From: Antoine Martin <antoine@nagafix.co.uk>
To: Blaisorblade <blaisorblade@yahoo.it>
Cc: uml@nagafix.co.uk,
UML devel <user-mode-linux-devel@lists.sourceforge.net>
Subject: [uml-devel] [PATCH] Re: 2.4.32-bs2
Date: Mon, 12 Dec 2005 02:24:26 +0000 [thread overview]
Message-ID: <1134354266.29209.116.camel@localhost.localdomain> (raw)
In-Reply-To: <200512112043.26863.blaisorblade@yahoo.it>
[-- Attachment #1: Type: text/plain, Size: 1988 bytes --]
> > I have no idea what I've done different!
I was using mrproper, but I've scripted the whole thing to test all the
combinations from scratch again (with a brand new tree each time - not
just mrproper) and the error only occurs when using the hardened gcc,
and I'm not too bothered about that (it's easy enough to switch, except
in SELinux enforcing mode - but that's another matter...)
So I guess that the patch is good enough as it is.
It is available here:
http://uml.nagafix.co.uk/uml-2.4.32-bs2.patch.bz2
it cumulates:
* 2.4.28-bs2
* offsets for 2.4.32 and removed the whitespace no-change in
drivers/char/tty_io.c.rej
* tiny patch (attached) for tt/Makefile to cope with duplicate symbols
in newer glibc
(done properly this time)
* mconsole-exec patch backported to 2.4:
http://uml.nagafix.co.uk/mconsole-exec.patch-2.4.bz2
(unlike in 2.6 there is no config option for it)
> > How would I go about building a 32-bit pcap enabled kernel in the
> > chroot? Do I copy the pcap.a from outside the chroot?
>
> No, pcap.a is a static library so you don't need that. The problem is that you
> need a matching glibc inside the chroot and outside for the libraries which
> are dlopen()ed by glibc, i.e. the whole libnss + helpers called
> by /etc/nsswitch.conf for instance. On my system the linker spits out a
> warning to say that ("You must have the same glibc used for compiling, even
> if you're linking statically").
Same here. I wish it could tell me which ones so I could place them in
the chroot without guessing or tracing it (we have discussed this a
while back)
So in this case, on a multilib amd64 system, I should be able to build a
chroot using a bind mount to preserve /lib32 and /usr/lib32 to guarantee
that the libs are exactly the same? (and add a 32bit bash, gcc, make...)
> > I'll send a patch when the whole thing works.
>
> The patch is for review, and please don't send any whole patch on the ML -
Sure, I meant just the diffs with 2.4.28-bs2!
Antoine
[-- Attachment #2: 2.4.32-bs2-tt-makefile.patch --]
[-- Type: text/x-patch, Size: 617 bytes --]
--- linux-2.4.32/arch/um/kernel/tt/Makefile 2005-12-13 03:29:03.000000000 +0000
+++ linux-2.4.32-bs2/arch/um/kernel/tt/Makefile 2005-12-13 03:27:52.000000000 +0000
@@ -26,7 +26,7 @@
include $(TOPDIR)/Rules.make
$(USER_OBJS) : %.o: %.c
- $(CC) $(CFLAGS_$@) $(USER_CFLAGS) -c -o $@ $<
+ $(CC) $(USER_CFLAGS) $(CFLAGS_$@) -c -o $@ $<
$(O_TARGET) : unmap_fin.o
@@ -34,6 +34,7 @@
$(CC) $(UNMAP_CFLAGS) -c -o $@ $<
unmap_fin.o : unmap.o
- ld -r -o $@ $< -lc -L/usr/lib
+ ld -r -o unmap_tmp.o unmap.o -lc -L/usr/lib
+ objcopy unmap_tmp.o unmap_fin.o -G switcheroo
clean :
prev parent reply other threads:[~2005-12-12 2:24 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-12-08 2:10 [uml-devel] 2.4.32-bs2 Antoine Martin
2005-12-11 16:47 ` [uml-devel] 2.4.32-bs2 Blaisorblade
2005-12-11 19:19 ` Antoine Martin
2005-12-11 19:43 ` Blaisorblade
2005-12-12 2:24 ` Antoine Martin [this message]
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=1134354266.29209.116.camel@localhost.localdomain \
--to=antoine@nagafix.co.uk \
--cc=blaisorblade@yahoo.it \
--cc=uml@nagafix.co.uk \
--cc=user-mode-linux-devel@lists.sourceforge.net \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox