netdev - Re: [PATCH net-next v3 3/6] tls: implement rekey for TLS1.3

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 for Android: free password hash cracker in your pocket

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

Message-ID: <ZNUkz7UNMPQVOr2M@vergenet.net>
Date: Thu, 10 Aug 2023 19:56:31 +0200
From: Simon Horman <horms@...nel.org>
To: Sabrina Dubroca <sd@...asysnail.net>
Cc: netdev@...r.kernel.org, Vadim Fedorenko <vfedorenko@...ek.ru>,
	Frantisek Krenzelok <fkrenzel@...hat.com>,
	Jakub Kicinski <kuba@...nel.org>,
	Kuniyuki Iwashima <kuniyu@...zon.com>,
	Apoorv Kothari <apoorvko@...zon.com>,
	Boris Pismenny <borisp@...dia.com>,
	John Fastabend <john.fastabend@...il.com>,
	Shuah Khan <shuah@...nel.org>, linux-kselftest@...r.kernel.org,
	Gal Pressman <gal@...dia.com>,
	Marcel Holtmann <marcel@...tmann.org>
Subject: Re: [PATCH net-next v3 3/6] tls: implement rekey for TLS1.3

On Wed, Aug 09, 2023 at 02:58:52PM +0200, Sabrina Dubroca wrote:
> This adds the possibility to change the key and IV when using
> TLS1.3. Changing the cipher or TLS version is not supported.
> 
> Once we have updated the RX key, we can unblock the receive side. If
> the rekey fails, the context is unmodified and userspace is free to
> retry the update or close the socket.
> 
> This change only affects tls_sw, since 1.3 offload isn't supported.
> 
> v2:
>  - reverse xmas tree
>  - turn the alt_crypto_info into an else if
>  - don't modify the context when rekey fails
> 
> v3:
>  - only call tls_sw_strparser_arm when setting the initial RX key, not
>    on rekeys
>  - update tls_sk_poll to not say the socket is readable when we're
>    waiting for a rekey, and wake up poll() when the new key is installed
>  - use unsafe_memcpy to make FORTIFY_SOURCE happy
> 
> Signed-off-by: Sabrina Dubroca <sd@...asysnail.net>

...

> diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c

...

> @@ -2873,14 +2911,24 @@ int tls_set_sw_offload(struct sock *sk, int tx)
>  
>  	ctx->push_pending_record = tls_sw_push_pending_record;
>  
> +	/* setkey is the last operation that could fail during a
> +	 * rekey. if it succeeds, we can start modifying the
> +	 * context.
> +	 */
>  	rc = crypto_aead_setkey(*aead, key, keysize);
> +	if (rc) {
> +		if (new_crypto_info)
> +			goto out;
> +		else
> +			goto free_aead;
> +	}
>  
> -	if (rc)
> -		goto free_aead;
> -
> -	rc = crypto_aead_setauthsize(*aead, prot->tag_size);
> -	if (rc)
> -		goto free_aead;
> +	if (!new_crypto_info) {
> +		rc = crypto_aead_setauthsize(*aead, prot->tag_size);
> +		if (rc) {
> +			goto free_aead;
> +		}

nit: no need for {} here.

> +	}
>  
>  	if (sw_ctx_rx) {
>  		tfm = crypto_aead_tfm(sw_ctx_rx->aead_recv);

...