Created attachment 292019 [details] webkit-gtk-1.2.3-fix-pool-sparc.patch.out See subject
I thought it was already dropped from it too :S Please, try to manually edit that ebuild and comment that line to skip the patch and see it it builds or not
(In reply to comment #1) > I thought it was already dropped from it too :S > > Please, try to manually edit that ebuild and comment that line to skip the > patch and see it it builds or not Done it doesn't build: make[1]: *** [Source/JavaScriptCore/wtf/libjavascriptcoregtk_1_0_la-ThreadingPthreads.lo] Error 1 make[1]: *** Waiting for unfinished jobs.... Will attach build.log soon
Created attachment 292115 [details] build.log [ebuild U ~] net-libs/webkit-gtk-1.6.1-r200 [1.2.7] USE="gstreamer* introspection* jit test (-aqua) -coverage -debug -spell% -webgl% (-doc%)" Portage 2.1.10.11 (default/linux/sparc/experimental/multilib/server, gcc-4.5.3, glibc-2.12.2-r0, 2.6.39-gentoo-r3 sparc64) ================================================================= System uname: Linux-2.6.39-gentoo-r3-sparc64-sun4u-with-gentoo-2.0.3 Timestamp of tree: Thu, 10 Nov 2011 07:45:01 +0000 distcc 3.1 sparc-unknown-linux-gnu [disabled] ccache version 2.4 [enabled] app-shells/bash: 4.1_p9 dev-lang/python: 2.7.2-r3 dev-util/ccache: 2.4-r9 dev-util/cmake: 2.8.4-r1 dev-util/pkgconfig: 0.26 sys-apps/baselayout: 2.0.3 sys-apps/openrc: 0.8.3-r1 sys-apps/sandbox: 2.4 sys-devel/autoconf: 2.13, 2.68 sys-devel/automake: 1.11.1 sys-devel/binutils: 2.20.1-r1 sys-devel/gcc: 4.5.3-r1 sys-devel/gcc-config: 1.4.1-r1 sys-devel/libtool: 2.4-r1 sys-devel/make: 3.82-r1 sys-kernel/linux-headers: 2.6.36.1 (virtual/os-headers) sys-libs/glibc: 2.12.2 Repositories: gentoo modified-ebuilds DuPol betagarden ACCEPT_KEYWORDS="sparc" ACCEPT_LICENSE="* -@EULA" CBUILD="sparc-unknown-linux-gnu" CFLAGS="-O2 -mcpu=ultrasparc -pipe -ggdb" CHOST="sparc-unknown-linux-gnu" CONFIG_PROTECT="/etc /usr/share/openvpn/easy-rsa" CONFIG_PROTECT_MASK="/etc/ca-certificates.conf /etc/env.d /etc/fonts/fonts.conf /etc/gconf /etc/gentoo-release /etc/revdep-rebuild /etc/sandbox.d /etc/terminfo" CXXFLAGS="-O2 -mcpu=ultrasparc -pipe -ggdb" DISTDIR="/usr/portage/distfiles" EMERGE_DEFAULT_OPTS="--ask --autounmask=n" FEATURES="assume-digests binpkg-logs ccache collision-protect distlocks ebuild-locks fixlafiles fixpackages multilib-strict news parallel-fetch protect-owned sandbox sfperms splitdebug strict test unknown-features-warn unmerge-logs unmerge-orphans userfetch userpriv usersandbox" FFLAGS="" GENTOO_MIRRORS="ftp://mirror.netcologne.de/gentoo/ ftp://gentoo.tiscali.nl/pub/mirror/gentoo/ ftp://mirror.cambrium.nl/pub/os/linux/gentoo/ ftp://ftp.join.uni-muenster.de/pub/linux/distributions/gentoo ftp://ftp6.uni-muenster.de/pub/linux/distributions/gentoo" LDFLAGS="-Wl,-O1 -Wl,--as-needed" LINGUAS="en de" MAKEOPTS="-j3" PKGDIR="/usr/portage/packages" PORTAGE_COMPRESS="" PORTAGE_CONFIGROOT="/" PORTAGE_RSYNC_EXTRA_OPTS="--timeout=500" 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="/home/portage" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/usr/local/portage/modified /var/lib/layman/DuPol /var/lib/layman/betagarden" SYNC="rsync://rsync.de.gentoo.org/gentoo-portage" USE="X acl admin bash-completion branding bzip2 cli coverpage cracklib crypt cups cxx dbus dri fortran gd git gpm iconv iproute2 javascript jpeg kerberos keyscrub logrotate loop-aes mime modules mudflap multilib nls nptl nptlonly openmp pam pcre png posix pppd readline sasl session sparc ssl sysfs syslog system-sqlite tcpd threads tiff truetype unicode userlocales xml xorg xvfb zlib" ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file hooks iec958 ioplug ladspa lfloat linear meter mmap_emul mulaw multi null plug rate route share shm softvol" APACHE2_MODULES="auth_basic auth_core authn_file authz_core authz_host authz_user dav dir log_config mime unixd" CALLIGRA_FEATURES="kexi words flow plan stage tables krita karbon braindump" CAMERAS="ptp2" COLLECTD_PLUGINS="df interface irq load memory rrdtool swap syslog" ELIBC="glibc" GPSD_PROTOCOLS="ashtech aivdm earthmate evermore fv18 garmin garmintxt gpsclock itrax mtk3301 nmea ntrip navcom oceanserver oldstyle oncore rtcm104v2 rtcm104v3 sirf superstar2 timing tsip tripmate tnt ubx" INPUT_DEVICES="evdev" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" LINGUAS="en de" PHP_TARGETS="php5-3" RUBY_TARGETS="ruby18" USERLAND="GNU" VIDEO_CARDS="mach64" XTABLES_ADDONS="quota2 psd pknock lscan length2 ipv4options ipset ipp2p iface geoip fuzzy condition tee tarpit sysrq steal rawnat logmark ipmark dhcpmac delude chaos account" Unset: CPPFLAGS, CTARGET, INSTALL_MASK, LANG, LC_ALL, PORTAGE_BUNZIP2_COMMAND, PORTAGE_COMPRESS_FLAGS
Probably old pool patch needs to be updated... but this is really a headache and I don't understand why upstream is so slow accepting patches :S Sadly, I don't have time now for taking care of this :(
(In reply to comment #4) > Probably old pool patch needs to be updated... but this is really a headache > and I don't understand why upstream is so slow accepting patches :S > > Sadly, I don't have time now for taking care of this :( This means that as this version and 1.4.3-r200 (bug #389945) don't compile right now, sparc has no usable version for security bug #254209 right now... Please decide, whether sparc keyword needs to be dropped on webkit-gtk with all the negative side-effects.
Can you try this patch? http://patch-tracker.debian.org/patch/series/dl/webkit/1.6.1-5/01_double-conversion.patch
(In reply to comment #6) > Can you try this patch? > http://patch-tracker.debian.org/patch/series/dl/webkit/1.6.1-5/01_double-conversion.patch It applies and compiles until it fails identically to bug #389945. Do you want the build log?
Created attachment 292453 [details, diff] webkit-gtk-1.6.1-__exchange_and_add-sparc.patch (In reply to comment #7) > It applies and compiles until it fails identically to bug #389945. Do you want > the build log? Can you try the attached patch (in addition to the one in comment #6)? I think it might solve the "undefined reference to __sync_fetch_and_add_4" problem.
(In reply to comment #8) > Created attachment 292453 [details, diff] [details, diff] > webkit-gtk-1.6.1-__exchange_and_add-sparc.patch > > (In reply to comment #7) > > It applies and compiles until it fails identically to bug #389945. Do you want > > the build log? > > Can you try the attached patch (in addition to the one in comment #6)? I think > it might solve the "undefined reference to __sync_fetch_and_add_4" problem. Unfortunately, it doesn't slve the problem. Here the relevant snippet from build.log: Tools/DumpRenderTree/Programs_DumpRenderTree-LayoutTestController.o: In function `__exchange_and_add': /usr/lib/gcc/sparc-unknown-linux-gnu/4.5.3/include/g++-v4/ext/atomicity.h:46: undefined reference to `__sync_fetch_and_add_4' /usr/lib/gcc/sparc-unknown-linux-gnu/4.5.3/include/g++-v4/ext/atomicity.h:46: undefined reference to `__sync_fetch_and_add_4' /usr/lib/gcc/sparc-unknown-linux-gnu/4.5.3/include/g++-v4/ext/atomicity.h:46: undefined reference to `__sync_fetch_and_add_4' /usr/lib/gcc/sparc-unknown-linux-gnu/4.5.3/include/g++-v4/ext/atomicity.h:46: undefined reference to `__sync_fetch_and_add_4' /usr/lib/gcc/sparc-unknown-linux-gnu/4.5.3/include/g++-v4/ext/atomicity.h:46: undefined reference to `__sync_fetch_and_add_4' Tools/DumpRenderTree/Programs_DumpRenderTree-LayoutTestController.o:/usr/lib/gcc/sparc-unknown-linux-gnu/4.5.3/include/g++-v4/ext/atomicity.h:46: more undefined references to `__sync_fetch_and_add_4 ' follow collect2: ld returned 1 exit status make[1]: *** [Programs/DumpRenderTree] Error 1 make[1]: *** Waiting for unfinished jobs.... If you want the complete build log, please ask for it. I need to compress it as it is too large to attach directly.
Maybe openbsd patches could help: http://www.openbsd.org/cgi-bin/cvsweb/ports/www/webkit/patches/ But I don't have a sparc machine to test :S
*** Bug 389945 has been marked as a duplicate of this bug. ***
(In reply to comment #10) > Maybe openbsd patches could help: > http://www.openbsd.org/cgi-bin/cvsweb/ports/www/webkit/patches/ > > But I don't have a sparc machine to test :S Just a quick note: I am trying with those patches, but up to now with no avail. Still failing the same way as reported before...
I guess you should try to build webkit-1.6.1 manually and, if still failing, report to upstream, as I have failed to find any other people supplying webkit for sparc
Here is another one for sparc and sparc64 :/ http://cvsweb.se.netbsd.org/cgi-bin/bsdweb.cgi/pkgsrc/www/webkit-gtk/patches/patch-ag?rev=1.4;content-type=text%2Fplain http://cvsweb.se.netbsd.org/cgi-bin/bsdweb.cgi/pkgsrc/www/webkit-gtk/patches/patch-ah?rev=1.4;content-type=text%2Fplain
I have narrowed the problem down: webkit-gtk-1.4.3-r200 builds as is with gcc-4.4.5 but fails with gcc-4.5.3-r1 :-/ (In reply to comment #14) > Here is another one for sparc and sparc64 :/ > http://cvsweb.se.netbsd.org/cgi-bin/bsdweb.cgi/pkgsrc/www/webkit-gtk/patches/patch-ag?rev=1.4;content-type=text%2Fplain > http://cvsweb.se.netbsd.org/cgi-bin/bsdweb.cgi/pkgsrc/www/webkit-gtk/patches/patch-ah?rev=1.4;content-type=text%2Fplain I can try with these as well, but maybe you want talk to members of toolchain herd first. Either gcc-4.5.3-r1 is broken or stricter compared to 4.4.5 (though I doubt that)
Will CC tolchain here then
Maybe would also be interesting to test with gcc-4.6
(In reply to comment #15) > I have narrowed the problem down: > webkit-gtk-1.4.3-r200 builds as is with gcc-4.4.5 but fails with gcc-4.5.3-r1 > :-/ > > (In reply to comment #14) > > Here is another one for sparc and sparc64 :/ > > http://cvsweb.se.netbsd.org/cgi-bin/bsdweb.cgi/pkgsrc/www/webkit-gtk/patches/patch-ag?rev=1.4;content-type=text%2Fplain > > http://cvsweb.se.netbsd.org/cgi-bin/bsdweb.cgi/pkgsrc/www/webkit-gtk/patches/patch-ah?rev=1.4;content-type=text%2Fplain > > I can try with these as well, but maybe you want talk to members of toolchain > herd first. Either gcc-4.5.3-r1 is broken or stricter compared to 4.4.5 (though > I doubt that) Anyway, try these patches also if possible please Thanks
One of the openbsd patches talk about an overoptimising bug on sparc64, maybe you can also try without "-O2"...
(In reply to comment #18) > Anyway, try these patches also if possible please (In reply to comment #19) > One of the openbsd patches talk about an overoptimising bug on sparc64, maybe > you can also try without "-O2"... I will try, but this will take a while. Please don't expect results before end of the week. Thanks.
Any news here please? Thanks
(In reply to comment #21) > Any news here please? Thanks I am still on gcc-4.4.5 and so far webkit-gtk-1.6.1-r201 compiles but the test suite fails. If I remember correctly, there is still a warning about increased required alignment of target type... I will attach the build log with -O2 later if it is failing as well.
(In reply to comment #22) > (In reply to comment #21) > > Any news here please? Thanks > > I am still on gcc-4.4.5 and so far webkit-gtk-1.6.1-r201 compiles but the test > suite fails. > If I remember correctly, there is still a warning about increased required > alignment of target type... > > I will attach the build log with -O2 later if it is failing as well. This is of course without -O2. Unfortunately, the box ran out of memory without -O2 while linking libwebkitgtk-1.0.la. I need to enlarge the swap partition but this takes a while because I need physical access to the box.
(In reply to comment #22) > (In reply to comment #21) > > Any news here please? Thanks > > I am still on gcc-4.4.5 and so far webkit-gtk-1.6.1-r201 compiles but the test > suite fails. > If I remember correctly, there is still a warning about increased required > alignment of target type... > > I will attach the build log with -O2 later if it is failing as well. I opened bug #399153 for the test failure. Will now try without -O2 and gcc-4.5.3-r1 and see whether it compiles.
The failure of undefined reference to `__sync_fetch_and_add_4' is due to this in the ebuild: # Sigbuses on SPARC with mcpu and co. use sparc && filter-flags "-mcpu=*" "-mvis" "-mtune=*" For some reason, in our stages we have -mcpu=ultrasparc. With that line, we filter out mcpu, meaning we use the default mcpu which is not a sparc64, which doesn't have those sync_fetch...functions. Removing that line makes webkit-gtk compile...but it sigbuses(unaligned accesses)...guess it needs more debugging. I'll check if debian has the same issue when i have time
This patch fixes it: http://patch-tracker.debian.org/patch/series/dl/webkit/1.6.3-2/05_sparc-needs-alignment.patch So please, remove the mcpu, mtune filtering, feel free to keep the mvis one, and apply this patch. Please do it for both revisions of 1.6.1 so i can do bug 393007, and also do it for 1.6.3
(In reply to comment #25) (In reply to comment #26) > 17 Mar 2012; Alexandre Rostovtsev <tetromino@gentoo.org> > webkit-gtk-1.6.1-r201.ebuild, webkit-gtk-1.6.1-r301.ebuild, > +files/webkit-gtk-1.6.1-sparc-needs-alignment.patch, > webkit-gtk-1.6.3-r200.ebuild, webkit-gtk-1.6.3-r300.ebuild: > Sparc fix by Raúl Porcel for bug #389963 Thank you for your investigation! The patch is applied and "-mcpu=*" and "-mtune=*" filtering is removed in all 1.6.x ebuilds. Can this bug be closed now?
Yes, I guess Alexandre, is upstream aware of this bug? Thanks :)
(In reply to comment #28) > Alexandre, is upstream aware of this bug? Probably, since the Debian patch was already applied upstream in webkit-gtk-1.7.x.
