| Trolltech Home | Qt4-preview-feedback Home | Recent Threads | All Threads | Author | Date | |
| All threads index page 1 | |
Trolls, I am still having trouble compiling qdesigner_propertysheet.cpp on Solaris 9 with SunStudio 8 fully patched. "CC -V" yields: CC: Sun C++ 5.5 Patch 113817-19 2006/10/13 The error message is now slightly different, but it is the same type of error as before: CC -c -I/code/dev-ms12/sunobjs/include -I/code/dev-ms12/sunobjs/include/mysql -O2 -mt -KPIC -DQT_EDITION=QT_EDITION_DESKTOP -DQDESIGNER_SDK_LIBRARY -DQDESIGNER_EXTENSION_LIBRARY -DQDESIGNER_UILIB_LIBRARY -DQDESIGNER_SHARED_LIBRARY -DQT_DESIGNER -DQT_NO_DEBUG -DQT_SCRIPT_LIB -DQT_XML_LIB -DQT_GUI_LIB -DQT_CORE_LIB -D_LARGEFILE64_SOURCE -D_LARGEFILE_SOURCE -DQT_SHARED -I/code2/Qt/qt-4.3.0-rc1/mkspecs/solaris-cc -I/code2/Qt/qt-4.3.0-rc1/tools/designer/src/lib -I../../../../include/QtCore -I../../../../include/QtCore -I../../../../include/QtGui -I../../../../include/QtGui -I../../../../include/QtXml -I../../../../include/QtXml -I../../../../include/QtScript -I../../../../include/QtScript -I../../../../include -I../../../../include -I/code2/Qt/qt-4.3.0-rc1/tools/designer/src/lib/extension -I/code2/Qt/qt-4.3.0-rc1/tools/designer/src/lib/sdk -I/code2/Qt/qt-4.3.0-rc1/tools/designer/src/lib/uilib -I/code2/Qt/qt-4.3.0-rc1/tools/designer/src/lib/shared -I.moc/release-shared -I.uic/release-shared -I. -o .obj/release-shared/qdesigner_propertysheet.o /code2/Qt/qt-4.3.0-rc1/tools/designer/src/lib/shared/qdesigner_propertysheet.cpp "/code2/Qt/qt-4.3.0-rc1/tools/designer/src/lib/shared/qdesigner_propertysheet.cpp", line 350: Error: The operand "___LQTB" cannot be assigned to. 1 Error(s) detected. *** Error code 1 The configure line is: /code2/Qt/qt-4.3.0-rc1/configure -v -prefix /code/qt/qt-4.3.0-rc1 -shared -fast -largefile -exceptions -accessibility -no-stl -plugin-sql-mysql -no-sql-sqlite -no-qt3support -system-zlib -no-gif -no-libtiff -system-libpng -no-libmng -system-libjpeg -openssl -no-rpath -no-optimized-qmake -no-nis -no-cups -no-iconv -no-pch -no-qdbus -no-nas-sound -opengl -sm -xshape -no-xinerama -no-xcursor -no-xfixes -no-xrandr -no-xrender -no-fontconfig -no-tablet -xkb -no-glib -no-separate-debug-info -platform solaris-cc -I/code/dev-ms12/sunobjs/include -I/code/dev-ms12/sunobjs/include/mysql -L/code/dev-ms12/sunobjs/lib -L/code/dev-ms12/sunobjs/lib/mysql -R/code/dev-ms12/sunobjs/lib -R/code/dev-ms12/sunobjs/lib/mysql This is a shadow-build situation, as I run configure from /code2/Qt/build-4.3.0/sol9-32. If there is any other information that is needed to debug this, please let me know and I'll provide it. Is there any possibility of getting this fixed before Qt-4.3.0 rolls out? Darin --- Darin Broady <dbroady1@xxxxxxxxx> wrote: > Hello all, > > I am attempting to compile Qt-4.3.0-beta on Solaris 9 with Sun Studio 8 fully > patched. It compiles almost everything until it gets to > src/lib/shared/qdesigner_propertySheet.cpp, where it spits out the following > error message: > > qdesigner_propertysheet.cpp, line 346: Error: The operation "___LTTB" connot > be assigned to. > > That particular string can be found no where in the source code. I've also > created a pre-processed file of said source and it doesn't contain that > string either. I've tried googling for ___LTTB and it comes up empty. > > Has anyone seen this before or have any idea on how to workaround it? > > Thanks. > > Darin Broady > To unsubscribe - send "unsubscribe" in the subject to qt4-preview-feedback-request@xxxxxxxxxxxxx
Hi, > line 350: Error: The operand "___LQTB" cannot be assigned to. What's exactly in line 350? It looks like the Solaris system headers contain a macro that replaces part of that line with "__LQTB". -- [ signature omitted ]
Dimitri,
The original function is:
bool QDesignerPropertySheet::isDynamicProperty(int index) const
{
if (m_info.value(index).defaultDynamic)
return false;
return isDynamic(index);
}
where line 350 is the if statement. I've preprocessed the file, and this
function is identical (except for spacing) in the .i file. So, macro is
inadvertantly replacing anythin here.
BTW, I've searched my entire Solaris system for "___LQTB" and it doesn't appear
anywhere.
Darin
--- Dimitri <dimitri@xxxxxxxxxxxxx> wrote:
> Hi,
>
> > line 350: Error: The operand "___LQTB" cannot be assigned to.
>
> What's exactly in line 350?
>
> It looks like the Solaris system headers contain a macro that replaces part
> of
> that line with "__LQTB".
>
> --
> Dimitri
>
> To unsubscribe - send "unsubscribe" in the subject to
> qt4-preview-feedback-request@xxxxxxxxxxxxx
>
>
Darin Broady
dbroady1@xxxxxxxxx
To unsubscribe - send "unsubscribe" in the subject to qt4-preview-feedback-request@xxxxxxxxxxxxx
Ooops, I mean "NO" macro is inadvertantly replacing anything here.
Darin
--- Darin Broady <dbroady1@xxxxxxxxx> wrote:
> Dimitri,
>
> The original function is:
>
> bool QDesignerPropertySheet::isDynamicProperty(int index) const
> {
> if (m_info.value(index).defaultDynamic)
> return false;
>
> return isDynamic(index);
> }
>
> where line 350 is the if statement. I've preprocessed the file, and this
> function is identical (except for spacing) in the .i file. So, macro is
> inadvertantly replacing anythin here.
>
> BTW, I've searched my entire Solaris system for "___LQTB" and it doesn't
> appear
> anywhere.
>
> Darin
>
> --- Dimitri <dimitri@xxxxxxxxxxxxx> wrote:
>
> > Hi,
> >
> > > line 350: Error: The operand "___LQTB" cannot be assigned to.
> >
> > What's exactly in line 350?
> >
> > It looks like the Solaris system headers contain a macro that replaces part
> > of
> > that line with "__LQTB".
> >
> > --
> > Dimitri
> >
> > To unsubscribe - send "unsubscribe" in the subject to
> > qt4-preview-feedback-request@xxxxxxxxxxxxx
> >
> >
>
>
> Darin Broady
> dbroady1@xxxxxxxxx
>
> To unsubscribe - send "unsubscribe" in the subject to
> qt4-preview-feedback-request@xxxxxxxxxxxxx
>
>
To unsubscribe - send "unsubscribe" in the subject to qt4-preview-feedback-request@xxxxxxxxxxxxx
Hi,
> The original function is:
>
> bool QDesignerPropertySheet::isDynamicProperty(int index) const
> {
> if (m_info.value(index).defaultDynamic)
> return false;
>
> return isDynamic(index);
> }
>
> where line 350 is the if statement. I've preprocessed the file, and this
> function is identical (except for spacing) in the .i file. So, macro is
> inadvertantly replacing anythin here.
An yet that's the only possible explanation. Have you preprocessed with the
*exact same* compiler options used by the Qt build system?
> BTW, I've searched my entire Solaris system for "___LQTB" and it doesn't appear
> anywhere.
Where exactly have you looked for "___LQTB"? The macro may be defined by the
compiler's private headers in addition to the system headers in /usr/include.
--
[ signature omitted ]
Dimitri <dimitri@xxxxxxxxxxxxx> wrote: Hi,
> The original function is:
>
> bool QDesignerPropertySheet::isDynamicProperty(int index) const
> {
> if (m_info.value(index).defaultDynamic)
> return false;
>
> return isDynamic(index);
> }
>
> where line 350 is the if statement. I've preprocessed the file, and this
> function is identical (except for spacing) in the .i file. So, macro is
> inadvertantly replacing anythin here.
An yet that's the only possible explanation. Have you preprocessed with the
*exact same* compiler options used by the Qt build system?
> BTW, I've searched my entire Solaris system for "___LQTB" and it doesn't appear
> anywhere.
Where exactly have you looked for "___LQTB"? The macro may be defined by the
compiler's private headers in addition to the system headers in /usr/include.
--
[ signature omitted ]
Hi,
> The way I created the preprocessed output was take the compile line from
> the terminal, put it in a file, change -c to -E and remove the -o
> <outputFile>, and redirect the output to a file.i. I can give you the
> shell scrip t that I put this in, if it would help, as well as the
> preprocessed file, if it will help.
That looks OK. I really don't know. Maybe you could make the pre-processed
file available on a Web/FTP site?
> As for looking for LQTB, here's the searches I did that didn't find it.
> find /usr/include -exec grep LQTB {} \; -print
> find /opt/WS8/SUNWspro -exec grep LQTB {} \; -print
> find /code/dev-ms12/sunobjs -exec grep LQTB {} \; -print
> All turned up empty.
>
> Any other ideas?
Not really I'm afraid. By the way, it used to be "___LTTB" and now it's
"___LQTB", isn't it?
Have you asked Sun about this error?
--
[ signature omitted ]
Dimitri <dimitri@xxxxxxxxxxxxx> wrote: Hi,
> The way I created the preprocessed output was take the compile line from
> the terminal, put it in a file, change -c to -E and remove the -o
> , and redirect the output to a file.i. I can give you the
> shell scrip t that I put this in, if it would help, as well as the
> preprocessed file, if it will help.
That looks OK. I really don't know. Maybe you could make the pre-processed
file available on a Web/FTP site?
> As for looking for LQTB, here's the searches I did that didn't find it.
> find /usr/include -exec grep LQTB {} \; -print
> find /opt/WS8/SUNWspro -exec grep LQTB {} \; -print
> find /code/dev-ms12/sunobjs -exec grep LQTB {} \; -print
> All turned up empty.
>
> Any other ideas?
Not really I'm afraid. By the way, it used to be "___LTTB" and now it's
"___LQTB", isn't it?
Have you asked Sun about this error?
--
[ signature omitted ]
Message 9 in thread
Hi, > I've compressed the preprocessed output and am attaching it > (qdesigner_propertysheet.i.gz). I was able to compile this pre-processed file using our Sun Studio 8 compiler: $ CC -c -O2 -mt -KPIC qdesigner_propertysheet.i $ $ CC -V CC: Sun C++ 5.5 Patch 113817-09 2004/08/03 $ What happens if you try the above commands? As you can see, our compiler does not seem to have the latest patches. Maybe some of the more recent ones are bad patches. I really think you should contact Sun about this issue. > As for the operand name, yes it did change. I can't find ei ther name > anywhere on my system. To be honest, it sounds like a compiler problem, > but it's like nothing I've ever seen. Have you (or someone at > Trolltech) compiled Qt-4.3.0 (beta or rc1) with SunStudio 8? I where > SunStudio 9 and 10 are regularly used, but nothing is mentioned about > SunStudio 8. I think we're mainly using Sun Studio 8 to test Qt. > As for contacting Sun, I have not done that yet. If the error message were not so cryptic, we could maybe try and work around the problem. I think it's best to contact Sun. -- [ signature omitted ]
Dimitri <dimitri@xxxxxxxxxxxxx> wrote: Hi, > I've compressed the preprocessed output and am attaching it > (qdesigner_propertysheet.i.gz). I was able to compile this pre-processed file using our Sun Studio 8 compiler: $ CC -c -O2 -mt -KPIC qdesigner_propertysheet.i $ $ CC -V CC: Sun C++ 5.5 Patch 113817-09 2004/08/03 $ What happens if you try the above commands? As you can see, our compiler does not seem to have the latest patches. Maybe some of the more recent ones are bad patches. I really think you should contact Sun about this issue. > As for the operand name, yes it did change. I can't find ei ther name > anywhere on my system. To be honest, it sounds like a compiler problem, > but it's like nothing I've ever seen. Have you (or someone at > Trolltech) compiled Qt-4.3.0 (beta or rc1) with SunStudio 8? I where > SunStudio 9 and 10 are regularly used, but nothing is mentioned about > SunStudio 8. I think we're mainly using Sun Studio 8 to test Qt. > As for contacting Sun, I have not done that yet. If the error message were not so cryptic, we could maybe try and work around the problem. I think it's best to contact Sun. -- [ signature omitted ]