[SciPy-dev] Bugs?

[Charles R Harris]

Correction,

The documentation for gcc3.4 says

-march=cpu-type
    Generate instructions for the machine type cpu-type. The choices
for cpu-type are the same as for -mtune. Moreover, specifying
-march=cpu-type implies -mtune=cpu-type.
-mcpu=cpu-type
    A deprecated synonym for -mtune.

So it looks like the flags should just be -march=athlon64.

I sympathize with anyone trying to set up the correct gcc flags. It
gives me a headache trying to keep track of all the gcc options.

Chuck



On 2/9/06, Charles R Harris <charlesr.harris at gmail.com> wrote:
> OK,
>
> Removing the build directory fixed things up. Why isn't this part of
> the normal build?
>
> A nit here, the cflags are:
>
>  '-pthread -fno-strict-aliasing -DNDEBUG -O2 -g -pipe -m32 -march=i386
> -mtune=pentium4 -D_GNU_SOURCE -fPIC -mcpu=athlon64
> -fomit-frame-pointers -ffast-math -fno-strict-aliasing -fPIC'
>
> With -mcpu=athlon64 the flags -march=i386 -mtune=pentium4 are
> unnecessary and possibly counterproductive, at least for recent
> compiler versions. The cpu flag alone sets both the instruction set
> and tuning for the athlon.
>
> On 2/9/06, Robert Kern <robert.kern at gmail.com> wrote:
> > Charles R Harris wrote:
> > > Hi,
> > >
> > > Running latest numpy from cvs I get things like:
> > >
> > >
> > >>>>arange(25)
> > >
> > > array([-1214459252, -1214447606, -1214446384, -1214019752, -1212353482,
> > >                  0,           0,           0,           0,           0,
> > >                  0,           0,           1,           2,           4,
> > >                105,       10000,     9898072,   166961568,   167979252,
> > >          167979252,   168017420,   690120039,  1029992041,  1919249698])
> > >
> > > And
> > >
> > >
> > >>>>a = arange(25)
> > >>>>a.sort(kind='merge')
> > >
> > > Traceback (most recent call last):
> > >   File "<stdin>", line 1, in ?
> > > TypeError: desired sort not supported for this type
> > >
> > > The other sorts, 'heap' and 'quick', work fine. Any thoughts?
> >
> > Revision 2085 works fine for me on OS X. Did you "rm -r ./build/" before
> > building? Some parts deep inside the C code changed recently.
> >
> > --
> > Robert Kern
> > robert.kern at gmail.com
> >
> > "In the fields of hell where the grass grows high
> >  Are the graves of dreams allowed to die."
> >   -- Richard Harter
> >
> > _______________________________________________
> > Scipy-dev mailing list
> > Scipy-dev at scipy.net
> > http://www.scipy.net/mailman/listinfo/scipy-dev
> >
>