home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.unix.aix
- Path: sparky!uunet!cs.utexas.edu!milano!cactus.org!chukran
- From: chukran@cactus.org (Rudy E. Chukran)
- Subject: Re: Virtual Memory curiosity
- Message-ID: <1992Nov19.064654.8005@cactus.org>
- Keywords: VM overflow response
- Organization: Capital Area Central Texas UNIX Society, Austin, Tx
- References: <BxM5C5.Jyw@csn.org> <1992Nov13.034308.8602@oucsace.cs.ohiou.edu> <BxnyG7.4Hv@news.cso.uiuc.edu>
- Distribution: usa
- Date: Thu, 19 Nov 1992 06:46:54 GMT
- Lines: 21
-
- In article <BxnyG7.4Hv@news.cso.uiuc.edu> shair@vnet.ibm.com writes:
- >Good idea, but not for AIX. Don't think AIX ever swaps the read-only stuff
- >out in the first place.
- If you only consider readonly persistent text segments, then these are
- never paged out because they are never dirtied. They get paged back in
- from the filesystem.
-
- However, to be complete, consider the shared text segment. The shared text
- segment consists of many sections mapped readonly from many shared objects
- in the filesystem. But, these are mapped into a working segment, and therefore
- the associated pages are paged to page space. So these pages are paged out,
- even though they are readonly.
-
- They are paged out at least once and consume page space pages. The part that
- I dont know for sure is if they are ever paged out a second time?
- My understanding is that a page is paged out only if it is dirty. If it
- is clean, then it is simple replaced. This holds for any non dirty page,
- regardless of whether is is readonly or writtable.
-
- Any corroboration from the VMM folks? Is Scott listening?
-
-