Date: Sun, 16 Nov 1997 02:42:20 -0500 (EST) From: "John S. Dyson" <toor@dyson.iquest.net> To: garbanzo@hooked.net (Alex) Cc: tom@sdf.com, hackers@FreeBSD.ORG Subject: Re: 256Meg Message-ID: <199711160742.CAA12053@dyson.iquest.net> In-Reply-To: <Pine.BSF.3.96.971115212150.267B-100000@zippy.dyn.ml.org> from Alex at "Nov 15, 97 09:22:38 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
Alex said: > > > On Fri, 14 Nov 1997, Tom wrote: > > > > > On Fri, 14 Nov 1997, dennis wrote: > > > > > Is there a maximum that FreeBSD can support? > > > > > > Dennis > > > > > > What? Filesystems? RAM? Something else? > > > > RAM... no problem. I'm running two 256MB RAM servers now. > > Actually, there's a limit of 4GB or so of ram, on the 486 (if you call > tha ta limit ;-) ), and AFAIK the P5, P6 and PII and clones as well. > Physically, the limit is 36Bits on a P6. It would require some mods to the pmap layer, and maybe some enhancements to the upper level VM code. One disadvantage with the extended 3 level translation mode is that the PTE's become twice as large. I seriously doubt that we'll need that on a P6, but on future Slot1/Slot2 processors, we might find that 4GB is a real limit, and have to accomodate the modified PTD/PTE format. Imagine a processor that is perhaps 2X to 5X as fast as a P6, in a multiprocessor config -- that would appear to be able to use more than the typical upper end of 1GByte of memory, and even more than the normally addressable 4Gbytes. I haven't given this alot of thought yet, but these issues are probably going to be important in the medium-term future (approx 1yr.) -- John dyson@freebsd.org jdyson@nc.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199711160742.CAA12053>