GCC 3.4

Message

johnnyboyro · Post by **johnnyboyro** » Sun May 02, 2004 1:25 pm

Is it possible that you are not root when emerging it?

jimbojetset · Post by **jimbojetset** » Sun May 02, 2004 1:39 pm

Can't compile gtkmm-1.2.9-r2

Code: Select all

In file included from ./../gtk--/base.h:34,
                 from ./../gtk--/object.h:30,
                 from wrap.cc:1:
./../gtk--/proxy.h: In member function `SigC::Slot0<R> Gtk::EmitProxySignal0<R, Obj, gObj, index, emit_func>::slot()':
./../gtk--/proxy.h:153: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h:153: error: (Each undeclared identifier is reported only once for each function it appears in.)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal0<R, Obj, gObj, index, emit_func>::emit()':
./../gtk--/proxy.h:160: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal0<R, Obj, gObj, index, emit_func>::operator()()':
./../gtk--/proxy.h:166: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `SigC::Slot1<R, P1> Gtk::EmitProxySignal1<R, P1, Obj, gObj, index, emit_func>::slot()':
./../gtk--/proxy.h:265: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal1<R, P1, Obj, gObj, index, emit_func>::emit(P1)':
./../gtk--/proxy.h:272: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal1<R, P1, Obj, gObj, index, emit_func>::operator()(P1)':
./../gtk--/proxy.h:278: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `SigC::Slot2<R, P1, P2> Gtk::EmitProxySignal2<R, P1, P2, Obj, gObj, index, emit_func>::slot()':
./../gtk--/proxy.h:377: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal2<R, P1, P2, Obj, gObj, index, emit_func>::emit(P1, P2)':
./../gtk--/proxy.h:384: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal2<R, P1, P2, Obj, gObj, index, emit_func>::operator()(P1, P2)':
./../gtk--/proxy.h:390: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `SigC::Slot3<R, P1, P2, P3> Gtk::EmitProxySignal3<R, P1, P2, P3, Obj, gObj, index, emit_func>::slot()':
./../gtk--/proxy.h:489: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal3<R, P1, P2, P3, Obj, gObj, index, emit_func>::emit(P1, P2, P3)':
./../gtk--/proxy.h:496: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal3<R, P1, P2, P3, Obj, gObj, index, emit_func>::operator()(P1, P2, P3)':
./../gtk--/proxy.h:502: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `SigC::Slot4<R, P1, P2, P3, P4> Gtk::EmitProxySignal4<R, P1, P2, P3, P4, Obj, gObj, index, emit_func>::slot()':
./../gtk--/proxy.h:601: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal4<R, P1, P2, P3, P4, Obj, gObj, index, emit_func>::emit(P1, P2, P3, P4)':
./../gtk--/proxy.h:608: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal4<R, P1, P2, P3, P4, Obj, gObj, index, emit_func>::operator()(P1, P2, P3, P4)':
./../gtk--/proxy.h:614: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `SigC::Slot5<R, P1, P2, P3, P4, P5> Gtk::EmitProxySignal5<R, P1, P2, P3, P4, P5, Obj, gObj, index, emit_func>::slot()':
./../gtk--/proxy.h:713: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal5<R, P1, P2, P3, P4, P5, Obj, gObj, index, emit_func>::emit(P1, P2, P3, P4, P5)':
./../gtk--/proxy.h:720: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal5<R, P1, P2, P3, P4, P5, Obj, gObj, index, emit_func>::operator()(P1, P2, P3, P4, P5)':
./../gtk--/proxy.h:726: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `SigC::Slot6<R, P1, P2, P3, P4, P5, P6> Gtk::EmitProxySignal6<R, P1, P2, P3, P4, P5, P6, Obj, gObj, index, emit_func>::slot()':
./../gtk--/proxy.h:825: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal6<R, P1, P2, P3, P4, P5, P6, Obj, gObj, index, emit_func>::emit(P1, P2, P3, P4, P5, P6)':
./../gtk--/proxy.h:832: error: `obj' undeclared (first use this function)
./../gtk--/proxy.h: In member function `R Gtk::EmitProxySignal6<R, P1, P2, P3, P4, P5, P6, Obj, gObj, index, emit_func>::operator()(P1, P2, P3, P4, P5, P6)':
./../gtk--/proxy.h:838: error: `obj' undeclared (first use this function)
In file included from ./../gtk--/object.h:30,
                 from wrap.cc:1:
./../gtk--/base.h: In member function `interf* G_List_Cpp_Iterator<impl, interf>::operator*() const':
./../gtk--/base.h:485: error: `wrap' is not a member of `Gtk'
make[4]: *** [wrap.lo] Error 1
make[4]: Leaving directory `/var/tmp/portage/gtkmm-1.2.9-r2/work/gtkmm-1.2.9/src/gtk--'
make[3]: *** [all-recursive] Error 1
make[3]: Leaving directory `/var/tmp/portage/gtkmm-1.2.9-r2/work/gtkmm-1.2.9/src/gtk--'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/var/tmp/portage/gtkmm-1.2.9-r2/work/gtkmm-1.2.9/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/var/tmp/portage/gtkmm-1.2.9-r2/work/gtkmm-1.2.9'
make: *** [all-recursive-am] Error 2

!!! ERROR: dev-cpp/gtkmm-1.2.9-r2 failed.
!!! Function src_compile, Line 41, Exitcode 2
!!! (no error message)

Any help appreciated

TheCoop · Post by **TheCoop** » Sun May 02, 2004 1:40 pm

johnnyboyro wrote:Is it possible that you are not root when emerging it?

no, its not possible. it compiles fine for 15mins before it bombs out, and i definately am root

4nykey · Post by **4nykey** » Sun May 02, 2004 2:46 pm

Well, while making in some dir it needs header '../nss/nss.h' which is there, but there's a text file called 'nss' in current dir and it somehow confuses gcc.
Altough cmdline has '-I..', it still locks. Workaround would be to '-I<absolute path>' or remove that 'nss' file, methinks.
There was another package with this sort of error, don't remeber which one.

thundersteele · Post by **thundersteele** » Sun May 02, 2004 2:52 pm

ecatmur wrote:It looks simple enough.

sm3600.h:279: __SM3600EXPORT__ TState GetScannerModel(unsigned short idVendor, unsigned short idProduct);

sm3600-scanmtek.c:68: __SM3600EXPORT__
TModel GetScannerModel(unsigned short idVendor,
unsigned short idProduct)

You need to patch sm3600.h to declare GetScannerModel as returning a TModel. Remember to submit your patches to bugzilla at gentoo.org and upstream.

Thanx, this was really an easy one. Now the more complicated part:
I hope this is a correct patch:
sane-backends-1.0.13-gcc34.patch

Code: Select all

diff -u -r sane-backends-1.0.13/backend/sm3600.h sane-backends.new/backend/sm3600.h
--- sane-backends-1.0.13/backend/sm3600.h	2003-09-12 16:18:58.000000000 +0200
+++ sane-backends.new/backend/sm3600.h	2004-05-02 16:30:19.429232120 +0200
@@ -276,7 +276,7 @@
 __SM3600EXPORT__ TState DoReset(TInstance *this);
 __SM3600EXPORT__ TState WaitWhileBusy(TInstance *this,int cSecs);
 __SM3600EXPORT__ TState WaitWhileScanning(TInstance *this,int cSecs);
-__SM3600EXPORT__ TState GetScannerModel(unsigned short idVendor, unsigned short idProduct);
+__SM3600EXPORT__ TModel GetScannerModel(unsigned short idVendor, unsigned short idProduct);
 
 #ifdef INSANE_VERSION
 __SM3600EXPORT__ TState DoLampSwitch(TInstance *this,int nPattern);

I just tried applying it to the original sources, and it changed them correctly. So I will submit it to bugs.gentoo.org

jimbojetset · Post by **jimbojetset** » Sun May 02, 2004 10:00 pm

[EDIT]Sorry me being dumb[/EDIT]

ecatmur · Post by **ecatmur** » Mon May 03, 2004 5:49 am

thundersteele wrote:Thanx, this was really an easy one. Now the more complicated part:
I hope this is a correct patch:
sane-backends-1.0.13-gcc34.patch
I just tried applying it to the original sources, and it changed them correctly. So I will submit it to bugs.gentoo.org

Looks good, well done.

When someone submits their first patch to Bugzilla, an angel gets his wings...

stonent · Post by **stonent** » Mon May 03, 2004 5:59 am

I submitted a one line patch to an ebuild... Does that count?

ecatmur · Post by **ecatmur** » Mon May 03, 2004 6:43 am

Oh, definitely. The only difference between a 1 line patch and a 1000 line patch is the amount of bloody-mindedness (and caffeine) that is needed to produce the latter. The one-line-patch problem can be just as much a test of your debugging skills as the bug that takes a 1000 line patch to fix. (the hours I've spent putting semicolons on default: statements and removing them from after function forward declarations, just for the sake of gcc 3.4...)

Ecco · Post by **Ecco** » Mon May 03, 2004 8:14 am

I decided to join in on the 'how to make your system bleeding edge yet only semi-functional'-fun and compiled gcc3.4, then glibc with nptl, and then my world -e

Worked out pretty well, specially with the patches provided in this thread

If anyone is using the hpijs printer drivers, it doesn't compile without a minor patch

Code: Select all

diff -u hpijs-1.5-orig/hpijs.cpp hpijs-1.5/hpijs.cpp
--- hpijs-1.5-orig/hpijs.cpp	2004-05-02 19:25:56.135556960 +0000
+++ hpijs-1.5/hpijs.cpp	2004-05-02 19:56:03.839744152 +0000
@@ -79,7 +79,10 @@
    {
       BOOL        bDevText;
       bug("unable to set Quality=%d, ColorMode=%d, MediaType=%d, err=%d\n", pSS->Quality, pSS->ColorMode, pSS->MediaType, r);
-      pSS->pPC->GetPrintModeSettings ((QUALITY_MODE)pSS->Quality, (MEDIATYPE)pSS->MediaType, (COLORMODE)pSS->ColorMode, bDevText);
+      apdk::QUALITY_MODE quality_mode = (QUALITY_MODE)pSS->Quality;
+      apdk::MEDIATYPE mediatype = (MEDIATYPE)pSS->MediaType;
+      apdk::COLORMODE colormode = (COLORMODE)pSS->ColorMode;
+      pSS->pPC->GetPrintModeSettings (quality_mode, mediatype, colormode, bDevText);
       bug("following will be used Quality=%d, ColorMode=%d, MediaType=%d\n", pSS->Quality, pSS->ColorMode, pSS->MediaType);
    }
 
@@ -211,7 +214,10 @@
 
          /* Got a valid device class, let's set some print mode defaults. */
          BOOL        bDevText;
-         pSS->pPC->GetPrintModeSettings ((QUALITY_MODE)pSS->Quality, (MEDIATYPE)pSS->MediaType, (COLORMODE)pSS->ColorMode, bDevText);
+	 apdk::QUALITY_MODE quality_mode = (QUALITY_MODE)pSS->Quality;
+	 apdk::MEDIATYPE mediatype = (MEDIATYPE)pSS->MediaType;
+	 apdk::COLORMODE colormode = (COLORMODE)pSS->ColorMode;           
+         pSS->pPC->GetPrintModeSettings (quality_mode, mediatype, colormode, bDevText);
       }
       else
       {

Now I just anxiously waiting for reiser4 to go into mm-sources

robmoss · Post by **robmoss** » Mon May 03, 2004 1:05 pm

Once reiser4 hits my kernel sources, I'm moving

I'm also having the binutils-2.15/gcc-3.4 problem with x.org. Has anyone managed to fix this yet? If so, let me know and I'll put it in my overlay.

garrison · Post by **garrison** » Mon May 03, 2004 1:19 pm

robmoss2k wrote:Once reiser4 hits my kernel sources, I'm moving

I'm also having the binutils-2.15/gcc-3.4 problem with x.org. Has anyone managed to fix this yet? If so, let me know and I'll put it in my overlay.

I've not followed xorg-x11 discussion (if any), so...
Kill that "#define SharedLibraryLoadFlags" line insertion in ebuild.
"-Wl,-z,defs" is the one causing unresolved symbols. I've commented out that line and it works for me (i.e. no runtime problems).

jimbojetset · Post by **jimbojetset** » Mon May 03, 2004 4:17 pm

sob-foobillard not compiling

Code: Select all

billard3d.c:30:21: GL/glut.h: No such file or directory
ball.c: In function `ball_subdivide_nonrec':
ball.c:463: warning: unused variable `rho'
ball.c: In function `draw_balls':
ball.c:1511: warning: unused variable `fresnel_init'
ball.c: At top level:
ball.c:57: warning: 'fresnel_vert_prog_bind' defined but not used
ball.c:58: warning: 'fresnel_vert_prog_world_coord_str' defined but not used
ball.c:117: warning: 'fresnel_vert_prog_obj_coord_str' defined but not used
ball.c:179: warning: 'col_null' defined but not used
ball.c:185: warning: 'col_refl4' defined but not used
ball.c:228: warning: 'in_array_old' defined but not used
billard3d.c: In function `DisplayFunc':
billard3d.c:3387: warning: implicit declaration of function `glutBitmapCharacter'
billard3d.c:3387: error: `GLUT_BITMAP_HELVETICA_18' undeclared (first use in this function)
billard3d.c:3387: error: (Each undeclared identifier is reported only once
billard3d.c:3387: error: for each function it appears in.)
billard3d.c: In function `Init':
billard3d.c:4733: warning: implicit declaration of function `gluBuild2DMipmaps'
billard3d.c: At top level:
billard3d.c:211: warning: 'half_full_names' defined but not used
make[2]: *** [billard3d.o] Error 1
make[2]: *** Waiting for unfinished jobs....
evaluate_move.c: In function `evaluate_last_move_snooker':
evaluate_move.c:513: warning: enumeration value `SN_DONE' not handled in switch
table.c: In function `create_table':
table.c:1796: warning: unused variable `pass'
table.c:1108: warning: unused variable `bump_cloth_init'
table.c:1111: warning: unused variable `bumpcloth'
table.c:1128: warning: unused variable `cm'
table.c:1155: warning: unused variable `t_gen_params'
table.c:1156: warning: unused variable `s_gen_params'
make[1]: *** [all] Error 2
make: *** [all-recursive] Error 1

!!! ERROR: games-sports/foobillard-2.9 failed.
!!! Function src_compile, Line 39, Exitcode 2
!!! (no error message)

My knowledge of C is poor to say the least but am willing to try-any help much appreciated

James

jimbojetset · Post by **jimbojetset** » Mon May 03, 2004 4:23 pm

not a gcc problem at all - dependency on glut not picked up in ebuild-emerged glut and it compiles fine

mekong · Post by **mekong** » Mon May 03, 2004 4:56 pm

TheCoop wrote:
johnnyboyro wrote:Is it possible that you are not root when emerging it?
no, its not possible. it compiles fine for 15mins before it bombs out, and i definately am root

Hi, do you use reiser4? check page 25 of this thread, there are people with same error on reiser4.

neenee · Post by **neenee** » Mon May 03, 2004 5:10 pm

mekong wrote:
TheCoop wrote:
johnnyboyro wrote:Is it possible that you are not root when emerging it?
no, its not possible. it compiles fine for 15mins before it bombs out, and i definately am root
Hi, do you use reiser4? check page 25 of this thread, there are people with same error on reiser4.

yes. and also people without problems and reiser4,
who will say the problem is not reiser4-related.

like me

robmoss · Post by **robmoss** » Mon May 03, 2004 5:26 pm

garrison wrote:
robmoss2k wrote:Once reiser4 hits my kernel sources, I'm moving

I'm also having the binutils-2.15/gcc-3.4 problem with x.org. Has anyone managed to fix this yet? If so, let me know and I'll put it in my overlay.
I've not followed xorg-x11 discussion (if any), so...
Kill that "#define SharedLibraryLoadFlags" line insertion in ebuild.
"-Wl,-z,defs" is the one causing unresolved symbols. I've commented out that line and it works for me (i.e. no runtime problems).

garrison, you're the man! I'll do this as soon as I've eaten something. I haven't eaten all day, and I've been awake for 9 hours...

Twix · Post by **Twix** » Tue May 04, 2004 2:06 am

Packages failed for me:

qt
hpijs

robmoss · Post by **robmoss** » Tue May 04, 2004 2:13 am

Twix wrote:Packages failed for me:

qt
hpijs

I *think* there was a fix somewhere in this thread for hpijs, but it may just have been someone reporting that it didn't work.

QT, however, appears to work for everyone else - what's your error message?

Ecco · Post by **Ecco** » Tue May 04, 2004 8:34 am

Twix wrote:Packages failed for me:

qt
hpijs

In fact, there's a patch for hpijs on this very page

robmoss · Post by **robmoss** » Tue May 04, 2004 12:51 pm

Ecco wrote:In fact, there's a patch for hpijs on this very page

Ah yes, silly me!

spb · Post by **spb** » Tue May 04, 2004 5:56 pm

GMP 4.1.2 doesn't compile as-is; I've submitted a patch to Bugzilla (#49996) to fix the compilation.
Thought it might be at home in this thread too.

Chineseyes · Post by **Chineseyes** » Tue May 04, 2004 8:39 pm

/usr/kde/3.2/lib/libartskde.so: undefined reference to `virtual thunk to Arts::SynthModule_stub::streamInit()'
/usr/kde/3.2/lib/libartskde.so: undefined reference to `virtual thunk to Arts::StdSynthModule::stop()'
/usr/kde/3.2/lib/libartskde.so: undefined reference to `virtual thunk to Arts::SynthModule_stub::streamEnd()'
/usr/kde/3.2/lib/libartskde.so: undefined reference to `virtual thunk to Arts::StdSynthModule::streamEnd()'
/usr/kde/3.2/lib/libartskde.so: undefined reference to `virtual thunk to Arts::SynthModule_stub::autoSuspend()'
/usr/kde/3.2/lib/libartskde.so: undefined reference to `virtual thunk to Arts::StdSynthModule::streamInit()'
/usr/kde/3.2/lib/libartskde.so: undefined reference to `virtual thunk to Arts::StdSynthModule::streamStart()'
/usr/kde/3.2/lib/libartskde.so: undefined reference to `virtual thunk to Arts::StdSynthModule::autoSuspend()'
/usr/kde/3.2/lib/libartskde.so: undefined reference to `virtual thunk to Arts::StdSynthModule::start()'
/usr/kde/3.2/lib/libartskde.so: undefined reference to `virtual thunk to Arts::SynthModule_stub::start()'
/usr/kde/3.2/lib/libartskde.so: undefined reference to `virtual thunk to Arts::SynthModule_stub::streamStart()'
/usr/kde/3.2/lib/libartskde.so: undefined reference to `virtual thunk to Arts::SynthModule_stub::stop()'
collect2: ld returned 1 exit status
make[4]: *** [artsbuilder] Error 1
make[4]: Leaving directory `/var/tmp/portage/kdemultimedia-3.2.2-r1/work/kdemultimedia-3.2.2/arts/builder'
make[3]: *** [all-recursive] Error 1
make[3]: Leaving directory `/var/tmp/portage/kdemultimedia-3.2.2-r1/work/kdemultimedia-3.2.2/arts/builder'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/var/tmp/portage/kdemultimedia-3.2.2-r1/work/kdemultimedia-3.2.2/arts'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/var/tmp/portage/kdemultimedia-3.2.2-r1/work/kdemultimedia-3.2.2'
make: *** [all] Error 2

!!! ERROR: kde-base/kdemultimedia-3.2.2-r1 failed.
!!! Function kde_src_compile, Line 128, Exitcode 2
!!! died running emake, kde_src_compile:make

anyone else get this error compiling with gcc 3.4 I tried recompiling kde and arts with minimal cflags and no matter what cflags I use same error in kdemultimedia

robmoss · Post by **robmoss** » Tue May 04, 2004 8:42 pm

I *think* there's a fix for that floating around in this thread somewhere, but don't hold me to that.

robmoss · Post by **robmoss** » Tue May 04, 2004 11:42 pm

Well, I just unmerged GCC 3.3.3. I have a pure 3.4.0 system! I can live without gnomesword until I get that working, but everything else works great...