[Gcl-devel] Re: [Axiom-developer] current bug

[Camm Maguire]

Greetings!  Just a followup -- this appeared to work!  gcl will build
axiom cvs with this setting at least until the message:

make[3]: Leaving directory `/fix/s/camm/axiom/axiom1/new/new/src/algebra'
make[2]: *** No rule to make target 
`/fix/s/camm/axiom/axiom1/new/new/src/input/Makefile.pamphlet', needed by 
`/fix/s/camm/axiom/axiom1/new/new/src/input/Makefile'.  Stop.
make[2]: Leaving directory `/fix/s/camm/axiom/axiom1/new/new/src'
make[1]: *** [srcdir] Error 2
make[1]: Leaving directory `/fix/s/camm/axiom/axiom1/new/new'
make: *** [all] Error 2

Take care,


Camm Maguire <address@hidden> writes:

> Greetings!  I just looked at this again, and have come to the
> conclusion that the issue is the limit on the C stack size.  After
> increasing the value stack size to taste in stacks.h, one must (at
> least) also redefine MAX_STACK_SIZE in the .h file upward from its
> default value of (1<<23) /* 8Mb */.  I'm trying a compile now with
> double the value.  One may also have to adjust other stack limits -- I
> hope to report soon.  I did verify though in gdb that the segfault in
> executing the command below occurred on attempting to push the stack
> pointer %esp past the 8Mb boundary, 0xbf800000.
> 
> Take care,
> 
> root <address@hidden> writes:
> 
> > If you have the ability to do 1+1 in an axiom interpreter
> > built from the current sources then you can also try tracking
> > the same bug i'm chasing at the moment. It should be the case
> > that if you try to compile the XPR domain from xpoly.spad thus:
> > 
> > )co xpoly.spad )con XPR
> > 
> > you will see a value stack overflow. 
> > 
> > The nature of the problem is that there is a line of the form:
> > 
> > if R has Field
> >  then ...
> > 
> > where, in general, Field could be replaced by other categories.
> > Algebra code that contains "if R has" conditionals will not compile.
> > 
> > This causes an infinite loop in the compiler apparently related
> > to RING.
> > 
> > You now have all of the basic information I have. If we can solve
> > this problem it is likely that we can get the rest of the algebra
> > to compile and we will have a running system.
> > 
> > Tim
> > address@hidden
> > address@hidden
> > 
> > 
> > _______________________________________________
> > Axiom-developer mailing list
> > address@hidden
> > http://mail.nongnu.org/mailman/listinfo/axiom-developer
> > 
> > 
> > 
> 
> -- 
> Camm Maguire                                          address@hidden
> ==========================================================================
> "The earth is but one country, and mankind its citizens."  --  Baha'u'llah
> 
> 
> _______________________________________________
> Axiom-developer mailing list
> address@hidden
> http://mail.nongnu.org/mailman/listinfo/axiom-developer
> 
> 
> 

-- 
Camm Maguire                                            address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah