experchange > fortran

GianLuigi Piacentini (01-30-19, 01:26 PM)
Hi all,

I tried to install LAPACK 3.8.0 in my system (gfortran 8.2 under Kubuntu
Linux 18.04)
Libraries (.a) files are generated, however the testing hangs as follows:
gfortran-8 -o xeigtstz zchkee.o zbdt01.o zbdt02.o zbdt03.o zbdt05.o
zchkbb.o zchkbd.o zchkbk.o zchkbl.o zchkec.o zchkgg.o zchkgk.o zchkgl.o
zchkhb.o zchkhs.o zchkst.o zchkst2stg.o zchkhb2stg.o zckcsd.o zckglm.o
zckgqr.o zckgsv.o zcklse.o zcsdts.o zdrges.o zdrgev.o zdrges3.o
zdrgev3.o zdrgsx.o zdrgvx.o zdrvbd.o zdrves.o zdrvev.o zdrvsg.o
zdrvsg2stg.o zdrvst.o zdrvst2stg.o zdrvsx.o zdrvvx.o zerrbd.o zerrec.o
zerred.o zerrgg.o zerrhs.o zerrst.o zget02.o zget10.o zget22.o zget23.o
zget24.o zget35.o zget36.o zget37.o zget38.o zget51.o zget52.o zget54.o
zglmts.o zgqrts.o zgrqts.o zgsvts3.o zhbt21.o zhet21.o zhet22.o zhpt21.o
zhst01.o zlarfy.o zlarhs.o zlatm4.o zlctes.o zlctsx.o zlsets.o zsbmv.o
zsgt01.o zslect.o zstt21.o zstt22.o zunt01.o zunt03.o dlafts.o dlahd2.o
dlasum.o dlatb9.o dstech.o dstect.o dsvdch.o dsvdct.o dsxt1.o alahdg.o
alasum.o alasvm.o alareq.o ilaenv.o xerbla.o xlaenv.o chkxer.o
.../../libtmglib.a ../../liblapack.a ../../librefblas.a
make[2]: Leaving directory
'/home/gg/knowledge_base/math/lib/LAPACK/lapack-3.8.0/TESTING/EIG'
NEP: Testing Nonsymmetric Eigenvalue Problem routines
../EIG/xeigtstz < nep.in > znep.out 2>&1
Makefile:463: recipe for target 'znep.out' failed
make[1]: Leaving directory
'/home/gg/knowledge_base/math/lib/LAPACK/lapack-3.8.0/TESTING'
Makefile:42: recipe for target 'lapack_testing' failed

Running ./EIG/xeigtstz < nep.in stand-alone from the TESTING directory
produces a segfault, with absolutely no output.

Any suggestion ?

As a companion question, I am interested in solving small (<10) system
of linear equations, just for personal use only. Found suitable routines
in Math-a-la-carte, SLATEC, NSWC, and LAPACK. Guess choosing which
library is mainly a matter of taste, or there are considerations I
forgot ?

Thanks in advance

Gigi
Juergen Reuter (01-30-19, 07:06 PM)
On 1/30/19 12:26, GianLuigi Piacentini wrote:

[...]

> make[2]: Leaving directory
> '/home/gg/knowledge_base/math/lib/LAPACK/lapack-3.8.0/TESTING/EIG'
> NEP: Testing Nonsymmetric Eigenvalue Problem routines
> ./EIG/xeigtstz < nep.in > znep.out 2>&1
> Makefile:463: recipe for target 'znep.out' failed


This is the error message in question:
../EIG/xeigtstz < nep.in > znep.out 2>&1
Makefile:463: recipe for target 'znep.out' failed

What happens actually if you run this command by hand?
../EIG/xeigtstz < nep.in > znep.out 2>&1
so in the directory EIG of the build directory, execute:
../xeigtstz < nep.in > znep.out 2>&1
GianLuigi Piacentini (01-30-19, 10:23 PM)
On 30/01/19 18:06, Juergen Reuter wrote:
> On 1/30/19 12:26, GianLuigi Piacentini wrote:
> [...]
> This is the error message in question:
> ./EIG/xeigtstz < nep.in > znep.out 2>&1
> Makefile:463: recipe for target 'znep.out' failed
> What happens actually if you run this command by hand?
> ./EIG/xeigtstz < nep.in > znep.out 2>&1
> so in the directory EIG of the build directory, execute:
> ./xeigtstz < nep.in > znep.out 2>&1 Dear Juergen, thanks for your reply.


If I "manually" run the command from directory TESTING the computer gives:

gg@gg-Studio-XPS-435MT:~/knowledge_base/math/lib/LAPACK/lapack-3.8.0/TESTING$
../EIG/xeigtstz < nep.in > znep.out 2>&1
Segmentation fault (core dumped)

running from directory EIG (and addressing nep.in in the parent directory):
gg@gg-Studio-XPS-435MT:~/knowledge_base/math/lib/LAPACK/lapack-3.8.0/TESTING/EIG$
../xeigtstz < ../nep.in > znep.out 2>&1
Segmentation fault (core dumped)>

So result is the same.

Gigi
GianLuigi Piacentini (01-31-19, 01:54 AM)
On 30/01/19 18:06, Juergen Reuter wrote:
> On 1/30/19 12:26, GianLuigi Piacentini wrote:
> [...]
> This is the error message in question:
> ./EIG/xeigtstz < nep.in > znep.out 2>&1
> Makefile:463: recipe for target 'znep.out' failed
> What happens actually if you run this command by hand?
> ./EIG/xeigtstz < nep.in > znep.out 2>&1
> so in the directory EIG of the build directory, execute:
> ./xeigtstz < nep.in > znep.out 2>&1 Dear Juergen,


thanks for your reply.

when manually running from directory TESTING I get:
gg@gg-Studio-XPS-435MT:~/knowledge_base/math/lib/LAPACK/lapack-3.8.0/TESTING$
../EIG/xeigtstz < nep.in > znep.out 2>&1
Segmentation fault (core dumped)

when manually running from directory EIG (and taking care of nep.in file
location) I get
g@gg-Studio-XPS-435MT:~/knowledge_base/math/lib/LAPACK/lapack-3.8.0/TESTING/EIG$
../xeigtstz < ../nep.in > znep.out 2>&1
Segmentation fault (core dumped)

So no difference: segfault.

Gigi
spectrum (01-31-19, 03:53 AM)
Some online search gives me this page



which mentions a stacksize issue (ulimit -s unlimited) for xeigtstz,
but I'm not sure if this is related to your case...

# Out of curiosity, is OpenBlas (available via apt-get) etc not sufficient?
GianLuigi Piacentini (01-31-19, 07:37 PM)
On 31/01/19 13:56, spectrum wrote:
[..]
> We can also check the symbolic links etc by
> $ locate libblas
> $ locate liblapack Dear Spectrum,


many thanks for your valuable informations.

I checked the link you gave me, and "ulimit -s unlimited" did the magic
and the make process succeeded !

I do not think that, given the size of my problems (say less than 100,
but more likely on the order of 10 equations), an optimized blas will
show appreciable difference, anyway I will try.

Thanks a lot
Gigi
Wolfman Jack (02-03-19, 09:28 PM)
Unless you're using some new features in the latest LAPACK, Kubuntu also has liblapack and liblapack-dev packaged, and then there's no need to build it yourself. It is also possible to use it with openblas.

On Thursday, January 31, 2019 at 6:37:13 PM UTC+1, GianLuigi Piacentini wrote:
[..]
GianLuigi Piacentini (02-04-19, 10:41 PM)
On 03/02/19 20:28, Wolfman Jack wrote:
> Unless you're using some new features in the latest LAPACK, Kubuntu also has liblapack and liblapack-dev packaged, and then there's no need to build it yourself. It is also possible to use it with openblas.


Many, many thanks.
I searched the Muon package manager for "lapack", and these did not come
out. So I compiled the whole from source.
Now, after your post, I searched for "liblapack", and, voila, the stuff
was found !

Gigi
[..]
robin.vowels (02-08-19, 09:38 AM)
On Wednesday, January 30, 2019 at 10:26:39 PM UTC+11, GianLuigi Piacentini wrote:

> As a companion question, I am interested in solving small (<10) system
> of linear equations, just for personal use only. Found suitable routines
> in Math-a-la-carte, SLATEC, NSWC, and LAPACK. Guess choosing which
> library is mainly a matter of taste, or there are considerations I
> forgot ?


For < 10 equations, a classic equation solver should be more than adequate.
Similar Threads