linux-kernel - Re: [PATCH v4 11/14] treewide: Prepare to remove VLA usage for AHASH_REQUEST_ON

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAGXu5jJp3Y4RZD+ZUYsWqf+sTbGJwdN3M=KOiPBn9VecYE4OgQ@mail.gmail.com>
Date:   Fri, 13 Jul 2018 20:07:10 -0700
From:   Kees Cook <keescook@...omium.org>
To:     Herbert Xu <herbert@...dor.apana.org.au>
Cc:     Arnd Bergmann <arnd@...db.de>,
        "Gustavo A. R. Silva" <gustavo@...eddedor.com>,
        Eric Biggers <ebiggers@...gle.com>,
        Alasdair Kergon <agk@...hat.com>,
        Giovanni Cabiddu <giovanni.cabiddu@...el.com>,
        Lars Persson <larper@...s.com>,
        Mike Snitzer <snitzer@...hat.com>,
        Rabin Vincent <rabinv@...s.com>,
        Tim Chen <tim.c.chen@...ux.intel.com>,
        "David S. Miller" <davem@...emloft.net>,
        Masahiro Yamada <yamada.masahiro@...ionext.com>,
        "open list:HARDWARE RANDOM NUMBER GENERATOR CORE" 
        <linux-crypto@...r.kernel.org>, qat-linux@...el.com,
        dm-devel@...hat.com,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v4 11/14] treewide: Prepare to remove VLA usage for AHASH_REQUEST_ON_STACK

On Thu, Jul 12, 2018 at 11:22 PM, Herbert Xu
<herbert@...dor.apana.org.au> wrote:
> On Thu, Jul 12, 2018 at 11:16:28PM -0700, Kees Cook wrote:
>>
>> Is this correct? It seems like you did the bulk of
>> AHASH_REQUEST_ON_STACK conversions in 2016. Can shash grow an sg
>> interface?
>
> shash does not need to grow an sg interface.  All users of
> AHASH_REQUEST_ON_STACK set the CRYPTO_ALG_ASYNC flag to zero
> when allocating the tfm.

On a plane today I started converting all these to shash. IIUC, it
just looks like this (apologies for whitespace damage):


 static int crypt_iv_essiv_init(struct crypt_config *cc)
 {
        struct iv_essiv_private *essiv = &cc->iv_gen_private.essiv;
-       AHASH_REQUEST_ON_STACK(req, essiv->hash_tfm);
-       struct scatterlist sg;
+       SHASH_DESC_ON_STACK(desc, essiv->hash_tfm);
        struct crypto_cipher *essiv_tfm;
        int err;

-       sg_init_one(&sg, cc->key, cc->key_size);
-       ahash_request_set_tfm(req, essiv->hash_tfm);
-       ahash_request_set_callback(req, CRYPTO_TFM_REQ_MAY_SLEEP, NULL, NULL);
-       ahash_request_set_crypt(req, &sg, essiv->salt, cc->key_size);
+       desc->tfm = essiv->hash_tfm;
+       desc->flags = CRYPTO_TFM_REQ_MAY_SLEEP;

-       err = crypto_ahash_digest(req);
-       ahash_request_zero(req);
+       err = crypto_shash_digest(desc, key, cc->key_size, essiv->salt);
+       shash_desc_zero(desc);
        if (err)
                return err;


(I left out all the s/ahash/shash/ in types and function declarations.)

Does this look like what you were thinking of for converting these
away from ahash? The only one I couldn't make sense of was in
drivers/crypto/inside-secure/safexcel_hash.c. I have no idea what's
happening there.

-Kees

-- 
Kees Cook
Pixel Security