Re: [PATCH 5/6] nfs: check all iov segments for correct memoryaccess rights

[Andrew Morton]

Chuck Lever <cel@xxxxxxxxxx> wrote:
>
> +/*
> + * Check:
> + * 1.  All bytes in the user buffers are properly accessible
> + * 2.  The resulting number of bytes won't overflow ssize_t
> + */

hm.

> +static ssize_t check_access_ok(int type, const struct iovec *iov, unsigned long nr_segs)
> +{
> +	ssize_t count = 0;
> +	ssize_t retval = -EINVAL;
> +	unsigned long seg;
> +
> +	for (seg = 0; seg < nr_segs; seg++) {
> +		void __user *buf = iov[seg].iov_base;
> +		ssize_t len = (ssize_t) iov[seg].iov_len;
> +
> +		if (len < 0)		/* size_t not fitting an ssize_t .. */
> +			goto out;

do_readv_writev() already checked for negative iov_len, and that's the more
appropriate place to do it, rather than duplicating it in each filesystem
(or forgetting to!)

So is this check really needed?

> +		if (unlikely(!access_ok(type, buf, len))) {
> +			retval = -EFAULT;
> +			goto out;
> +		}

Now what's up here?  Why does NFS, at this level, care about the page's
virtual address?  get_user_pages() will handle that?

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/