Home | Docs | Forums | Lists | Bugs | Planet | Store | GMN | Get Gentoo!
Not eligible to see or edit group visibility for this bug.
View Bug Activity | Format For Printing | XML | Clone This Bug
Building qt-3.3.8 on mips results in uic throwing a SIGBUS at us. /var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/bin/uic -L /var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/plugins pixmapfunction.ui -o pixmapfunction.h /var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/bin/uic -L /var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/plugins createtemplate.ui -o createtemplate.h /var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/bin/uic -L /var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/plugins actioneditor.ui -o actioneditor.h /var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/bin/uic -L /var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/plugins projectsettings.ui -o projectsettings.h /var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/bin/uic -L /var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/plugins finddialog.ui -o finddialog.h /var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/bin/uic -L /var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/plugins replacedialog.ui -o replacedialog.h make[3]: *** [pixmapfunction.h] Bus error make[3]: *** Deleting file `pixmapfunction.h' make[3]: *** Waiting for unfinished jobs.... make[3]: *** [createtemplate.h] Bus error make[3]: *** Deleting file `createtemplate.h' make[3]: Leaving directory `/var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/tools/designer/designer' make[2]: *** [sub-designer] Error 2 make[2]: Leaving directory `/var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/tools/designer' make[1]: *** [sub-designer] Error 2 make[1]: Leaving directory `/var/tmp/portage/x11-libs/qt-3.3.8-r4/work/qt-x11-free-3.3.8/tools' make: *** [sub-tools] Error 2 * * ERROR: x11-libs/qt-3.3.8-r4 failed. The attached patch from Debian was originally used to fix the same error on HPPA, but it works well enough for us. http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=342545#165 Upstream status is unknown, but considering it's Debian and a cornercase arch I bet they've never seen this patch. It fixes a legitimate problem in qlocale.cpp though so they would probably be interested. According to Redhatter, we've been previously working around this by building with -mips1.
Created an attachment (id=143841) [details] qt-3.3.8-mips.patch Fix the bogus assumption in src/tools/qlocale.cpp that a char[] can be cast to a double *. Steve Langasek <vorlon@paer.debian.org>
mneme ~ # emerge --info Portage 2.1.4.4 (default-linux/mips/2007.1-dev/generic-be/o32, gcc-4.1.2, glibc-2.6.1-r0, 2.6.23.14-mipsgit-20080206 mips64) ================================================================= System uname: 2.6.23.14-mipsgit-20080206 mips64 R5000 V2.1 FPU V1.0 Timestamp of tree: Thu, 14 Feb 2008 01:00:04 +0000 distcc 2.18.3 mips-unknown-linux-gnu (protocols 1 and 2) (default port 3632) [enabled] ccache version 2.4 [enabled] app-shells/bash: 3.2_p33 dev-lang/python: 2.5.1-r5 dev-python/pycrypto: 2.0.1-r6 dev-util/ccache: 2.4-r7 sys-apps/baselayout: 1.12.11.1 sys-apps/sandbox: 1.2.18.1-r2 sys-devel/autoconf: 2.61-r1 sys-devel/automake: 1.6.3, 1.7.9-r1, 1.9.6-r2, 1.10.1 sys-devel/binutils: 2.17-r2, 2.18-r1 sys-devel/gcc-config: 1.4.0-r4 sys-devel/libtool: 1.5.26 virtual/os-headers: 2.6.24 ACCEPT_KEYWORDS="mips ~mips" CBUILD="mips-unknown-linux-gnu" CFLAGS="-O2 -march=r5000 -mabi=32 -fomit-frame-pointer -pipe" CHOST="mips-unknown-linux-gnu" CONFIG_PROTECT="/etc" CONFIG_PROTECT_MASK="/etc/env.d /etc/fonts/fonts.conf /etc/gconf /etc/revdep-rebuild /etc/terminfo /etc/udev/rules.d" CXXFLAGS="-O2 -march=r5000 -mabi=32 -fomit-frame-pointer -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoconfig ccache digest distcc distlocks fixpackages metadata-transfer parallel-fetch sfperms splitdebug strict unmerge-orphans userfetch" GENTOO_MIRRORS="http://gentoo.arcticnetwork.ca/source/ http://gentoo.osuosl.org http://distfiles.gentoo.org" LANG="en_US.utf8" LDFLAGS="-Wl,-O1" LINGUAS="en en_US" MAKEOPTS="-j6" PKGDIR="/usr/portage/packages" PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --stats --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/home/dirtyepic/overlay" SYNC="rsync://rsync.gentoo.org/gentoo-portage" USE="X akode bash-completion berkdb bitmap-fonts bzip2 cairo cli cracklib crypt fam fbcon fftw flac gdbm gif gmp gnutls gtk iconv imlib isdnlog jpeg libwww lzo midi mips mp3 ncurses network-cron nptl nptlonly ogg opengl openmp pam pcre pdf perl pic png pppd python qt3 readline reflection sdl session sndfile spell spl sqlite ssl svg tcpd theora threads tiff truetype truetype-fonts type1-fonts unicode urandom vim-syntax vorbis xml xorg xv zlib" ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file hooks iec958 ioplug ladspa lfloat linear meter mulaw multi null plug rate route share shm softvol" APACHE2_MODULES="actions alias auth_basic authn_alias authn_anon authn_dbm authn_default authn_file authz_dbm authz_default authz_groupfile authz_host authz_owner authz_user autoindex cache dav dav_fs dav_lock deflate dir disk_cache env expires ext_filter file_cache filter headers include info log_config logio mem_cache mime mime_magic negotiation rewrite setenvif speling status unique_id userdir usertrack vhost_alias" ELIBC="glibc" INPUT_DEVICES="mouse keyboard" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" LINGUAS="en en_US" USERLAND="GNU" VIDEO_CARDS="fbdev" Unset: CPPFLAGS, CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LC_ALL, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS
It's been building fine on HPPA without that patch. Maybe it's related to a specific gcc version. I can't think of another reason why it just works for Gentoo and not for Debian.
OK, I get the picture now. On Debian's test build hosts, unaligned access results in a SIGBUS being sent and causes a build failure. The bug mentioned in the description explains that the test builds failed for that reason. Gentoo/HPPA kernels do not respond to unaligned access events with a SIGBUS being returned, but simply note it in dmesg quietly and get on with life, like this: Mar 3 20:25:00 elmer fcron[13578]: Job php /var/www/localhost/htdocs/cacti/cmd. php >> /var/log/cacti_cmd 2>&1 started for user root (pid 13579) Mar 3 20:25:00 elmer [60756.184000] php(13580): unaligned access to 0x4005604d at ip=0x004f4ebf Mar 3 20:25:00 elmer [60756.184000] php(13580): unaligned access to 0x4005604d at ip=0x004f5037 Mar 3 20:25:00 elmer [60756.184000] php(13580): unaligned access to 0x4005604e at ip=0x004f48af Mar 3 20:25:00 elmer [60756.184000] php(13580): unaligned access to 0x4005604e at ip=0x004f5037 Mar 3 20:25:04 elmer fcron[13578]: Job php /var/www/localhost/htdocs/cacti/cmd. php >> /var/log/cacti_cmd 2>&1 completed
Applied, thanks :)
