_NONE;
> +	goto encrypt;
> +
> +encrypt:

I guess the above goto is redundant?

> +	if (crypto_key) {
> +		u8 iv[UBIFS_CRYPTO_KEYSIZE];
> +
> +		memset(iv, 0, UBIFS_CRYPTO_KEYSIZE);
> +		ubifs_aes_crypt(out_buf, *out_len, crypto_key,
> +				UBIFS_CRYPTO_KEYSIZE, iv, UBIFS_CRYPTO_KEYSIZE);
> +	}
>  }
> 
>  /**
> @@ -149,7 +211,7 @@ no_compr:
>   * The length of the uncompressed data is returned in @out_len. This functions
>   * returns %0 on success or a negative error code on failure.
>   */
> -int ubifs_decompress(const void *in_buf, int in_len, void *out_buf,
> +int ubifs_decompress(void *in_buf, int in_len, void *out_buf,
>  		     int *out_len, int compr_type, u8 *crypto_key)

Please, write a fat "WARNING" note in the comment and tell that this
function modifies the input buffer.

> +/* Size of 128 bits in bytes */
> +#define AES_KEYSIZE_128 16

If you have no plans to support keys larger than 128 just kill this
constant please.

-- 
Best Regards,
Artem Bityutskiy

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

  reply	other threads:[~2012-04-02 14:33 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-02-09 15:24 [patch] Adding Secure Deletion to UBIFS Joel Reardon
2012-02-13 16:54 ` Artem Bityutskiy
2012-02-23 14:59   ` Joel Reardon
2012-02-23 15:29     ` [patch] Add encryption key parameter to compress/decompress functions Joel Reardon
2012-03-09  7:17       ` Artem Bityutskiy
2012-03-19 16:54         ` [patch] Add design document for UBIFS secure deletion Joel Reardon
2012-03-20 20:10           ` Randy Dunlap
2012-03-21 13:26             ` Joel Reardon
2012-03-21 16:20               ` Artem Bityutskiy
2012-03-21 16:10           ` Artem Bityutskiy
2012-03-23 13:50             ` Joel Reardon
2012-03-23 15:38               ` Artem Bityutskiy
2012-03-23 16:38                 ` Joel Reardon
2012-03-26 15:03                   ` Artem Bityutskiy
2012-02-29 17:09     ` [patch] Adding Secure Deletion to UBIFS Artem Bityutskiy
2012-03-15 14:48     ` [patch] Remove notion of key schemes Joel Reardon
2012-03-16 12:43       ` Artem Bityutskiy
2012-03-16 12:51       ` Artem Bityutskiy
2012-03-16 13:34         ` Joel Reardon
2012-03-16 13:41           ` Artem Bityutskiy
2012-03-16 15:02             ` Joel Reardon
2012-03-19 14:56               ` Artem Bityutskiy
2012-02-20 20:15 ` [patch] Move CRC computation to separate function Joel Reardon
2012-02-29 16:10   ` Artem Bityutskiy
2012-03-19 22:46     ` Joel Reardon
2012-03-23 14:09       ` Artem Bityutskiy
2012-03-23 16:45         ` Joel Reardon
2012-03-23 16:51           ` Artem Bityutskiy
2012-03-25 20:38             ` Joel Reardon
2012-03-26 15:34               ` Artem Bityutskiy
2012-03-25 21:11             ` [patch] Add a encryption key parameter to the compress / decompress function Joel Reardon
2012-03-25 21:38               ` [patch] Add cryptographic functionality when a key is passed to the compress / decompress functions Joel Reardon
2012-03-27  8:33                 ` Artem Bityutskiy
2012-03-29 14:39                   ` [patch] UBIFS: " Joel Reardon
2012-04-02 14:36                     ` Artem Bityutskiy [this message]
2012-04-02 14:48                       ` Joel Reardon
2012-04-02 14:57                         ` Artem Bityutskiy
2012-04-02 14:58                           ` Joel Reardon
2012-04-03 10:29                           ` Joel Reardon
2012-04-03 10:41                             ` Guillaume LECERF
2012-04-03 11:35                               ` Joel Reardon
2012-04-12 14:05                                 ` Artem Bityutskiy
2012-03-27  8:27               ` [patch] Add a encryption key parameter to the compress / decompress function Artem Bityutskiy
2012-03-29 14:11                 ` [patch] UBIFS: " Joel Reardon
2012-04-02 14:02                   ` Artem Bityutskiy
2012-02-29 17:25 ` [patch] Adding Secure Deletion to UBIFS Artem Bityutskiy
2012-03-01 13:41   ` Joel Reardon
2012-03-09  7:36     ` Artem Bityutskiy
2012-03-09 19:29       ` Joel Reardon
2012-03-12 13:30         ` Artem Bityutskiy
2012-03-12 13:34           ` Joel Reardon
2012-03-12 13:36           ` Artem Bityutskiy
2012-03-12 13:37             ` Joel Reardon
2012-03-14 10:20             ` Joel Reardon
2012-03-14 10:27               ` Artem Bityutskiy

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=1333377383.22146.14.camel@sauron.fi.intel.com \
    --to=dedekind1@gmail.com \
    --cc=joel@clambassador.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mtd@lists.infradead.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).