Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 11 Nov 1995 16:34:53 -0600
From:      rkw@dataplex.net (Richard Wackerbarth)
To:        Terry Lambert <terry@lambert.org>
Cc:        current@freebsd.org, root@deadline.snafu.de, hackers@freebsd.org
Subject:   Re: LKM's still wont compile in -current
Message-ID:  <v02130502accad38d4171@[199.183.109.242]>

next in thread | raw e-mail | index | archive | help
>> >===> lkm
>> >===> lkm/atapi
>> >cc -O2 -m486 -pipe -nostdinc -I. -I/sys -DATAPI -DATAPI_MODULE
>>-DKERNEL -I/usr/src/lkm/atapi/../../sys -W -Wcomment -Wredundant-decls -c
>>/usr/src/lkm/atapi/../../sys/i386/isa/atapi.c
>> >In file included from /sys/sys/conf.h:205,
>> >                 from /usr/src/lkm/atapi/../../sys/i386/isa/atapi.c:912:
>> >./machine/conf.h:14: ioconf.h: No such file or directory
>> >*** Error code 1
>>
>> >This __still__ happens when doing a make world.
>>
>> Install bsd.kmod.mk.
>
>Thank you.  This provides an example of broken code for maintaining
>-stable and -current on the same machine other than the "config" program
>itself.

Now can I sell "you" (the holdouts, not you Terry) on getting rid of the
absolute paths?
EVERYTHING needs to be referenced relative to the ROOT of the build tree.
SRC, OBJ location, TOOLS, INCludes, MKfiles, etc. EVERYTHING.
The default definitions can revert to the rooted definitions for the very
common "special case" of compiling an update to the running system. However
the general case is still that of building a totally different system to be
run on another machine, perhaps even with a different kind of cpu.

----
Richard Wackerbarth
rkw@dataplex.net





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?v02130502accad38d4171>