linux-kernel - Re: [PATCH] tpm2_load

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]

Message-ID: <80e5b6d5ef4472bd1490dc5ddf3175cfcc119418.camel@linux.ibm.com>
Date:   Wed, 16 Jun 2021 07:49:52 -0700
From:   James Bottomley <jejb@...ux.ibm.com>
To:     Dhiraj Shah <find.dhiraj@...il.com>
Cc:     Jarkko Sakkinen <jarkko@...nel.org>,
        Mimi Zohar <zohar@...ux.ibm.com>,
        David Howells <dhowells@...hat.com>,
        James Morris <jmorris@...ei.org>,
        "Serge E. Hallyn" <serge@...lyn.com>,
        linux-integrity@...r.kernel.org, keyrings@...r.kernel.org,
        linux-security-module@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] tpm2_load_command leaks memory

On Thu, 2021-06-10 at 10:49 +0100, Dhiraj Shah wrote:
> tpm2_key_decode allocates memory which is stored in blob and it's not
> freed.
> 
> Signed-off-by: Dhiraj Shah <find.dhiraj@...il.com>
> ---
>  security/keys/trusted-keys/trusted_tpm2.c | 41 +++++++++++++++----
> ----
>  1 file changed, 27 insertions(+), 14 deletions(-)
> 
> diff --git a/security/keys/trusted-keys/trusted_tpm2.c
> b/security/keys/trusted-keys/trusted_tpm2.c
> index 0165da386289..52dd43bb8cdb 100644
> --- a/security/keys/trusted-keys/trusted_tpm2.c
> +++ b/security/keys/trusted-keys/trusted_tpm2.c
> @@ -378,22 +378,31 @@ static int tpm2_load_cmd(struct tpm_chip *chip,
>  	}
>  
>  	/* new format carries keyhandle but old format doesn't */
> -	if (!options->keyhandle)
> -		return -EINVAL;
> +	if (!options->keyhandle) {
> +		rc = -EINVAL;
> +		goto err;
> +	}

This one is unnecessary ... for the old format there's nothing to free.

>  	/* must be big enough for at least the two be16 size counts */
> -	if (payload->blob_len < 4)
> -		return -EINVAL;
> +	if (payload->blob_len < 4) {
> +		rc = -EINVAL;
> +		goto err;
> +	}
>  
>  	private_len = get_unaligned_be16(blob);
>  
>  	/* must be big enough for following public_len */
> -	if (private_len + 2 + 2 > (payload->blob_len))
> -		return -E2BIG;
> +	if (private_len + 2 + 2 > (payload->blob_len)) {
> +		rc = -E2BIG;
> +		goto err;
> +	}
>  
>  	public_len = get_unaligned_be16(blob + 2 + private_len);
> -	if (private_len + 2 + public_len + 2 > payload->blob_len)
> -		return -E2BIG;
> +
> +	if (private_len + 2 + public_len + 2 > payload->blob_len) {
> +		rc = -E2BIG;
> +		goto err;
> +	}
>  
>  	pub = blob + 2 + private_len + 2;
>  	/* key attributes are always at offset 4 */
> @@ -406,13 +415,16 @@ static int tpm2_load_cmd(struct tpm_chip *chip,
>  		payload->migratable = 1;
>  
>  	blob_len = private_len + public_len + 4;
> -	if (blob_len > payload->blob_len)
> -		return -E2BIG;
>  
> -	rc = tpm_buf_init(&buf, TPM2_ST_SESSIONS, TPM2_CC_LOAD);
> -	if (rc)
> -		return rc;
> +	if (blob_len > payload->blob_len) {
> +		rc = -E2BIG;
> +		goto err;
> +	}
>  
> +	if (tpm_buf_init(&buf, TPM2_ST_SESSIONS, TPM2_CC_LOAD) != 0)

You didn't compile this, did you?  There's no opening brace here ...

James