Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Message-ID: <e362242edea8931f045beea1228de99b6572aa89.camel@chromium.org>
Subject: Re: [PATCH] ima: add the ability to query ima for the hash of a
 given file.
From:   Florent Revest <revest@chromium.org>
To:     Lakshmi Ramasubramanian <nramas@linux.microsoft.com>,
        linux-integrity@vger.kernel.org
Cc:     kpsingh@chromium.org, mjg59@google.com, zohar@linux.ibm.com,
        linux-kernel@vger.kernel.org,
        linux-security-module@vger.kernel.org,
        Florent Revest <revest@google.com>
Date:   Mon, 06 Jan 2020 17:10:35 +0100
In-Reply-To: <8f4d9c4e-735d-8ba9-b84a-4f341030e0cf@linux.microsoft.com>
References: <20191220163136.25010-1-revest@chromium.org>
         <8f4d9c4e-735d-8ba9-b84a-4f341030e0cf@linux.microsoft.com>
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.30.5-1.1 
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Fri, 2019-12-20 at 08:48 -0800, Lakshmi Ramasubramanian wrote:
> On 12/20/2019 8:31 AM, Florent Revest wrote:
> 
> >   
> > +/**
> > + * ima_file_hash - return the stored measurement if a file has
> > been hashed.
> > + * @file: pointer to the file
> > + * @buf: buffer in which to store the hash
> > + * @buf_size: length of the buffer
> > + *
> > + * On success, output the hash into buf and return the hash
> > algorithm (as
> > + * defined in the enum hash_algo).
> > + * If the hash is larger than buf, then only size bytes will be
> > copied. It
> > + * generally just makes sense to pass a buffer capable of holding
> > the largest
> > + * possible hash: IMA_MAX_DIGEST_SIZE
> 
> If the given buffer is smaller than the hash length, wouldn't it be 
> better to return the required size and a status indicating the buffer
> is not enough. The caller can then call back with the required
> buffer.
> 
> If the hash is truncated the caller may not know if the hash is
> partial or not.

I agree with Mimi's answer that the caller would know based on the
returned hash algorithm.

> > + *
> > + * If IMA is disabled or if no measurement is available, return
> > -EOPNOTSUPP.
> > + * If the parameters are incorrect, return -EINVAL.
> > + */
> > +int ima_file_hash(struct file *file, char *buf, size_t buf_size)
> > +{
> > +	struct inode *inode;
> > +	struct integrity_iint_cache *iint;
> > +	size_t copied_size;
> > +
> > +	if (!file || !buf)
> > +		return -EINVAL;
> > +
> > +	if (!ima_policy_flag)
> > +		return -EOPNOTSUPP;
> > +
> > +	inode = file_inode(file);
> > +	iint = integrity_iint_find(inode);
> > +	if (!iint)
> > +		return -EOPNOTSUPP;
> > +
> > +	mutex_lock(&iint->mutex);
> > +	copied_size = min_t(size_t, iint->ima_hash->length, buf_size);
> > +	memcpy(buf, iint->ima_hash->digest, copied_size);
> > +	mutex_unlock(&iint->mutex);
> > +
> > +	return iint->ima_hash->algo;
> 
> Should the hash algorithm be copied from iinit->ima_hash to a local 
> variable while holding the mutex and that one returned?
> 
> I assume iinit->mutex  is taken to ensure iinit->ima_hash is not
> removed while this function is accessing it.

Ah! Good catch, thank you :)