Messages in this thread | | | Date | Thu, 13 Nov 2014 15:35:56 -0800 | Subject | Re: [PATCH] devcoredump: provide a one-way disable function | From | Kees Cook <> |
| |
On Thu, Nov 13, 2014 at 2:36 PM, Johannes Berg <johannes@sipsolutions.net> wrote: > On Thu, 2014-11-13 at 13:56 -0800, Kees Cook wrote: > >> > + /* >> > + * This essentially makes the attribute write-once, since you can't >> > + * go back to not having it disabled. This is intentional, it serves >> > + * as a system lockdown feature. >> > + */ >> > + if (tmp != 1) >> > + return -EINVAL; >> >> Just a nit, but writing "0" is valid if devcd_disabled = false? > > I thought about that too, but what would the point be? The only > operation you ever can/want to do is writing "1" to it to disable the > framework.
Yup, seems like not a useful thing to do, but figured I'd point it out just in case. I think the patch is fine as-is.
-Kees
-- Kees Cook Chrome OS Security
| |