Re: [PATCH v2 2/2] crypto: Remove unnecessary memzero_explicit()

From: Waiman Long <longman@redhat.com>
To: "Michal Suchánek" <msuchanek@suse.de>
Cc: Christophe Leroy <christophe.leroy@c-s.fr>,
	Andrew Morton <akpm@linux-foundation.org>,
	David Howells <dhowells@redhat.com>,
	Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>,
	James Morris <jmorris@namei.org>,
	"Serge E. Hallyn" <serge@hallyn.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Joe Perches <joe@perches.com>,
	Matthew Wilcox <willy@infradead.org>,
	David Rientjes <rientjes@google.com>,
	linux-mm@kvack.org, keyrings@vger.kernel.org,
	linux-kernel@vger.kernel.org, x86@kernel.org,
	linux-crypto@vger.kernel.org, linux-s390@vger.kernel.org,
	linux-pm@vger.kernel.org,
	linux-stm32@st-md-mailman.stormreply.com,
	linux-arm-kernel@lists.infradead.org,
	linux-amlogic@lists.infradead.org,
	linux-mediatek@lists.infradead.org,
	linuxppc-dev@lists.ozlabs.org,
	virtualization@lists.linux-foundation.org,
	netdev@vger.kernel.org, intel-wired-lan@lists.osuosl.org,
	linux-ppp@vger.kernel.org, wireguard@lists.zx2c4.com,
	linux-wireless@vger.kernel.org, devel@driverdev.osuosl.org,
	linux-scsi@vger.kernel.org, target-devel@vger.kernel.org,
	linux-btrfs@vger.kernel.org, linux-cifs@vger.kernel.org,
	samba-technical@lists.samba.org, linux-fscrypt@vger.kernel.org,
	ecryptfs@vger.kernel.org, kasan-dev@googlegroups.com,
	linux-bluetooth@vger.kernel.org, linux-wpan@vger.kernel.org,
	linux-sctp@vger.kernel.org, linux-nfs@vger.kernel.org,
	tipc-discussion@lists.sourceforge.net, cocci@systeme.lip6.fr,
	linux-security-module@vger.kernel.org,
	linux-integrity@vger.kernel.org
Subject: Re: [PATCH v2 2/2] crypto: Remove unnecessary memzero_explicit()
Date: Tue, 14 Apr 2020 15:37:51 -0400	[thread overview]
Message-ID: <578fe9b6-1ccd-2698-60aa-96c3f2dd2c31@redhat.com> (raw)
In-Reply-To: <20200414191601.GZ25468@kitsune.suse.cz>

On 4/14/20 3:16 PM, Michal Suchánek wrote:
> On Tue, Apr 14, 2020 at 12:24:36PM -0400, Waiman Long wrote:
>> On 4/14/20 2:08 AM, Christophe Leroy wrote:
>>>
>>> Le 14/04/2020 à 00:28, Waiman Long a écrit :
>>>> Since kfree_sensitive() will do an implicit memzero_explicit(), there
>>>> is no need to call memzero_explicit() before it. Eliminate those
>>>> memzero_explicit() and simplify the call sites. For better correctness,
>>>> the setting of keylen is also moved down after the key pointer check.
>>>>
>>>> Signed-off-by: Waiman Long <longman@redhat.com>
>>>> ---
>>>>   .../allwinner/sun8i-ce/sun8i-ce-cipher.c      | 19 +++++-------------
>>>>   .../allwinner/sun8i-ss/sun8i-ss-cipher.c      | 20 +++++--------------
>>>>   drivers/crypto/amlogic/amlogic-gxl-cipher.c   | 12 +++--------
>>>>   drivers/crypto/inside-secure/safexcel_hash.c  |  3 +--
>>>>   4 files changed, 14 insertions(+), 40 deletions(-)
>>>>
>>>> diff --git a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c
>>>> b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c
>>>> index aa4e8fdc2b32..8358fac98719 100644
>>>> --- a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c
>>>> +++ b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c
>>>> @@ -366,10 +366,7 @@ void sun8i_ce_cipher_exit(struct crypto_tfm *tfm)
>>>>   {
>>>>       struct sun8i_cipher_tfm_ctx *op = crypto_tfm_ctx(tfm);
>>>>   -    if (op->key) {
>>>> -        memzero_explicit(op->key, op->keylen);
>>>> -        kfree(op->key);
>>>> -    }
>>>> +    kfree_sensitive(op->key);
>>>>       crypto_free_sync_skcipher(op->fallback_tfm);
>>>>       pm_runtime_put_sync_suspend(op->ce->dev);
>>>>   }
>>>> @@ -391,14 +388,11 @@ int sun8i_ce_aes_setkey(struct crypto_skcipher
>>>> *tfm, const u8 *key,
>>>>           dev_dbg(ce->dev, "ERROR: Invalid keylen %u\n", keylen);
>>>>           return -EINVAL;
>>>>       }
>>>> -    if (op->key) {
>>>> -        memzero_explicit(op->key, op->keylen);
>>>> -        kfree(op->key);
>>>> -    }
>>>> -    op->keylen = keylen;
>>>> +    kfree_sensitive(op->key);
>>>>       op->key = kmemdup(key, keylen, GFP_KERNEL | GFP_DMA);
>>>>       if (!op->key)
>>>>           return -ENOMEM;
>>>> +    op->keylen = keylen;
>>> Does it matter at all to ensure op->keylen is not set when of->key is
>>> NULL ? I'm not sure.
>>>
>>> But if it does, then op->keylen should be set to 0 when freeing op->key. 
>> My thinking is that if memory allocation fails, we just don't touch
>> anything and return an error code. I will not explicitly set keylen to 0
>> in this case unless it is specified in the API documentation.
> You already freed the key by now so not touching anything is not
> possible. The key is set to NULL on allocation failure so setting keylen
> to 0 should be redundant. However, setting keylen to 0 is consisent with
> not having a key, and it avoids the possibility of leaking the length
> later should that ever cause any problem.

OK, I can change it to clear the key length when the allocation failed
which isn't likely.

Cheers,
Longman