Re: The /tmp and modules_install saga

Peter T. Breuer (ptb@it.uc3m.es)
Fri, 16 Oct 1998 17:23:37 +0200 (MET DST)


"A month of sundays ago David Woodhouse wrote:"
>
>
> ptb@it.uc3m.es said:
> > OK. I still don't think you need any tmp files. If it were me doing
> > it, i'd 1) mark all .o's executable to start. 2) unmark them
> > executable as I copy them. 3) fill misc with the .o's still executable
> > at the end.
>
> This doesn't work if your kernel tree is read-only. Which'd screw me because

It's not conceptually the kernel tree that is being written, but the
site in which the modules were built that is being written, which could
be anywhere, but wait ...

> that's how I distribute kernels. The 'slave' workstations just mount my
> development tree by NFS and run 'make modules_install'

Aha. OK. I do something like that too. Except I make a package and
they run installpkg, or equivalent. They get the package remotely from
the server via nfs mount of a package distribution area.

Well, either use shell variables to hold the file lists as people
suggested or: run through all source objects at the end of the
modules_install; if they're listed in one of the _MODULE files,
(that's a [ -n "`grep -w $module *_MODULE`" ]) skip, otherwise
copy to misc. I think it's worth the effort to avoid building
tmp files in /lib.

> ---- ---- ----
> David Woodhouse David.Woodhouse@mvhi.com Office: (+44) 1223 810302
> Project Leader, Process Information Systems Mobile: (+44) 976 658355
> Axiom (Cambridge) Ltd., Swaffham Bulbeck, Cambridge, CB5 0NA, UK.
> finger dwmw2@ferret.lmh.ox.ac.uk for PGP key.

Peter

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/