Date: Sat, 10 Jun 2006 00:45:41 +0100 (BST) From: Robert Watson <rwatson@FreeBSD.org> To: Scott Long <scottl@samsco.org> Cc: Mikhail Teterin <mi+mx@aldan.algebra.com>, fs@freebsd.org Subject: Re: Space-saving of UFS1 Message-ID: <20060610004447.A26068@fledge.watson.org> In-Reply-To: <4489BD63.7060309@samsco.org> References: <20060609065656.31225.qmail@web30313.mail.mud.yahoo.com> <200606091313.04913.mi%2Bmx@aldan.algebra.com> <4489ADC9.3090809@samsco.org> <200606091330.10007.mi%2Bmx@aldan.algebra.com> <4489BD63.7060309@samsco.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 9 Jun 2006, Scott Long wrote: > The inode size was extended from 128 bytes to 256 bytes to allow for 64-bit > block pointers. This includes 12 direct block pointers and one pointer for > each of the single, double, and triple indirect blocks. That didn't fill > left some extra space in the 256 bytes, so ACL size info and block pointers > were put in there. However, ACLs are just a side effect of the larger size, > not the sole reason. And, ACLs are not actually stored in the inode, only > block pointers to them are. While the technical statements above are correct, actually, the extended attribute storage was the primary motivation for getting UFS2 development kicked off. Since it required rolling the file system layout, we did 64-bit support at the same time, dropped back in the birth time, etc. Robert N M Watson
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060610004447.A26068>