Re: toplevel Makefile bug and simple fix

Peter Samuelson (peter@wire.cadcamlab.org)
Tue, 9 Nov 1999 01:37:55 -0600 (CST)


[paulsch@us.ibm.com]
> I fully understand what is happening here.. If I didn't, I wouldn't
> have said anything in the first place... I can leave /usr/src/linux
> pointing to some kernel source tree, but it won't be the right one
> for what I'm building..

And again: it won't matter. split-include.c does *not* need a specific
version of kernel headers installed. Any helper program that *does*
depend on kernel constants should use -I$(HPATH). (I would expect to
find such in places like arch/*/boot/.) This one *doesn't*.

So you can just put back your symlink or whatever, it'll be fine. I
compile multiple kernel versions all the time too; my <linux/*.h>
doesn't ever point to current kernel source and it's never bitten me.

> > > I don't see why it should be a big deal just to add -I$(HPATH) to
> > > HOSTCC in the Makefile...
> > Well, to me it's the principle of the thing -- not only do I
> > believe that it's not a kernel bug, I also disagree with
> > overloading $(CC) (or $(HOSTCC)) with what are really CFLAGS.
> But $(CC) already does this... So the point is moot...

Not to me. I recently sent Michael Chastain a patch to fix the $(CC)
thing. He agrees with the intent and the approach and has indicated
that he will likely send it on to Linus some time soon. Perhaps this
explains why I really can't agree with your suggestion to do more or
less the reverse.

> I'm not trying to start any arguments here (I sure as hell wasn't
> expecting the thread that resulted).

I know.

-- 
Peter Samuelson
<sampo.creighton.edu!psamuels>

- 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/