From: "Paweł Sikora" <pluto@ds14.agh.edu.pl>
To: linux-kernel@vger.kernel.org
Subject: [2.6.6] (1) broken stack? | (2) broken kbuild-out-of-tree?
Date: Wed, 19 May 2004 19:16:48 +0200 [thread overview]
Message-ID: <200405191916.48687.pluto@ds14.agh.edu.pl> (raw)
[-- Attachment #1: Type: text/plain, Size: 2293 bytes --]
Hi,
i'm a part of the pld-linux (www.pld-linux.org) team,
and i have two problems with 2.6.6 kernel.
1).
kernel + gcc340[-fstack-protector]
patches:
http://cvs.pld-linux.org/cgi-bin/cvsweb/SOURCES/Attic/kernel-ssp.patch?rev=1.1.2.2
http://www.trl.ibm.com/projects/security/ssp/
netfilter/iptables snap from 2004/05/18.
testcase:
[root]-[~] # strace -o iptables.log iptables -L
Kernel panic: propolice detects c0521f25 at function <NULL>.
without -fstack-protector `iptables -L` works fine.
is this a hidden kernel/netfilter bug or a gcc bug?
any ideas?
info:
[root]-[~] # uname -a
Linux vmx 2.6.6-grsec #4 (...) i686 Celeron_(Coppermine) unknown PLD Linux
2).
with the 2.6.[4,5] kernel i use a simple loop (in the .spec files) to build
external kernel modules (based on the different configs (eg. -smp, -uni))...
[ eg. nvidia driver ]
cd usr/src/nv/
cp Makefile.kbuild Makefile
for cfg in %{?with_dist_kernel:%{?with_smp:smp} up} \
%{!?with_dist_kernel:nondist}; do
if [ ! -r "%{_kernelsrcdir}/config-$cfg" ]; then
exit 1
fi
%{__make} -C %{_kernelsrcdir} mrproper \
SUBDIRS=$PWD O=$PWD V=1 \
RCS_FIND_IGNORE="-name built -prune o" \
rm -rf include
install -d include/{linux,config}
ln -sf %{_kernelsrcdir}/config-$cfg .config
ln -sf %{_kernelsrcdir}/include/linux/autoconf-$cfg.h \
include/linux/autoconf.h
touch include/config/MARKER
%{__make} -C %{_kernelsrcdir} modules \
SUBDIRS=$PWD O=$PWD V=1
mv nvidia.ko built/nvidia-$cfg.ko
done
loop works with 2.6.[4,5], but it doesn't work
(after s/make mrproper/make clean/) with 2.6.6.
under the 2.6.6:
- it doesn't build required scripts outside the kernel's tree.
(i must add `make basic_scripts)
- it doesn't build modpost.
- it doesn't build proper modules.
(for -smp and -uni config it produces modules which have the same md5
and eg. nvidia's driver freezes my x11(x.org)
IMO it(kernel's makefile) ignores linked .config).
who and why change kernel's makefile in the 2.6.6 ?
--
If you think of MS-DOS as mono, and Windows as stereo,
then Linux is Dolby Digital and all the music is free...
unofficial PLD(nptl): http://pld-nptl.ds14.agh.edu.pl/
[-- Attachment #2: iptables.log --]
[-- Type: text/x-log, Size: 2312 bytes --]
execve("/usr/sbin/iptables", ["iptables", "-L"], [/* 20 vars */]) = 0
uname({sys="Linux", node="vmx", ...}) = 0
brk(0) = 0x805675c
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x4001c000
open("/etc/ld.so.preload", O_RDONLY) = -1 ENOENT (No such file or directory)
open("/etc/ld.so.cache", O_RDONLY) = 3
fstat64(3, {st_mode=S_IFREG|0644, st_size=60461, ...}) = 0
mmap2(NULL, 60461, PROT_READ, MAP_PRIVATE, 3, 0) = 0x4001d000
close(3) = 0
open("/lib/libdl.so.2", O_RDONLY) = 3
read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0000\36\0"..., 512) = 512
fstat64(3, {st_mode=S_IFREG|0755, st_size=10004, ...}) = 0
mmap2(NULL, 12936, PROT_READ|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0x4002c000
mmap2(0x4002f000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 3, 0x2) = 0x4002f000
close(3) = 0
open("/lib/libnsl.so.1", O_RDONLY) = 3
read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\0<\0\000"..., 512) = 512
fstat64(3, {st_mode=S_IFREG|0755, st_size=81420, ...}) = 0
mmap2(NULL, 88736, PROT_READ|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0x40030000
mmap2(0x40043000, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 3, 0x13) = 0x40043000
mmap2(0x40044000, 6816, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x40044000
close(3) = 0
open("/lib/libc.so.6", O_RDONLY) = 3
read(3, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\340O\1"..., 512) = 512
fstat64(3, {st_mode=S_IFREG|0755, st_size=1398488, ...}) = 0
mmap2(NULL, 1408588, PROT_READ|PROT_EXEC, MAP_PRIVATE, 3, 0) = 0x40046000
mmap2(0x40193000, 36864, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED, 3, 0x14c) = 0x40193000
mmap2(0x4019c000, 7756, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x4019c000
close(3) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x4019e000
set_thread_area({entry_number:-1 -> 6, base_addr:0x4019e4f0, limit:1048575, seg_32bit:1, contents:0, read_exec_only:0, limit_in_pages:1, seg_not_present:0, useable:1}) = 0
munmap(0x4001d000, 60461) = 0
socket(PF_INET, SOCK_RAW, IPPROTO_RAW) = 3
getsockopt(3, SOL_IP, 0x40 /* IP_??? */
reply other threads:[~2004-05-19 17:23 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=200405191916.48687.pluto@ds14.agh.edu.pl \
--to=pluto@ds14.agh.edu.pl \
--cc=linux-kernel@vger.kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Powered by JetHome