[bug #48807] gsvd function needs matlab compatible interface

--===============1561491888686833974==
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: base64
Content-Disposition: inline

UGxlYXNlIHVzZSB0aGUgYnVnIHRyYWNrZXIgdG8gcG9zdCB1cGRhdGVzIHRvIGEgYnVnIHJlcG9y
dC4gIFRoZSBtYWlsaW5nIGxpc3QgaXMgaW50ZW5kZWQgYXMgYSByZWFkLW9ubHkgbm90aWZpY2F0
aW9uIHN0cmVhbS4gIEluZm8gcG9zdGVkIHRvIHRoaXMgbWFpbGluZyBsaXN0IGFkZHJlc3Mgd29u
J3QgYXBwZWFyIGluIHRoZSB0cmFja2VyIGRhdGFiYXNlIHdoZXJlIGl0IGlzIG1vc3QgdXNlZnVs
LgoK

--===============1561491888686833974==
Content-Type: MULTIPART/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; BOUNDARY="8323329-1726956429-1782296428=:1050108"
Content-Disposition: inline

--8323329-1726956429-1782296428=:1050108
Content-Type: TEXT/plain; CHARSET=utf-8
Content-Transfer-Encoding: QUOTED-PRINTABLE
Content-Disposition: inline

Follow-up Comment #30, bug #48807 (group octave):=0A=0AOne potential Matlab=
 compatibility issue. With posted patch I see:=0A=0AM =3D 4; P =3D 2; N =3D=
 8;=0AA =3D [1:8; 1:8; 1:8; 1:8];=0AB =3D [8:-1:1; 8:-1:1];=0A[U, V, X, C, =
S] =3D gsvd (A, B)=0A=0A=0AU =3D=0A=0A  -8.6603e-01  -5.0000e-01  -5.6156e-=
17  -2.5375e-17=0A   2.8868e-01  -5.0000e-01   8.1650e-01  -9.0181e-17=0A  =
 2.8868e-01  -5.0000e-01  -4.0825e-01  -7.0711e-01=0A   2.8868e-01  -5.0000=
e-01  -4.0825e-01   7.0711e-01=0A=0AV =3D=0A=0A  -0.7071  -0.7071=0A  -0.70=
71   0.7071=0A=0AX =3D=0A=0A         0         0         0         0       =
  0         0  -11.3137=0A-2.0000=0A         0         0         0         =
0         0         0   -9.8995=0A-4.0000=0A         0         0         0 =
        0         0         0   -8.4853=0A-6.0000=0A         0         0   =
      0         0         0         0   -7.0711=0A-8.0000=0A         0     =
    0         0         0         0         0   -5.6569=0A-10.0000=0A      =
   0         0         0         0         0         0   -4.2426=0A-12.0000=
=0A         0         0         0         0         0         0   -2.8284=
=0A-14.0000=0A         0         0         0         0         0         0 =
  -1.4142=0A-16.0000=0A=0AC =3D=0A=0A        0        0        0        0  =
      0        0   0.0000        0=0A        0        0        0        0  =
      0        0        0   1.0000=0A        0        0        0        0  =
      0        0        0        0=0A        0        0        0        0  =
      0        0        0        0=0A=0AS =3D=0A=0A   0   0   0   0   0   0=
   1   0=0A   0   0   0   0   0   0   0   0=0A=0A=0A=0ASo identity C^2+S^2 =
=3D I does not hold:=0A=0A=0Aoctave:8> C'*C + S'*S=0Aans =3D=0A=0A   0   0 =
  0   0   0   0   0   0=0A   0   0   0   0   0   0   0   0=0A   0   0   0  =
 0   0   0   0   0=0A   0   0   0   0   0   0   0   0=0A   0   0   0   0   =
0   0   0   0=0A   0   0   0   0   0   0   0   0=0A   0   0   0   0   0   0=
   1   0=0A   0   0   0   0   0   0   0   1=0A=0A=0A=0Abut Matlab may "chea=
t" and fill it up. Can anyone confirm one way or another?=0A=0ADmitri.=0A--=
 =0A=0A=0A=0A=0A    _______________________________________________________=
=0A=0AReply to this item at:=0A=0A  <https://savannah.gnu.org/bugs/?48807>=
=0A=0A_______________________________________________=0AMessage sent via Sa=
vannah=0Ahttps://savannah.gnu.org/=0A
--8323329-1726956429-1782296428=:1050108
Content-Type: APPLICATION/pgp-signature; name=signature.asc

-----BEGIN PGP SIGNATURE-----

iHUEABYIAB0WIQQk97aszIMMAvLLwm6qLAuaBUf3TgUCajuvbAAKCRCqLAuaBUf3
ToPnAP9Jx9RG0Cld5asFGpoPIaFxjEOq3CwmVcAb2X7R8UMd9wD/RO21BTYtg8mD
3kT8tNSQK52qG62CR8Lh9shvg9M3iw4=
=7snJ
-----END PGP SIGNATURE-----

--8323329-1726956429-1782296428=:1050108--



--===============1561491888686833974==--