I have a case (old laptop with a somewhat non-standard x86) where the
only way to build Atlas spkg is to hack the spkg-install-script to re-
run make several times.
The non-hacked version does not work since I upgraded to gcc-4.5.1 -
it does the prescribed 5 attempts, and always fails due to highly
varying MFLOPS in the tuning phase, even if I don't run X11.
------------------------
build()
{
# build library
cd $CUR/ATLAS-build
make
make
make
make
make
make
---------------------------
3 or 4 times are usually enough.
(that's of course a crude hack, the right one would be to see if it
was a tuning error that resulted in failures, and repeat make based on
this, some finite number of times)
Is it just me, or someone saw something similar?
Dima
--
To post to this group, send an email to [email protected]
To unsubscribe from this group, send an email to
[email protected]
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
