linux-kernel - Re: s390 crypto: explicitly memzero stack key material in aes

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] [day] [month] [year] [list]

Date:   Tue, 25 Feb 2020 15:39:46 +0100
From:   Harald Freudenberger <freude@...ux.ibm.com>
To:     Torsten Duwe <duwe@....de>
Cc:     Herbert Xu <herbert@...dor.apana.org.au>,
        "David S. Miller" <davem@...emloft.net>,
        Heiko Carstens <heiko.carstens@...ibm.com>,
        Vasily Gorbik <gor@...ux.ibm.com>,
        Christian Borntraeger <borntraeger@...ibm.com>,
        linux-crypto@...r.kernel.org, linux-s390@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: s390 crypto: explicitly memzero stack key material in aes_s390.c

On 21.02.20 17:55, Torsten Duwe wrote:
> From: Torsten Duwe <duwe@...e.de>
>
> s390 crypto: explicitly memzero stack key material in aes_s390.c
>
> aes_s390.c has several functions which allocate space for key material on
> the stack and leave the used keys there. It is considered good practice
> to clean these locations before the function returns.
>
> Signed-off-by: Torsten Duwe <duwe@...e.de>
> ---
> This popped up during our FIPS certification.
> It's obviously a good idea not to leave key material on the stack.
>
> --- a/arch/s390/crypto/aes_s390.c
> +++ b/arch/s390/crypto/aes_s390.c
> @@ -392,6 +392,7 @@ static int cbc_aes_crypt(struct blkciphe
>  		ret = blkcipher_walk_done(desc, walk, nbytes - n);
>  	}
>  	memcpy(walk->iv, param.iv, AES_BLOCK_SIZE);
> +	memzero_explicit(&param, sizeof(param));
>  	return ret;
>  }
>
> @@ -576,6 +577,8 @@ static int xts_aes_crypt(struct blkciphe
>  			 walk->dst.virt.addr, walk->src.virt.addr, n);
>  		ret = blkcipher_walk_done(desc, walk, nbytes - n);
>  	}
> +	memzero_explicit(&pcc_param, sizeof(pcc_param));
> +	memzero_explicit(&xts_param, sizeof(xts_param));
>  	return ret;
>  }
>
>
Thanks Torsten, I've picked this patch. It will go upstream via the s390 subsystem with the next kernel merge window.
regards Harald Freudenberger