> > On Feb 18, 2018, at 9:00 AM, <simh-requ...@trailing-edge.com> > <simh-requ...@trailing-edge.com> wrote: > > Date: Sun, 18 Feb 2018 11:08:24 +0100 > From: Johnny Billquist <b...@softjar.se <mailto:b...@softjar.se>> > To: Larry Baker <ba...@usgs.gov <mailto:ba...@usgs.gov>> > Cc: SIMH <simh@trailing-edge.com <mailto:simh@trailing-edge.com>> > Subject: Re: [Simh] Simh Digest, Vol 169, Issue 58 > Message-ID: <9314564c-2452-fd85-bc18-075301e9c...@softjar.se > <mailto:9314564c-2452-fd85-bc18-075301e9c...@softjar.se>> > Content-Type: text/plain; charset=utf-8; format=flowed >
Johnny, > Memory mapped overlays potentially wasted lots of space. The "problem" > is that each overlay level must be page aligned, which means 8K aligned. > It also means that you cannot have more than 7 levels, since you only > have 8 pages. Compare to disk based overlays, which can pack the levels > right next to each other, and it should be obvious that you potentially > loose a lot of memory with memory mapped overlays. I was confused why you think memory is wasted. It actually is not. It is true that the program address space can only be mapped by an entire window. Thus, you have to be clever about packing the 8K address window. However, there is no requirement that an entire address window be mapped. The real memory allocated is only the portion that is mapped. The system memory is packed tightly, just the same for regular contiguous (on RSX at least) allocations for any program. Real memory is not wasted. Larry Baker US Geological Survey 650-329-5608 ba...@usgs.gov
_______________________________________________ Simh mailing list Simh@trailing-edge.com http://mailman.trailing-edge.com/mailman/listinfo/simh
