Date: Tue, 03 Oct 2006 09:03:45 -0500 From: Eric Anderson <anderson@centtech.com> To: =?ISO-8859-1?Q?Dag-Erling_Sm=F8rgrav?= <des@des.no> Cc: freebsd-fs@FreeBSD.ORG, Oliver Fromme <olli@lurza.secnetix.de> Subject: Re: Hi: Porting Cramfs on FreeBSD Message-ID: <45226DC1.3080302@centtech.com> In-Reply-To: <86y7rxh8v5.fsf@dwp.des.no> References: <200609271744.k8RHipTS032655@lurza.secnetix.de> <451AC81E.5070803@centtech.com> <86y7rxh8v5.fsf@dwp.des.no>
next in thread | previous in thread | raw e-mail | index | archive | help
On 10/03/06 07:38, Dag-Erling Smørgrav wrote: > Eric Anderson <anderson@centtech.com> writes: >> No, only inode+direntries need to fit into memory. So an FS with >> 1million inodes might take .5MB of memory (estimate). > > How are you going to do that? Whenever you want to read a file, > you'll have to start decompressing the entire tarball from the > beginning, since it's a single gzip stream. You can't just seek to an > arbitrary position in the stream and start decompressing there. > > DES First, that .5MB estimate was off by quite a bit - I'm not sure what I was calculating when I did that, but I know I was tired. :) Anyway, I'm not currently supporting compressed archives, but obviously they would be very useful and are on my roadmap. Tim Kientzle gave me some ideas, and I'll ping him again when I get to that point.' Eric -- ------------------------------------------------------------------------ Eric Anderson Sr. Systems Administrator Centaur Technology Anything that works is better than anything that doesn't. ------------------------------------------------------------------------
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?45226DC1.3080302>