linux-hardening - Re: [RESEND] evm: Refector struct evm

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: <61ffe974-8e15-6682-d53a-90df89dcb28a@huawei.com>
Date:   Wed, 25 May 2022 09:50:16 +0800
From:   "Guozihua (Scott)" <guozihua@...wei.com>
To:     "Gustavo A. R. Silva" <gustavoars@...nel.org>
CC:     <linux-integrity@...r.kernel.org>, <zohar@...ux.ibm.com>,
        <dmitry.kasatkin@...il.com>, <linux-hardening@...r.kernel.org>
Subject: Re: [RESEND] evm: Refector struct evm_xattr

On 2022/5/18 13:13, Gustavo A. R. Silva wrote:
> On Wed, May 18, 2022 at 09:14:39AM +0800, GUO Zihua wrote:
>> struct evm_xattr is only used for EVM_XATTR_HMAC type evm digest and is
>> glued together one flexible array and one fixed length array. The
>> original intention seems to be shortening the code as the "data" field
>> would always be a SHA1 digest.
>>
>> This implementation is not complying with GCC's specification about
>> flexible array and spars yield the following warning:
> 
> The sparse warning has nothing to do with any GCC specification. It's
> perfectly fine to apply the sizeof operator to a struct-with-flex-array.
> However, it might be suspicious if the intention is to also get the
> _dynamic_ size of the flexible array, because in that case the size of
> the flex array is always zero. See the example below:
> 
> 	struct foo {
> 		uint8_t len;
> 		struct boo data[];
> 	};
> 
> 	sizeof(struct foo) == 1
> 
> Also, you sent this patch twice in the last 24 hours. Give the maintainers
> time to review your patch (usually a couple of weeks) before resending it.
> 
>>
>> security/integrity/evm/evm_main.c:852:30: warning: using sizeof on a flexible
>> structure
>> security/integrity/evm/evm_main.c:862:32: warning: using sizeof on a flexible
>> structure
> 
> Regarding the warnings above, please take a look at my response to your
> other patch (the same applies in this case). :)
> 
> Thanks
> --
> Gustavo
> 
>>
>> Fix it by:
>> 1. Remove struct evm_xattr and use struct evm_ima_xattr_data directly.
>> 2. Get array size with struct_size instead of sizeof.
>>
>> Reference: https://github.com/KSPP/linux/issues/174
>>
>> Fixes: 6be5cc5246f80 ("evm: add support for different security.evm data types")
>> Signed-off-by: GUO Zihua <guozihua@...wei.com>
>> ---
>>   security/integrity/evm/evm_main.c | 14 ++++++++------
>>   security/integrity/integrity.h    |  6 ------
>>   2 files changed, 8 insertions(+), 12 deletions(-)
>>
>> diff --git a/security/integrity/evm/evm_main.c b/security/integrity/evm/evm_main.c
>> index 7d87772f0ce6..f2c4501a287a 100644
>> --- a/security/integrity/evm/evm_main.c
>> +++ b/security/integrity/evm/evm_main.c
>> @@ -211,7 +211,8 @@ static enum integrity_status evm_verify_hmac(struct dentry *dentry,
>>   	/* check value type */
>>   	switch (xattr_data->type) {
>>   	case EVM_XATTR_HMAC:
>> -		if (xattr_len != sizeof(struct evm_xattr)) {
>> +		if (xattr_len != struct_size(*xattr_data, data,
>> +					     SHA1_DIGEST_SIZE)) {
>>   			evm_status = INTEGRITY_FAIL;
>>   			goto out;
>>   		}
>> @@ -842,24 +843,25 @@ int evm_inode_init_security(struct inode *inode,
>>   				 const struct xattr *lsm_xattr,
>>   				 struct xattr *evm_xattr)
>>   {
>> -	struct evm_xattr *xattr_data;
>> +	struct evm_ima_xattr_data *xattr_data;
>>   	int rc;
>>   
>>   	if (!(evm_initialized & EVM_INIT_HMAC) ||
>>   	    !evm_protected_xattr(lsm_xattr->name))
>>   		return 0;
>>   
>> -	xattr_data = kzalloc(sizeof(*xattr_data), GFP_NOFS);
>> +	xattr_data = kzalloc(struct_size(*xattr_data, data,
>> +			     SHA1_DIGEST_SIZE), GFP_NOFS);
>>   	if (!xattr_data)
>>   		return -ENOMEM;
>>   
>> -	xattr_data->data.type = EVM_XATTR_HMAC;
>> -	rc = evm_init_hmac(inode, lsm_xattr, xattr_data->digest);
>> +	xattr_data->type = EVM_XATTR_HMAC;
>> +	rc = evm_init_hmac(inode, lsm_xattr, xattr_data->data);
>>   	if (rc < 0)
>>   		goto out;
>>   
>>   	evm_xattr->value = xattr_data;
>> -	evm_xattr->value_len = sizeof(*xattr_data);
>> +	evm_xattr->value_len = struct_size(*xattr_data, data, SHA1_DIGEST_SIZE);
>>   	evm_xattr->name = XATTR_EVM_SUFFIX;
>>   	return 0;
>>   out:
>> diff --git a/security/integrity/integrity.h b/security/integrity/integrity.h
>> index 3510e413ea17..91b16d620dd9 100644
>> --- a/security/integrity/integrity.h
>> +++ b/security/integrity/integrity.h
>> @@ -86,12 +86,6 @@ struct evm_ima_xattr_data {
>>   	u8 data[];
>>   } __packed;
>>   
>> -/* Only used in the EVM HMAC code. */
>> -struct evm_xattr {
>> -	struct evm_ima_xattr_data data;
>> -	u8 digest[SHA1_DIGEST_SIZE];
>> -} __packed;
>> -
>>   #define IMA_MAX_DIGEST_SIZE	64
>>   
>>   struct ima_digest_data {
>> -- 
>> 2.36.0
>>
> .
Thanks Gustavo!

I'm still pretty new to the community so your feedback is much appreciated!

-- 
Best
GUO Zihua