Re: [Gems-users] problem(s) compiling Ruby under Linux (SuSe 9.3)

["ayu" <ayu@xxxxxxxxxxxxx>]

Mike, you're right, it wasn't a compiler problem. The problem turned out
to be in ~/gems/scripts/calc_host.sh file. The first line, #!/bin/csh
-f, for some reason doesn't work, even after switching to tcsh. So all
the Makefiles that contain 

HOST_TYPE := $(shell $(GEMS_ROOT)/scripts/calc_host.sh)

generates that error. The quick work around was to hardcode it as 

HOST_TYPE := x86-linux

Thanks for your help, as well as Min Xu's. With some prodding from you
guys I was able to pinpoint the problem. 

Of course, with that, I have other subsequent problems. Maybe if you can
point me in the right direction I can figure it out as well. Here's a
sample of it (since the errors generated is long):

ls: generated/MOESI_CMP_token/*.C: No such file or directory
Ruby: Building protocol: MOESI_CMP_token
Ruby: Making slicc
cd ../slicc; make
make[1]: Entering directory `/home/troll/gems/slicc'
SLICC: Running flex and bison
make generated/generated
make[2]: Entering directory `/home/troll/gems/slicc'
make[2]: `generated/generated' is up to date.
make[2]: Leaving directory `/home/troll/gems/slicc'
SLICC: Building slicc executable
make x86-linux/bin/slicc.exec
make[2]: Entering directory `/home/troll/gems/slicc'
/usr/bin/g++ -o x86-linux/obj/parser.o -c generated/parser.c
-DYY_NO_UNPUT -Wall -W -Wwrite-strings -Woverloaded-virtual -Wno-unused 
-march=i686 -g -ggdb -g3 -O2 -finline-functions
-DNO_VECTOR_BOUNDS_CHECKS -DMULTIFACET_NO_OPT_WARN   -I. -I../common
-Igenerated -Isymbols -Igenerator -Iast    
In file included from main.h:69,
                 from ast/ASTs.h:65,
                 from parser/parser.y:65:
ast/DeclListAST.h:69:17: AST.h: No such file or directory
In file included from ast/DeclListAST.h:70,
                 from main.h:69,
                 from ast/ASTs.h:65,
                 from parser/parser.y:65:
ast/DeclAST.h:69:17: AST.h: No such file or directory
In file included from ast/DeclAST.h:70,
                 from ast/DeclListAST.h:70,
                 from main.h:69,
                 from ast/ASTs.h:65,
                 from parser/parser.y:65:
ast/PairListAST.h:69:17: AST.h: No such file or directory
In file included from ast/PairListAST.h:70,
                 from ast/DeclAST.h:70,
                 from ast/DeclListAST.h:70,
                 from main.h:69,
                 from ast/ASTs.h:65,
                 from parser/parser.y:65:
ast/PairAST.h:69:17: AST.h: No such file or directory
In file included from ast/PairListAST.h:70,
                 from ast/DeclAST.h:70,
                 from ast/DeclListAST.h:70,
                 from main.h:69,
                 from ast/ASTs.h:65,
                 from parser/parser.y:65:
ast/PairAST.h:71: error: syntax error before `{' token
ast/PairAST.h:75: error: syntax error before `,' token
ast/PairAST.h:76: error: syntax error before `,' token
ast/PairAST.h:79: error: destructors must be member functions
ast/PairAST.h:82: error: non-member function `std::string key()' cannot
have `
   const' method qualifier
ast/PairAST.h: In function `std::string key()':
ast/PairAST.h:82: error: `m_key_ptr' undeclared (first use this
function)
ast/PairAST.h:82: error: (Each undeclared identifier is reported only
once for 
   each function it appears in.)
ast/PairAST.h: At global scope:
ast/PairAST.h:83: error: non-member function `std::string value()'
cannot have 
   `const' method qualifier
ast/PairAST.h: In function `std::string value()':
ast/PairAST.h:83: error: `m_value_ptr' undeclared (first use this
function)
ast/PairAST.h: At global scope:
ast/PairAST.h:85: error: virtual outside class declaration

Thanks,
Andy


----- Original Message Follows -----
> No, I don't think that is your problem.  We've used gcc 3.3.x before
> and your error messages don't look at all like a compiler version
> mismatch.
> 
> 
> >
> > I think I might have found out what the problem is. In
> > ~/common/Makefile.common (as well as ~/ruby/Makefile) it looks like
> > you're using gcc 3.4.1. I'm using gcc 3.3.5. Is it absolutely
> > essential to use gcc 3.4.1 (or later)? I'm also using make 3.80. I'd
> > like to make sure before I go through the hassle of getting the
> > latest copy of gcc. Thanks.
> >
> > ----- Original Message Follows -----
> > > It is something to do with your environment.  You can run make in
> > > debug mode
> > >
> > > make -d PROTOCOL=MOESI_CMP_token >& make.output
> > >
> > > However the output will be huge.
> > >
> > > Also ensure sure your version of make is GNU something or other:
> > >
> > > maya(4)% make --version
> > > GNU Make version 3.79.1, by Richard Stallman and Roland McGrath.
> > > Built for i686-pc-linux-gnu
> > > Copyright (C) 1988, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99,
> > >         2000 Free Software Foundation, Inc.
> > > This is free software; see the source for copying conditions.
> > > There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR
> > > A PARTICULAR PURPOSE.
> > >
> > > --Mike
> > >
> > >
> > > > I am not sure. We are using Tao linux here. Maybe your search
> > > > path is different?
> > > > The error message says it is the shell's problem.
> > > >
> > > > Can you set SHELL variable in the Makefile?
> > > >
> > > > ayu wrote:
> > > >
> > > > >I've tried running it on tcsh and it produced the same exact
> > > > errors. Any >other possible source of error?
> > > > >
> > > > >----- Original Message Follows -----
> > > > >
> > > > >
> > > > >>What shell are you using? Most of us use tcsh, not csh, here.
> > > > >>
> > > _______________________________________________
> > > Gems-users mailing list
> > > Gems-users@xxxxxxxxxxx
> > > https://lists.cs.wisc.edu/mailman/listinfo/gems-users
> > _______________________________________________
> > Gems-users mailing list
> > Gems-users@xxxxxxxxxxx
> > https://lists.cs.wisc.edu/mailman/listinfo/gems-users
> >