From: Henry Nestler <Henry.Ne@Arcor.de>
To: linux-kernel@vger.kernel.org
Cc: Roland McGrath <roland@redhat.com>, Sam Ravnborg <sam@ravnborg.org>
Subject: kbuild: LDFLAGS_MODULE unusable for external module builds (2.6.23-rc2)
Date: Tue, 25 Sep 2007 11:49:44 +0200 [thread overview]
Message-ID: <46F8D9B8.3090806@Arcor.de> (raw)
In reference of git 114f51577724b782a30f4f5ceaee9880de93d776:
> kbuild: use LDFLAGS_MODULE only for .ko links
>
> Sam Ravnborg pointed out that Documentation/kbuild/makefiles.txt already
> says this is what it's for. This patch makes the reality live up to the
> documentation. This fixes the problem of LDFLAGS_BUILD_ID getting into too
> many places.
LDFLAGS_MODULE is not usable in module build out of kernel tree since
2.6.23-rc2. LDFLAGS_$@ should use, but does never work. - Not for
external module builds with the option "M=..."
What macro should set for linker parameters of foo.o ? I'm not shure.
Currently found, that only LDFLAGS is usable. But, found LDFLAGS in
modpost calls, there also exist a --start-group/--end-group. Is it right
to use it for external module?
Here examples and the outputs. Host and target is an i686 SMP Linux.
mkdir /tmp/test; cd /tmp/test
cat >Makefile <<EOF
obj-m := name.o
name-objs := foo.o
# Use multipe defines, hope any would work
LDFLAGS_MODULE += --start-group
LDFLAGS_name += --start-group
LDFLAGS_name.o += --start-group
EOF
cat >foo.c <<EOF
#include <linux/module.h>
#include <linux/init.h>
int init_module(void) {return 0;}
void cleanup_module(void) {return;}
EOF
#2.6.22 works. LDFLAGS_MODULE was used:
make V=1 -C src/linux-2.6.22 M=$PWD | grep " ld "
ld -m elf_i386 -m elf_i386 -r --start-group -o /tmp/name.o /tmp/foo.o
ld -m elf_i386 -m elf_i386 -r -o /tmp/name.ko /tmp/name.o /tmp/name.mod.o
#2.6.23-rc7 does not work. Nothing from LDFLAGS can seen in linker line:
make V=1 -C src/linux-2.6.23-rc7 M=$PWD | grep " ld "
ld -m elf_i386 -m elf_i386 -r -o /tmp/name.o /tmp/foo.o
ld -m elf_i386 -r -m elf_i386 -o /tmp/name.ko /tmp/name.o
/tmp/name.mod.o
------
Perhaps anybody can give better idea for Makefile with
recursive depens of libraries? What I'm missing, is a setup for
--start-group AND the --end-group arrount the lib.a files. Currently
--start-groups is somether before the first object and has no --end-group.
obj-y := dir1/ dir2/ dir3/
obj-m := name.ko
name-objs = foo.o dir1/lib.a dir2/lib.a dir3/lib.a
LDFLAGS += --start-group # Deal with recursive depens of libraries
--
Henry
next reply other threads:[~2007-09-25 9:49 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-09-25 9:49 Henry Nestler [this message]
2007-09-25 19:22 ` kbuild: LDFLAGS_MODULE unusable for external module builds (2.6.23-rc2) Sam Ravnborg
2007-09-26 9:56 ` Henry Nestler
2007-09-26 11:01 ` Sam Ravnborg
2007-09-26 11:31 ` Henry Nestler
2007-09-27 8:21 ` Henry Nestler
2007-09-27 8:26 ` Sam Ravnborg
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=46F8D9B8.3090806@Arcor.de \
--to=henry.ne@arcor.de \
--cc=linux-kernel@vger.kernel.org \
--cc=roland@redhat.com \
--cc=sam@ravnborg.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.