home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.unix.sysv386
- Path: sparky!uunet!virtech!cpcahil
- From: cpcahil@virtech.uucp (Conor P. Cahill)
- Subject: Re: >16MB on a 486
- Message-ID: <BuJIw0.AE3@virtech.uucp>
- Organization: Virtual Technologies Inc.
- References: <BuFF20.1tq@constant.demon.co.uk> <2659@aegis.or.jp> <BuILEt.1MG@virtech.uucp> <1992Sep13.105315.15555@husc3.harvard.edu>
- Date: Sun, 13 Sep 1992 23:35:59 GMT
- Lines: 28
-
- maziere1@husc8.harvard.edu (David Mazieres) writes:
-
- >In article <BuILEt.1MG@virtech.uucp> cpcahil@virtech.uucp (Conor P. Cahill) writes:
- >>It cannot be done with ISC 2.2 or 3.0 and the testing that I did with 2.0.2
- >>made me believe that regarless of the setting of this file the
- >>system would start swapping at 16MB...
-
- >This maybe a stupid question, but what's wrong with doing:
- > "MEMRANGE=0-640K:0,1M-16M:0,16M-64M:1"
-
- >The man page says of the 0/1 flag:
- > "0 indicates no special properties and 1 indicates memory for which
- > DMA is not allowed."
-
- First of, this option is not available in 2.2 (as far as I remember) nor
- in 3.0. In the testing that I did with 2.0.2, even setting 16M-64M:1
- *seemed* to have no effect on whether or not the system started swapping
- when I reached 16MB of data in processes (using a test process I wrote
- which allocates 1MB of data and then pokes bytes into it in random
- locations). This is something I tried about a year and a half ago and
- I don't have 2.0.2 running anywhere at this point to review the
- test, but I am pretty sure that those were the results I observed.
-
- --
- *** SENTINEL(tm) The ultimate Debugging Environment - email for more info ***
-
- Conor P. Cahill (703)430-9247 cpcahil@virtech.vti.com
- Virtual Technologies, Inc. 46030 Manekin Plaza Dulles, VA 20166
-