Date: Sat, 3 Jul 1999 11:43:18 +0930 From: Greg Lehey <grog@lemis.com> To: Mike Smith <mike@smith.net.au> Cc: Mike Smith <msmith@FreeBSD.org>, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/i386/i386 machdep.c Message-ID: <19990703114318.M87392@freebie.lemis.com> In-Reply-To: <199907030112.SAA02288@dingo.cdrom.com>; from Mike Smith on Fri, Jul 02, 1999 at 06:12:35PM -0700 References: <19990703103111.F87392@freebie.lemis.com> <199907030112.SAA02288@dingo.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Friday, 2 July 1999 at 18:12:35 -0700, Mike Smith wrote: >> On Friday, 2 July 1999 at 17:39:31 -0700, Mike Smith wrote: >>>> On Friday, 2 July 1999 at 13:33:35 -0700, Mike Smith wrote: >>>>> msmith 1999/07/02 13:33:35 PDT >>>>> >>>>> Modified files: >>>>> sys/i386/i386 machdep.c >>>>> Log: >>>>> Lightly overhaul the memory sizing code again. >>>>> >>>>> - The kernel environment variable 'hw.physmem' can be used to set the >>>>> amount of physical memory space, based at 0, that FreeBSD will use. >>>>> Any memory detected over this limit is ignored. Documentation for >>>>> this is available under 'help set tunables' in the loader. >>>>> >>>>> - In the case where system memory size can't be accurately determined, >>>>> hw.physmem is used as a best-guess memory size, but speculative >>>>> probing will be used to determine actual memory size if any of the >>>>> guesses or hints are 16M or more. >>>> >>>> Don't these two paragraphs contradict each other? >>> >>> No. In the first case, if more than hw.physmem is detected, any excess >>> is ignored. In the second, if it's not possible to determine how much >>> memory there is, hw.physmem is used as a working guess in conjunction >>> with a speculative probe. >> >> So depending on factors beyond your control, it either limits the size >> of memory used, or it doesn't? > > No; the system that you use it on is a factor under your control. Up to a point. If I have a machine for which the memory size cannot be accurately determined, but I know it's 144 MB, and I want to limit memory to 32 MB, how do I do it? >> It seems that you could have found a less confusing and more >> controllable way to do it. > > I guess you're not feeling very constructive today. Bad guess. > There are a number of sets of existing behaviour that I didn't want > to change just yet, since I wanted some more cycles on this code > first. OK, that's a valid viewpoint. So this is still work in progress? > There's all sorts of silliness regarding the 16M and 64M points that > don't yet quite add up. Ultimately, hw.physmem will always mean "do > not ever use more memory than this (but if you can't work it out > yourself, try to find this much)". OK, accepted. I suppose it's worth putting this information in the commit messages: "another step to a sane method of determining memory size". Greg -- See complete headers for address, home page and phone numbers finger grog@lemis.com for PGP public key To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990703114318.M87392>