Re: [PATCH] memory cgroup enhancements [1/5] force_empty for memory cgroup

[David Rientjes <rientjes@google.com>]

On Wed, 17 Oct 2007, KAMEZAWA Hiroyuki wrote:

> > > +static ssize_t mem_force_empty_read(struct cgroup *cont,
> > > +				struct cftype *cft,
> > > +				struct file *file, char __user *userbuf,
> > > +				size_t nbytes, loff_t *ppos)
> > > +{
> > > +	static const char buf[2] = "0";
> > > +	return simple_read_from_buffer((void __user *)userbuf, nbytes,
> > > +			ppos, buf, strlen(buf));
> > 
> > Reading memory.force_empty is pretty useless, so why allow it to be read 
> > at all?
> 
> Ah, yes. How can I make this as read-only ? just remove .read ?
> 
> Thanks,

You mean make it write-only?  Typically it would be as easy as only 
specifying a mode of S_IWUSR so that it can only be written to, the 
S_IFREG is already provided by cgroup_add_file().

Unfortunately, cgroups do not appear to allow that.  It hardcodes
the permissions of 0644 | S_IFREG into the cgroup_create_file() call from 
cgroup_add_file(), which is a bug.  Cgroup files should be able to be 
marked as read-only or write-only depending on their semantics.

So until that bug gets fixed and you're allowed to pass your own file 
modes to cgroup_add_files(), you'll have to provide the read function.

		David

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>