linux-kernel - Re: [PATCH net] net/tls: fix slab-out-of-bounds bug in decrypt

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]

Date:   Wed, 30 Mar 2022 09:39:25 -0700
From:   Jakub Kicinski <kuba@...nel.org>
To:     Ziyang Xuan <william.xuanziyang@...wei.com>
Cc:     <borisp@...dia.com>, <john.fastabend@...il.com>,
        <daniel@...earbox.net>, <davem@...emloft.net>, <pabeni@...hat.com>,
        <netdev@...r.kernel.org>, <vakul.garg@....com>,
        <davejwatson@...com>, <linux-kernel@...r.kernel.org>,
        Vadim Fedorenko <vfedorenko@...ek.ru>
Subject: Re: [PATCH net] net/tls: fix slab-out-of-bounds bug in
 decrypt_internal

On Wed, 30 Mar 2022 16:50:09 +0800 Ziyang Xuan wrote:
> The memory size of tls_ctx->rx.iv for AES128-CCM is 12 setting in
> tls_set_sw_offload(). The return value of crypto_aead_ivsize()
> for "ccm(aes)" is 16. So memcpy() require 16 bytes from 12 bytes
> memory space will trigger slab-out-of-bounds bug as following:
> 
> ==================================================================
> BUG: KASAN: slab-out-of-bounds in decrypt_internal+0x385/0xc40 [tls]
> Read of size 16 at addr ffff888114e84e60 by task tls/10911
> 
> Call Trace:
>  <TASK>
>  dump_stack_lvl+0x34/0x44
>  print_report.cold+0x5e/0x5db
>  ? decrypt_internal+0x385/0xc40 [tls]
>  kasan_report+0xab/0x120
>  ? decrypt_internal+0x385/0xc40 [tls]
>  kasan_check_range+0xf9/0x1e0
>  memcpy+0x20/0x60
>  decrypt_internal+0x385/0xc40 [tls]
>  ? tls_get_rec+0x2e0/0x2e0 [tls]
>  ? process_rx_list+0x1a5/0x420 [tls]
>  ? tls_setup_from_iter.constprop.0+0x2e0/0x2e0 [tls]
>  decrypt_skb_update+0x9d/0x400 [tls]
>  tls_sw_recvmsg+0x3c8/0xb50 [tls]
> 
> Allocated by task 10911:
>  kasan_save_stack+0x1e/0x40
>  __kasan_kmalloc+0x81/0xa0
>  tls_set_sw_offload+0x2eb/0xa20 [tls]
>  tls_setsockopt+0x68c/0x700 [tls]
>  __sys_setsockopt+0xfe/0x1b0

Interesting, are you running on non-x86 platform or with some crypto
accelerator? I wonder why we're not hitting it with KASAN and the
selftest we have.

> Reserve MAX_IV_SIZE memory space for iv to be compatible with all
> ciphers. And do iv and salt copy like done in tls_do_encryption().
> 
> Fixes: f295b3ae9f59 ("net/tls: Add support of AES128-CCM based ciphers")
> Signed-off-by: Ziyang Xuan <william.xuanziyang@...wei.com>
> ---
>  net/tls/tls_sw.c | 10 +++-------
>  1 file changed, 3 insertions(+), 7 deletions(-)
> 
> diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c
> index 0024a692f0f8..6b858f995b23 100644
> --- a/net/tls/tls_sw.c
> +++ b/net/tls/tls_sw.c
> @@ -1456,7 +1456,7 @@ static int decrypt_internal(struct sock *sk, struct sk_buff *skb,
>  	aead_size = sizeof(*aead_req) + crypto_aead_reqsize(ctx->aead_recv);
>  	mem_size = aead_size + (nsg * sizeof(struct scatterlist));
>  	mem_size = mem_size + prot->aad_size;
> -	mem_size = mem_size + crypto_aead_ivsize(ctx->aead_recv);
> +	mem_size = mem_size + MAX_IV_SIZE;

This change is not strictly required for the patch, right?
Can we drop it, and perhaps send as an optimization separately later?

>  	/* Allocate a single block of memory which contains
>  	 * aead_req || sgin[] || sgout[] || aad || iv.
> @@ -1493,12 +1493,8 @@ static int decrypt_internal(struct sock *sk, struct sk_buff *skb,
>  		kfree(mem);
>  		return err;
>  	}
> -	if (prot->version == TLS_1_3_VERSION ||
> -	    prot->cipher_type == TLS_CIPHER_CHACHA20_POLY1305)
> -		memcpy(iv + iv_offset, tls_ctx->rx.iv,
> -		       crypto_aead_ivsize(ctx->aead_recv));
> -	else
> -		memcpy(iv + iv_offset, tls_ctx->rx.iv, prot->salt_size);
> +	memcpy(iv + iv_offset, tls_ctx->rx.iv,
> +	       prot->iv_size + prot->salt_size);

If the IV really is 16B then we're passing 4 bytes of uninitialized
data at the end of the buffer, right?

>  	xor_iv_with_seq(prot, iv + iv_offset, tls_ctx->rx.rec_seq);
>