Bugzilla Bug 189738

Once bug #189725 is taken care of, please carry on stabilizing
sci-libs/lapack-refence-3.1.1-r1. It will be first and safer choice for new
virtual/lapack.

You need to switch to the new reference with "eselect blas set reference". The
src_test includes standard lapack tests and should be enough. 

Sébastien

I wondered why no one was reporting: I simply forgot to CC arches. duh.

Anyway, same thing as for the blas-reference packages, you need lapack-docs and
eselect-lapack to keyword/stable first.

Created an attachment (id=130046) [details]
AMD64 lapack-reference-3.1.1-r1 build log

AMD64:

eselect-lapack: don't yet have a complete lapack install to test with

lapack-docs-3.1.1:
during emerge, following is displayed, otherwise all ok:
>>> Unpacking lapack-man-3.1.1.tgz to /var/tmp/portage/app-doc/lapack-docs-3.1.1/work
>>> Unpacking lapackqref.ps to /var/tmp/portage/app-doc/lapack-docs-3.1.1/work
unpack lapackqref.ps: file format not recognized. Ignoring.
>>> Source unpacked.
>>> Compiling source in /var/tmp/portage/app-doc/lapack-docs-3.1.1/work/lapack-3.1.1/manpages ...
>>> Source compiled.

lapack-reference-3.1.1-r1:
Error during emerge:

gfortran   zchkab.o zdrvab.o zerrab.o zget08.o alaerh.o alahd.o  aladhd.o
alareq.o chkxer.o zget02.o zlarhs.o zlatb4.o zsbmv.o  xerbla.o \
        ../../tmglib_LINUX.a ../../SRC/.libs/liblapack.a -lblas    -o
../xlintstzc
gfortran   -c zerrtz.f -o zerrtz.o
make[1]: Leaving directory
`/var/tmp/portage/sci-libs/lapack-reference-3.1.1-r1/work/lapack-lite-3.1.1/TESTING/LIN'
gfortran   -c zerrvx.f -o zerrvx.o
Testing REAL LAPACK linear equation routines
./xlintsts < stest.in > stest.out 2>&1
make: *** [stest.out] Error 127
gfortran   -c zdrvgg.f -o zdrvgg.o
make: *** Waiting for unfinished jobs....
gfortran   -c zdrvsg.f -o zdrvsg.o
...
make[1]: Leaving directory
`/var/tmp/portage/sci-libs/lapack-reference-3.1.1-r1/work/lapack-lite-3.1.1/TESTING/LIN'

!!! ERROR: sci-libs/lapack-reference-3.1.1-r1 failed.
Call stack:
  ebuild.sh, line 1638:   Called dyn_test
  ebuild.sh, line 1047:   Called qa_call 'src_test'
  ebuild.sh, line 44:   Called src_test
  lapack-reference-3.1.1-r1.ebuild, line 73:   Called die

!!! lapack-reference tests failed.

full build log is attached

Portage 2.1.2.12 (default-linux/amd64/2007.0/desktop, gcc-4.1.2, glibc-2.5-r4,
2.6.22-gentoo-r5 x86_64)
=================================================================
System uname: 2.6.22-gentoo-r5 x86_64 AMD Athlon(tm) 64 X2 Dual Core Processor
4600+
Gentoo Base System release 1.12.9
Timestamp of tree: Tue, 04 Sep 2007 21:50:01 +0000
app-shells/bash:     3.2_p17
dev-java/java-config: 1.3.7, 2.0.33-r1
dev-lang/python:     2.4.4-r4
dev-python/pycrypto: 2.0.1-r6
sys-apps/baselayout: 1.12.9-r2
sys-apps/sandbox:    1.2.17
sys-devel/autoconf:  2.13, 2.61
sys-devel/automake:  1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2, 1.10
sys-devel/binutils:  2.17
sys-devel/gcc-config: 1.3.16
sys-devel/libtool:   1.5.24
virtual/os-headers:  2.6.21
ACCEPT_KEYWORDS="amd64"
CBUILD="x86_64-pc-linux-gnu"
CFLAGS="-march=athlon64 -O2 -msse3 -pipe"
CHOST="x86_64-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/share/X11/xkb"
CONFIG_PROTECT_MASK="/etc/env.d /etc/env.d/java/ /etc/gconf /etc/revdep-rebuild
/etc/terminfo"
CXXFLAGS="-march=athlon64 -O2 -msse3 -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="buildpkg collision-protect distlocks fixpackages metadata-transfer
multilib-strict sandbox sfperms strict test userpriv"
GENTOO_MIRRORS="http://gentoo.virginmedia.com http://distfiles.gentoo.org"
LINGUAS="en en_GB"
MAKEOPTS="-j3"
PKGDIR="/usr/portage/packages"
PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress
--force --whole-file --delete --delete-after --stats --timeout=180
--exclude=/distfiles --exclude=/local --exclude=/packages
--filter=H_**/files/digest-*"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/portage/local/layman/enlightenment /usr/portage/local"
SYNC="rsync://rsync.europe.gentoo.org/gentoo-portage"
USE="3dnow X Xaw3d a52 aac acpi alsa amd64 audiofile berkdb bitmap-fonts
branding bzip2 cairo caps cddb cdparanoia cdr cli cracklib cups curl dbus dri
dvd dvdr dvdread emboss encode evo exif expat fam ffmpeg firefox foomaticdb
fortran gdbm gif glitz gs gtk guile hal iconv icu imagemagick imlib isdnlog
ithreads java javascript jpeg libsamplerate libwww lirc lm_sensors logrotate
mad midi mmap mmx mmxext mp3 msn mudflap ncurses nls nptl nptlonly nsplugin
offensive ogg openal opengl openmp pam pango pcre pdf perl png ppds pppd python
qt3support readline reflection sdl session sndfile spl sse sse2 ssl
startup-notification svg symlink tcl tcpd test threads tiff tk truetype
truetype-fonts type1-fonts unicode usb v4l v4l2 vorbis wifi x264 xcb xine xml
xorg xosd xpm xscreensaver xv zlib" ALSA_CARDS="usb-audio"
ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file
hooks iec958 ioplug ladspa lfloat linear meter mulaw multi null plug rate route
share shm softvol" ELIBC="glibc" INPUT_DEVICES="keyboard mouse evdev"
KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001
mtxorb ncurses text" LINGUAS="en en_GB" LIRC_DEVICES="mceusb2" USERLAND="GNU"
VIDEO_CARDS="nvidia"
Unset:  CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL, LDFLAGS,
PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS

Hi Simon,

Could you show the stest.out file that fails? 
Also, could you tell me which blas you are compiling with (result of "eselect
blas show" and "pkg-config --libs blas"), which version of it, and what was the
fortran compiler/version used for blas?

Sebastien

Hi Simon,

Could you also please try with MAKEOPTS="-j1"; make's error 127
indicates that it can't find a program it needs to run the test.

Thanks,
Markus

# pkg-config --libs blas
Package blas was not found in the pkg-config search path.
Perhaps you should add the directory containing `blas.pc'
to the PKG_CONFIG_PATH environment variable
No package 'blas' found

# eselect blas show
lib: reference
# eselect blas set reference
Implementation "reference" already active for library directory "lib"!
Failed to switch to implementation "reference" for library directory "lib"!
!!! Error: One or more actions have failed!
Killed

lapack-reference fails to get even past configure here, because it can't find
blas.pc. There is no </usr/lib/pkgconfig/blas.pc>, only a
</usr/lib/blas/reference/blas.pc>. Emerging (src_compile and src_test)
lapack-reference works fine when I copy that file to </usr/lib/pkgconfig/>,
though.

> # pkg-config --libs blas
> Package blas was not found in the pkg-config search path.

You probably hit the eselect bug #189942 then. In order to carry on testing,
you can just remove the /etc/env.d/blas/$(get_libdir)/config and do a "eselect
blas set reference" again. Then running the normal test procedure should be
fine.

(In reply to comment #6)
> > # pkg-config --libs blas
> > Package blas was not found in the pkg-config search path.
> 
> You probably hit the eselect bug #189942 then. In order to carry on testing,
> you can just remove the /etc/env.d/blas/$(get_libdir)/config and do a "eselect
> blas set reference" again. Then running the normal test procedure should be
> fine.

Yes, that fixed it.

Now I'm feeling stupid, because repoman tells me this:

   sci-libs/lapack-reference/lapack-reference-3.1.1-r1.ebuild:
hppa(default-linux/hppa/2007.0/desktop) ['sci-libs/blas-atlas']

sci-libs/blas-reference is stable for HPPA, so why does it require blas-atlas?

same error with MAKEOPTS="-j1"

# cat stest.out
./xlintsts: error while loading shared libraries: libblas.so.0: cannot open
shared object file: No such file or directory

# eselect blas show
lib64: reference

# pkg-config --libs blas
-lblas

# equery f blas-reference
[ Searching for packages matching blas-reference... ]
* Contents of sci-libs/blas-reference-20070226:
/etc
/etc/env.d
/etc/env.d/blas
/etc/env.d/blas/lib64
/etc/env.d/blas/lib64/reference
/usr
/usr/lib64
/usr/lib64/blas
/usr/lib64/blas/reference
/usr/lib64/blas/reference/blas.pc
/usr/lib64/blas/reference/libblas.a
/usr/lib64/blas/reference/libblas.la
/usr/lib64/blas/reference/libblas.so -> libblas.so.0.0.0
/usr/lib64/blas/reference/libblas.so.0 -> libblas.so.0.0.0
/usr/lib64/blas/reference/libblas.so.0.0.0

from lapack-reference output (same compiler used for blas-reference)
 * You need one of these Fortran Compilers: g77 gfortran ifc
 * Installed are:  gfortran

# gcc -v  
Using built-in specs.
Target: x86_64-pc-linux-gnu
Configured with: /var/tmp/portage/sys-devel/gcc-4.1.2/work/gcc-4.1.2/configure
--prefix=/usr --bindir=/usr/x86_64-pc-linux-gnu/gcc-bin/4.1.2
--includedir=/usr/lib/gcc/x86_64-pc-linux-gnu/4.1.2/include
--datadir=/usr/share/gcc-data/x86_64-pc-linux-gnu/4.1.2
--mandir=/usr/share/gcc-data/x86_64-pc-linux-gnu/4.1.2/man
--infodir=/usr/share/gcc-data/x86_64-pc-linux-gnu/4.1.2/info
--with-gxx-include-dir=/usr/lib/gcc/x86_64-pc-linux-gnu/4.1.2/include/g++-v4
--host=x86_64-pc-linux-gnu --build=x86_64-pc-linux-gnu --disable-altivec
--enable-nls --without-included-gettext --with-system-zlib --disable-checking
--disable-werror --enable-secureplt --disable-libunwind-exceptions
--enable-multilib --enable-libmudflap --disable-libssp --disable-libgcj
--enable-languages=c,c++,fortran --enable-shared --enable-threads=posix
--enable-__cxa_atexit --enable-clocale=gnu
Thread model: posix
gcc version 4.1.2 (Gentoo 4.1.2 p1.0.1)

Can't do it on alpha/ia64 until bug 159002 is fixed

(In reply to comment #9)
> Can't do it on alpha/ia64 until bug 159002 is fixed
> 

The current bug does not hinge on #159002 if I am not mistaken. 
lapack-reference does not require blas-atlas and should be quite happy 
with blas-reference which is keyworded for alpha/ia64. Bug #159002 
is an upstream issue and it may very well be that we have to drop the 
atlas stuff from alpha/ia64 in the future if upstream discontinues support 
for these arches; but this is not clear at this point.

As a matter of fact, one of the main points of the blas/cblas/lapack
reorganization is to move away from the atlas ebuilds as defaults
and replace them by the reference ebuilds. So the sooner we can get
this done the better.


Thanks,
Markus

(In reply to comment #10)
armin # emerge -av lapack-reference

These are the packages that would be merged, in order:

Calculating dependencies |
!!! All ebuilds that could satisfy "sci-libs/blas-atlas" have been masked.
!!! One of the following masked packages is required to complete your request:
- sci-libs/blas-atlas-3.6.0 (masked by: missing keyword)
- sci-libs/blas-atlas-3.7.34 (masked by: missing keyword)
- sci-libs/blas-atlas-3.7.32 (masked by: missing keyword)
- sci-libs/blas-atlas-3.7.31 (masked by: missing keyword)
- sci-libs/blas-atlas-3.7.11 (masked by: missing keyword)
- sci-libs/blas-atlas-3.6.0-r1 (masked by: missing keyword)
- sci-libs/blas-atlas-3.6.0-r2 (masked by: missing keyword)
- sci-libs/blas-atlas-3.7.11-r1 (masked by: missing keyword)

For more information, see MASKED PACKAGES section in the emerge man page or 
refer to the Gentoo Handbook.
(dependency required by "sci-libs/lapack-reference-3.1.1-r1" [ebuild])

from sci-libs/lapack-reference-3.1.1-r1:
DEPEND="virtual/blas
        app-admin/eselect-lapack"

from profiles/base/virtuals:
virtual/blas                            sci-libs/blas-atlas

(In reply to comment #10)
> As a matter of fact, one of the main points of the blas/cblas/lapack
> reorganization is to move away from the atlas ebuilds as defaults
> and replace them by the reference ebuilds. So the sooner we can get
> this done the better.

The only way I can keyword =sci-libs/lapack-reference-3.1.1-r1 ~hppa or hppa (I
was going for the latter, actually) is to change the virtuals default in the
same go:

gentoo/cvs/gentoo-x86/profiles # diff -u base/virtuals{.,}
--- base/virtuals.      2007-09-06 17:00:04.000000000 +0200
+++ base/virtuals       2007-09-06 17:00:24.000000000 +0200
@@ -14,7 +14,7 @@
 virtual/baselayout                     sys-apps/baselayout
 virtual/bittorrent                     net-p2p/bittorrent
 virtual/blackbox                       x11-wm/blackbox
-virtual/blas                           sci-libs/blas-atlas
+virtual/blas                           sci-libs/blas-reference
 virtual/cblas                          sci-libs/blas-atlas
 virtual/cdrtools                       app-cdr/cdrkit
 virtual/commonlisp                     dev-lisp/sbcl
...
gentoo/cvs/gentoo-x86/sci-libs/lapack-reference # cvs diff
Index: lapack-reference-3.1.1-r1.ebuild
===================================================================
RCS file:
/var/cvsroot/gentoo-x86/sci-libs/lapack-reference/lapack-reference-3.1.1-r1.ebuild,v
retrieving revision 1.2
diff -u -B -r1.2 lapack-reference-3.1.1-r1.ebuild
--- lapack-reference-3.1.1-r1.ebuild    21 Aug 2007 17:19:57 -0000      1.2
+++ lapack-reference-3.1.1-r1.ebuild    6 Sep 2007 15:03:42 -0000
@@ -13,7 +13,7 @@

 SLOT="0"
 IUSE="doc"
-KEYWORDS="~amd64 ~x86"
+KEYWORDS="~amd64 hppa ~x86"

 DEPEND="virtual/blas
        app-admin/eselect-lapack"

Do I have permission to make these changes immediately or does it break the
tree elsewhere?

I think Sébastien has to have the final word on this since
he's been doing most of the blas/lapack stuff; I don't think
there should be any problems switching the virtual/blas 
in profiles from sci-libs/blas-atlas to sci-libs/blas-reference
since this is exactly what the new-style virtuals will do, but
I might have missed something. Any opinios from the other
sci folks on this.

Markus

BTW, this is exactly what we did a while back in freebsd's profile to enable it
to work and be keyworded.

(In reply to comment #14)
> BTW, this is exactly what we did a while back in freebsd's profile to enable it
> to work and be keyworded.
> 

Ahh, thanks for pointing this out; that's good to know. The only "problem" in 
switching the default virtual to *reference could be a handful (if at all) of
packages 
that in addition to blas also depend on cblas which blas-atlas happens to
provide 
but blas-reference doesn't. But these are packages that we need to track down
anyway and tag with virtual/cblas so it shouldn't keep us from switching the
blas
virtual. Let's see what Sébastien has to say.

Markus

Actually, we changed cblas in bsd too:

virtual/blas                    sci-libs/blas-reference
virtual/cblas                   sci-libs/blas-reference

This seems to work, but let me know if this is problematic...

(In reply to comment #16)
> virtual/cblas                   sci-libs/blas-reference

Yes, this should be

virtual/cblas                   sci-libs/cblas-reference

which is keyworded for your arch AFAIK. Since cblas is only
needed by a handful of sci-packages it probably hasn't 
cropped up yet but in contrast to blas-atlas, blas-reference
does not provide any cblas support.

Markus

Hmm, well, even though it is keyworded, I cannot get cblas to emerge.  It fails
during configure with:

checking for BLAS... gnome-config: not found
configure: error: Package requirements (blas) were not met:

No package 'blas' found


I have not looked into it further, but I wonder if I had trouble, and that was
why I did not set the virtual to cblas.  But I agree that it should be set that
way.  If you have ideas off the top of your head how to address the above, let
me know.

OK, so to have a smoother stabilization process and not pull out the blas-atlas
libraries, you should swap virtual/blas in your corresponding profile, but only
when blas-reference-20070226 is stable. virtual/cblas should has never existed
on the tree, it will be introduced only with new style virtual. You can remove
it, not a single ebuild depend on it.

-virtual/blas                           sci-libs/blas-atlas
+virtual/blas                           sci-libs/blas-reference
-virtual/cblas                          sci-libs/blas-atlas

Now there seems to have another problem reported on comment #8:

> ./xlintsts: error while loading shared libraries: libblas.so.0: cannot open
> shared object file: No such file or directory

Most likely the /usr/lib*/libblas.so.0 link is gone. I wondered how it
happened. It might be due to the eselect bug #190041, but it had no reason to
happen on a system where blas-reference wasn't installed previously. In this
case, the hack
rm -f /etc/env.d/blas/lib*/config && eselect blas set reference
should fix it temporarly to carry on testing.

What a mess this bug has turned into.

Stable for HPPA.

> lapack-reference-3.1.1-r1:
> Error during emerge:
> 
> gfortran   zchkab.o zdrvab.o zerrab.o zget08.o alaerh.o alahd.o  aladhd.o
> alareq.o chkxer.o zget02.o zlarhs.o zlatb4.o zsbmv.o  xerbla.o \
>         ../../tmglib_LINUX.a ../../SRC/.libs/liblapack.a -lblas    -o
> ../xlintstzc
> gfortran   -c zerrtz.f -o zerrtz.o
> make[1]: Leaving directory
> `/var/tmp/portage/sci-libs/lapack-reference-3.1.1-r1/work/lapack-lite-3.1.1/TESTING/LIN'
> gfortran   -c zerrvx.f -o zerrvx.o
> Testing REAL LAPACK linear equation routines
> ./xlintsts < stest.in > stest.out 2>&1
> make: *** [stest.out] Error 127
> gfortran   -c zdrvgg.f -o zdrvgg.o
> make: *** Waiting for unfinished jobs....
> gfortran   -c zdrvsg.f -o zdrvsg.o
> ...
> make[1]: Leaving directory
> `/var/tmp/portage/sci-libs/lapack-reference-3.1.1-r1/work/lapack-lite-3.1.1/TESTING/LIN'
> 
> !!! ERROR: sci-libs/lapack-reference-3.1.1-r1 failed.
> Call stack:
>   ebuild.sh, line 1638:   Called dyn_test
>   ebuild.sh, line 1047:   Called qa_call 'src_test'
>   ebuild.sh, line 44:   Called src_test
>   lapack-reference-3.1.1-r1.ebuild, line 73:   Called die
> 
> !!! lapack-reference tests failed.
> 

I get the same test error as well on alpha/ia64/x86

> rm -f /etc/env.d/blas/lib*/config && eselect blas set referenc

this fixes it for me. After this, all tests pass on AMD64 & emerge succeeds.
Its up to devs to decide what to stabilise now :P

x86 stable

Removed virtual/cblas from the bsd virtuals as suggested above.

On alpha i get this:

      ../../tmglib_LINUX.a ../../SRC/.libs/liblapack.a -lblas    -o ../xlintsts
make[1]: Leaving directory
`/var/tmp/portage/sci-libs/lapack-reference-3.1.1-r1/work/lapack-lite-3.1.1/TESTING/LIN'
Testing REAL LAPACK linear equation routines
./xlintsts < stest.in > stest.out 2>&1
/bin/sh: line 1: 26668 Floating point exception./xlintsts < stest.in >
stest.out 2>&1
make: *** [stest.out] Error 136
 * 
 * ERROR: sci-libs/lapack-reference-3.1.1-r1 failed.
 * Call stack:
 *   ebuild.sh, line 1654:   Called dyn_test

ia64/sparc stable

(In reply to comment #25)

> ./xlintsts < stest.in > stest.out 2>&1
> /bin/sh: line 1: 26668 Floating point exception./xlintsts < stest.in >
> stest.out 2>&1
> make: *** [stest.out] Error 136

Could you post the stest.out and tell me:
- which fortran compiler (and which version)
- eselect blas show

> ia64/sparc stable

Did you forget to commit?

(In reply to comment #26)
> Could you post the stest.out and tell me:
It's empty

> - which fortran compiler (and which version)
gcc-4.1.2

> - eselect blas show
lib: reference
> 
> > ia64/sparc stable
> 
> Did you forget to commit?
> 
Right, fixed

ppc64 stable

> ppc64 stable

forgot to commit app-doc/lapack-docs?


other arches, could you check again? the blas/cblas virtual transition is done,
and we would like to do the same for lapack.

(In reply to comment #29)
> > ppc64 stable
> 
> forgot to commit app-doc/lapack-docs?

whoops.. fixed

~ppc keywords added. leaving us on cc for stabling in 30 days.

alpha is stable, even if the tests fail, but apps work fine against it. And
lapack-atlas doesn't compile

AMD64:
Compiles fine
No collision
Testing goes fine

So I'd say let roll on amd64.. 

emerge --info:

Portage 2.1.3.16 (default-linux/amd64/2007.0/no-multilib, gcc-4.1.2,
glibc-2.6.1-r0, 2.6.22-gentoo-r8 x86_64)
=================================================================
System uname: 2.6.22-gentoo-r8 x86_64 AMD Turion(tm) 64 Mobile Technology MT-28
Timestamp of tree: Sun, 04 Nov 2007 02:20:01 +0000
distcc 2.18.3 x86_64-pc-linux-gnu (protocols 1 and 2) (default port 3632)
[enabled]
app-shells/bash:     3.2_p17
dev-java/java-config: 1.3.7, 2.0.33-r1
dev-lang/python:     2.4.4-r6
dev-python/pycrypto: 2.0.1-r6
sys-apps/baselayout: 1.12.9-r2
sys-apps/sandbox:    1.2.18.1-r2
sys-devel/autoconf:  2.61-r1
sys-devel/automake:  1.4_p6, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2, 1.10
sys-devel/binutils:  2.18-r1
sys-devel/gcc-config: 1.3.16
sys-devel/libtool:   1.5.24
virtual/os-headers:  2.6.22-r2
ACCEPT_KEYWORDS="amd64"
CBUILD="x86_64-pc-linux-gnu"
CFLAGS="-march=k8 -msse3 -O2 -pipe"
CHOST="x86_64-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/kde/3.5/env /usr/kde/3.5/share/config
/usr/kde/3.5/shutdown /usr/share/X11/xkb /usr/share/config"
CONFIG_PROTECT_MASK="/etc/env.d /etc/env.d/java/ /etc/fonts/fonts.conf
/etc/gconf /etc/revdep-rebuild /etc/terminfo /etc/texmf/web2c
/etc/udev/rules.d"
CXXFLAGS="-march=k8 -msse3 -O2 -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="collision-protect distcc distlocks metadata-transfer multilib-strict
sandbox sfperms strict test unmerge-orphans userfetch userpriv"
GENTOO_MIRRORS="http://ftp.snt.utwente.nl/pub/os/linux/gentoo
ftp://ftp.snt.utwente.nl/pub/os/linux/gentoo
ftp://mirror.scarlet-internet.nl/pub/gentoo http://gentoo.tiscali.nl/
ftp://gentoo.tiscali.nl/pub/mirror/gentoo/ "
LINGUAS="en"
MAKEOPTS="-j4"
PKGDIR="/usr/portage/packages"
PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress
--force --whole-file --delete --delete-after --stats --timeout=180
--exclude=/distfiles --exclude=/local --exclude=/packages
--filter=H_**/files/digest-*"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
SYNC="rsync://godfather/gentoo-portage"
USE="X acpi alsa amd64 bash-completion bitmap-fonts bzip2 cli cracklib crypt
cups dbus dri fontconfig fortran gdbm gif hal highlight history iconv isdnlog
jpeg jpeg2k latex midi mmx mudflap ncurses nls nowebdav nptl nptlonly nsplugin
ogg opengl openmp oss pcre perl png pppd python qt3 readline reflection session
spl sse sse2 ssl tcpd test tiff truetype truetype-fonts type1-fonts unicode
vim-syntax vorbis xml xorg xv" ALSA_CARDS="ali5451 als4000 atiixp atiixp-modem
bt87x ca0106 cmipci emu10k1x ens1370 ens1371 es1938 es1968 fm801 hda-intel
intel8x0 intel8x0m maestro3 trident usb-audio via82xx via82xx-modem ymfpci"
ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file
hooks iec958 ioplug ladspa lfloat linear meter mulaw multi null plug rate route
share shm softvol" ELIBC="glibc" INPUT_DEVICES="keyboard mouse evdev"
KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001
mtxorb ncurses text" LINGUAS="en" USERLAND="GNU" VIDEO_CARDS="sis"
Unset:  CPPFLAGS, CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL,
LDFLAGS, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS,
PORTDIR_OVERLAY

ping amd64. 

lapack-reference-3.1.1-r1 now emerges successfully for me, and numeric compiles
against it. I say go with it as well

I can't tell, is amd64 still needed on this bug?  please readd us if true(and
clue us in as to what you want done)

ppc stable

x86-fbsd, only you left to keyword it

Keyworded ~x96-fbsd

closing then.