Re: [PATCH 1/2] fs: make do_mkdirat() take struct filename

[Al Viro <viro@zeniv.linux.org.uk>]

On Sun, Jan 24, 2021 at 09:38:19PM -0700, Jens Axboe wrote:
> On 11/15/20 9:45 PM, Dmitry Kadashev wrote:
> > Pass in the struct filename pointers instead of the user string, and
> > update the three callers to do the same. This is heavily based on
> > commit dbea8d345177 ("fs: make do_renameat2() take struct filename").
> > 
> > This behaves like do_unlinkat() and do_renameat2().
> 
> Al, are you OK with this patch? Leaving it quoted, though you should
> have the original too.

> > -static long do_mkdirat(int dfd, const char __user *pathname, umode_t mode)
> > +long do_mkdirat(int dfd, struct filename *name, umode_t mode)
> >  {
> >  	struct dentry *dentry;
> >  	struct path path;
> >  	int error;
> >  	unsigned int lookup_flags = LOOKUP_DIRECTORY;
> >  
> > +	if (IS_ERR(name))
> > +		return PTR_ERR(name);
> > +
> >  retry:
> > -	dentry = user_path_create(dfd, pathname, &path, lookup_flags);
> > -	if (IS_ERR(dentry))
> > -		return PTR_ERR(dentry);
> > +	name->refcnt++; /* filename_create() drops our ref */
> > +	dentry = filename_create(dfd, name, &path, lookup_flags);
> > +	if (IS_ERR(dentry)) {
> > +		error = PTR_ERR(dentry);
> > +		goto out;
> > +	}

No.  This is going to be a source of confusion from hell.  If anything,
you want a variant of filename_create() that does not drop name on
success.  With filename_create() itself being an inlined wrapper
for it.