-
-
perhaps a cleaner way would be to add a new PG_* flag in page-flags.h and set that on pages allocated in the VMA. require madvise to happen while the VMA has no pages. and then do the wiping in the put_page() machinery.
-
So yeah, ideally this is something that propagates down to the pages rather than just being a property of the VMA. Hmm.
- 2 more replies
New conversation -
-
-
Doesn't the page_count() check prevent that? Of course, it also means that if the page is referenced by something other than the process and the process exits first, the page won't be wiped
-
your patch looks at page_mapcount(), not page_count()/page_ref_count(). struct page has two refcounts, one for user mappings (->_mapcount), one for kernel references (->_refcount)
- 1 more reply
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.