Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 41091 - Lyx doesn't build; probably because of new compiler gcc 3.3.2
Summary: Lyx doesn't build; probably because of new compiler gcc 3.3.2
Status: RESOLVED WORKSFORME
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: Current packages (show other bugs)
Hardware: x86 Linux
: High normal (vote)
Assignee: Text-Markup Team (OBSOLETE)
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2004-02-10 03:11 UTC by Sebastian
Modified: 2004-02-23 08:37 UTC (History)
0 users

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


Attachments
weired log (log,228.80 KB, text/plain)
2004-02-22 10:01 UTC, Sebastian
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Sebastian 2004-02-10 03:11:32 UTC
Hello!

I tried two versions of Lyx, namely lyx 1.3.2-r1 and lyx 1.3.3. Bot failed. During compilation of the 1.3.2.-r1 ebuild there was often a message saying that I should run some configure tests (I don't know what that means). During the compilation of 1.3.3 these messages didn't occur, but it failed to build just like the earlier ebuild. 
This is the first time I tried to install lyx, but I think that these errors appear because my Gentoo recently switched gcc from 3.2.x to 3.3.2. 

I can't really paste to stuff from make, but both ebuild broke up the same way and here's the rest of it:

g++: Internal error: Segmentation fault (program cc1plus)
Please submit a (full) bug report to bugs.gentoo.org

If you're interrested in more of this I'll restart the compilation in an X console so I can paste it.

Thx
Sebastian

Reproducible: Always
Steps to Reproduce:
1. emerge --update --deep world
2. emerge lyx
3.

Actual Results:  
Lyx ebuilds break

Expected Results:  
Lyx ebuild works like a charm (like always) :D
Comment 1 Sebastian 2004-02-10 03:50:57 UTC
Here's what compilation of 1.3.2-r1 produces:

Often sth like this:
g++ -DHAVE_CONFIG_H -I. -I. -I../../../src -I../../../images -I../../../src -I../../../src/frontends -I../../../src/frontends/controllers -I../../../boost -isystem /usr/X11R6/include -O2 -march=pentium3 -pipe -mmmx -msse -mfpmath=sse -c Dialogs.C -MT Dialogs.lo -MD -MP -MF .deps/Dialogs.TPlo
In file included from ../../../boost/boost/config.hpp:35,
                 from ../../../boost/boost/signals/signal_template.hpp:23,
                 from ../../../boost/boost/signals/signal0.hpp:30,
                 from ../../../src/frontends/Dialogs.h:22,
                 from Dialogs_impl.h:18,
                 from Dialogs.C:19:
../../../boost/boost/config/compiler/gcc.hpp:57:7: warning: #warning "Unknown compiler version - please run the configure tests and report the results"
echo timestamp > ColorHandler.lo

At the end:

g++ -DHAVE_CONFIG_H -I. -I. -I../../../src -I../../../images -I../../../src -I../../../src/frontends -I../../../src/frontends/controllers -I../../../boost -isystem /usr/X11R6/include -O2 -march=pentium3 -pipe -mmmx -msse -mfpmath=sse -c Dialogs2.C -MT Dialogs2.lo -MD -MP -MF .deps/Dialogs2.TPlo
In file included from ../../../boost/boost/config.hpp:35,
                 from ../../../boost/boost/signals/signal_template.hpp:23,
                 from ../../../boost/boost/signals/signal0.hpp:30,
                 from ../../../src/frontends/Dialogs.h:22,
                 from Dialogs_impl.h:18,
                 from Dialogs2.C:19:
../../../boost/boost/config/compiler/gcc.hpp:57:7: warning: #warning "Unknown compiler version - please run the configure tests and report the results"
../../../boost/boost/type_traits/is_class.hpp: In instantiation of `boost::detail::is_function_impl<boost::_bi::bind_t<, boost::_mfi::mf0<, ControlInset<, ::string> >, boost::_bi::list1<boost::_bi::value<ControlInset<, ::string>*> > > >':
../../../boost/boost/type_traits/is_class.hpp:65:   instantiated from `boost::is_function<boost::_bi::bind_t<, boost::_mfi::mf0<, ControlInset<, ::string> >, boost::_bi::list1<boost::_bi::value<ControlInset<, ::string>*> > > >'
../../../boost/boost/type_traits/is_class.hpp:65:   instantiated from `boost::de
tail::is_class_impl<boost::_bi::bind_t<, boost::_mfi::mf0<, ControlInset<, ::string> >, boost::_bi::list1<boost::_bi::value<ControlInset<, ::string>*> > > >'
../../../boost/boost/type_traits/is_stateless.hpp:31:   instantiated from `boost::is_class<boost::_bi::bind_t<, boost::_mfi::mf0<, ControlInset<, ::string> >, boost::_bi::list1<boost::_bi::value<ControlInset<, ::string>*> > > >'
../../../boost/boost/type_traits/is_stateless.hpp:31:   instantiated from `boost::detail::is_stateless_impl<boost::_bi::bind_t<, boost::_mfi::mf0<, ControlInset<, ::string> >, boost::_bi::list1<boost::_bi::value<ControlInset<, ::string>*> > > >'
../../../boost/boost/function/function_base.hpp:124:   instantiated from `boost::is_stateless<boost::_bi::bind_t<, boost::_mfi::mf0<, ControlInset<, ::string> >, boost::_bi::list1<boost::_bi::value<ControlInset<, ::string>*> > > >'
../../../boost/boost/function/function_base.hpp:124:   instantiated from `get_function_tag<boost::_bi::bind_t<, boost::_mfi::mf0<, ControlInset<, ::string> >, boost::_bi::list1<boost::_bi::value<ControlInset<, ::string>*> > > >'

Internal compiler error: Error reporting routines re-entered.
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://bugs.gentoo.org/> for instructions.
g++: Internal error: Segmentation fault (program cc1plus)
Please submit a full bug report.
See <URL:http://bugs.gentoo.org/> for instructions.
{standard input}: Assembler messages:
{standard input}:1813: Warning: end of file not at end of a line; newline inserted
{standard input}:3014: Error: unknown pseudo-op: `.l'
make[5]: *** [Dialogs.lo] Fehler 1
make[5]: *** Warte auf noch nicht beendete Prozesse...
The bug is not reproduceable, so it is likely a hardware or OS problem
make[5]: *** [Dialogs2.lo] Fehler 1
make[5]: Leaving directory `/var/tmp/portage/lyx-1.3.2-r1/work/lyx-1.3.2/src/frontends/xforms'
make[4]: *** [all-recursive] Fehler 1
make[4]: Leaving directory `/var/tmp/portage/lyx-1.3.2-r1/work/lyx-1.3.2/src/frontends/xforms'
make[3]: *** [all-recursive] Fehler 1
make[3]: Leaving directory `/var/tmp/portage/lyx-1.3.2-r1/work/lyx-1.3.2/src/frontends'
make[2]: *** [all-recursive] Fehler 1
make[2]: Leaving directory `/var/tmp/portage/lyx-1.3.2-r1/work/lyx-1.3.2/src'
make[1]: *** [all] Fehler 2
make[1]: Leaving directory `/var/tmp/portage/lyx-1.3.2-r1/work/lyx-1.3.2/src'
make: *** [all-recursive] Fehler 1

!!! ERROR: app-office/lyx-1.3.2-r1 failed.
!!! Function src_compile, Line 71, Exitcode 2
!!! emake failed


My USE flags are these: 
USE="-* lcms tiff wmf gif png spell jpeg moznoirc moznocompose acpi motif gtk2 cups perl python vim-with-x zlib xml2 truetype x86 nls mmx sse ncurses gdbm readline berkdb ssl X pam crypt tcpd"

CFLAGS="-O2 -march=pentium3 -pipe -mmmx -msse -mfpmath=sse"

I'll post the output of emergin 1.3.3 in some minutes or so.

Thx
Sebastian
Comment 2 Sebastian 2004-02-10 04:48:52 UTC
Sorry,

I tried emerging 1.3.3 twice within X, but in the end it always used large amounts of swap and locked my X up. So the box frooze and I couldn't cut'n'paste. I have only 128MB RAM on this P3 700MHz laptop but that shouldn't be the problem (at least I hope so).

Thx
Sebastian
Comment 3 Mamoru KOMACHI (RETIRED) gentoo-dev 2004-02-20 11:32:30 UTC
Hum, which version of boost do you have?
Comment 4 Sebastian 2004-02-21 06:46:29 UTC
Hmm, I don't have boost. And when I do an "emerge lyx -pv" a package named "boost" is not shown.

Sebastian
Comment 5 Mamoru KOMACHI (RETIRED) gentoo-dev 2004-02-21 12:28:02 UTC
I've just committed lyx-1.3.4.ebuild. Could you try this
and report back if the problem still persists?
Comment 6 Sebastian 2004-02-22 09:59:10 UTC
Hello Mamoru,

I just tried 1.3.4. The first time it didn't build. Errors where the same as in the past. The next time I emerged boost first, but lyx just stuck to his own boost and failed again. I read somewhere that old versions of boost have problems with newer python.
I tried dev-libs/boost-1.30.2 and there were a lot of errors. I then emerged boost-1.31.0_alpha2-r1.ebuild which worked out. But n go with lyx as mentioned above. I tried buidling in X which crashed after a while. So no log again. I got a console log but it's kind of wired. 
I am sorry that this doesn't work here. I guess it's my fault, too, because I often use emerge --update --deep world. 
I have Python 2.3.3 and gcc-Version 3.3.2 20031218 (Gentoo Linux 3.3.2-r5, propolice-3.3-7), what versions do you have?

Take care
Sebastian 
Comment 7 Sebastian 2004-02-22 10:01:07 UTC
Created attachment 26114 [details]
weired log

Oops, I forgot. Here's what I got from emerge
/usr/portage/app-office/lyx/lyx-1.3.4.ebuild > log

Sebastian
Comment 8 Mamoru KOMACHI (RETIRED) gentoo-dev 2004-02-22 10:29:13 UTC
Hum, it's weird. I have gcc 3.3.2 20040119 (Gentoo Linux 3.3.2-r7, 
propolice-3.3-7) and my configuration is

Configuration
  Host type:                      i686-pc-linux-gnu
  Special build flags:            xforms-image-loader
  C   Compiler:                   gcc
  C   Compiler flags:             -g -O2
  C++ Compiler:                   g++ (3.3.2)
  C++ Compiler flags:             -O2 -mcpu=pentium3 -march=i586 -funroll-loops -fomit-frame-pointer -pipe
  Linker flags:
  Frontend:                       xforms
    libXpm version:               4.11
    libforms version:             1.0.0
  LyX binary dir:                 /usr/bin
  LyX files dir:                  /usr/share/lyx

and both USE="qt" and USE="-qt" worked fine here.

For the first thing, please check your local time by date command and
adjust it by ntpdate if you are in doubt. Second, if you set
CFLAGS/CXXFLAGS just "-O -pipe" and emerge lyx, does it compile fine?
Third, if you add some extra swap by

# dd if=/dev/zero of=/some/where/swap.img bs=1k count=512 (creates 512MB of disk image)
# mkswap /some/where/swap.img
# swapon /some/where/swap.img

and try emerge, does it solve X freeze durning emerge?  I have 386MB
of ram and 512MB of swap space to compile, so it may make some
difference. You may want to add an extra ram as the last resort ;)
Comment 9 Sebastian 2004-02-22 12:32:01 UTC
Hello again,

switching FLAGS didn't work, neither did adding 512MB of swap, also swap usage peaked at ~380MB. In the end g++ segfaults as before.
I'm done with it here. I have not enough time for this. I'll simply use latex as is.

Anyway, thanks man!

Sebastian  
Comment 10 Mamoru KOMACHI (RETIRED) gentoo-dev 2004-02-23 08:37:31 UTC
No worries. I mark the bug as WORKSFORME (because lyx emerges fine
with both with and without Qt here ...), so please reopen it bug
when you have time.