Re: fixing slink boot-floppies for sparc
Quoting Eric Delaunay <delaunay@lix.polytechnique.fr>:
> About wrong /dev/fb0 permissions, could someone investigate why X does not find
> the device if not world writable ? I was not able to track it down to the
> point (strace is segfaulting early on X, Xsun, startx, ...).
> Even with Xsun running suid I get the following error: "no screen found".
I ran startx under strace and the problem seems to be that
/usr/bin/X11/Xsun is testing access permissions for /dev/fb[012] using
the access() system call, which uses the real user ID instead of the
effective user ID. I have never worried about X with my PC and
therefore I don't know how it works, but I'd say the approach taken by
Xsun is not the correct one. But again, I'm not very knowledgeable
about how X works. Anybody cares to comment on this?
Regarding the /dev/mouse symlink problem, I'd say it's better to keep
the symlink (and bear the work to manage it). In this way applications
don't have to worry about where the mouse is, whatever platform it's
running in. True, there aren't that many applications using the moouse
directly but why run the risk? After all, one more symlink is not a
problem (famous last words).
--
Hermano Cabral
Reply to: