Date: Mon, 9 Feb 1998 02:37:20 +0000 (GMT) From: Terry Lambert <tlambert@primenet.com> To: reilly@zeta.org.au (Andrew Reilly) Cc: mike@smith.net.au, tlambert@primenet.com, dyson@FreeBSD.ORG, abial@nask.pl, freebsd-current@FreeBSD.ORG, jkh@FreeBSD.ORG Subject: Re: Custom init(8) (and some ideas) Message-ID: <199802090237.TAA22548@usr06.primenet.com> In-Reply-To: <199802082219.JAA03349@gurney.reilly.home> from "Andrew Reilly" at Feb 9, 98 09:19:11 am
next in thread | previous in thread | raw e-mail | index | archive | help
> If this is likely to become a popular game (paging from compressed > binaries), would it be useful to build a special "binary gzipper" that > split and separately compressed independent 4k pages? I believe that > the V.42 compression in modems forces blocking at 256-byte boundaries, > to minimise the amount that needs to be re-transmitted on errors, and > that still manages reasonable error rates. You would need to build a > table of pointers to each of the blocks to avoid scanning the whole > thing, but there could be some gain. Is there an obvious reason for > this not to work? If you could be *guaranteed* that the compressed imaged of 8k would become at most 4k, then you could do page doubling. The problem is that you need a VM object for the executable to grab pages off of on page boundries, and you *also* need to grab pages off of disk in page boundries. The closest thing to this currently is the vn? devices configured by vnconfig. They just *happen* to present 4k as 4k; they could just as easily present a smaller amount as 4k (after uncompression). You would *still* have the block-boundry issues, though. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199802090237.TAA22548>