Re: Failed to build octave-6.4.0 and octave's configuration.

Markus Mützel <[email protected]> Mon, 17 Jan 2022 16:25:40 +0100
Newsgroups gmane.comp.gnu.octave.general
Message-ID <trinity-83d71ec6-1d88-4164-ae02-1f569c248e2d-1642433140918@3c-app-gmx-bs63>
Your point 1:

This will be fixed in Octave 7 (which will probably released soon).

See also: GNU Octave - Bugs: bug #61738, octave-6.4.0 fails to build with... [Savannah]

If you don't need these features, consider configuring with `--without-sundials_ida --without-sundials_nvecserial` in the meantime.

Your point 2:

It looks like some optional dependencies aren't installed when you configure Octave. In case you need these features, check the detailed configure output which failing tests are causing that these optional features are disabled.

Which operating system (and which distribution) are you using?

Markus

Gesendet: Montag, 17. Januar 2022 um 11:14 Uhr

Von: "Pei Jia" <[email protected]>

An: [email protected]

Betreff: Failed to build octave-6.4.0 and octave's configuration.

Hi, all:

1. I tried to build octave-6.4.0 with the newest sundials-6.1.0, but failed with the following ERROR messages. It looks Octave is NOT compatible with the newest sundials yet?

libinterp/dldfcn/__ode15__.cc: In member function ‘void octave::IDA::set_up(const ColumnVector&)’:

libinterp/dldfcn/__ode15__.cc:420:54: error: too few arguments to function ‘_generic_SUNMatrix* SUNDenseMatrix(sunindextype, sunindextype, SUNContext)’

420 | m_sunJacMatrix = SUNDenseMatrix (m_num, m_num);

| ^

In file included from /usr/local/include/sunlinsol/sunlinsol_dense.h:36,

from libinterp/dldfcn/__ode15__.cc:62:

/usr/local/include/sunmatrix/sunmatrix_dense.h:79:27: note: declared here

79 | SUNDIALS_EXPORT SUNMatrix SUNDenseMatrix(sunindextype M, sunindextype N, SUNContext sunctx);

| ^~~~~~~~~~~~~~

libinterp/dldfcn/__ode15__.cc:424:64: error: too few arguments to function ‘_generic_SUNLinearSolver* SUNLinSol_Dense(N_Vector, SUNMatrix, SUNContext)’

424 | m_sunLinearSolver = SUNLinSol_Dense (yy, m_sunJacMatrix);

| ^

In file included from libinterp/dldfcn/__ode15__.cc:62:

/usr/local/include/sunlinsol/sunlinsol_dense.h:58:33: note: declared here

58 | SUNDIALS_EXPORT SUNLinearSolver SUNLinSol_Dense(N_Vector y, SUNMatrix A, SUNContext sunctx);

| ^~~~~~~~~~~~~~~

libinterp/dldfcn/__ode15__.cc: In static member function ‘static _generic_N_Vector* octave::IDA::ColToNVec(const ColumnVector&, long int)’:

libinterp/dldfcn/__ode15__.cc:508:34: error: too few arguments to function ‘_generic_N_Vector* N_VNew_Serial(sunindextype, SUNContext)’

508 | N_Vector v = N_VNew_Serial (n);

| ^

In file included from libinterp/dldfcn/__ode15__.cc:47:

/usr/local/include/nvector/nvector_serial.h:85:26: note: declared here

85 | SUNDIALS_EXPORT N_Vector N_VNew_Serial(sunindextype vec_length, SUNContext sunctx);

| ^~~~~~~~~~~~~

libinterp/dldfcn/__ode15__.cc: In member function ‘void octave::IDA::initialize()’:

libinterp/dldfcn/__ode15__.cc:531:24: error: too few arguments to function ‘void* IDACreate(SUNContext)’

531 | m_mem = IDACreate ();

| ^

In file included from libinterp/dldfcn/__ode15__.cc:51:

/usr/local/include/ida/ida.h:107:23: note: declared here

107 | SUNDIALS_EXPORT void *IDACreate(SUNContext sunctx);

| ^~~~~~~~~

libinterp/dldfcn/__ode15__.cc: In member function ‘octave_value_list octave::IDA::integrate(int, const ColumnVector&, const ColumnVector&, const ColumnVector&, int, bool, bool, const octave_value&, bool, ColumnVector&, bool, const octave_value&)’:

libinterp/dldfcn/__ode15__.cc:690:48: error: too few arguments to function ‘_generic_N_Vector* N_VNew_Serial(sunindextype, SUNContext)’

690 | N_Vector dky = N_VNew_Serial (m_num);

| ^

In file included from libinterp/dldfcn/__ode15__.cc:47:

/usr/local/include/nvector/nvector_serial.h:85:26: note: declared here

85 | SUNDIALS_EXPORT N_Vector N_VNew_Serial(sunindextype vec_length, SUNContext sunctx);

| ^~~~~~~~~~~~~

libinterp/dldfcn/__ode15__.cc: In member function ‘bool octave::IDA::interpolate(int&, Matrix&, ColumnVector&, int, realtype, bool, bool, const octave_value&, ColumnVector&, bool, const octave_value&, ColumnVector&, Matrix&, ColumnVector&, ColumnVector&, ColumnVector&, ColumnVector&, int&, ColumnVector&)’:

libinterp/dldfcn/__ode15__.cc:840:40: error: too few arguments to function ‘_generic_N_Vector* N_VNew_Serial(sunindextype, SUNContext)’

840 | N_Vector dky = N_VNew_Serial (m_num);

| ^

In file included from libinterp/dldfcn/__ode15__.cc:47:

/usr/local/include/nvector/nvector_serial.h:85:26: note: declared here

85 | SUNDIALS_EXPORT N_Vector N_VNew_Serial(sunindextype vec_length, SUNContext sunctx);

| ^~~~~~~~~~~~~

libinterp/dldfcn/__ode15__.cc:842:41: error: too few arguments to function ‘_generic_N_Vector* N_VNew_Serial(sunindextype, SUNContext)’

842 | N_Vector dkyp = N_VNew_Serial (m_num);

| ^

In file included from libinterp/dldfcn/__ode15__.cc:47:

/usr/local/include/nvector/nvector_serial.h:85:26: note: declared here

85 | SUNDIALS_EXPORT N_Vector N_VNew_Serial(sunindextype vec_length, SUNContext sunctx);

| ^~~~~~~~~~~~~

CXX libinterp/operators/liboctinterp_la-op-bm-bm.lo

CXX libinterp/operators/liboctinterp_la-op-bm-sbm.lo

CXX libinterp/operators/liboctinterp_la-op-cdm-cdm.lo

CXX libinterp/operators/liboctinterp_la-op-cdm-cm.lo

CXX libinterp/operators/liboctinterp_la-op-cdm-cs.lo

CXX libinterp/operators/liboctinterp_la-op-cdm-dm.lo

make[2]: *** [Makefile:19680: libinterp/dldfcn/__ode15___la-__ode15__.lo] Error 1

make[2]: *** Waiting for unfinished jobs....

make[2]: Leaving directory '....../octave-6.4.0'

make[1]: *** [Makefile:27541: all-recursive] Error 1

make[1]: Leaving directory '....../octave-6.4.0'

make: *** [Makefile:11166: all] Error 2

2. When I configure octave, I always get the following warnings:

configure: WARNING: Qhull library found, but does not seem to work properly. This will result in loss of functionality for some geometry functions. Please try recompiling the library with -fno-strict-aliasing.

configure: WARNING: SUNDIALS IDA library not configured with SUNLINSOL_KLU or sunlinksol_klu.h is not usable. The solvers ode15i and ode15s will not support the sparse Jacobian feature.

configure:

configure: NOTE: Libraries or auxiliary programs may be skipped if they are not found

configure: NOTE: OR if they are missing required features on your system.

Any suggestions? Thank you ...

--

Dr. Pei Jia

CTO of Longer Vision Technology (Canada) Ltd.

Email: [email protected]

Cell Number: +1 778-863-5816

---------- We are transitioning to a web based forum for community help discussions at https://octave.discourse.group/c/help


----------
We are transitioning to a web based forum
for community help discussions at
https://octave.discourse.group/c/help