make[1]: *** [Makefile:106: install-symbolic-link] Segmentation fault It's sln that fails: #0 0x00000344 in ?? () #1 0x00011740 in __libc_setup_tls () at libc-tls.c:180 #2 0x00011174 in __libc_start_main (main=0x230, argc=1, argv=0xffffcdb4, init=0x218, fini=0xf8, rtld_fini=0x0, stack_end=0xffffceee) at libc-start.c:189 #3 0x00010654 in _start () at ../sysdeps/sparc/sparc32/start.S:87 180 memcpy (_dl_static_dtv[2].pointer.val, initimage, filesz); (gdb) p initimage $1 = (void *) 0xb6d08 (gdb) p filesz $2 = 16 (gdb) p _dl_static_dtv $3 = 0xb9dd4 <_dl_static_dtv> (gdb) p _dl_static_dtv[0] $5 = {counter = 62, pointer = {val = 0x3e, to_free = 0x0}} (gdb) p _dl_static_dtv[1] $6 = {counter = 0, pointer = {val = 0x0, to_free = 0x0}} (gdb) p _dl_static_dtv[2] $7 = {counter = 772100, pointer = {val = 0xbc804, to_free = 0x0}} (gdb) p _dl_static_dtv[3] $8 = {counter = 0, pointer = {val = 0x0, to_free = 0x0}} (gdb) p _dl_static_dtv[4] $9 = {counter = 0, pointer = {val = 0x0, to_free = 0x0}} (gdb) x/4xh 0xb6d08 0xb6d08: 0x000b 0x8414 0x000b 0x8414 Portage 2.3.8 (python 3.5.4-final-0, default/linux/sparc/17.0, gcc-6.4.0, glibc-2.25-r9, 4.14.2 sparc64) ================================================================= System Settings ================================================================= System uname: Linux-4.14.2-sparc64-sun4v-with-gentoo-2.3 KiB Mem: 33133616 total, 8210048 free KiB Swap: 0 total, 0 free Timestamp of repository gentoo: Wed, 13 Dec 2017 13:15:01 +0000 Head commit of repository gentoo: 8c957bd55d3acac6af02f13cc9b526ad6f1029b9 sh bash 4.3_p48-r1 ld GNU ld (Gentoo 2.28.1 p1.0) 2.28.1 app-shells/bash: 4.3_p48-r1::gentoo dev-lang/perl: 5.24.3::gentoo dev-lang/python: 2.7.14::gentoo, 3.4.5::gentoo, 3.5.4-r1::gentoo dev-util/pkgconfig: 0.29.2::gentoo sys-apps/baselayout: 2.3::gentoo sys-apps/openrc: 0.34.11::gentoo sys-apps/sandbox: 2.10-r4::gentoo sys-devel/autoconf: 2.69::gentoo sys-devel/automake: 1.15.1-r1::gentoo sys-devel/binutils: 2.28.1::gentoo sys-devel/gcc: 5.4.0-r3::gentoo, 6.4.0::gentoo sys-devel/gcc-config: 1.8-r1::gentoo sys-devel/libtool: 2.4.6-r3::gentoo sys-devel/make: 4.2.1::gentoo sys-kernel/linux-headers: 4.4::gentoo (virtual/os-headers) sys-libs/glibc: 2.25-r9::gentoo Repositories: gentoo location: /usr/portage sync-type: rsync sync-uri: rsync://rsync.gentoo.org/gentoo-portage priority: -1000 ACCEPT_KEYWORDS="sparc" ACCEPT_LICENSE="* -@EULA" CBUILD="sparc-unknown-linux-gnu" CFLAGS="-O2 -mcpu=niagara2 -pipe" CHOST="sparc-unknown-linux-gnu" CONFIG_PROTECT="/etc /usr/share/gnupg/qualified.txt" CONFIG_PROTECT_MASK="/etc/ca-certificates.conf /etc/dconf /etc/env.d /etc/fonts/fonts.conf /etc/gconf /etc/gentoo-release /etc/revdep-rebuild /etc/sandbox.d /etc/terminfo /etc/texmf/language.dat.d /etc/texmf/language.def.d /etc/texmf/updmap.d /etc/texmf/web2c" CXXFLAGS="-O2 -mcpu=niagara2 -pipe" DISTDIR="/usr/portage/distfiles" FCFLAGS="" FEATURES="assume-digests binpkg-logs config-protect-if-modified distlocks ebuild-locks fixlafiles merge-sync news parallel-fetch preserve-libs protect-owned sandbox sfperms strict unknown-features-warn unmerge-logs unmerge-orphans userfetch userpriv usersandbox usersync xattr" FFLAGS="" GENTOO_MIRRORS="ftp://ftp.wh2.tu-dresden.de/pub/mirrors/gentoo ftp://ftp.join.uni-muenster.de/pub/linux/distributions/gentoo" LANG="en_US.UTF-8" LDFLAGS="-Wl,-O1 -Wl,--as-needed" MAKEOPTS="-j20" PKGDIR="/usr/portage/packages" PORTAGE_CONFIGROOT="/" PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --omit-dir-times --compress --force --whole-file --delete --stats --human-readable --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages --exclude=/.git" PORTAGE_TMPDIR="/var/tmp" USE="acl berkdb big-endian bzip2 cli cracklib crypt cxx dri fortran gdbm iconv ipv6 modules ncurses nls nptl openmp pam pcre readline session sparc ssl tcpd unicode xattr zlib" APACHE2_MODULES="authn_core authz_core socache_shmcb unixd 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 cgi cgid 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" CALLIGRA_FEATURES="kexi words flow plan sheets stage tables krita karbon braindump author" COLLECTD_PLUGINS="df interface irq load memory rrdtool swap syslog" ELIBC="glibc" GPSD_PROTOCOLS="ashtech aivdm earthmate evermore fv18 garmin garmintxt gpsclock isync itrax mtk3301 nmea ntrip navcom oceanserver oldstyle oncore rtcm104v2 rtcm104v3 sirf skytraq superstar2 timing tsip tripmate tnt ublox ubx" INPUT_DEVICES="libinput keyboard mouse" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" LIBREOFFICE_EXTENSIONS="presenter-console presenter-minimizer" OFFICE_IMPLEMENTATION="libreoffice" PHP_TARGETS="php5-6 php7-0" POSTGRES_TARGETS="postgres9_5" PYTHON_SINGLE_TARGET="python3_5" PYTHON_TARGETS="python2_7 python3_5" RUBY_TARGETS="ruby22" USERLAND="GNU" VIDEO_CARDS="fbdev glint mga r128 radeon dummy v4l" 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: CC, CPPFLAGS, CTARGET, CXX, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LC_ALL, PORTAGE_BUNZIP2_COMMAND, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS ================================================================= Package Settings ================================================================= sys-libs/glibc-2.25-r9::gentoo was built with the following: USE="nscd rpc -audit -caps -debug -gd (-hardened) (-multilib) -profile (-selinux) -suid -systemtap (-vanilla)" CFLAGS="-mcpu=niagara2 -pipe -fcall-used-g6 -O2 -fno-strict-aliasing" CXXFLAGS="-mcpu=niagara2 -pipe -fcall-used-g6 -O2 -fno-strict-aliasing" Reproducible: Always
Created attachment 509968 [details] build.log
The primary suspect is gcc[pie] introduced by 17.0 profile.
I too am seeing sln segfaulting during the install phase. I tried the usual -j1 & -sandbox since those have given me trouble in the past. So at least another report of failure.
(In reply to James McMechan from comment #3) > I too am seeing sln segfaulting during the install phase. > I tried the usual -j1 & -sandbox since those have given me trouble in the > past. > So at least another report of failure. Can you post 'emerge --info' for affected system?
Created attachment 510224 [details] emerge info for a T5140 attached emerge info I have been trying various things. -j1, -sandbox, binutils-2.29.1-r1 but even at -j64 it takes about 25min
Same problem here, after switching to the 17.0 profile. Building with -j1 gives the same outcome. This is a T5120 box. # emerge --info Portage 2.3.8 (python 2.7.14-final-0, default/linux/sparc/17.0, gcc-6.4.0, glibc-2.25-r9, 4.9.49-gentoo-r1 sparc64) ================================================================= System uname: Linux-4.9.49-gentoo-r1-sparc64-sun4v-with-gentoo-2.3 KiB Mem: 8150424 total, 4470120 free KiB Swap: 48184 total, 48184 free Timestamp of repository gentoo: Wed, 13 Dec 2017 05:45:01 +0000 Head commit of repository gentoo: 618a5ee9ca432020e1ce66f1724611b456029bfe sh bash 4.3_p48-r1 ld GNU ld (Gentoo 2.28.1 p1.0) 2.28.1 app-shells/bash: 4.3_p48-r1::gentoo dev-lang/perl: 5.24.3::gentoo dev-lang/python: 2.7.14-r1::gentoo, 3.4.5-r1::gentoo, 3.5.4-r1::gentoo dev-util/cmake: 3.8.2::gentoo dev-util/pkgconfig: 0.29.2::gentoo sys-apps/baselayout: 2.3::gentoo sys-apps/openrc: 0.34.11::gentoo sys-apps/sandbox: 2.10-r4::gentoo sys-devel/autoconf: 2.69::gentoo sys-devel/automake: 1.15.1-r1::gentoo sys-devel/binutils: 2.28.1::gentoo sys-devel/gcc: 6.4.0::gentoo, 7.2.0::gentoo sys-devel/gcc-config: 1.8-r1::gentoo sys-devel/libtool: 2.4.6-r3::gentoo sys-devel/make: 4.2.1::gentoo sys-kernel/linux-headers: 4.4::gentoo (virtual/os-headers) sys-libs/glibc: 2.25-r9::gentoo Repositories: gentoo location: /usr/portage sync-type: rsync sync-uri: rsync://mirrors.tuna.tsinghua.edu.cn/gentoo-portage priority: -1000 jsteward location: /home/jsteward/jsteward masters: gentoo gentoo-zh location: /var/lib/layman/gentoo-zh sync-type: laymansync sync-uri: git://github.com/microcai/gentoo-zh.git masters: gentoo priority: 50 stuff location: /var/lib/layman/stuff sync-type: laymansync sync-uri: https://github.com/istitov/stuff.git masters: gentoo priority: 50 ACCEPT_KEYWORDS="sparc" ACCEPT_LICENSE="* -@EULA" CBUILD="sparc-unknown-linux-gnu" CFLAGS="-O2 -mcpu=niagara2 -mtune=niagara2 -pipe" CHOST="sparc-unknown-linux-gnu" CONFIG_PROTECT="/etc /usr/share/gnupg/qualified.txt" CONFIG_PROTECT_MASK="/etc/ca-certificates.conf /etc/env.d /etc/gconf /etc/gentoo-release /etc/php/apache2-php7.0/ext-active/ /etc/php/apache2-php7.1/ext-active/ /etc/php/cgi-php7.0/ext-active/ /etc/php/cgi-php7.1/ext-active/ /etc/php/cli-php7.0/ext-active/ /etc/php/cli-php7.1/ext-active/ /etc/revdep-rebuild /etc/sandbox.d /etc/terminfo" CXXFLAGS="-O2 -mcpu=niagara2 -mtune=niagara2 -pipe" DISTDIR="/usr/portage/distfiles" FCFLAGS="" FEATURES="assume-digests binpkg-logs config-protect-if-modified distlocks ebuild-locks fixlafiles merge-sync news parallel-fetch preserve-libs protect-owned sandbox sfperms strict unknown-features-warn unmerge-logs unmerge-orphans userfetch userpriv usersandbox usersync xattr" FFLAGS="" GENTOO_MIRRORS="https://mirrors.tuna.tsinghua.edu.cn/gentoo" LANG="en_US.utf8" LDFLAGS="-Wl,-O1 -Wl,--as-needed" MAKEOPTS="-j32" PKGDIR="/usr/portage/packages" PORTAGE_CONFIGROOT="/" PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --omit-dir-times --compress --force --whole-file --delete --stats --human-readable --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages --exclude=/.git" PORTAGE_TMPDIR="/var/tmp" USE="acl berkdb big-endian bzip2 cli cracklib crypt cxx dri fortran gdbm iconv ipv6 modules ncurses nls nptl openmp pam pcre php readline session sparc ssl tcpd unicode xattr zlib" APACHE2_MODULES="authn_core authz_core socache_shmcb unixd 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 cgi cgid 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" CALLIGRA_FEATURES="kexi words flow plan sheets stage tables krita karbon braindump author" COLLECTD_PLUGINS="df interface irq load memory rrdtool swap syslog" ELIBC="glibc" GPSD_PROTOCOLS="ashtech aivdm earthmate evermore fv18 garmin garmintxt gpsclock isync itrax mtk3301 nmea ntrip navcom oceanserver oldstyle oncore rtcm104v2 rtcm104v3 sirf skytraq superstar2 timing tsip tripmate tnt ublox ubx" INPUT_DEVICES="libinput keyboard mouse" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" LIBREOFFICE_EXTENSIONS="presenter-console presenter-minimizer" OFFICE_IMPLEMENTATION="libreoffice" PHP_TARGETS="php7-0" POSTGRES_TARGETS="postgres9_5" PYTHON_SINGLE_TARGET="python3_5" PYTHON_TARGETS="python2_7 python3_5" RUBY_TARGETS="ruby22" USERLAND="GNU" VIDEO_CARDS="fbdev glint mga r128 radeon dummy v4l" 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: CC, CPPFLAGS, CTARGET, CXX, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LC_ALL, PORTAGE_BUNZIP2_COMMAND, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS
Can anyone give me instructions how to disable pie for that profile locally so I can test if that is the issue?
(In reply to Rolf Eike Beer from comment #7) > Can anyone give me instructions how to disable pie for that profile locally > so I can test if that is the issue? Something like this https://bugs.gentoo.org/641216#c3 should work. """ For that you would need to un-force USE=pic on gcc as: /etc/portage/profile/package.use.force: sys-devel/gcc -pie and then build USE=-pie gcc. """ Another thing to try is apply patches from https://bugs.gentoo.org/641474 which should switch to using correct startup files for static+default-pie case as it certainly affects 'sln' but I'm not sure it it's enough.
I have picked only the second patch from bug 641474, which did not help. Also the binutils patch did not help. Building gcc with [-pie] and the same patches _did_ help.
Did not find too much but a few hints: Reproducible in qemu-user and hangs at memcpy() call in __libc_setup_tls: https://sourceware.org/git/?p=glibc.git;a=blob;f=csu/libc-tls.c;h=1f8ddaf5430d5de3a59e3b9fe6505200119a80d2;hb=HEAD#l181 $ qemu-sparc32plus -d in_asm elf/sln ... IN: __libc_setup_tls 0x000000000001162c: add %l2, %i1, %g1 0x0000000000011630: add %g1, -1, %g1 0x0000000000011634: mov %g0, %y 0x0000000000011638: udiv %g1, %i1, %l4 0x000000000001163c: smul %l4, %i1, %l4 0x0000000000011640: sub %i5, %l4, %o0 0x0000000000011644: clr [ %i4 + 0x14 ] 0x0000000000011648: mov %i3, %o2 0x000000000001164c: add %i0, %o0, %o0 0x0000000000011650: st %l4, [ %l3 + 0x23c ] 0x0000000000011654: mov %i2, %o1 0x0000000000011658: st %o0, [ %i4 + 0x10 ] 0x000000000001165c: add %i0, %i5, %i5 0x0000000000011660: call 0xb43e4 0x0000000000011664: add %i4, 8, %i4 -------------- IN: 0x00000000000b43e4: b,a %icc, 0x2e0 It's an early initialization call. Objdump shows exactly the same addresses (handly to debug): $ objdump -d -S -r elf/sln memcpy (_dl_static_dtv[2].pointer.val, initimage, filesz); 11660: 40 02 8b 61 call b43e4 <_GLOBAL_OFFSET_TABLE_+0x3e4> Call into GOT looks a bit fishy. Same code before being linked into sln: $ objdump -d -r -S csu/libc-tls.o 15c: ba 06 00 1d add %i0, %i5, %i5 memcpy (_dl_static_dtv[2].pointer.val, initimage, filesz); 160: 40 00 00 00 call 160 <__libc_setup_tls+0x160> 160: R_SPARC_WPLT30 memcpy 164: b8 07 20 08 add %i4, 8, %i4
Created attachment 511216 [details, diff] glibc-2.26-no-memcpy-ifunc.patch __libc_setup_tls() is called before redirectors for IFUNC are configured. I'm not sure how first pass initializes those: https://sourceware.org/git/?p=glibc.git;a=blob;f=csu/libc-start.c;h=34dd1252605519dfa034c836d5c0cb732bc39243;hb=HEAD#l197 Assuming those contain garbage. Attachted a workaround to avoid memcpy() from IFUNC in __libc_setup_tls(). qemu still hangs early but at least it gets further through __libc_setup_tls().
One of the workarounds is to compile glibc with disabled multiarch: EXTRA_ECONF=--disable-multi-arch emerge -v1 glibc
(In reply to Sergei Trofimovich from comment #12) > One of the workarounds is to compile glibc with disabled multiarch: > EXTRA_ECONF=--disable-multi-arch emerge -v1 glibc Does not seem to be enough as well.
Filed https://sourceware.org/PR22638 upstream to get some guidance on safety of IFUNC usage in early startup code.
(In reply to Sergei Trofimovich from comment #14) > Filed https://sourceware.org/PR22638 upstream to get some guidance on safety > of IFUNC usage in early startup code. H.J. and Anreas pointed out IFUNCs get set up before TLS setup thus are safe to use. The problem is in assembly code for sparc which assumes PIC more is used only for DSOs. And that breaks with --enable-default-pie. Thus most of '#ifdef SHARED' should be changed to '#ifdef PIC'. The good thing is glibc-master changed IFUNC selection from assembly level to C level (and fixing PIC Once and For all there). There is a single patch needed on top of master to make sparc work: https://sourceware.org/ml/libc-alpha/2017-09/msg01140.html For glibc-2.26 it's less easy. I see a few avenues: 1. do nothing and mask <glibc-2.27 on 17.0 sparc profiles 2. backport a huge batch of patches porting IFUNC selection from assembly to C (some of the patches remove code): Adhemerval Zanella (25): nptl: Add tests for internal pthread_mutex_t offsets nptl: Define __PTHREAD_MUTEX_{NUSERS_AFTER_KIND,USE_UNION} sparc: Implement memcpy/mempcpy ifunc selection in C sparc: Implement memset/bzero ifunc selection in C sparc: Assume VIS3 support sparc: refactor sparc64 signbit{f} selector to C sparc: refactor sparc64 isnan{f} selector to C sparc: refactor sparc64 isinf{f} selector to C sparc: refactor sparc64 finite{f} selector to C sparc: refactor sparc64 nearbyint{f} selector to C sparc: refactor sparc64 lrint{f} selector to C sparc: refactor sparc64 rint{f} selector to C sparc: refactor sparc64 __mpn_mul_1 selector to C sparc: refactor sparc64 __mpn_sub_n selector to C sparc: refactor sparc64 __mpn_addmul_1 selector to C sparc: refactor sparc64 __mpn_submul_1 selector to C sparc: refactor sparc64 __mpn_add_n selector to C sparc: refactor sparc32 copysign selector to C sparc: refactor sparc32 fabs{f} selector to C sparc: refactor sparc32 llrint{f} selector to C sparc: refactor sparc32 rint{f} selector to C sparc: refactor sparc32 nearbyint{f} selector to C sparc: refactor cpu_relax to C sparc: Fix sparv9 multiarch build sparc: Remove ununsed ifunc assembly macros (and likely more) 3. do a targeted "sed 's/SHARED/PIC/g'" on a bunch of "sysdeps/sparc/**.S" files.
(In reply to Sergei Trofimovich from comment #15) > (In reply to Sergei Trofimovich from comment #14) > > Filed https://sourceware.org/PR22638 upstream to get some guidance on safety > > of IFUNC usage in early startup code. > > H.J. and Anreas pointed out IFUNCs get set up before TLS setup thus are safe > to use. The problem is in assembly code for sparc which assumes PIC more is > used only for DSOs. And that breaks with --enable-default-pie. > > Thus most of '#ifdef SHARED' should be changed to '#ifdef PIC'. The good > thing is glibc-master changed IFUNC selection from assembly level to C level > (and fixing PIC Once and For all there). There is a single patch needed on > top of master to make sparc work: > https://sourceware.org/ml/libc-alpha/2017-09/msg01140.html > > For glibc-2.26 it's less easy. I see a few avenues: > 1. do nothing and mask <glibc-2.27 on 17.0 sparc profiles > 2. backport a huge batch of patches porting IFUNC selection from assembly to > C (some of the patches remove code): > Adhemerval Zanella (25): > nptl: Add tests for internal pthread_mutex_t offsets > nptl: Define __PTHREAD_MUTEX_{NUSERS_AFTER_KIND,USE_UNION} > sparc: Implement memcpy/mempcpy ifunc selection in C > sparc: Implement memset/bzero ifunc selection in C > sparc: Assume VIS3 support > sparc: refactor sparc64 signbit{f} selector to C > sparc: refactor sparc64 isnan{f} selector to C > sparc: refactor sparc64 isinf{f} selector to C > sparc: refactor sparc64 finite{f} selector to C > sparc: refactor sparc64 nearbyint{f} selector to C > sparc: refactor sparc64 lrint{f} selector to C > sparc: refactor sparc64 rint{f} selector to C > sparc: refactor sparc64 __mpn_mul_1 selector to C > sparc: refactor sparc64 __mpn_sub_n selector to C > sparc: refactor sparc64 __mpn_addmul_1 selector to C > sparc: refactor sparc64 __mpn_submul_1 selector to C > sparc: refactor sparc64 __mpn_add_n selector to C > sparc: refactor sparc32 copysign selector to C > sparc: refactor sparc32 fabs{f} selector to C > sparc: refactor sparc32 llrint{f} selector to C > sparc: refactor sparc32 rint{f} selector to C > sparc: refactor sparc32 nearbyint{f} selector to C > sparc: refactor cpu_relax to C > sparc: Fix sparv9 multiarch build > sparc: Remove ununsed ifunc assembly macros > (and likely more) > 3. do a targeted "sed 's/SHARED/PIC/g'" on a bunch of "sysdeps/sparc/**.S" > files. Seems like masking <glibc-2.27 will cause problems as libnsl:0 (which is a dependency of opensp, then a dependency of openjade, then a dependency of po4a, which is required for man-db ( nls ? >=app-text/po4a-0.45 ) ) requires <glibc-2.26.
> 3. do a targeted "sed 's/SHARED/PIC/g'" on a bunch of "sysdeps/sparc/**.S" > files. Following this, I did this as a work-around: # ebuild /usr/portage/sys-libs/glibc/glibc-2.25-r9.ebuild fetch unpack # find /var/tmp/portage/sys-libs/glibc-2.25-r9/work/glibc-2.25/sysdeps/sparc/ -name \*.S -exec sed -i 's/def SHARED/def PIC/' {} \; -print # ebuild /usr/portage/sys-libs/glibc/glibc-2.25-r9.ebuild compile install # ebuild /usr/portage/sys-libs/glibc/glibc-2.25-r9.ebuild qmerge This gave me a succesful install and so far, nothing has broken. Re-emerging @world now, but that will likely take a day or two on this thing. (Netra T1 200)
Queued the fix for 2.27 as: https://github.com/gentoo/glibc/commit/fec498315526476aaf05260677d58f5ac5adaf00
Created attachment 517822 [details, diff] Patch for seraph@xs4all.nl sed changes Following seraph@xs4all.nl method I have prepared a patch. I have done full "emerge -e @world" twice successfully. First I followed seraph's method from comment 17 then I made a patch to put in /etc/portage/patches/sys-libs/glibc-2.25-r9/ So I could do my @world rebuilds without pause. Two full cycles using the patch so seraph's method looks good. I am not up to tinkering with ebuild's to the extent of running find/sed between steps. Thank you. Enjoy, Jim
Fixed in sys-libs/glibc-2.27-r3 Should we maybe also backport this to our 2.26 ?!
> > Should we maybe also backport this to our 2.26 ?! Added to gentoo/2.26 branch as dec42d1257717054c08b8293d799e640e94df1d7
(In reply to Andreas K. Hüttel from comment #21) > > > > Should we maybe also backport this to our 2.26 ?! > > Added to gentoo/2.26 branch as > dec42d1257717054c08b8293d799e640e94df1d7 It was not chear what "this" means. Clickable link: https://github.com/gentoo/glibc/commit/dec42d1257717054c08b8293d799e640e94df1d7 At the risk od stating the obvious. 2.26 will require a lot more fixes as described in #c15 than just start.S tweak.
This seems not to be in 2.26-r7.
(In reply to Rolf Eike Beer from comment #23) > This seems not to be in 2.26-r7. 2.26-r7 still fails in exactly the same way for me. My earlier work-around still works on this version though.
2.27 is now available in ~arch, and this is fixed there.
glibc-2.27 is stable on sparc and shows signs of life. Declaring victory.
