LANANA: To Pending Device Number Registrants
Quoted 9 lines · Linus Torvalds
Quoted 4 lines · Alexander Viro
Driver can export a tree and we mount it on fb0. After that you have the whole set - yes,
/dev/fb0/colourspace, etc. - no problem. And no need to do mknod, BTW. Yes, we'll need to use/dev/fb0/framefor frame itself. BFD...Actually, we can just continue to use "
/dev/fb0", which would continue to work the way ti has always worked.
Quoted 6 lines
It's a mistake to think that a directory has to be a directory. Or to think that a device node has to be a device node. It's perfectly ok to just think of it as namespaces. So opening
/dev/fb0continues to open the "master fd", whatever that means (in this case, the actual frame buffer). The namespaces _under_/dev/fb0would be the control channels, or in fact _anything_ that the frame buffer driver wants to expose.
Why not take it a step further than just devices? This is a perfect model for supporting named forks.
In fact, I believe this is how MacOS X is exposing HFS+ named forks to the UNIX side of things. (HFS+ supports not just the old style Macintosh data and resource forks but an arbitrary number of named forks.) So: you open "foo", you get what an older MacOS would consider the "data" fork. Open "foo/rsrc" and you get the resource fork. Open "foo/joesfork" and you get whatever Joe wanted to use a named fork for.
Quoted 2 lines · Timothy A. Seufert
Why not take it a step further than just devices? This is a perfect model for supporting named forks.
Because unlike devices you need to be able to tar the tree up, copy it, edit, etc.
Please, don't mix _that_ flamewar into the thread, OK?
Quoted 20 lines · Timothy A. Seufert
Quoted 9 lines · Linus Torvalds
Quoted 4 lines · Alexander Viro
Driver can export a tree and we mount it on fb0. After that you have the whole set - yes,
/dev/fb0/colourspace, etc. - no problem. And no need to do mknod, BTW. Yes, we'll need to use/dev/fb0/framefor frame itself. BFD...Actually, we can just continue to use "
/dev/fb0", which would continue to work the way ti has always worked.Quoted 6 lines
It's a mistake to think that a directory has to be a directory. Or to think that a device node has to be a device node. It's perfectly ok to just think of it as namespaces. So opening
/dev/fb0continues to open the "master fd", whatever that means (in this case, the actual frame buffer). The namespaces _under_/dev/fb0would be the control channels, or in fact _anything_ that the frame buffer driver wants to expose.Why not take it a step further than just devices? This is a perfect model for supporting named forks.
Quoted 6 lines
In fact, I believe this is how MacOS X is exposing HFS+ named forks to the UNIX side of things. (HFS+ supports not just the old style Macintosh data and resource forks but an arbitrary number of named forks.) So: you open "foo", you get what an older MacOS would consider the "data" fork. Open "foo/rsrc" and you get the resource fork. Open "foo/joesfork" and you get whatever Joe wanted to use a named fork for.
How to find out which forks are available? UNIX semantics for directories are that a directory is just a file that you open to get the directory contents. But if your `directory' is a file, you'll get the file's contents if you open it. Or should we open wigth O_DIRECTORY to get at the fork directory?
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 2 lines · Timothy A. Seufert
Why not take it a step further than just devices? This is a perfect model for supporting named forks.
Because this only works on filesystems where directories can't themselves have named forks (or streams, or whatever you wish to call them) associated with them.
Read the archives on this issue :)
Mo.
- --
Mo McKinlay [email protected] http://ekto.org
Read http://www.ietf.org/rfc/rfc1855.txt
- ------------------------------------------------------------------------- GnuPG/PGP Key: pub 1024D/76A275F9 2000-07-22
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.5 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iEYEARECAAYFAjsCod0ACgkQRcGgB3aidflJyQCfdjh+o8vZvzZLfowM5QoLGICy tmAAoMIF4DcyqTep43Hd2/9X9tfeIH9X
=cKMC
-----END PGP SIGNATURE-----