Re: [sage-devel] Where is sage/tests/parallel.pyx?

2015-02-15 Thread François Bissey
On 02/16/15 10:50, Antonio Rojas wrote: Hi, In module_list.py there is reference to a sage/tests/parallel.pyx extension, but there is no such file in the sources. This breaks compiling the Sage library with cython 0.22, which seems to be stricter than previous versions in dealing with missing s

[sage-devel] Where is sage/tests/parallel.pyx?

2015-02-15 Thread Antonio Rojas
Hi, In module_list.py there is reference to a sage/tests/parallel.pyx extension, but there is no such file in the sources. This breaks compiling the Sage library with cython 0.22, which seems to be stricter than previous versions in dealing with missing sources. -- You received this message b

Re: [sage-devel] bug report: "interval graph recognition"

2015-02-15 Thread Nathann Cohen
Fixed. Thanks Dima for the review ! http://trac.sagemath.org/ticket/17787 Nathann -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googleg

[sage-devel] Re: OpenBLAS beats MK for larger problems

2015-02-15 Thread maldun
I wouldn't see it that negative, since till now Intel's MKL was considered unbeatable in performance on Intel processors. On Sunday, February 15, 2015 at 3:47:41 PM UTC+1, Volker Braun wrote: > > Blas $X beats $Y on an aging cpu for a relatively unimportant set of > operations (specifically, mat

[sage-devel] Re: OpenBLAS beats MK for larger problems

2015-02-15 Thread Volker Braun
Blas $X beats $Y on an aging cpu for a relatively unimportant set of operations (specifically, matrix-vector). On Sunday, February 15, 2015 at 3:37:25 PM UTC+1, maldun wrote: > > Hi! > > Maybe this may be interesting for future consideration with respect which > blas implementation to choose:

[sage-devel] Re: OpenBLAS beats MK for larger problems

2015-02-15 Thread maldun
Sorry I mean MKL, somehow the I lost the L ... -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send

[sage-devel] OpenBLAS beats MK for larger problems

2015-02-15 Thread maldun
Hi! Maybe this may be interesting for future consideration with respect which blas implementation to choose: https://github.com/xianyi/OpenBLAS/issues/322 It appears that OpenBLAS seems to match more and more MKL performance. -- You received this message because you are subscribed to the Goog