Re: [patch] mlock-as-nonroot revisted

[Andrea Arcangeli]

On Tue, Aug 03, 2004 at 07:07:30PM -0700, Chris Wright wrote:
> I think it's no different from current behaviour.  Which for fs based
> allocations is only guarded by fsuid and max_huge_pages.

sorry, what does prevent people from creating a 2M file, filling it with
pagefaults, closing creating another one and so on? Either "this"
accounting is applied to hugetlb files too or it's insecure.

Now if they didn't change the creation of the hugetlb files to be
allowed if the rlimit is !=0 then it wouldn't be insecure, but they
changed that in the first patch.

could you post a final patch once you're ready so that I will apply and
verify that I can exploit it? If it doesn't break it'll be still messed
up w.r.t. chown semantics (a true quota would get that right).
-
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/