Re: [RFC PATCH v2 06/16] fs: split off need_remove_file_privs() do_remove_file_privs()

[Stefan Roesch <shr@fb.com>]

On 5/17/22 6:18 AM, Christian Brauner wrote:
> On Mon, May 16, 2022 at 09:47:08AM -0700, Stefan Roesch wrote:
>> This splits off the function need_remove_file_privs() from the function
>> do_remove_file_privs() from the function file_remove_privs().
>>
>> No intended functional changes in this patch.
>>
>> Signed-off-by: Stefan Roesch <shr@fb.com>
>> ---
> 
> Just a few nits...
> 
>>  fs/inode.c | 57 ++++++++++++++++++++++++++++++++++++------------------
>>  1 file changed, 38 insertions(+), 19 deletions(-)
>>
>> diff --git a/fs/inode.c b/fs/inode.c
>> index 9d9b422504d1..a6d70a1983f8 100644
>> --- a/fs/inode.c
>> +++ b/fs/inode.c
>> @@ -2010,17 +2010,8 @@ static int __remove_privs(struct user_namespace *mnt_userns,
>>  	return notify_change(mnt_userns, dentry, &newattrs, NULL);
>>  }
>>  
>> -/*
>> - * Remove special file priviledges (suid, capabilities) when file is written
>> - * to or truncated.
>> - */
>> -int file_remove_privs(struct file *file)
>> +static int need_file_remove_privs(struct inode *inode, struct dentry *dentry)
> 
> I'd rather call this file_needs_remove_privs()?
> 
Renamed to file_needs_remove_privs()

>>  {
>> -	struct dentry *dentry = file_dentry(file);
>> -	struct inode *inode = file_inode(file);
>> -	int kill;
>> -	int error = 0;
>> -
>>  	/*
>>  	 * Fast path for nothing security related.
>>  	 * As well for non-regular files, e.g. blkdev inodes.
>> @@ -2030,16 +2021,37 @@ int file_remove_privs(struct file *file)
>>  	if (IS_NOSEC(inode) || !S_ISREG(inode->i_mode))
>>  		return 0;
>>  
>> -	kill = dentry_needs_remove_privs(dentry);
>> -	if (kill < 0)
>> -		return kill;
>> -	if (kill)
>> -		error = __remove_privs(file_mnt_user_ns(file), dentry, kill);
>> +	return dentry_needs_remove_privs(dentry);
>> +}
>> +
>> +static int do_file_remove_privs(struct file *file, struct inode *inode,
>> +				struct dentry *dentry, int kill)
> 
> and that __file_remove_privs() which matches the rest of the file since
> here we don't have a lot of do_* but rather __* convention afaict.
> 

Renamed the function to __file_remove_privs().

>> +{
>> +	int error = 0;
>> +
>> +	error = __remove_privs(file_mnt_user_ns(file), dentry, kill);
>>  	if (!error)
>>  		inode_has_no_xattr(inode);
>>  
>>  	return error;
>>  }
>> +
>> +/*
>> + * Remove special file privileges (suid, capabilities) when file is written
>> + * to or truncated.
>> + */
>> +int file_remove_privs(struct file *file)
> 
> This is a generic comment, not aimed specifically at your change but we
> really need to get better at kernel-doc...
> 
> Since you're already touching this code could you at least to the
> exported function you're modifying add sm like:
> 

I added the kernel documentation.

> /**
>  * file_remove_privs - remove special file privileges (suid, capabilities) 
>  * @file: file to remove privileges from
>  * 
>  * When file is modified by a write or truncation ensure that special
>  * file privileges are removed.
>  *
>  * Return: 0 on success, negative errno on failure.
>  */
> int file_remove_privs(struct file *file)
> 
> This will then render on kernel.org/doc see e.g. lookup_one():
> https://www.kernel.org/doc/html/latest/filesystems/api-summary.html?highlight=lookup_one#c.lookup_one
> 
>> +{
>> +	struct dentry *dentry = file_dentry(file);
>> +	struct inode *inode = file_inode(file);
>> +	int kill;
>> +
>> +	kill = need_file_remove_privs(inode, dentry);
>> +	if (kill <= 0)
>> +		return kill;
>> +
>> +	return do_file_remove_privs(file, inode, dentry, kill);
>> +}
>>  EXPORT_SYMBOL(file_remove_privs);
>>  
>>  /**
>> @@ -2093,15 +2105,22 @@ EXPORT_SYMBOL(file_update_time);
>>  /* Caller must hold the file's inode lock */
>>  int file_modified(struct file *file)
> 
> Similar I'd add sm like:
> 

I added the kernel documentation.

> /**
>  * file_modified - handle mandated vfs changes when modifying a file
>  * @file: file that was was modified
>  * 
>  * When file has been modified ensure that special
>  * file privileges are removed and time settings are updated.
>  *
>  * Context: Caller must hold the file's inode lock.
>  *
>  * Return: 0 on success, negative errno on failure.
>  */
> int file_remove_privs(struct file *file)
> 
>>  {
>> -	int err;
>> +	int ret;
>> +	struct dentry *dentry = file_dentry(file);
>> +	struct inode *inode = file_inode(file);
>>  
>>  	/*
>>  	 * Clear the security bits if the process is not being run by root.
>>  	 * This keeps people from modifying setuid and setgid binaries.
>>  	 */
>> -	err = file_remove_privs(file);
>> -	if (err)
>> -		return err;
>> +	ret = need_file_remove_privs(inode, dentry);
>> +	if (ret < 0) {
>> +		return ret;
>> +	} else if (ret > 0) {
>> +		ret = do_file_remove_privs(file, inode, dentry, ret);
>> +		if (ret)
>> +			return ret;
>> +	}
> 
> The else-if branch looks a bit unorthodox to me. I'd probably rather
> make this:
> 
> 	ret = need_file_remove_privs(inode, dentry);
> 	if (ret < 0)
> 		return ret;
> 	
> 	if (ret > 0) {
> 		ret = do_file_remove_privs(file, inode, dentry, ret);
> 		if (ret)
> 			return ret;
> 	}
>>  
>>  	if (unlikely(file->f_mode & FMODE_NOCMTIME))
>>  		return 0;

I replaced the else if.

>> -- 
>> 2.30.2
>>