Re: [PATCH net-next v2] crypto: af_alg/hash: Fix recvmsg() after sendmsg(MSG_MORE)

[Herbert Xu <herbert@gondor.apana.org.au>]

On Fri, Jun 16, 2023 at 12:10:32PM +0100, David Howells wrote:
> If an AF_ALG socket bound to a hashing algorithm is sent a zero-length
> message with MSG_MORE set and then recvmsg() is called without first
> sending another message without MSG_MORE set to end the operation, an oops
> will occur because the crypto context and result doesn't now get set up in
> advance because hash_sendmsg() now defers that as long as possible in the
> hope that it can use crypto_ahash_digest() - and then because the message
> is zero-length, it the data wrangling loop is skipped.
> 
> Fix this by handling zero-length sends at the top of the hash_sendmsg()
> function.  If we're not continuing the previous sendmsg(), then just ignore
> the send (hash_recvmsg() will invent something when called); if we are
> continuing, then we finalise the request at this point if MSG_MORE is not
> set to get any error here, otherwise the send is of no effect and can be
> ignored.
> 
> Whilst we're at it, remove the code to create a kvmalloc'd scatterlist if
> we get more than ALG_MAX_PAGES - this shouldn't happen.
> 
> Fixes: c662b043cdca ("crypto: af_alg/hash: Support MSG_SPLICE_PAGES")
> Reported-by: syzbot+13a08c0bf4d212766c3c@syzkaller.appspotmail.com
> Link: https://lore.kernel.org/r/000000000000b928f705fdeb873a@google.com/
> Reported-by: syzbot+14234ccf6d0ef629ec1a@syzkaller.appspotmail.com
> Link: https://lore.kernel.org/r/000000000000c047db05fdeb8790@google.com/
> Reported-by: syzbot+4e2e47f32607d0f72d43@syzkaller.appspotmail.com
> Link: https://lore.kernel.org/r/000000000000bcca3205fdeb87fb@google.com/
> Reported-by: syzbot+472626bb5e7c59fb768f@syzkaller.appspotmail.com
> Link: https://lore.kernel.org/r/000000000000b55d8805fdeb8385@google.com/
> Signed-off-by: David Howells <dhowells@redhat.com>
> Reported-and-tested-by: syzbot+6efc50cc1f8d718d6cb7@syzkaller.appspotmail.com
> cc: Herbert Xu <herbert@gondor.apana.org.au>
> cc: "David S. Miller" <davem@davemloft.net>
> cc: Eric Dumazet <edumazet@google.com>
> cc: Jakub Kicinski <kuba@kernel.org>
> cc: Paolo Abeni <pabeni@redhat.com>
> cc: Jens Axboe <axboe@kernel.dk>
> cc: Matthew Wilcox <willy@infradead.org>
> cc: linux-crypto@vger.kernel.org
> cc: netdev@vger.kernel.org
> ---
>  crypto/algif_hash.c |   38 +++++++++++++++++++++++++-------------
>  1 file changed, 25 insertions(+), 13 deletions(-)
> 
> diff --git a/crypto/algif_hash.c b/crypto/algif_hash.c
> index dfb048cefb60..0ab43e149f0e 100644
> --- a/crypto/algif_hash.c
> +++ b/crypto/algif_hash.c
> @@ -76,13 +76,30 @@ static int hash_sendmsg(struct socket *sock, struct msghdr *msg,
>  
>  	lock_sock(sk);
>  	if (!continuing) {
> -		if ((msg->msg_flags & MSG_MORE))
> -			hash_free_result(sk, ctx);
> +		/* Discard a previous request that wasn't marked MSG_MORE. */
> +		hash_free_result(sk, ctx);

Please revert this change as I explained in the other message.

> +		if (!msg_data_left(msg))
> +			goto done; /* Zero-length; don't start new req */

This is still broken in the case of a zero-length message with
MSG_MORE set.  Here you will short-circuit out without ever calling
crypto_ahash_init.  However, hash_recvmsg will directly call
crypto_ahash_final on this, which is undefined.

Cheers,
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt