Re: [PATCH RFC v2 3/4] mm/mempolicy: fix the check of nodemask from user

[Yisheng Xie]

On 2017/10/31 17:30, Vlastimil Babka wrote:
> On 10/27/2017 12:14 PM, Yisheng Xie wrote:
>> +	/*
>> +	 * When the user specified more nodes than supported just check
>> +	 * if the non supported part is all zero.
>> +	 *
>> +	 * If maxnode have more longs than MAX_NUMNODES, check
>> +	 * the bits in that area first. And then go through to
>> +	 * check the rest bits which equal or bigger than MAX_NUMNODES.
>> +	 * Otherwise, just check bits [MAX_NUMNODES, maxnode).
>> +	 */
>>  	if (nlongs > BITS_TO_LONGS(MAX_NUMNODES)) {
>>  		for (k = BITS_TO_LONGS(MAX_NUMNODES); k < nlongs; k++) {
>> -			unsigned long t;
>>  			if (get_user(t, nmask + k))
>>  				return -EFAULT;
>>  			if (k == nlongs - 1) {
>> @@ -1294,6 +1301,16 @@ static int get_nodes(nodemask_t *nodes, const unsigned long __user *nmask,
>>  		endmask = ~0UL;
>>  	}
>>  
>> +	if (maxnode > MAX_NUMNODES && MAX_NUMNODES % BITS_PER_LONG != 0) {
>> +		unsigned long valid_mask = endmask;
>> +
>> +		valid_mask &= ~((1UL << (MAX_NUMNODES % BITS_PER_LONG)) - 1);
> 
> I'm not sure if the combination with endmask works in this case:
> 
> 0      BITS_PER_LONG  2xBITS_PER_LONG
> |____________|____________|
>        |             |
>   MAX_NUMNODES      maxnode
> 
> endmask will contain bits between 0 and maxnode

In the case, BITS_TO_LONGS(maxnode) > BITS_TO_LONGS(MAX_NUMNODES), right?
And after checking BITS_PER_LONG to 2xBITS_PER_LONGïendmask will set to
"~0UL". e.g. endmask will be 0xffff ffff ffff ffff if
unsigned long is 64bit.

Then the valid_mask will just contain bits MAX_NUMNODES to BITS_PER_LONG.

Thanks
Yisheng Xie

> but here we want to check bits between MAX_NUMNODES and BITS_PER_LONG
> and endmask should not be mixed up with that?
> 
> 
> Vlastimil
>