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;
Subject: Re: [PATCH v2 1/2] fork: extend clone3() to support CLONE_SET_TID
From:   Dmitry Safonov <0x7f454c46@gmail.com>
To:     Adrian Reber <areber@redhat.com>,
        Christian Brauner <christian@brauner.io>,
        Eric Biederman <ebiederm@xmission.com>,
        Pavel Emelianov <xemul@virtuozzo.com>,
        Jann Horn <jannh@google.com>, Oleg Nesterov <oleg@redhat.com>
Cc:     linux-kernel@vger.kernel.org, Andrei Vagin <avagin@gmail.com>,
        Mike Rapoport <rppt@linux.ibm.com>,
        Radostin Stoyanov <rstoyanov1@gmail.com>
References: <20190731161223.2928-1-areber@redhat.com>
 <417a9682-c7fc-fec4-3510-81a8aa7cd0af@gmail.com>
Message-ID: <6ccdda8e-73a3-c8e9-1f37-f89bf688c0fa@gmail.com>
Date:   Wed, 31 Jul 2019 17:56:42 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101
 Thunderbird/60.7.2
MIME-Version: 1.0
In-Reply-To: <417a9682-c7fc-fec4-3510-81a8aa7cd0af@gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 7/31/19 5:49 PM, Dmitry Safonov wrote:
> Hi Adrian,
> 
> On 7/31/19 5:12 PM, Adrian Reber wrote:
> [..]
>> @@ -2530,14 +2530,12 @@ noinline static int copy_clone_args_from_user(struct kernel_clone_args *kargs,
>>  					      struct clone_args __user *uargs,
>>  					      size_t size)
>>  {
>> +	struct pid_namespace *pid_ns = task_active_pid_ns(current);
>>  	struct clone_args args;
>>  
>>  	if (unlikely(size > PAGE_SIZE))
>>  		return -E2BIG;
>>  
>> -	if (unlikely(size < sizeof(struct clone_args)))
>> -		return -EINVAL;
>> -
> 
> It might be better to validate it still somehow, but I don't insist.
> 
> [..]
>> @@ -2578,11 +2580,16 @@ noinline static int copy_clone_args_from_user(struct kernel_clone_args *kargs,
>>  
>>  static bool clone3_args_valid(const struct kernel_clone_args *kargs)
>>  {
>> -	/*
>> -	 * All lower bits of the flag word are taken.
>> -	 * Verify that no other unknown flags are passed along.
>> -	 */
>> -	if (kargs->flags & ~CLONE_LEGACY_FLAGS)
>> +	/* Verify that no other unknown flags are passed along. */
>> +	if (kargs->flags & ~(CLONE_LEGACY_FLAGS | CLONE_SET_TID))
>> +		return false;
>> +
>> +	/* Fail if set_tid is set without CLONE_SET_TID */
>> +	if (kargs->set_tid && !(kargs->flags & CLONE_SET_TID))
>> +		return false;
>> +
>> +	/* Also fail if set_tid is invalid */
>> +	if ((kargs->set_tid <= 0) && (kargs->flags & CLONE_SET_TID))
>>  		return false;
> 
> Sorry for not mentioning it on v1, but I've noticed it only now:
> you check kargs->set_tid even with the legacy-sized kernel_clone_args,
> which is probably some random value on a task's stack?

Self-correction: On kernel stack in copy_clone_args_from_user().
Which may probably be considered as a security leak..
Sorry again for not spotting it in v1.

Thanks,
          Dmitry