Gentoo Archives: gentoo-embedded

From: "Peter S. Mazinger" <ps.m@×××.net>
To: gentoo-embedded@l.g.o
Cc: Mike Frysinger <vapier@g.o>
Subject: Re: [gentoo-embedded] hardened gcc-4.1 and uclibc
Date: Wed, 04 Oct 2006 07:16:57
Message-Id: Pine.LNX.4.44.0610030811150.23267-100000@lnx.bridge.intra
In Reply to: Re: [gentoo-embedded] hardened gcc-4.1 and uclibc by Natanael Copa
1 On Sun, 24 Sep 2006, Natanael Copa wrote:
2
3 > On Sat, 2006-09-23 at 10:55 -0400, Mike Frysinger wrote:
4 > > On Saturday 23 September 2006 10:40, Natanael Copa wrote:
5 > > > I get this:
6 > > >
7 > > > Calculating dependencies... done!
8 > > > [ebuild NS ] sys-devel/gcc-4.1.1 USE="(-altivec) -bootstrap -build
9 > > > -doc -fortran -gcj -gtk (-hardened) -ip28 -ip32r10k -mudflap (-multilib)
10 > > > -multislot (-n32) (-n64) (-nls) -nocxx -objc -objc++ -objc-gc -test
11 > > > -vanilla" 0 kB
12 > >
13 > > prob because i added gcc/hardened to default-linux/package.use.mask
14 > >
15 > > should be fixed now
16 >
17 > Its still not fixed. Note the (-hardened)
18 >
19 > al-1.5 / # emerge -pv gcc
20 >
21 > These are the packages that would be merged, in order:
22 >
23 > Calculating dependencies... done!
24 > [ebuild NS ] sys-devel/gcc-4.1.1 USE="(-altivec) -bootstrap -build
25 > -doc -fortran -gcj -gtk (-hardened) -ip28 -ip32r10k -mudflap (-multilib)
26 > -multislot (-n32) (-n64) (-nls) -nocxx -objc -objc++ -objc-gc -test
27 > -vanilla" 0 kB
28 >
29 > Total size of downloads: 0 kB
30 > al-1.5 / # emerge -pv \<gcc-4
31 >
32 > These are the packages that would be merged, in order:
33 >
34 > Calculating dependencies... done!
35 > [ebuild R ] sys-devel/gcc-3.4.6-r1 USE="(-altivec) -bootstrap
36 > -boundschecking -build -doc -fortran -gcj -gtk (-hardened*) -ip28
37 > -ip32r10k (-multilib) -multislot (-n32) (-n64) (-nls) -nocxx -nopie
38 > -nossp -objc -test% -vanilla" 12,330 kB
39 >
40 > Total size of downloads: 12,330 kB
41 >
42 >
43 > I removed it from default-linux/package.use.mask manually but the
44 > compile died with a stacksmashing attack. I'm not sure if I should
45 > report it on bugzilla, since it says that its not supported.
46 >
47 > stage1/xgcc -Bstage1/ -B/usr/i386-gentoo-linux-uclibc/bin/ -c -march=i386 -pipe -O2 -fprofile-generate -DIN_GCC -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long -Wno-variadic-macros -Wold-style-definition -Wmissing-format-attribute -DHAVE_CONFIG_H -I. -I. -I/var/tmp/portage/gcc-4.1.1/work/gcc-4.1.1/gcc -I/var/tmp/portage/gcc-4.1.1/work/gcc-4.1.1/gcc/. -I/var/tmp/portage/gcc-4.1.1/work/gcc-4.1.1/gcc/../include -I/var/tmp/portage/gcc-4.1.1/work/gcc-4.1.1/gcc/../libcpp/include genrtl.c -o genrtl.o
48 > stage1/xgcc -Bstage1/ -B/usr/i386-gentoo-linux-uclibc/bin/ -c -march=i386 -pipe -O2 -fprofile-generate -DIN_GCC -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long -Wno-variadic-macros -Wold-style-definition -Wmissing-format-attribute -DHAVE_CONFIG_H -I. -I. -I/var/tmp/portage/gcc-4.1.1/work/gcc-4.1.1/gcc -I/var/tmp/portage/gcc-4.1.1/work/gcc-4.1.1/gcc/. -I/var/tmp/portage/gcc-4.1.1/work/gcc-4.1.1/gcc/../include -I/var/tmp/portage/gcc-4.1.1/work/gcc-4.1.1/gcc/../libcpp/include /var/tmp/portage/gcc-4.1.1/work/gcc-4.1.1/gcc/ggc-common.c -o ggc-common.o
49 > stage1/cc1: stack smashing attack in function ix86_split_to_parts()
50 > xgcc: Internal error: Killed (program cc1)
51 > Please submit a full bug report.
52 > See <URL:http://bugs.gentoo.org/> for instructions.
53 > make[2]: *** [ggc-common.o] Error 1
54 > make[2]: Leaving directory `/var/tmp/portage/gcc-4.1.1/work/build/gcc'
55 > make[1]: *** [stageprofile_build] Error 2
56 > make[1]: Leaving directory `/var/tmp/portage/gcc-4.1.1/work/build/gcc'
57 > make: *** [profiledbootstrap] Error 2
58
59 hardened for gcc-4.x is not done at all (not even for a glibc env),
60 the above bug can be solved though by adapting the patch applied to
61 gcc-3.4, search toolchain.eclass for ix86_split_to_parts (that comment is
62 wrong there and the conditions as well, the patch is needed in any case,
63 if the building compiler is hardened.
64
65 Peter
66
67 --
68 Peter S. Mazinger <ps dot m at gmx dot net> ID: 0xA5F059F2
69 Key fingerprint = 92A4 31E1 56BC 3D5A 2D08 BB6E C389 975E A5F0 59F2
70
71 --
72 gentoo-embedded@g.o mailing list