Re: [PATCH 1/3] extend get/setrlimit to support setting rlimits external to a process (v3)

[Marcin Slusarz <marcin.slusarz@gmail.com>]

Neil Horman wrote:
>>> +static ssize_t proc_pid_limit_write(struct file *file, const char __user *buf,
>>> +		size_t count, loff_t *ppos)
>>> +{
>>> +	char *buffer;
>>> +	char *element, *vmc, *vmm;
>>> +	struct rlimit new_rlim;
>>> +	unsigned long flags;
>>> +	int i;
>>> +	int index = -1;
>>> +	size_t wcount = 0;
>>> +	struct task_struct *task = get_proc_task(file->f_path.dentry->d_inode);
>>> +
>>> +
>>> +	if (*ppos != 0)
>>> +		goto out;
>>> +
>>> +	if (count > 128)
>>> +		goto out;
>>> +	buffer = kzalloc(128, GFP_KERNEL);
>>> +
>>> +	if (!buffer)
>>> +		goto out;
>> element, vmc, vmm are not initialized and you kfree them in this case
>>
> Yep, I'll fix that
> 
>>> +
>>> +	element = kzalloc(sizeof(buffer), GFP_KERNEL);
>>> +	vmc = kzalloc(sizeof(buffer), GFP_KERNEL);
>>> +	vmm = kzalloc(sizeof(buffer), GFP_KERNEL);
>> sizeof(buffer) == 4 or 8 - pretty short buffer...
>>
>>> +
>>> +	if (!element || !vmm || !vmc)
>>> +		goto out_free;
>>> +
>>> +	wcount = count - copy_from_user(buffer, buf, count);
>>> +	if (wcount < count)
>>> +		goto out_free;
>> copy_from_user usage usually looks like:
>> if (copy_from_user()) {
>> 	ret = -EFAULT;
>> 	goto err;
>> }
>> you don't seem to use copy_from_user return value for anything useful
>>
> I did at one point, a few versions ago, that can likely be removed now.
> 
>>> +
>>> +	i = sscanf(buffer, "%s %s %s", element, vmc, vmm);
>> what if user pass strings longer than size of buffers?
>>
> You fail the write, theres a check at the top of the function for that.  By the
> time you get here, buffer is guaranteed to be no more than 128 bytes.

But you allocated only 4/8 bytes (depending on size of void*) for element, vmc, vmm.
It will overflow for string like "xxxxxxxxxxxxxxxxxxx y z".

Marcin