Summary: | tpm-emulator-0.3 has linking trouble with hardened libgmp | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Daniel Black (RETIRED) <dragonheart> |
Component: | [OLD] Unspecified | Assignee: | Daniel Black (RETIRED) <dragonheart> |
Status: | VERIFIED WONTFIX | ||
Severity: | normal | CC: | pageexec, solar |
Priority: | High | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
URL: | https://developer.berlios.de/bugs/?func=detailbug&bug_id=8055&group_id=2491 | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Daniel Black (RETIRED)
2006-06-30 06:30:26 UTC
$ emerge --info Portage 2.1.1_pre1-r5 (!/home/dan/gentoo/gentoo-x86/profiles/hardened/x86/2.6, gcc-4.1.0/hardened, glibc-2.3.6-r4, 2.6.16-gentoo-r9 i686) ================================================================= System uname: 2.6.16-gentoo-r9 i686 AMD Athlon(tm) Gentoo Base System version 1.12.1 distcc 2.18.3 i686-pc-linux-gnu (protocols 1 and 2) (default port 3632) [disabled] ccache version 2.3 [disabled] dev-lang/python: 2.3.5-r2, 2.4.3-r1 dev-python/pycrypto: 2.0.1-r5 dev-util/ccache: 2.3 dev-util/confcache: [Not Present] sys-apps/sandbox: 1.2.18.1 sys-devel/autoconf: 2.13, 2.59-r7 sys-devel/automake: 1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2 sys-devel/binutils: 2.17 sys-devel/gcc-config: 2.0.0_rc1 sys-devel/libtool: 1.5.22 virtual/os-headers: 2.6.11-r5 ACCEPT_KEYWORDS="x86 ~x86" AUTOCLEAN="yes" CBUILD="i686-pc-linux-gnu" CFLAGS="-march=athlon-xp -O2 -pipe" CHOST="i686-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/kde/3.5/env /usr/kde/3.5/share/config /usr/kde/3.5/shutdown /usr/lib/X11/xkb /usr/lib/mozilla/defaults/pref /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/" CONFIG_PROTECT_MASK="/etc/env.d /etc/eselect/compiler /etc/gconf /etc/revdep-rebuild /etc/terminfo" CXXFLAGS="-march=athlon-xp -O2 -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoconfig collision-protect cvs digest keepwork metadata-transfer nostrip sandbox sfperms sign strict stricter test userpriv usersandbox verify-rdepend" GENTOO_MIRRORS="" LDFLAGS=" -Wl,--as-needed" MAKEOPTS="-j3" PKGDIR="/usr/portage/packages" PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --delete-after --stats --timeout=180 --exclude='/distfiles' --exclude='/local' --exclude='/packages'" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/home/dan/gentoo/gentoo-x86" SYNC="rsync://rsync.au.gentoo.org/gentoo-portage" USE="3dnow 3dnowex X aac acpi alsa avi bash-completion berkdb bitmap-fonts bzip2 caps crypt cups curl divx4linux dlloader ecc encode ethereal extensions fastcgi fla flac fortran gd gmp gnutls gphoto2 gstreamer gtk gtk2 hardened hpn i8x0 ilbc imagemagick imlib ipv6 javacomm jpeg kde kdeenablefinal kerberos libgda lzo mbox mmx mp3 mpeg multitarget mysql nptl nptlonly ntlm ogg oggvorbis openal opengl operanom2 pam perl php pic png postgres ppds python qt quicktime readline samba sdl sguil slp smime socks5 sox spell sse ssl tiff true-type truetype type1-fonts usb userlocales vhosts vorbis x86 xine xinerama xml xml2 xorg zlib elibc_glibc input_devices_mouse input_devices_keyboard kernel_linux userland_GNU" Unset: CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL, LINGUAS, PORTAGE_RSYNC_EXTRA_OPTS hmm, i thought hardened had some provision for compiling kernel modules so that ssp would be disabled... solar? There is no such thing as a hardened-gcc-4.x compiler so whatever dragonheart is encountering is a result of old libs leaking into his non-hardened gcc-4.x setup. Also yes hardened prevents ssp from being enabled on kernel code using a {!D__KERNEL__: macro The root problem here has to be that the module is stupid and is linking with userland code when it should not. (In reply to comment #3) > The root problem here has to be that the module is stupid and is linking > with userland code when it should not. yeah, on a second look i agree, this is an upstream issue, using userland code in the kernel is a big no-no (think of potential issues like PIC, TLS, and here, ssp). might be safer to simply mask this package altogether until the code is properly fixed. reported upstream as per URL. Thanks people. I didn't think of the PIC or TLS problems (because I don't understand them enough). I've done a check for SSP in the ebuild. I probably won't mask this as it may work for most people. I'll look at fixing the problem when I do 0.4 (just released) by recompiling the libgmp build with -D__KERNEL__ (if upstream haven't fixed it). opps ment to close as UPSTREAM. |