Date: Tue, 4 Jan 2011 19:55:58 +0200 From: Kostik Belousov <kostikbel@gmail.com> To: Gleb Kurtsou <gleb.kurtsou@gmail.com> Cc: freebsd-fs@freebsd.org Subject: Re: [rfc] 64-bit inode numbers Message-ID: <20110104175558.GR3140@deviant.kiev.zoral.com.ua> In-Reply-To: <20101201091203.GA3933@tops> References: <20101201091203.GA3933@tops>
next in thread | previous in thread | raw e-mail | index | archive | help
--G6ArjEZjY3m60389 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Dec 01, 2010 at 11:12:03AM +0200, Gleb Kurtsou wrote: > Hi, >=20 > I've been working on adding support for 64 bit ino_t and 32 bit nlink_t. > I have a patchset which is still work in progress, but I wasn't able to > find time to continue the project for a month already. >=20 > Notes: > * Keep using 32 bit inodes in UFS and UFS boot code: UFS is a critical > component and switching it to 64bit ino_t won't improve anything > * Don't use nlink_t in UFS on-disk structs, introduce ufs_ino_t > * Deprecate incomplete set of NetBSD emulation nstat* syscalls. Remove > COMPAT_FREEBSD32 support which was never functional, mark as COMPAT8 > * Implement both kernel level and libc compatibility shims, support > COMPAT_FREEBSD32 > * Add d_off (now unused) to struct dirent to facilitate removing cookies > argument from VOP_READDIR. (both OpenSolaris and Linux do the same) > * Libc compatibility shims for struct dirent users >=20 >=20 > Tar archive of individual patches to make review easier. First several > patches are bug fixes and can be commited as they are: > https://github.com/downloads/glk/freebsd-ino64/freebsd-ino64-patch.tgz >=20 > The same but as a single patch: > https://github.com/downloads/glk/freebsd-ino64/freebsd-ino64.big.patch.gz >=20 > Patches are against recent CURRENT: svn r215808. >=20 > Code is also hosted on gitorious: > http://gitorious.org/~glk/glk-freebsd-ino64 >=20 I think some more comments for each patch in the set, in addition to the one-line title, would be useful. No need to add regen patches, they only confuse the reader. Just add a note to other patches where the regen is needed. I have big doubts about 0009, since struct inoref is not on-disk struct. My impression is that the issue of extending ino_t to 64 bit is much bigger then presented in your patch. E.g. FTSENT (include/fts.h) explicitely include ino_t member. As result, there are more ABI changes that handled. Or, did I missed this in the patchset ? Might be, libarchive and libufs are also affected. Not sure about struct pidfh from libutil. --G6ArjEZjY3m60389 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (FreeBSD) iEYEARECAAYFAk0jXy4ACgkQC3+MBN1Mb4jJdwCgrBItWMmKku/o4Pj/9BryZ1t3 Fc0An3CH4ZVYJy68v4hHkQIhcBh2LCO0 =v9lM -----END PGP SIGNATURE----- --G6ArjEZjY3m60389--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20110104175558.GR3140>