Re: memory pages nulling when releasing

看板FB_security作者時間19年前 (2006/06/19 16:45), 編輯推噓0(000)
留言0則, 0人參與, 最新討論串9/27 (看更多)
--X1bOJ3K7DJ5YkBrT Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sun, 2006-Jun-18 13:39:03 -0700, R. B. Riddick wrote: Instead of >zero'ing pages immediately after the process does not need them >anymore, it would be much better, to keep the system safe >(especially: security relevant software patches; and (even more) >physical safety) The Unix model provides security as long as you don't bypass the access controls by (eg) reading /dev/mem. The OS only needs to explicitly zero a page if it is handing it back to a process without otherwise initialising it. There's no need to zero a page if it's going to be used to satisfy a pagein request. FreeBSD tries to reduce the effective overhead of page zeroing by zeroing them in the idle loop and keeping a cache of pre-zeroed pages for handing out to processes. --=20 Peter Jeremy --X1bOJ3K7DJ5YkBrT Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.3 (FreeBSD) iD8DBQFElmPF/opHv/APuIcRAr5/AJ0WHP5ubWmzXfcZXjNxKVb6XtNnlACfVLup TfjBJSpkNjCnQXOplaTo0Dw= =xxZa -----END PGP SIGNATURE----- --X1bOJ3K7DJ5YkBrT--
文章代碼(AID): #14bcGV00 (FB_security)
討論串 (同標題文章)
文章代碼(AID): #14bcGV00 (FB_security)