Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 80118 - Internal compiler error in try_split at emit-rtl.c
Summary: Internal compiler error in try_split at emit-rtl.c
Status: RESOLVED DUPLICATE of bug 79852
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: [OLD] GCC Porting (show other bugs)
Hardware: x86 Linux
: High normal (vote)
Assignee: Please assign to toolchain
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-01-30 12:23 UTC by Andreas Nilsson
Modified: 2005-07-17 13:06 UTC (History)
8 users (show)

See Also:
Package list:
Runtime testing required: ---


Attachments
The preprocessed source asked for (ccnkFLuy.out,302.77 KB, application/octet-stream)
2005-01-30 12:27 UTC, Andreas Nilsson
Details
GCC Preprocessed source (ccKEjNGC.out,133.11 KB, text/plain)
2005-02-09 02:56 UTC, Alexander Skwar
Details
Perl preprocessed source (cc3E0tDE.out,392.44 KB, text/plain)
2005-02-09 02:57 UTC, Alexander Skwar
Details
To #28: Preprocessed source (ccBdWrPp.out,197.92 KB, application/octet-stream)
2005-04-14 13:34 UTC, Myles Goodwin
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Andreas Nilsson 2005-01-30 12:23:03 UTC
Im trying to rebuild the toolchain via emerge -e system, which comes along fine to libperl-5.8.6. There emerge chrashes with:

`sh  cflags "optimize='-O2 -march=pentium2 -fomit-frame-pointer -pipe'" perlapi.o` -fPIC perlapi.c
          CCCMD =  i686-pc-linux-gnu-gcc -DPERL_CORE -c -fno-strict-aliasing -pipe -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -march=pentium2 -fomit-frame-pointer -pipe  -Wall
`sh  cflags "optimize='-O2 -march=pentium2 -fomit-frame-pointer -pipe'" numeric.o` -fPIC numeric.c
          CCCMD =  i686-pc-linux-gnu-gcc -DPERL_CORE -c -fno-strict-aliasing -pipe -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -O2 -march=pentium2 -fomit-frame-pointer -pipe  -Wall
numeric.c: In function `S_mulexp10':
numeric.c:815: internal compiler error: in try_split, at emit-rtl.c:3343
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://bugs.gentoo.org/> for instructions.
Preprocessed source stored into /var/tmp/portage/libperl-5.8.6/temp/ccnkFLuy.out file, please attach this to your bugreport.
make: *** [numeric.o] Error 1

!!! ERROR: sys-devel/libperl-5.8.6 failed.
!!! Function src_compile, Line 218, Exitcode 2
!!! Unable to make libperl.so
!!! If you need support, post the topmost build error, NOT this status message.

router ~ #                               


Reproducible: Always
Steps to Reproduce:
1. Do "emerge -e system" or emerge libperl, sometimes "emerge gcc"
2. 
3.

Actual Results:  
Emerge crashes with above errors

Expected Results:  
A stable toolchain

As you can see below this error occur on a pII 233MHz, which has 64 Mb ram, 120
GB hdd, which I have running as a router/file-printer-server for my home
network. It also runs an rsync daemon to provide rsync to my internal
gentoo-machines. It also shares /usr/portage/distfiles/ via nfs. Also there is
no attached screen, all commands are run over ssh. 

As this is additional, i feel i must mention that the same version of the above
listed programs compile cleanly on my other machines, although, they run a
differnt kernel and are a p4 and p3.

Here is the emerge info details:

Gentoo Base System version 1.6.8
Portage 2.0.51-r15 (default-linux/x86/2004.3, gcc-3.4.3,
glibc-2.3.4.20041102-r0, 2.6.11-rc2 i686)
=================================================================
System uname: 2.6.11-rc2 i686 Pentium II (Klamath)
Python:              dev-lang/python-2.3.4 [2.3.4 (#1, Jan 30 2005, 08:28:32)]
distcc 2.18.3 i686-pc-linux-gnu (protocols 1 and 2) (default port 3632) [disabled]
ccache version 2.3 [enabled]
dev-lang/python:     2.3.4
sys-devel/autoconf:  2.59-r6, 2.13
sys-devel/automake:  1.7.9-r1, 1.8.5-r3, 1.5, 1.4_p6, 1.6.3, 1.9.4
sys-devel/binutils:  2.15.92.0.2-r2
sys-devel/libtool:   1.5.10-r4
virtual/os-headers:  2.6.8.1-r2
ACCEPT_KEYWORDS="x86 ~x86"
AUTOCLEAN="yes"
CFLAGS="-O2 -march=pentium2 -fomit-frame-pointer -pipe"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3/share/config
/usr/share/config /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-O2 -march=pentium2 -fomit-frame-pointer -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoaddcvs autoconfig ccache distlocks sandbox sfperms"
GENTOO_MIRRORS="ftp.du.se/pub/os/gentoo http://ds.thn.htu.se/linux/gentoo
http://mirror.pudas.net/gentoo"
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
SYNC="rsync://rsync.gentoo.org/gentoo-portage"
USE="x86 acl apache2 apm arts avi bash-completion berkdb bitmap-fonts crypt cups
encode f77 font-server foomaticdb fortran ftp gdbm gif gpm hardened hardenedphp
imlib ipv6 java jikes jpeg kerberos krb4 libg++ libwww mad mikmod motif mpeg
mysql ncurses nls ntpl ntplonly oggvorbis opengl oss pam pdflib perl png ppds
python quicktime readline samba sdl spell ssl svaga svga tcpd tiff truetype
truetype-fonts type1-fonts usb xml xml2 xmms xv zlib"
Unset:  ASFLAGS, CBUILD, CTARGET, LANG, LC_ALL, LDFLAGS, PORTDIR_OVERLAY
Comment 1 Andreas Nilsson 2005-01-30 12:27:29 UTC
Created attachment 49975 [details]
The preprocessed source asked for

The preprocessed sorces asked for
Comment 2 Alexander Skwar 2005-02-09 02:56:44 UTC
Adding perl@gentoo.org to CC, since this build problem affects the perl package as well.

Here's my emerge info:

11:45:41 askwar@alturo:~ $ emerge info
Portage 2.0.51-r15 (default-linux/x86/2004.3, gcc-3.4.3, glibc-2.3.4.20041102-r0, 2.6.10-gentoo-r7.PHP2.07 i686)
=================================================================
System uname: 2.6.10-gentoo-r7.PHP2.07 i686 Intel(R) Celeron(R) CPU 2.00GHz
Gentoo Base System version 1.6.8
Python:              dev-lang/python-2.3.4 [2.3.4 (#1, Dec 31 2004, 10:25:08)]
ccache version 2.3 [enabled]
dev-lang/python:     2.3.4
sys-devel/autoconf:  2.59-r6, 2.13
sys-devel/automake:  1.8.5-r2, 1.5, 1.4_p6, 1.6.3, 1.7.9, 1.9.4
sys-devel/binutils:  2.15.92.0.2-r2
sys-devel/libtool:   1.5.10-r3
virtual/os-headers:  2.6.8.1-r2
ACCEPT_KEYWORDS="x86 ~x86"
AUTOCLEAN="yes"
CFLAGS="-march=pentium4 -O2 -pipe -fomit-frame-pointer"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3/share/config /usr/share/config /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-O2 -mcpu=i686 -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoaddcvs autoconfig ccache collision-protect distlocks sandbox sfperms"
GENTOO_MIRRORS="http://php2.ath.cx/~askwar/gentoo-files/ http://ftp-stud.fht-esslingen.de/pub/Mirrors/gentoo/ ftp://mirrors.sec.informatik.tu-darmstadt.de/gentoo/ ftp://sunsite.informatik.rwth-aachen.de/pub/Linux/gentoo ftp://ftp.wh2.tu-dresden.de/pub/mirrors/gentoo"
LANG="de_DE.UTF-8"
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
SYNC="rsync://rsync.europe.gentoo.org/gentoo-portage"
USE="x86 acl apache2 apm arts async avi bash-completion bzlib crypt cscope curl curlwrappers diet djbfft encode exif fam flash font-server foomaticdb ftp fwdzone gd gdbm gif hal hardened iconv imagemagick imap imlib immqt-bc ithreads javascript jpeg libedit libg++ libwww lzo lzw lzw-tiff mad maildir mailwrapper mbox mmap mmx mng mysql ncurses nls no-old-linux noantlr nobcel nobeanutils nobsh nocommonslogging nocommonsnet nojdepend nojsch nojython nolog4j nooro noregexp norhino noxalan noxerces nptl offensive oggvorbis opengl oss pam parse-clocks pcntl pcre pdflib perl pic pie png posix postgres pwdb python qt quicktime readline recode sasl sdl shared slang sockets spell sse ssl sysvipc tcpd tiff truetype truetype-fonts type1-fonts unicode vhosts xfs xml2 xmms xv zlib linguas_de"
Unset:  ASFLAGS, CBUILD, CTARGET, LC_ALL, LDFLAGS, PORTDIR_OVERLAY
Comment 3 Alexander Skwar 2005-02-09 02:56:53 UTC
Created attachment 50805 [details]
GCC Preprocessed source

Same error here as well:

gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC	 -W -Wall -Wwrite-strings
-Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long   
-DHAVE_CONFIG_H    -I. -I.
-I/var/tmp/portage/gcc-3.4.3.20050110/work/gcc-3.4.3/gcc
-I/var/tmp/portage/gcc-3.4.3.20050110/work/gcc-3.4.3/gcc/.
-I/var/tmp/portage/gcc-3.4.3.20050110/work/gcc-3.4.3/gcc/../include 
/var/tmp/portage/gcc-3.4.3.20050110/work/gcc-3.4.3/gcc/hashtable.c -o
hashtable.o
/var/tmp/portage/gcc-3.4.3.20050110/work/gcc-3.4.3/gcc/hashtable.c: In function
`approx_sqrt':
/var/tmp/portage/gcc-3.4.3.20050110/work/gcc-3.4.3/gcc/hashtable.c:310:
interner Compiler-Fehler: in try_split, bei emit-rtl.c:3314
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://bugs.gentoo.org/> for instructions.
Preprocessed source stored into
/var/tmp/portage/gcc-3.4.3.20050110/temp/ccKEjNGC.out file, please attach this
to your bugreport.
make[2]: *** [hashtable.o] Fehler 1
make[2]: Leaving directory `/var/tmp/portage/gcc-3.4.3.20050110/work/build/gcc'

make[1]: *** [stage1_build] Fehler 2
make[1]: Leaving directory `/var/tmp/portage/gcc-3.4.3.20050110/work/build/gcc'

make: *** [profiledbootstrap] Fehler 2

!!! ERROR: sys-devel/gcc-3.4.3.20050110 failed.
!!! Function gcc_do_make, Line 1086, Exitcode 2
!!! make failed with profiledbootstrap
!!! If you need support, post the topmost build error, NOT this status message.

									       
		   ...done!
>>> emerge (1 of 1) sys-devel/gcc-3.4.3.20050110 to /
Comment 4 Alexander Skwar 2005-02-09 02:57:05 UTC
Created attachment 50806 [details]
Perl preprocessed source

Perl is failing to build as well for me:

`sh  cflags "optimize='-march=pentium4 -O2 -pipe -fomit-frame-pointer'"
numeric.o` -fPIC numeric.c
	  CCCMD =  gcc -DPERL_CORE -c -D_REENTRANT -D_GNU_SOURCE
-DTHREADS_HAVE_PIDS -fno-strict-aliasing -pipe -D_LARGEFILE_SOURCE
-D_FILE_OFFSET_BITS=64 -I/usr/include/gdbm -march=pentium4 -O2 -pipe
-fomit-frame-pointer  -Wall
numeric.c: In function `S_mulexp10':
numeric.c:815: internal compiler error: in try_split, at emit-rtl.c:3314
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://bugs.gentoo.org/> for instructions.
Preprocessed source stored into
/var/tmp/portage/perl-5.8.6-r2/temp/cc3E0tDE.out file, please attach this to
your bugreport.
make: *** [numeric.o] Error 1

!!! ERROR: dev-lang/perl-5.8.6-r2 failed.
Comment 5 chrisp 2005-03-20 16:53:32 UTC
I had these same  problems after I tried moving to vanilla gcc-3.4.3.20050110 after upgrading from gcc 3.3.4 to gcc-3.4.3.20050110 hardened.  After I found that hardened wouldn't work for me, I tried moving to a non-hardened profile and had these problems (even after recompiling my entire toolchain   other than gcc - which wouldn't compile due to the previous error).  I finally worked around it by setting gcc to 3.3.4 (it was still installed) and recompiling 3.4.3.20050110 with it.  Once it compiled I set gcc to 3.4.3.20050110 and recompiled my entire toolchain with:  emerge linux26-headers glibc && emerge glibc binutils gcc && emerge binutils gcc

Finally I had gcc configured properly.  I've since done an emerge -uD xorg-x11 and everything built properly (although I did have to manhandle opengl-update to unblock a mask).  I assume perl will build properly as well.  

Things are stable at this point, but I think I'll do emerge -e world twice to ensure that everything is built under the same profile and wait until gcc 4 becomes stable before going through this again..

I'm pretty sure that this is caused by either reverting from hardened or upgrading gcc without a proper toolchain rebuild (or a combination of the two).  In either case, I think this method will work for those that find themselves in this situation.
Comment 6 Andreas Nilsson 2005-03-24 06:04:15 UTC
I don't rember how I solved the problem, but you could try somethind like this:

prompt # STAGE1_CFLAGS="-O2 -march=your_arch -mtune=your_arch -pipe"

Best regards
Andreas Nilsson
Comment 7 Charles Pence 2005-03-26 07:35:20 UTC
I'm getting exactly the same libperl error as the original poster, but with a different setup:

- Previously stable gcc-33.4.3.20050110 without the hardened use flag
- Recompiled with the hardened useflag with a _proper_ toolchain rebuild:

emerge -av gcc && emerge -av glibc binutils gcc && emerge -av glibc binutils gcc portage

- Proceeded to emerge -e system, which bails out at libperl.

So I think this is almost certainly a bug in hardened on 3.4.3.20050110.  I'm trying to back down to 3.4.3-r1 and I'll report back whether or not that works.
Comment 8 steveb 2005-03-29 14:35:37 UTC
I have exactly the same error, when I try to rebuild my toolchain:
gcc -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H -DGENERATOR_FILE  -o genconfig \
        genconfig.o rtl.o read-rtl.o bitmap.o ggc-none.o min-insn-modes.o gensupport.o insn-conditions.o print-rtl1.o \
        errors.o ../libiberty/libiberty.a
./genconfig /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/config/i386/i386.md > tmp-config.h
/bin/sh /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/move-if-change tmp-config.h insn-config.h
echo timestamp > s-config
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/c-objc-common.c -o c-objc-common.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/c-dump.c -o c-dump.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  \
  -DHOST_MACHINE=\"i686-pc-linux-gnu\" -DTARGET_MACHINE=\"i686-pc-linux-gnu\" \
  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/c-pch.c -o c-pch.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/cpplib.c -o cpplib.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/cpplex.c -o cpplex.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/cppmacro.c -o cppmacro.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/cppexp.c -o cppexp.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/cppfiles.c -o cppfiles.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/cpptrad.c -o cpptrad.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/cpphash.c -o cpphash.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/cpperror.c -o cpperror.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/cppinit.c -o cppinit.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/cppcharset.c -o cppcharset.o
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -pedantic -Wno-long-long    -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/. -I/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/../include  /var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/hashtable.c -o hashtable.o
/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/hashtable.c: In function `approx_sqrt':
/var/tmp/portage/gcc-3.4.3.20050110-r1/work/gcc-3.4.3/gcc/hashtable.c:310: internal compiler error: in try_split, at emit-rtl.c:3343
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://bugs.gentoo.org/> for instructions.
Preprocessed source stored into /var/tmp/portage/gcc-3.4.3.20050110-r1/temp/ccPktGPP.out file, please attach this to your bugreport.
make[2]: *** [hashtable.o] Error 1
make[2]: Leaving directory `/var/tmp/portage/gcc-3.4.3.20050110-r1/work/build/gcc'
make[1]: *** [stage1_build] Error 2
make[1]: Leaving directory `/var/tmp/portage/gcc-3.4.3.20050110-r1/work/build/gcc'
make: *** [profiledbootstrap] Error 2

!!! ERROR: sys-devel/gcc-3.4.3.20050110-r1 failed.
!!! Function gcc_do_make, Line 1138, Exitcode 2
!!! make failed with profiledbootstrap
!!! If you need support, post the topmost build error, NOT this status message.



I will try now to switch to the gcc-3.3.x version and rebuild my toolchain with it and then switch back to 3.4.x and do the rebuild again. Will post my results here again.
Comment 9 Alexander Skwar 2005-03-29 21:15:56 UTC
I reverted back to gcc-3.3.3-r6, which was still present from when I installed the server. With this gcc, I could build Perl and all the other packages just fine. This old gcc was *NOT* hardened.

Then, I updated to:

gcc-Version 3.4.3 20041125 (Gentoo Hardened Linux 3.4.3-r1, ssp-3.4.3-0, pie-8.7.7)

With this *HARDENED* gcc, building Perl no longer works (error in emit-rtl.c).

To me, it seems like a hardened toolchain doesn't work.
Comment 10 Andreas Nilsson 2005-03-30 02:16:56 UTC
Did you try with STAGE1_CFLAGS="-O2 -march=your_arch -mtune=your_arch -pipe" ?
Comment 11 steveb 2005-03-30 12:53:51 UTC
To Comment #9:
I can NOT confirm that. I have now used gcc 3.4.3 hardened to compile 3.3.5, then I have switched to 3.3.5 hardened and then recompiled 3.4.3 hardened (sys-apps/texinfo gettext sys-devel/binutils sys-devel/gcc sys-apps/baselayout sys-libs/zlib virtual/libc) and after that I have switched to 3.4.3 hardened and recompiled again 3.4.3 and as well libperl and perl. Strictly following this path I was able to get everything compiled. Currently I can confirm, that this works for the following CPU's/arch:

- Intel Pentium Classic @ 150 MHz with and without MMX (Ahhhgrr! This one was real hell to wait and wait and wait and wait to get any result. Well... finaly it works)

- Intel Pentium Pro @ 200 MHz (Quad System)

- Intel Pentium II @ 333 MHz (Dual System)

- Intel Pentium III @ 933 MHz, 550 MHz and 1 GHz

- AMD Athlon Tbird @ 1400 MHz

- AMD Athlon XP 2400+ @ 2000 MHz



cheers

SteveB
Comment 12 Alexander Ivanchev 2005-03-30 21:19:54 UTC
Same thing here - as a result, this one is pretty serious cause the following fail to build at my end:

gst-plugins - thus no gnome
open office
ghostscript

That's really weird and IMO was introduced either by linux-headers or glibc upgrade since i've had a working hardened gcc-3.4.3.20050110-r1 before that (which now fails to rebuild)

CHOST="i686-pc-linux-gnu"
CFLAGS="-march=athlon-xp -O3 -fomit-frame-pointer -pipe -fstack-protector"
CXXFLAGS="${CFLAGS}"
Comment 13 Peter S. Mazinger 2005-03-30 22:35:43 UTC
remove -fstack-protector from CFLAGS and add USE=hardened
I am curious if all of you having this problem, have fstack-protector in CFLAGS
Comment 14 Alexander Ivanchev 2005-03-31 00:25:14 UTC
Peter,

OK, I can try that, but thing is, I've always had -fstack-protector and no probs with the hardened toolchain - this very same version of gcc, which fails (gcc gcc-3.4.3.20050110-r1) IS my present compiler, previously compiled WITH +hardened AND -fstack-protector...
Comment 15 Alexander Skwar 2005-03-31 01:10:35 UTC
Re: comment #13:
08:44:13 askwar@alturo:~ $ emerge info | grep CFLAGS
CFLAGS="-march=pentium4 -O2 -pipe -fomit-frame-pointer"

Re: comment #10:
I did not have STAGE1_CFLAGS set.

I now have rebuilt gcc-Version 3.4.3 20041125 with USE=-hardened and also rebuild binutils and glibc. After doing so, I was able to build perl.
Comment 16 Peter S. Mazinger 2005-03-31 06:11:47 UTC
correct usage is to remove -fPIC/-fpic/-fPIE/-fpie/-fstack-protector/
-fstack-protector-all from CFLAGS and rely only on USE="pic pie hardened"
predefined by any hardened profile
Comment 17 Charles Pence 2005-03-31 07:50:03 UTC
To #13:  No, I do not, nor have I ever, had -fstack-protector in CFLAGS.
Comment 18 Peter S. Mazinger 2005-03-31 09:36:44 UTC
the bug shown here is similar to some other that I have provided a patch applied
in toolchain.eclass (present since 3.4.0, not solved by ssp-creator)
My presumption that the updated gcc-3.4.3.2005 does not have a proper ssp patch
gcc-3.4.3-r1 is the last I used (and use), due to too much trouble w/ the newer
version, I would say, try to rebuild first gcc w/o hardened (3.4.3-r1) and then
rebuild it w/ hardened
Comment 19 Alexander Ivanchev 2005-03-31 13:06:20 UTC
Peter,

I've emerged gcc-3.4.3-r1 from the binary package, recompiled binutils and glibc, tried re-emering the gcc-3.4.3.2005 - same thing - I've even switched to [1] i386-pc-linux-gnu-3.3.4, and again same thing...

What are the odds linux-headers26/glibc could be the cause, that's the only recent thing I've recently upgraded (to linux-headers-2.6.8.1-r4 & glibc-2.3.4.20050125-r1)..? ... or am I missing the point here?
Comment 20 Peter S. Mazinger 2005-03-31 13:49:26 UTC
try emerging gcc-3.4.3-r1 (not 2005xxxx), I suppose the forward port of the ssp
patch causes this
Comment 21 Peter S. Mazinger 2005-03-31 13:51:51 UTC
I do not use the 2.6 headers, so can't really tell if that is the problem
Comment 22 Alexander Ivanchev 2005-04-02 03:18:23 UTC
FYI, I managed to get my toolchain fixed - it did turn out to be a gcc problem...

I got a gcc from here:
http://dev.gentoo.org/~avenj/bins/i686/

With this one gcc-3.4.3.2005 compiled fine (finally!)

Lessons learned - make good use of buildpkgs and keep 'em for awhile too...
Comment 23 Mark Loeser (RETIRED) gentoo-dev 2005-04-02 23:35:46 UTC

*** This bug has been marked as a duplicate of 79852 ***
Comment 24 Benno Schulenberg 2005-04-09 06:08:11 UTC
At #22: Please recompile gcc-3.4.3.2005 with itself.  Then again.  And then a _third_ time...  Prediction: failure.
Comment 25 Myles Goodwin 2005-04-12 13:30:16 UTC
To #24:  Your predication came true for me.  First time I built it using gcc-3.3 then used 3.4.3-20050110-r2 to build itself (second time).  It completed successfully, then tried it a third time and it choked at the same point that's being mentioned here.  Something about emit-rtl.c that it just does not like.  I'm seeing the same error pop up in libstdc++-v3-3.3.4.  Is there a solution out there?  I'm seeing duplicates of the same problem all over the site.
Comment 26 steveb 2005-04-12 13:45:04 UTC
Am I the only one not having anymore this problem? I did all the stuff I mentioned in comment #11 and on top of that I emerged linux-headers-2.6.11. I did recompiled sys-devel/gcc-3.4.3.20050110-r2 serval times and no problem so far.
Comment 27 Benno Schulenberg 2005-04-14 04:44:25 UTC
At comment #26: How many times did you do an 'emerge -e system' with gcc-3.4.3 selected?  If you can do that three times in a row without a fault, then you're clean.  (Try this only on your fastest system :)
Comment 28 Myles Goodwin 2005-04-14 13:32:18 UTC
to #11:  tried compiling 3.3.5 with both hardened and vanilla 3.4.3.20050110-r2.  it bombed at the same point both times:

/var/tmp/portage/gcc-3.3.5.20050130-r2/work/gcc-3.3.5/gcc/line-map.h:47: warning: type of bit-field `reason' is a GCC extension
In file included from /var/tmp/portage/gcc-3.3.5.20050130-r2/work/gcc-3.3.5/gcc/cppmain.c:26:
/var/tmp/portage/gcc-3.3.5.20050130-r2/work/gcc-3.3.5/gcc/cpplib.h:180: warning: type of bit-field `type' is a GCC extension
/var/tmp/portage/gcc-3.3.5.20050130-r2/work/gcc-3.3.5/gcc/cpplib.h:487: warning: type of bit-field `type' is a GCC extension
gcc -c -DEFAULT_PIE_SSP    -O -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wtraditional -pedantic -Wno-long-long   -DHAVE_CONFIG_H    -I. -I. -I/var/tmp/portage/gcc-3.3.5.20050130-r2/work/gcc-3.3.5/gcc -I/var/tmp/portage/gcc-3.3.5.20050130-r2/work/gcc-3.3.5/gcc/. -I/var/tmp/portage/gcc-3.3.5.20050130-r2/work/gcc-3.3.5/gcc/config -I/var/tmp/portage/gcc-3.3.5.20050130-r2/work/gcc-3.3.5/gcc/../include /var/tmp/portage/gcc-3.3.5.20050130-r2/work/gcc-3.3.5/gcc/hashtable.c -o hashtable.o
/var/tmp/portage/gcc-3.3.5.20050130-r2/work/gcc-3.3.5/gcc/hashtable.c: In function `approx_sqrt':
/var/tmp/portage/gcc-3.3.5.20050130-r2/work/gcc-3.3.5/gcc/hashtable.c:328: internal compiler error: in try_split, at emit-rtl.c:3343
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://bugs.gentoo.org/> for instructions.
Preprocessed source stored into /var/tmp/portage/gcc-3.3.5.20050130-r2/temp/ccBdWrPp.out file, please attach this to your bugreport.
make[2]: *** [hashtable.o] Error 1

**I'll attach the preprocessed source right after this post**

My emerge info:

Gentoo Base System version 1.4.16
Portage 2.0.51.19 (default-linux/x86/2005.0, gcc-3.4.3-20050110-vanilla, glibc-2.3.4.20050125-r1, 2.6.11-gentoo-r4 i686)
=================================================================
System uname: 2.6.11-gentoo-r4 i686 AMD Athlon(tm) XP 2600+
Python:              dev-lang/python-2.3.4-r1 [2.3.4 (#1, Apr  2 2005, 16:28:04)]
dev-lang/python:     2.3.4-r1
sys-devel/autoconf:  2.59-r6, 2.13
sys-devel/automake:  1.7.9-r1, 1.8.5-r3, 1.5, 1.4_p6, 1.6.3, 1.9.4
sys-devel/binutils:  2.15.92.0.2-r7
sys-devel/libtool:   1.5.14
virtual/os-headers:  2.6.8.1-r2
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CFLAGS="-march=athlon-xp -O3 -mtune=athlon-xp -pipe -fforce-addr -momit-leaf-frame-pointer -fomit-frame-pointer -ftracer -m128bit-long-double -mfpmath=sse -mmmx -m3dnow -msse"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3/share/config /usr/lib/X11/xkb /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/ /var/qmail/control"
CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d"
CXXFLAGS="-march=athlon-xp -O3 -mtune=athlon-xp -pipe -fforce-addr -momit-leaf-frame-pointer -fomit-frame-pointer -ftracer -m128bit-long-double -mfpmath=sse -mmmx -m3dnow -msse -fvisibility-inlines-hidden"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoaddcvs autoconfig ccache distlocks sandbox sfperms"
GENTOO_MIRRORS="http://distfiles.gentoo.org http://distro.ibiblio.org/pub/Linux/distributions/gentoo"
MAKEOPTS="-j2"
PKGDIR="/usr/portage/packages"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage"
SYNC="rsync://rsync.gentoo.org/gentoo-portage"
USE="fortran gcj gtk hardened nls objc -boundschecking"
Unset:  ASFLAGS, CBUILD, CTARGET, LANG, LC_ALL, LDFLAGS, LINGUAS

I do not seem to be having this problem on my Averatec Centrino Notebook.  Could this problem be hardware-specific?

**CLEARLY NOT RESOLVED.  PLEASE RE-OPEN***
Comment 29 Myles Goodwin 2005-04-14 13:34:56 UTC
Created attachment 56293 [details]
To #28: Preprocessed source

To #28: Preprocessed Source
Comment 30 Alexander Ivanchev 2005-04-14 13:36:32 UTC
#24 - I assume, you're right - I was just pointing out a work around for people left with no working toolchain - the bug is still valid.