Messages in this thread | | | Date | Mon, 23 Aug 2010 12:54:09 -0700 | From | Jeremy Fitzhardinge <> | Subject | Re: [RFC] mlock/stack guard interaction fixup |
| |
On 08/23/2010 12:26 PM, Peter Zijlstra wrote: > On Mon, 2010-08-23 at 12:23 -0700, Jeremy Fitzhardinge wrote: >> On 08/23/2010 12:07 PM, Peter Zijlstra wrote: >>> mlock() simply avoids major faults, nothing more. >>> >>> I think both page migration and page-out for shared pages where some >>> maps are !mlocked can cause unmaps and thus minor faults. >>> >>> mlock and dirty do not interact, they will still be cleaned/written out >>> as normal. >> So mlock is useless for preventing secret stuff from being written to disk. > Well, if you put your sekrit in a file map, sure. > > Use a mmap(MAP_ANONYMOUS|MAP_LOCK) and madvise(MADV_DONTFORK) for your > sekrits.
Won't dirty anonymous pages also get written to swap?
J
| |