| 1 |
Quoting Thomas Kahle <tomka@g.o>: |
| 2 |
|
| 3 |
> On 09:42 Thu 01 Nov 2012, fbissey@××××××××××××.nz wrote: |
| 4 |
>> Quoting Thomas Kahle <tomka@g.o>: |
| 5 |
>> |
| 6 |
>> > One more thing, This is fixed in 1.6 tree of numpy, where the |
| 7 |
>> > call to gfortran uses what is in the site config. |
| 8 |
>> > Will sage move to a higher version of numpy at some point? |
| 9 |
>> > |
| 10 |
>> > |
| 11 |
>> Yes! The problem that prevent us to upgrade numpy is fixed in numpy-1.7. The |
| 12 |
>> release cannot come quickly enough for us. |
| 13 |
> |
| 14 |
> OK. Any idea when this will happen? |
| 15 |
> |
| 16 |
>> For numpy 1.5 would by any chance gfortran be called with -fexternal-blas at |
| 17 |
>> some point? |
| 18 |
> |
| 19 |
> Can't find '-fexternal' in build.log. |
| 20 |
> |
| 21 |
>> On my system with python 3.2 lapack was not properly detected, |
| 22 |
>> that's why dotblas hasn't been built. Numpy 1.5.1 with python 3.2 used an |
| 23 |
>> internal implementation of the lapack functions it uses. |
| 24 |
>> |
| 25 |
>> This probably explains why scipy doesn't compile with numpy 1.5 and |
| 26 |
>> python 3.2. |
| 27 |
> |
| 28 |
> And this where all this started for me ... |
| 29 |
> |
| 30 |
> Anyway, the solution of my broken numpy,scipy is to wait for 1.7 or |
| 31 |
> rollback to a reference-lapack that installs itself as liblapack? |
| 32 |
> |
| 33 |
|
| 34 |
If you don't use numpy with python 3.x you could simply create a file |
| 35 |
/etc/portage/env/dev-python/numpy with USE_PYTHON="2.7" in it. |
| 36 |
That would skip the offending problem quite nicely without locking you |
| 37 |
in a particular blas implementation. |
| 38 |
|
| 39 |
Francois |
Archives