Date: Sat, 29 Aug 1998 00:45:26 -0400 (EDT) From: Simon Shapiro <shimon@simon-shapiro.org> To: Matthew Jacob <mjacob@feral.com> Cc: Doug Rabson <dfr@nlsystems.com>, freebsd-alpha@FreeBSD.ORG Subject: Re: Alpha Install Message-ID: <XFMail.980829004526.shimon@simon-shapiro.org> In-Reply-To: <Pine.LNX.4.02.9808282032370.7208-100000@feral-gw>
next in thread | previous in thread | raw e-mail | index | archive | help
Matthew Jacob, On 29-Aug-98 you wrote:
>
>
> On Sat, 29 Aug 1998, Simon Shapiro wrote:
>
> >
> > Matthew Jacob, On 29-Aug-98 you wrote:
> > >
> > >
> > > Oh, dear, that should work. I've not installed a FreeBSD alpha
> > > system so I can't check this. ispmbox.h isn't getting included.
> > >
> > > I don't think that __FreeBSD__ must be getting defined in this case.
> >
> > I added to the Makefile (in compile/... - ugly but config is not there
> > yet), -D__FreeBSD__=3. This carries the day a bit further.
> >
> > Also, dod I mention that /usr/share/mk/bsd/kern.mk is missing from
> > NetBSD
> > and the kernel's makefile sort of insists on it?
> >
> > > Under NetBSD this file lives in dev/ic, but under FreeBSD it's
> > > under dev/isp.
> >
> > All the ifdefs are there, but the define is not.
> >
> > > What compiler are you using? If the NetBSD one, you'd better
> > > make some define changes 'coz the isp driver lives in both
> > > worlds (as well as others) and assumes the layouts as driven
> > > by correct defines!
> >
> > It gets uglier than that. In pci/if_de.c there is a cure set of:
> >
> > #ifdef __some_os__
> > #define YUCK "foo/bar.h"
> > ...
> > #endif
> >
> > ...
> >
> > #include YUCK
> >
> >
> > This does not impress the NetBSD compiler one bit. It wuite downright
> > offends the bugger. I ugnly-hacked it to be a bit more explicit.
> >
> > I would have not started this, were I not been told to compile the
> > FreeBSD
> > kernel under NetBSD :-) As such, it may yet compile tonight.
> >
> > Ifyou want access to that Alpha box, send me your vipw passwd line so I
> > add
> > you as a user to the machine, while I still have it.
>
> Oh, thanks- I have plenty of Alphas, but I haven't made the time
> to switch to FreeBSD on one of them. I was spending all of today
> with one rebuild NetWorker/NetBSD/Alpha.
>
> I'll look at it tomorrow.
Good. thanx!
While at it, the next breakages are:
* In alpha/alpha/clock.c some functions are missing prototypes.
* The make soon afterwards blows up in:
sh ../../kern/makedevops.sh -c ; cc -c -O -Wreturn-type -Wcomment
-Wredundant-decls -Wimplicit -Wnested-externs -Wstrict-prototypes
-Wmissing-prototypes -Wpointer-arith -Winline -Wuninitialized -ansi
-nostdinc -I- -I. -I../.. -I../../../include -DKERNEL -include
opt_global.h -U__NetBSD__ -D__FreeBSD__=3 -mno-fp-regs -Wa,-mev56
`basename .m`.c
usage: makedevops.sh [-c|-h] srcfile
*** Error code 1
Looks like parsing the Makefile variable NORMAL_M is not successful.
The Makefile fragment I suspect is:
DRIVER_S= ${CC} -c -x assembler-with-cpp -DLOCORE ${COPTS} $<
PROFILE_C= ${CC} -c ${CFLAGS} ${PARAM} $<
NORMAL_M= sh $S/kern/makedevops.sh -c $<;\
${CC} -c ${CFLAGS} ${PROF} `basename $< .m`.c
# ${ALPHA}/alpha/setdef0.c and ${ALPHA}/alpha/setdef1.c are intentionally
BTW, what is $< ? Never noticed one of these...
Simon
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-alpha" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.980829004526.shimon>