| 1 |
On 09:42 Thu 01 Nov 2012, fbissey@××××××××××××.nz wrote: |
| 2 |
> Quoting Thomas Kahle <tomka@g.o>: |
| 3 |
> |
| 4 |
> > One more thing, This is fixed in 1.6 tree of numpy, where the |
| 5 |
> > call to gfortran uses what is in the site config. |
| 6 |
> > Will sage move to a higher version of numpy at some point? |
| 7 |
> > |
| 8 |
> > |
| 9 |
> Yes! The problem that prevent us to upgrade numpy is fixed in numpy-1.7. The |
| 10 |
> release cannot come quickly enough for us. |
| 11 |
|
| 12 |
OK. Any idea when this will happen? |
| 13 |
|
| 14 |
> For numpy 1.5 would by any chance gfortran be called with -fexternal-blas at |
| 15 |
> some point? |
| 16 |
|
| 17 |
Can't find '-fexternal' in build.log. |
| 18 |
|
| 19 |
> On my system with python 3.2 lapack was not properly detected, |
| 20 |
> that's why dotblas hasn't been built. Numpy 1.5.1 with python 3.2 used an |
| 21 |
> internal implementation of the lapack functions it uses. |
| 22 |
> |
| 23 |
> This probably explains why scipy doesn't compile with numpy 1.5 and |
| 24 |
> python 3.2. |
| 25 |
|
| 26 |
And this where all this started for me ... |
| 27 |
|
| 28 |
Anyway, the solution of my broken numpy,scipy is to wait for 1.7 or |
| 29 |
rollback to a reference-lapack that installs itself as liblapack? |
| 30 |
|
| 31 |
Thanks, |
| 32 |
Thomas |
| 33 |
|
| 34 |
-- |
| 35 |
Thomas Kahle |
| 36 |
http://dev.gentoo.org/~tomka/ |
Archives