Is this our bug, or do we need to inform the swig guys ? Ali
On Jul 21, 2011, at 7:17 PM, Gabriel Michael Black wrote: > While trying to get gcc 4.6.0 working, I found that 2.0.4 introduced a bug > that prevented some things from compiling. It produced errors that were > different from yours, but you might still want to go back to 2.0.3 or 2.0.2. > > Gabe > > Quoting Carole-Jean Wu <[email protected]>: > >> Hi Nil, >> >> I use SWIG version 2.0.4.... >> >> Just did >> scons -c >> scons RUBY=true build/X86_FS/gem5.opt >> >> scons: Building targets ... >> [ CXX] X86_FS/python/swig/event_wrap.cc -> .o >> cc1plus: warnings being treated as errors >> build/X86_FS/python/swig/event_wrap.cc: In function 'void >> SWIG_Python_SetSwigThis(PyObject*, PyObject*)': >> build/X86_FS/python/swig/event_wrap.cc:2172: error: deprecated conversion >> from string constant to 'char*' >> build/X86_FS/python/swig/event_wrap.cc: In function 'void >> EventQueue_schedule(EventQueue*, Event*, Tick)': >> build/X86_FS/python/swig/event_wrap.cc:2980: error: '$self' was not declared >> in this scope >> build/X86_FS/python/swig/event_wrap.cc: In function 'void >> EventQueue_deschedule(EventQueue*, Event*)': >> build/X86_FS/python/swig/event_wrap.cc:2983: error: '$self' was not declared >> in this scope >> scons: *** [build/X86_FS/python/swig/event_wrap.o] Error 1 >> scons: building terminated because of errors. >> >> >> Still getting errors :( >> >> On Thu, Jul 21, 2011 at 7:50 PM, Nilanjan Goswami >> <[email protected]>wrote: >> >>> Hi Carole, >>> >>> Make sure that your SWIG version is 1.3 or above. I use 2.0.2. If you still >>> see error, clean the object files and then try building. >>> >>> Thanks, >>> Nil >>> >>> On Jul 21, 2011, at 7:41 PM, Carole-Jean Wu wrote: >>> >>> Thanks, Nil. >>> >>> I am hitting the next weird error message. >>> >>> build/X86_FS/python/swig/event_wrap.cc: In function 'void >>> EventQueue_deschedule(EventQueue*, Event*)': >>> build/X86_FS/python/swig/event_wrap.cc:2983: error: '$self' was not >>> declared in this scope >>> >>> The source code says: >>> >>> SWIGINTERN void EventQueue_schedule(EventQueue *self,Event *event,Tick >>> when){ >>> // Any python event that are scheduled must have their >>> >>> >>> // internal object's refcount incremented so that the object >>> >>> >>> // sticks around while it is in the event queue. >>> >>> >>> PythonEvent *pyevent = dynamic_cast<PythonEvent *>(event); >>> if (pyevent) >>> pyevent->incref(); >>> $self->schedule(event, when); >>> } >>> >>> >>> Any idea what the error message is about? >>> >>> thanks, >>> Carole >>> >>> On Thu, Jul 21, 2011 at 6:31 PM, Nilanjan Goswami >>> <[email protected]>wrote: >>> >>>> I have resolved the issue with gcc 4.4.5 version. Make sure that you >>>> mention CC=/path/to/gcc4.4.5 and CXX=/path/to/g++4.4.5 . I modified the >>>> .bashrc file to create a symbolic link to gcc4.4.5 initially. For some >>>> reason it was still pointing to default gcc (version 4.1.2). I exported CC >>>> and CXX variables with path to 4.4.5 gcc/g++ binaries. Now things works >>>> fine. However, it fails compile .opt version of the simulator due to memory >>>> shortage issue. >>>> >>>> --Nil >>>> >>>> >>>> On Jul 21, 2011, at 6:08 PM, Carole-Jean Wu wrote: >>>> >>>> I am still having trouble compiling gem5 with gcc 4.1.2. Has this issue >>>> resolved? >>>> >>>> thanks, >>>> Carole >>>> >>>> On Tue, Jul 19, 2011 at 1:45 PM, Nilanjan Goswami >>>> <[email protected]>wrote: >>>> >>>>> Hi, >>>>> >>>>> I have tried with different versions of gcc (4.4) and the problem >>>>> persists on cluster. Surprisingly it works on my local system. >>>>> >>>>> Thanks, >>>>> Nil >>>>> >>>>> On Jul 17, 2011, at 2:03 AM, Gabriel Michael Black wrote: >>>>> >>>>> > I think somebody else had a similar problem with gcc 4.1.2 but with >>>>> ALPHA_FS. I'm planning to look into it, but in the mean time you can try a >>>>> different version of gcc. >>>>> > >>>>> > Gabe >>>>> > >>>>> > Quoting Nilanjan Goswami <[email protected]>: >>>>> > >>>>> >> Hi, >>>>> >> >>>>> >> I am experiencing following compilation error in HPC cluster. The same >>>>> code works fine in local system. >>>>> >> >>>>> >> $ scons RUBY=true build/X86_FS/gem5.opt >>>>> >> * >>>>> >> [ CXX] X86_FS/mem/ruby/buffers/MessageBuffer.cc -> .o* >>>>> >> build/X86_FS/mem/ruby/buffers/MessageBuffer.cc: In member function >>>>> 'void MessageBuffer::printStats(std::ostream&)': >>>>> >> build/X86_FS/mem/ruby/buffers/MessageBuffer.cc:436: error: ambiguous >>>>> overload for 'operator<<' in 'std::operator<< [with _Traits = >>>>> std::char_traits<char>](((std::basic_ostream<char, std::char_traits<char> >>>>> >&)((std::ostream*)out)), ((const char*)"MessageBuffer: ")) << >>>>> ((MessageBuffer*)this)->MessageBuffer::m_name' >>>>> >> build/X86_FS/mem/ruby/buffers/MessageBuffer.hh:198: note: candidates >>>>> are: std::ostream& operator<<(std::ostream&, const MessageBuffer&) >>>>> >> build/X86_FS/base/stl_helpers.hh:86: note: >>>>> std::ostream& m5::stl_helpers::operator<<(std::ostream&, const C<T, A>&) >>>>> [with T = char, C = std::basic_string, A = std::char_traits<char>] >>>>> >> >>>>> /usr/lib/gcc/x86_64-redhat-linux/4.1.2/../../../../include/c++/4.1.2/bits/basic_string.h:2395: >>>>> note: std::basic_ostream<_CharT, _Traits>& >>>>> std::operator<<(std::basic_ostream<_CharT, _Traits>&, const >>>>> std::basic_string<_CharT, _Traits, _Alloc>&) [with _CharT = char, _Traits >>>>> = >>>>> std::char_traits<char>, _Alloc = std::allocator<char>] >>>>> >> build/X86_FS/base/cprintf_formats.hh: In function 'void >>>>> cp::_format_string(std::ostream&, const T&, cp::Format&) [with T = >>>>> std::basic_string<char, std::char_traits<char>, std::allocator<char> >]': >>>>> >> build/X86_FS/base/cprintf_formats.hh:356: instantiated from here >>>>> >> build/X86_FS/base/cprintf_formats.hh:202: error: ambiguous overload >>>>> for 'operator<<' in 'foo << data' >>>>> >> build/X86_FS/mem/ruby/buffers/MessageBuffer.hh:198: note: candidates >>>>> are: std::ostream& operator<<(std::ostream&, const MessageBuffer&) >>>>> >> build/X86_FS/base/stl_helpers.hh:86: note: >>>>> std::ostream& m5::stl_helpers::operator<<(std::ostream&, const C<T, A>&) >>>>> [with T = char, C = std::basic_string, A = std::char_traits<char>] >>>>> >> >>>>> /usr/lib/gcc/x86_64-redhat-linux/4.1.2/../../../../include/c++/4.1.2/bits/basic_string.h:2395: >>>>> note: std::basic_ostream<_CharT, _Traits>& >>>>> std::operator<<(std::basic_ostream<_CharT, _Traits>&, const >>>>> std::basic_string<_CharT, _Traits, _Alloc>&) [with _CharT = char, _Traits >>>>> = >>>>> std::char_traits<char>, _Alloc = std::allocator<char>] >>>>> >> build/X86_FS/base/cprintf_formats.hh:211: error: ambiguous overload >>>>> for 'operator<<' in 'out << std::basic_stringstream<_CharT, _Traits, >>>>> _Alloc>::str() const [with _CharT = char, _Traits = >>>>> std::char_traits<char>, >>>>> _Alloc = std::allocator<char>]()' >>>>> >> >>>>> >> >>>>> >> Thanks, >>>>> >> Nil >>>>> >> >>>>> > >>>>> > >>>>> > _______________________________________________ >>>>> > gem5-users mailing list >>>>> > [email protected] >>>>> > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >>>>> >>>>> _______________________________________________ >>>>> gem5-users mailing list >>>>> [email protected] >>>>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >>>>> >>>> >>>> _______________________________________________ >>>> gem5-users mailing list >>>> [email protected] >>>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >>>> >>>> >>>> >>>> _______________________________________________ >>>> gem5-users mailing list >>>> [email protected] >>>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >>>> >>> >>> _______________________________________________ >>> gem5-users mailing list >>> [email protected] >>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >>> >>> >>> >>> _______________________________________________ >>> gem5-users mailing list >>> [email protected] >>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >>> >> > > > _______________________________________________ > gem5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > _______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
