>> That case was considered a bug which we fixed, so it's no longer
>> relevant; the mode of a symbolic link is now always ignored
>> (the owner is relevant in sticky directories and for quota
>> purposes)
>
>Code pointer?  I'd hate to update a now-useless lchmod() just to
>confirm that it is indeed now useless.

http://cvs.opensolaris.org/source/xref/on/usr/src/uts/common/fs/tmpfs/tmp_subr.c#74

I changed how this worked in '99  with:

4167560 symlinks and other non-files can be removed by anyone from sticky 
directories

The original BSD sticky semantics were, of course, that you needed to
be owner; SV changed this to "write access also", but forgot about the
special semantics of symlinks (mode ignored), pipes, sockets (things
carrying a context of sorts).

>That check is a bit of a disappointment, unless there is no other
>equally efficient way of doing whatever it's meant to do, or unless
>something actually requires that check as such.  IMO, resurrecting a
>nameless temporary file could be a very handy capability.
>
>That the hard links to /proc/$$/fd/$N don't work if $N is on zfs...is that
>just a limitation for now, or is that intentional and likely to stay that way?

I don't know what causes that.

Casper
_______________________________________________
opensolaris-code mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code

Reply via email to