Re: Debian on Fuloong (was: porterboxes vs. qemu)
- To: Daniel Clark <dclark@pobox.com>, fxzhang <fxzhang@ict.ac.cn>, debian-mips@lists.debian.org, Wu Zhangjin <wuzhangjin@gmail.com>, Dilinger <dilinger@queued.net>, liushiwei@gmail.com
- Subject: Re: Debian on Fuloong (was: porterboxes vs. qemu)
- From: Andreas Barth <aba@not.so.argh.org>
- Date: Wed, 10 Feb 2010 23:29:08 +0100
- Message-id: <20100210222908.GR27216@mails.so.argh.org>
- Mail-followup-to: Andreas Barth <aba@not.so.argh.org>, Daniel Clark <dclark@pobox.com>, fxzhang <fxzhang@ict.ac.cn>, debian-mips@lists.debian.org, Wu Zhangjin <wuzhangjin@gmail.com>, Dilinger <dilinger@queued.net>, liushiwei@gmail.com
- In-reply-to: <20100207181146.GH27216@mails.so.argh.org>
- References: <658802.61908.qm@web33008.mail.mud.yahoo.com> <20100204082720.6142b0de@epicfail.kgoetz.id.au> <4B6A3835.1000501@ict.ac.cn> <20100207013916.GA11393@mails.so.argh.org> <20100207151707.GA14886@mails.so.argh.org> <968838231002070804u3c5eda36u4efa1d91fbb5c22f@mail.gmail.com> <20100207181146.GH27216@mails.so.argh.org>
* Andreas Barth (aba@not.so.argh.org) [100207 19:11]:
> > You need to use Lemote's git repo for 2.6.32
> > -http://dev.lemote.com/code/rt4ls - I believe as of 2.6.33 all the
> > (important at least) changes have made it into mainline.
>
> I have tried it with an 2.6.33-rc7-kernel with these configs:
> http://alius.ayous.org/~aba/config-2.6.33-rc7-dsa-mipsel-lemote
> http://alius.ayous.org/~aba/config-2.6.33-rc7-dsa-mipsel-lemote.1
>
> This kernel ends at the moment where it displays the registers at the
> vga screen.
Based on the kernel config from
ftp://www.bjlx.org.cn/loongson/squeeze/kernel/linux-image-2.6.32-fuloong2e-bfs_2.6.32.3-rt4ls-20100108_mipsel.deb
I compiled an 2.6.33-rc7+-kernel that booted successfuly. The kernel
config is on http://alius.ayous.org/~aba/config-2.6.33-rc7-dsa-mipsel-lemote.2
using the git tree from from
git://git.linux-mips.org/pub/scm/linux-queue.git. I'll now try with a
"default" upstream kernel as well.
> The machine works with the heavily patched 2.6.31 kernel (which has a
> few patches from lemote). However, with that kernel (built with both
> new binutils options, used as -Wa,.. and -Wl,...) the machine stops
> at always the same step while building binutils in a standard debian
> unstable chroot:
> make[3]: Entering directory `/build/buildd-binutils_2.20-5~mipselaba.1-mipsel-1AL2Gb/binutils-2.20/builddir-single/libiberty'
> if [ x"-fPIC" != x ] && [ ! -d pic ]; then \
> mkdir pic; \
> else true; fi
> touch stamp-picdir
> if [ x"-fPIC" != x ]; then \
> gcc -c -DHAVE_CONFIG_H -g -O2 -I. -I../../libiberty/../include -W -Wall -Wwrite-strings -Wc++-compat -Wstrict-prototypes -pedantic -fPIC ../../libiberty/regex.c -o pic/regex.o; \
> else true; fi
Running that kernel, I also get one more command, till:
if [ x"-fPIC" != x ] && [ ! -d pic ]; then \
mkdir pic; \
else true; fi
touch stamp-picdir
if [ x"-fPIC" != x ]; then \
gcc -c -DHAVE_CONFIG_H -g -O2 -I. -I../../libiberty/../include -W -Wall -Wwrite-strings -Wc++-compat -Wstrict-prototypes -pedantic -fPIC ../../libiberty/regex.c -o pic/regex.o; \
else true; fi
gcc -c -DHAVE_CONFIG_H -g -O2 -I. -I../../libiberty/../include -W -Wall -Wwrite-strings -Wc++-compat -Wstrict-prototypes -pedantic ../../libiberty/regex.c -o regex.o
if [ x"-fPIC" != x ]; then \
gcc -c -DHAVE_CONFIG_H -g -O2 -I. -I../../libiberty/../include -W -Wall -Wwrite-strings -Wc++-compat -Wstrict-prototypes -pedantic -fPIC ../../libiberty/cplus-dem.c -o pic/cplus-dem.o; \
else true; fi
> For building any of the kernels, I used -mfix-loongson2f-nop and
> -mfix-loongson2f-jump, built with binutils patched with
> http://sourceware.org/ml/binutils/2009-11/msg00387.html .
For review, I put the patched binutils on
http://alius.ayous.org/~aba/binutils_2.20-5~mipselaba.1.dsc
> Do I need to use -mfix-ls2f-kernel from
> http://archlinux-for-loongson.googlegroups.com/web/binutils-2.19.1-loongson2f-3.patch
> as well? I understood it as that patch is already included in the
> newer binutils patch.
Do I need that patch? Do I miss something else?
Cheers,
Andi
Reply to: