Bug#609287: atlas: may be Debian experimental could get some fresh upstream's unstable release (e.g. 3.9.32)

[Yaroslav Halchenko]

FWIW... I decided not to pollute bugspace but could not stay silent so
just commenting on here:

on powerpc(pescetti, sid chroot), while trying to unittest nibabel:

Program received signal SIGILL, Illegal instruction.
0x0f577bbc in ATL_dnrm2_xp1yp0aXbX () from /usr/lib/libblas.so.3gf
(gdb) bt
#0  0x0f577bbc in ATL_dnrm2_xp1yp0aXbX () from /usr/lib/libblas.so.3gf
#1  0x0f5777e8 in ATL_dnrm2 () from /usr/lib/libblas.so.3gf
#2  0x0f490a90 in atl_f77wrap_dnrm2_ () from /usr/lib/libblas.so.3gf
#3  0x0f87b4e0 in dnrm2_ () from /usr/lib/libblas.so.3gf
#4  0x0ed3dee0 in dlarfg_ () from /usr/lib/liblapack.so.3gf
#5  0x0edcf5cc in dsytd2_ () from /usr/lib/liblapack.so.3gf
Backtrace stopped: frame did not save the PC

if need to reproduce -- just try to build  python-nibabel package there

On Thu, 13 Jan 2011, Sylvestre Ledru wrote:
> I agree it would be nice to have this release. However, when I tried to
> build it for an upload in Debian (the work is available in the svn of
> atlas/debian science as a branch), but it failed because of the bug
> #605370 ...
> I will have a look on these two bugs when Squeeze is out.

-- 
=------------------------------------------------------------------=
Keep in touch                                     www.onerussian.com
Yaroslav Halchenko                 www.ohloh.net/accounts/yarikoptic