From Jorge.Osorio at fysik.uu.se Mon Feb 10 15:45:08 2003 From: Jorge.Osorio at fysik.uu.se (Jorge Mario Osorio =?iso-8859-1?Q?Guill=E9n?=) Date: Mon, 10 Feb 2003 15:45:08 +0100 Subject: [Pw_forum] error: nc already filled Message-ID: <3E47BAF4.28F834C4@fysik.uu.se> Dear All, I am a new user of the PWSCF code and I succeed with calculations for cubic systems. But, doing a calculation for an orthorhombic system, with the following reciprocal axes as the ouput file prints out: reciprocal axes: (cart. coord. in units 2 pi/a_0) b(1) = ( 1.0000 -1.4625 0.0000 ) b(2) = ( 1.0000 1.4625 0.0000 ) b(3) = ( 0.0000 0.0000 1.0740 ) . I got some troubles. Actually, I used more precision in the vectors of the reciprocal lattice in order to generate the q-point grid (1.462525295 and 1.073961926). Some of the q-points are: 0.0000000000 0.0000000000 0.2684904815 0.0000000000 0.0000000000 1.0739619260 0.5000000000 0.3656313238 0.2684904815 0.2500000000 -0.3656313238 0.2684904815 0.2500000000 -0.3656313238 0.5369809630 0.2500000000 -0.7312626475 0.5369809630 0.5000000000 -0.3656313238 0.2684904815 1.0000000000 -0.3656313238 0.2684904815 etc., etc., As you can see, they are commensurable with the reciprocal lattice. The dynamical matrix calculation for all these q-points has been done. But, when I run the q2r.x program to get the interatomic foces constants in real space I got the following: q= 0.00000000 0.00000000 0.00000000 reading dyn.mat. from file ti.dyn_0.00_0.00_0.26 nqs= 2 q= 0.00000000 0.00000000 0.26849048 1 1 1 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from init : error # 1 nc already filled %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% or for example q= 0.00000000 0.00000000 0.00000000 reading dyn.mat. from file ti.dyn_0.05_0.36_0.26 nqs= 4 q= 0.50000000 0.36563132 0.26849048 1 1 1 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from init : error # 1 nc already filled %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Has someone the same kind of error before? How could I solve this? Is it related with the q-point grid? Any good sugestion that you can give me it is very welcome. With best, Jorge Osorio -- Jorge Mario Osorio Guillen |Telephone: (46-018)-471 35 67 Condensed Matter Theory Group|Fax: (46-018)-471 35 24 Department of Physics |e-mail: Jorge.Osorio at fysik.uu.se Uppsala University |www.fysik4.fysik.uu.se Box 530 | SE-751 21 Uppsala-Sweden | -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20030210/f407e5b6/attachment.htm From bungaro at physics.rutgers.edu Mon Feb 10 16:04:16 2003 From: bungaro at physics.rutgers.edu (CLAUDIA BUNGARO) Date: Mon, 10 Feb 2003 10:04:16 -0500 (EST) Subject: [Pw_forum] error: nc already filled In-Reply-To: <3E47BAF4.28F834C4@fysik.uu.se> References: <3E47BAF4.28F834C4@fysik.uu.se> Message-ID: Dear Jorge, in the input file for q2r.x, do you have the file with the dynamical matrix at gamma first in the list? (see Calculation of Interatomic Force Constants in real space in the on-line manual) Claudia On Mon, 10 Feb 2003, Jorge Mario Osorio [iso-8859-1] Guill?n wrote: > Dear All, > > I am a new user of the PWSCF code and I succeed with calculations for > > cubic > systems. But, doing a calculation for an orthorhombic system, with the > > following > reciprocal axes as the ouput file prints out: > > reciprocal axes: (cart. coord. in units 2 pi/a_0) > b(1) = ( 1.0000 -1.4625 0.0000 ) > b(2) = ( 1.0000 1.4625 0.0000 ) > b(3) = ( 0.0000 0.0000 1.0740 ) . > > I got some troubles. Actually, I used more precision in the vectors of > the > reciprocal > lattice in order to generate the q-point grid (1.462525295 and > 1.073961926). Some of > the q-points are: > > 0.0000000000 0.0000000000 0.2684904815 > 0.0000000000 0.0000000000 1.0739619260 > 0.5000000000 0.3656313238 0.2684904815 > 0.2500000000 -0.3656313238 0.2684904815 > 0.2500000000 -0.3656313238 0.5369809630 > 0.2500000000 -0.7312626475 0.5369809630 > 0.5000000000 -0.3656313238 0.2684904815 > 1.0000000000 -0.3656313238 0.2684904815 > etc., etc., > > As you can see, they are commensurable with the reciprocal lattice. > > The dynamical matrix calculation for all these q-points has been done. > But, > when I run > the q2r.x program to get the interatomic foces constants in real space > I > got the > following: > > q= 0.00000000 0.00000000 0.00000000 > reading dyn.mat. from file ti.dyn_0.00_0.00_0.26 > nqs= 2 > q= 0.00000000 0.00000000 0.26849048 > 1 1 1 > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from init : error # 1 > nc already filled > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > or for example > > q= 0.00000000 0.00000000 0.00000000 > reading dyn.mat. from file ti.dyn_0.05_0.36_0.26 > nqs= 4 > q= 0.50000000 0.36563132 0.26849048 > 1 1 1 > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from init : error # 1 > nc already filled > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > Has someone the same kind of error before? > > How could I solve this? > > Is it related with the q-point grid? > > Any good sugestion that you can give me it is very welcome. > > With best, > > Jorge Osorio > > > > -- > Jorge Mario Osorio Guillen |Telephone: (46-018)-471 35 67 > Condensed Matter Theory Group|Fax: (46-018)-471 35 24 > Department of Physics |e-mail: Jorge.Osorio at fysik.uu.se > Uppsala University |www.fysik4.fysik.uu.se > Box 530 | > SE-751 21 Uppsala-Sweden | > > > From Jorge.Osorio at fysik.uu.se Mon Feb 10 16:22:45 2003 From: Jorge.Osorio at fysik.uu.se (Jorge Mario Osorio =?iso-8859-1?Q?Guill=E9n?=) Date: Mon, 10 Feb 2003 16:22:45 +0100 Subject: [Pw_forum] error: nc already filled References: <3E47BAF4.28F834C4@fysik.uu.se> Message-ID: <3E47C3C5.EB2A5784@fysik.uu.se> Dear Claudia, Of course, the gamma point is the first. Jorge CLAUDIA BUNGARO wrote: > Dear Jorge, > > in the input file for q2r.x, do you have the file > with the dynamical matrix at gamma first in the list? > (see Calculation of Interatomic Force Constants > in real space in the on-line manual) > > Claudia > > On Mon, 10 Feb 2003, Jorge Mario Osorio [iso-8859-1] Guill?n wrote: > > > Dear All, > > > > I am a new user of the PWSCF code and I succeed with calculations for > > > > cubic > > systems. But, doing a calculation for an orthorhombic system, with the > > > > following > > reciprocal axes as the ouput file prints out: > > > > reciprocal axes: (cart. coord. in units 2 pi/a_0) > > b(1) = ( 1.0000 -1.4625 0.0000 ) > > b(2) = ( 1.0000 1.4625 0.0000 ) > > b(3) = ( 0.0000 0.0000 1.0740 ) . > > > > I got some troubles. Actually, I used more precision in the vectors of > > the > > reciprocal > > lattice in order to generate the q-point grid (1.462525295 and > > 1.073961926). Some of > > the q-points are: > > > > 0.0000000000 0.0000000000 0.2684904815 > > 0.0000000000 0.0000000000 1.0739619260 > > 0.5000000000 0.3656313238 0.2684904815 > > 0.2500000000 -0.3656313238 0.2684904815 > > 0.2500000000 -0.3656313238 0.5369809630 > > 0.2500000000 -0.7312626475 0.5369809630 > > 0.5000000000 -0.3656313238 0.2684904815 > > 1.0000000000 -0.3656313238 0.2684904815 > > etc., etc., > > > > As you can see, they are commensurable with the reciprocal lattice. > > > > The dynamical matrix calculation for all these q-points has been done. > > But, > > when I run > > the q2r.x program to get the interatomic foces constants in real space > > I > > got the > > following: > > > > q= 0.00000000 0.00000000 0.00000000 > > reading dyn.mat. from file ti.dyn_0.00_0.00_0.26 > > nqs= 2 > > q= 0.00000000 0.00000000 0.26849048 > > 1 1 1 > > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > from init : error # 1 > > nc already filled > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > > > or for example > > > > q= 0.00000000 0.00000000 0.00000000 > > reading dyn.mat. from file ti.dyn_0.05_0.36_0.26 > > nqs= 4 > > q= 0.50000000 0.36563132 0.26849048 > > 1 1 1 > > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > from init : error # 1 > > nc already filled > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > > > Has someone the same kind of error before? > > > > How could I solve this? > > > > Is it related with the q-point grid? > > > > Any good sugestion that you can give me it is very welcome. > > > > With best, > > > > Jorge Osorio > > > > > > > > -- > > Jorge Mario Osorio Guillen |Telephone: (46-018)-471 35 67 > > Condensed Matter Theory Group|Fax: (46-018)-471 35 24 > > Department of Physics |e-mail: Jorge.Osorio at fysik.uu.se > > Uppsala University |www.fysik4.fysik.uu.se > > Box 530 | > > SE-751 21 Uppsala-Sweden | > > > > > > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- Jorge Mario Osorio Guillen |Telephone: (46-018)-471 35 67 Condensed Matter Theory Group|Fax: (46-018)-471 35 24 Department of Physics |e-mail: Jorge.Osorio at fysik.uu.se Uppsala University |www.fysik4.fysik.uu.se Box 530 | SE-751 21 Uppsala-Sweden | -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20030210/1c52aeb9/attachment.htm From eyvaz_isaev at yahoo.com Mon Feb 10 16:27:36 2003 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 10 Feb 2003 07:27:36 -0800 (PST) Subject: [Pw_forum] error: nc already filled In-Reply-To: <3E47BAF4.28F834C4@fysik.uu.se> Message-ID: <20030210152736.13167.qmail@web80410.mail.yahoo.com> Dear Jorge, I guess there was some error during nscf calculations. The another reason for this could be the use of different basis vectors (or symmetry) for phonon and q2r programs. Please check also nr1, nr2, and nr3 in your q2r.in file. They should be the same for q-points generation. (So, the answer is "Yes" for your question). If the problem remains, please do not hesitate to send your input file to isaev at fysik.uu.se and let me know about it. Regards, Eyvaz. --- Jorge Mario Osorio Guill?n wrote: > Dear All, > > I am a new user of the PWSCF code and I succeed > with calculations for > > cubic > systems. But, doing a calculation for an > orthorhombic system, with the > > following > reciprocal axes as the ouput file prints out: > > reciprocal axes: (cart. coord. in units 2 > pi/a_0) > b(1) = ( 1.0000 -1.4625 0.0000 ) > b(2) = ( 1.0000 1.4625 0.0000 ) > b(3) = ( 0.0000 0.0000 1.0740 ) . > > I got some troubles. Actually, I used more > precision in the vectors of > the > reciprocal > lattice in order to generate the q-point grid > (1.462525295 and > 1.073961926). Some of > the q-points are: > > 0.0000000000 0.0000000000 0.2684904815 > 0.0000000000 0.0000000000 1.0739619260 > 0.5000000000 0.3656313238 0.2684904815 > 0.2500000000 -0.3656313238 0.2684904815 > 0.2500000000 -0.3656313238 0.5369809630 > 0.2500000000 -0.7312626475 0.5369809630 > 0.5000000000 -0.3656313238 0.2684904815 > 1.0000000000 -0.3656313238 0.2684904815 > etc., etc., > > As you can see, they are commensurable with the > reciprocal lattice. > > The dynamical matrix calculation for all these > q-points has been done. > But, > when I run > the q2r.x program to get the interatomic foces > constants in real space > I > got the > following: > > q= 0.00000000 0.00000000 0.00000000 > reading dyn.mat. from file ti.dyn_0.00_0.00_0.26 > nqs= 2 > q= 0.00000000 0.00000000 0.26849048 > 1 1 1 > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from init : error # 1 > nc already filled > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > or for example > > q= 0.00000000 0.00000000 0.00000000 > reading dyn.mat. from file ti.dyn_0.05_0.36_0.26 > nqs= 4 > q= 0.50000000 0.36563132 0.26849048 > 1 1 1 > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from init : error # 1 > nc already filled > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > Has someone the same kind of error before? > > How could I solve this? > > Is it related with the q-point grid? > > Any good sugestion that you can give me it is very > welcome. > > With best, > > Jorge Osorio > > > > -- > Jorge Mario Osorio Guillen |Telephone: > (46-018)-471 35 67 > Condensed Matter Theory Group|Fax: (46-018)-471 35 > 24 > Department of Physics |e-mail: > Jorge.Osorio at fysik.uu.se > Uppsala University |www.fysik4.fysik.uu.se > Box 530 | > SE-751 21 Uppsala-Sweden | > > > __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com From giannozz at nest.sns.it Mon Feb 10 17:03:03 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 10 Feb 2003 17:03:03 +0100 Subject: [Pw_forum] error: nc already filled In-Reply-To: <20030210152736.13167.qmail@web80410.mail.yahoo.com> References: <20030210152736.13167.qmail@web80410.mail.yahoo.com> Message-ID: <200302101703.03715.giannozz@nest.sns.it> Hi > I guess there was some error during nscf calculations. unlikely > another reason for this could be the use of > different basis vectors (or symmetry) for phonon and > q2r programs. as above > Please check also nr1, nr2, and nr3 in your q2r.in file. I vote for this. The q2r code does not check if you provided the right values for nr1, nr2, nr3 (there is no simple way to implement this) and it does not even check if you provided any value at all. In the latter case, unpredictable behaviour may follow. From the error message, it looks like nr1=nr2=nr3=0 . > They should be the same for q-points generation. exactly: I guess 4, 4, 4, in this case Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050509412 Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 From Jorge.Osorio at fysik.uu.se Mon Feb 10 17:47:04 2003 From: Jorge.Osorio at fysik.uu.se (Jorge Mario Osorio =?iso-8859-1?Q?Guill=E9n?=) Date: Mon, 10 Feb 2003 17:47:04 +0100 Subject: [Pw_forum] error: nc already filled References: <20030210152736.13167.qmail@web80410.mail.yahoo.com> <200302101703.03715.giannozz@nest.sns.it> Message-ID: <3E47D788.BC0551D6@fysik.uu.se> Hej, In fact, I used a small fortran program to generate the q-point grid. The inputs are the reciprocal axes (cart. coord. in units 2 pi/a_0) and nr1, nr2, nr3. I used nr1=4, nr2=4, nr3=4 to generate q-points and in the q2r.inp file as well. So, I guess it is not the problem. But, if someone want to exchange with me a q-points generator program in order to compare the grid that I have and to allow me to check that my troubles are not coming from this. Also, other possible sources of this problem that I can check are welcome. With best, Jorge Paolo Giannozzi wrote: > Hi > > > I guess there was some error during nscf calculations. > > unlikely > > > another reason for this could be the use of > > different basis vectors (or symmetry) for phonon and > > q2r programs. > > as above > > > Please check also nr1, nr2, and nr3 in your q2r.in file. > > I vote for this. The q2r code does not check if you provided > the right values for nr1, nr2, nr3 (there is no simple way > to implement this) and it does not even check if you provided > any value at all. In the latter case, unpredictable behaviour > may follow. From the error message, it looks like > nr1=nr2=nr3=0 . > > > They should be the same for q-points generation. > > exactly: I guess 4, 4, 4, in this case > > Paolo > > -- > Paolo Giannozzi e-mail: giannozz at nest.sns.it > Scuola Normale Superiore Phone: +39/050509412 > Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 > I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum -- Jorge Mario Osorio Guillen |Telephone: (46-018)-471 35 67 Condensed Matter Theory Group|Fax: (46-018)-471 35 24 Department of Physics |e-mail: Jorge.Osorio at fysik.uu.se Uppsala University |www.fysik4.fysik.uu.se Box 530 | SE-751 21 Uppsala-Sweden | -------------- next part -------------- An HTML attachment was scrubbed... URL: /pipermail/attachments/20030210/8fa3c138/attachment.htm From bungaro at physics.rutgers.edu Mon Feb 10 17:55:03 2003 From: bungaro at physics.rutgers.edu (CLAUDIA BUNGARO) Date: Mon, 10 Feb 2003 11:55:03 -0500 (EST) Subject: [Pw_forum] error: nc already filled In-Reply-To: <3E47D788.BC0551D6@fysik.uu.se> References: <20030210152736.13167.qmail@web80410.mail.yahoo.com> <200302101703.03715.giannozz@nest.sns.it> <3E47D788.BC0551D6@fysik.uu.se> Message-ID: Another possible source of error: in order to compute the real space IFC you need to compute the dynamical matrices only on the q-points which belong to the irreducible wedge of the crystal. If you have done this already you can ignore this email. There is not yet a separate q-point mesh generator, but if you want you can use pw.x to generate a non-shifted k-point grid specifying the automatic generation of k-points seeting in pw.in nks=0 and nk1,nk2,nk3, k1,k2,k3 4 4 4 0 0 0 you will find the k-pints in the irreducible wedge of the crystal in the standard output file of pw.x the nr1, nr2, and nr3 in q2r.in are equal to nk1,nk2,nk3. Claudia __________________________________________________ Claudia Bungaro Dept. of Physics and Astronomy, Rutgers University 136 Frelinghuysen Road, Piscataway, NJ 08854-8019 tel: +1 732 445 4197 fax: +1 732 445 4343 email: bungaro at physics.rutgers.edu __________________________________________________ On Mon, 10 Feb 2003, Jorge Mario Osorio [iso-8859-1] Guill?n wrote: > Hej, > > In fact, I used a small fortran program to generate the q-point grid. > The inputs are the reciprocal axes (cart. coord. in units 2 pi/a_0) and > nr1, nr2, nr3. I used nr1=4, nr2=4, nr3=4 to generate q-points and > in the q2r.inp file as well. So, I guess it is not the problem. > > But, if someone want to exchange with me a q-points generator program in > order to compare the grid that I have and to allow me to check that my > troubles are not coming from this. Also, other possible sources of this > problem that I can check are welcome. > > With best, > > Jorge > > > > Paolo Giannozzi wrote: > > > Hi > > > > > I guess there was some error during nscf calculations. > > > > unlikely > > > > > another reason for this could be the use of > > > different basis vectors (or symmetry) for phonon and > > > q2r programs. > > > > as above > > > > > Please check also nr1, nr2, and nr3 in your q2r.in file. > > > > I vote for this. The q2r code does not check if you provided > > the right values for nr1, nr2, nr3 (there is no simple way > > to implement this) and it does not even check if you provided > > any value at all. In the latter case, unpredictable behaviour > > may follow. From the error message, it looks like > > nr1=nr2=nr3=0 . > > > > > They should be the same for q-points generation. > > > > exactly: I guess 4, 4, 4, in this case > > > > Paolo > > > > -- > > Paolo Giannozzi e-mail: giannozz at nest.sns.it > > Scuola Normale Superiore Phone: +39/050509412 > > Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 > > I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > Jorge Mario Osorio Guillen |Telephone: (46-018)-471 35 67 > Condensed Matter Theory Group|Fax: (46-018)-471 35 24 > Department of Physics |e-mail: Jorge.Osorio at fysik.uu.se > Uppsala University |www.fysik4.fysik.uu.se > Box 530 | > SE-751 21 Uppsala-Sweden | > > > From eyvaz_isaev at yahoo.com Mon Feb 10 17:59:12 2003 From: eyvaz_isaev at yahoo.com (Eyvaz Isaev) Date: Mon, 10 Feb 2003 08:59:12 -0800 (PST) Subject: [Pw_forum] error: nc already filled In-Reply-To: <3E47D788.BC0551D6@fysik.uu.se> Message-ID: <20030210165912.2607.qmail@web80403.mail.yahoo.com> Dear Jorge, You don't need another program to generate q-points. Just put "nr1 nr2 nr2" mesh in scf.in file, and then start pw.x program. You can kill the job when q-points are generated. Usually it takes a couple of minuites. Then you shoud edit your output file in order to get q-points in the form acceptable for nscf and q2r calculations. Regards, Eyvaz. --- Jorge Mario Osorio Guill?n wrote: > Hej, > > In fact, I used a small fortran program to generate > the q-point grid. > The inputs are the reciprocal axes (cart. coord. in > units 2 pi/a_0) and > nr1, nr2, nr3. I used nr1=4, nr2=4, nr3=4 to > generate q-points and > in the q2r.inp file as well. So, I guess it is not > the problem. > > But, if someone want to exchange with me a q-points > generator program in > order to compare the grid that I have and to allow > me to check that my > troubles are not coming from this. Also, other > possible sources of this > problem that I can check are welcome. > > With best, > > Jorge > > > > Paolo Giannozzi wrote: > > > Hi > > > > > I guess there was some error during nscf > calculations. > > > > unlikely > > > > > another reason for this could be the use of > > > different basis vectors (or symmetry) for phonon > and > > > q2r programs. > > > > as above > > > > > Please check also nr1, nr2, and nr3 in your > q2r.in file. > > > > I vote for this. The q2r code does not check if > you provided > > the right values for nr1, nr2, nr3 (there is no > simple way > > to implement this) and it does not even check if > you provided > > any value at all. In the latter case, > unpredictable behaviour > > may follow. From the error message, it looks like > > nr1=nr2=nr3=0 . > > > > > They should be the same for q-points generation. > > > > exactly: I guess 4, 4, 4, in this case > > > > Paolo > > > > -- > > Paolo Giannozzi e-mail: > giannozz at nest.sns.it > > Scuola Normale Superiore Phone: +39/050509412 > > Piazza dei Cavalieri 7 Fax: > +39/050509417, 050563513 > > I-56126 Pisa, Italy Office: Lab. NEST, > Via della Faggiola 19 > > > > _______________________________________________ > > Pw_forum mailing list > > Pw_forum at pwscf.org > > http://www.democritos.it/mailman/listinfo/pw_forum > > -- > Jorge Mario Osorio Guillen |Telephone: > (46-018)-471 35 67 > Condensed Matter Theory Group|Fax: (46-018)-471 35 > 24 > Department of Physics |e-mail: > Jorge.Osorio at fysik.uu.se > Uppsala University |www.fysik4.fysik.uu.se > Box 530 | > SE-751 21 Uppsala-Sweden | > > > __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com From degironc at sissa.it Mon Feb 10 23:34:01 2003 From: degironc at sissa.it (Stefano de Gironcoli) Date: Mon, 10 Feb 2003 23:34:01 +0100 (CET) Subject: [Pw_forum] error: nc already filled In-Reply-To: <3E47BAF4.28F834C4@fysik.uu.se> Message-ID: On Mon, 10 Feb 2003, Jorge Mario Osorio Guill?n wrote: Dear Jorge Osorio, the problem you report looks like you are feeding q2r.x with a q vector that it already knows. only non equivalent points in the BZ of the crystal must be calculated by the phonon code, the resulting dynamical matrix files will contain also the matrices for the remaining equivalent q-points. It is possible that you have calculated more points than are needed. for instance I see that the point > 0.0000000000 0.0000000000 1.0739619260 which is equivalent to the gamma point is present in your grid. when feeded to the q2r.x program, it will correctly complain for this duplication. check your grid. look for the string " q =" in all your dynamical matrix files there should be one matching for each point in the grid plus an additinal matching per file since the results of the diagonalization of the first point of the star of q vectors is also writen in the file. therefore grep -e "q =" ti.dyn_* | wc should return a number of lines equal to nr1*nr2*nr3 + #files if not your grid is either incomplete or (should be your case) redundant. Obviously even if the number is correct the grid could still be wrong, however this is a simple check to make. regards, Stefano de Gironcoli > Dear All, > > I am a new user of the PWSCF code and I succeed with calculations for > > cubic > systems. But, doing a calculation for an orthorhombic system, with the > > following > reciprocal axes as the ouput file prints out: > > reciprocal axes: (cart. coord. in units 2 pi/a_0) > b(1) = ( 1.0000 -1.4625 0.0000 ) > b(2) = ( 1.0000 1.4625 0.0000 ) > b(3) = ( 0.0000 0.0000 1.0740 ) . > > I got some troubles. Actually, I used more precision in the vectors of > the > reciprocal > lattice in order to generate the q-point grid (1.462525295 and > 1.073961926). Some of > the q-points are: > > 0.0000000000 0.0000000000 0.2684904815 > 0.0000000000 0.0000000000 1.0739619260 > 0.5000000000 0.3656313238 0.2684904815 > 0.2500000000 -0.3656313238 0.2684904815 > 0.2500000000 -0.3656313238 0.5369809630 > 0.2500000000 -0.7312626475 0.5369809630 > 0.5000000000 -0.3656313238 0.2684904815 > 1.0000000000 -0.3656313238 0.2684904815 > etc., etc., > > As you can see, they are commensurable with the reciprocal lattice. > > The dynamical matrix calculation for all these q-points has been done. > But, > when I run > the q2r.x program to get the interatomic foces constants in real space > I > got the > following: > > q= 0.00000000 0.00000000 0.00000000 > reading dyn.mat. from file ti.dyn_0.00_0.00_0.26 > nqs= 2 > q= 0.00000000 0.00000000 0.26849048 > 1 1 1 > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from init : error # 1 > nc already filled > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > or for example > > q= 0.00000000 0.00000000 0.00000000 > reading dyn.mat. from file ti.dyn_0.05_0.36_0.26 > nqs= 4 > q= 0.50000000 0.36563132 0.26849048 > 1 1 1 > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from init : error # 1 > nc already filled > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > Has someone the same kind of error before? > > How could I solve this? > > Is it related with the q-point grid? > > Any good sugestion that you can give me it is very welcome. > > With best, > > Jorge Osorio > > > > -- > Jorge Mario Osorio Guillen |Telephone: (46-018)-471 35 67 > Condensed Matter Theory Group|Fax: (46-018)-471 35 24 > Department of Physics |e-mail: Jorge.Osorio at fysik.uu.se > Uppsala University |www.fysik4.fysik.uu.se > Box 530 | > SE-751 21 Uppsala-Sweden | > > > From mutombo at fzu.cz Thu Feb 13 10:01:12 2003 From: mutombo at fzu.cz (mutombo) Date: Thu, 13 Feb 2003 10:01:12 +0100 Subject: [Pw_forum] "pp.x error #29" Message-ID: <3E4B5ED8.4317795E@fzu.cz> Hi, I got the following error message when running pp.x module. What is error # 29 Regards, Pingo %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from read_file : error # 29 problem reading pbpun %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% stopping ... 2 From mutombo at fzu.cz Thu Feb 13 10:20:18 2003 From: mutombo at fzu.cz (mutombo) Date: Thu, 13 Feb 2003 10:20:18 +0100 Subject: [Pw_forum] "pp.x error #29" References: <3E4B5ED8.4317795E@fzu.cz> Message-ID: <3E4B6352.1164C44E@fzu.cz> Hello, Please disregard my previous email. I found a mistake in the input file. Best wishes, Pingo mutombo wrote: > Hi, > I got the following error message when running pp.x module. What > is error # 29 > Regards, > Pingo > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > from read_file : error # 29 > problem reading pbpun > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > stopping ... > 2 > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From ferretti.andrea at unimore.it Fri Feb 14 13:35:10 2003 From: ferretti.andrea at unimore.it (Andrea Ferretti) Date: Fri, 14 Feb 2003 12:35:10 +0000 (UTC) Subject: [Pw_forum] FILPUN error Message-ID: dear users, I have the following problem: I did a self consistent calculation (iswitch=0) to generate SCF potential, then I used this potential to generate bands (iswitch=-1) as usual. Now if I do NOT want PWSCF to write a punch file (I do not put the flag " filpun='namefile', ") everything is ok and the job ends well. Otherwise, if I want to write a punch file (for example to use it in PP.x to study |\psi|^2 of KS eigenvectors), I put " filpun='namefile', " in the input file and I get the following error: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% from Efermi : error # 1 unexpected error %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% and the output file seems to be really strange (for example the number of eigenvectors for each k-point is different and some lines are missing) An idea for this problem with output file could be that, while the job is stopping for a problem, it is not able (I'm running on IMB SP4 using 16 CPU) to write everything as well as it should (just missing lines and so on....) what do you think about this ? anyway the first problem about why I get this error message when I ask to write a punch file is still unanswered (to me).... any help is wellcome best regards Andrea -- Andrea Ferretti INFM National Research Center on nanoStructures and bioSystems at Surfaces (S3) Dipartimento di Fisica, Universita' di Modena e Reggio Emilia Via Campi 213/A I-41100 Modena, Italy Tel: +39 059 2055283 Fax: +39 059 367488 E-mail: ferretti.andrea at unimore.it URL: http://www.nanoscience.unimo.it From degironc at sissa.it Fri Feb 14 14:21:12 2003 From: degironc at sissa.it (Stefano de Gironcoli) Date: Fri, 14 Feb 2003 14:21:12 +0100 (CET) Subject: [Pw_forum] FILPUN error In-Reply-To: Message-ID: I think there may be a problem with the reconstruction of the charge density: In a band structure calculation it should not be needed but maybe if you give a non empty filpun the code maybe think you want to do some phonon calculation and get confused because the weights of the kpoints are rubbish... As a temporary fix try to specify iswitch=-2 (with q vector 0,0,0). It should compute correctly the charge. I hope so. stefano On Fri, 14 Feb 2003, Andrea Ferretti wrote: > > dear users, > I have the following problem: > > I did a self consistent calculation (iswitch=0) to generate SCF potential, > then I used this potential to generate bands (iswitch=-1) as usual. > Now if I do NOT want PWSCF to write a punch file (I do not put the flag > " filpun='namefile', ") everything is ok and the job ends well. > Otherwise, if I want to write a punch file (for example to use it in PP.x > to study |\psi|^2 of KS eigenvectors), I put " filpun='namefile', " in the > input file and I get the following error: > > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > from Efermi : error # 1 > unexpected error > > %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% > > and the output file seems to be really strange (for example the number of > eigenvectors for each k-point is different and some lines are missing) > > An idea for this problem with output file could be that, while the job is > stopping for a problem, it is not able (I'm running on IMB SP4 using 16 > CPU) to write > everything as well as it should (just missing lines and so on....) > > what do you think about this ? > > anyway the first problem about why I get this error message when I ask to > write a punch file is still unanswered (to me).... > > any help is wellcome > best regards > > Andrea > > From jinhm at ihpc.a-star.edu.sg Mon Feb 17 03:14:15 2003 From: jinhm at ihpc.a-star.edu.sg (Hong Mei Jin) Date: Mon, 17 Feb 2003 10:14:15 +0800 Subject: [Pw_forum] install error Message-ID: <3E504577.889F146E@ihpc.a-star.edu.sg> Dear users, We are new users of PWSCF. We have some problems when install the program. As the PWSCF is using f90 as request the lapack to be on f90 also, but the BLAS we are having is for f77, we tried to search for the f90 but not successful, so is there any way to obtain the BLAS's library in f90 version ? Any help is welcome. Thanks Hongmei From marzari at MIT.EDU Mon Feb 17 05:36:26 2003 From: marzari at MIT.EDU (Nicola Marzari) Date: Sun, 16 Feb 2003 23:36:26 -0500 Subject: [Pw_forum] install error References: <3E504577.889F146E@ihpc.a-star.edu.sg> Message-ID: <3E5066CA.3090602@mit.edu> Dear Hongmei, as far as I know the blas are written in f77. Once they are compiled onto a library, they can be linked to any f90-compiled object. One caveat is that some compilers can be case-sensitive, or can add a underscore to the calling sequence (so that a subroutine xxxx becomes actually xxxx_ ) It would help if you'd let us know (offline - we can resort to the list if we get stuck) the machine onto which you are compiling PWSCF, and the error messages you get. nicola --------------------------------------------------------------------- Prof Nicola Marzari Department of Materials Science and Engineering 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA tel 617.4522758 fax 617.2586534 marzari at mit.edu http://nnn.mit.edu From jinhm at ihpc.a-star.edu.sg Mon Feb 17 09:28:53 2003 From: jinhm at ihpc.a-star.edu.sg (Hong Mei Jin) Date: Mon, 17 Feb 2003 16:28:53 +0800 Subject: [Pw_forum] Re: Pw_forum digest, Vol 1 #22 - 2 msgs References: <20030217063506.2137.70450.Mailman@democritos.sissa.it> Message-ID: <3E509D45.68AB44CF@ihpc.a-star.edu.sg> Dear Nicola, We notice that too, but it doesn't seems can link to the libblas.a We are using Absoft fortran. We tried with UpperCase or lowerCase with or without _ all the combination, but none seem works. The machine specs are as following: Intel Pentium III 1GHz with Redhat Linux 7.3. I have attached the error messages in two documents. Thank you very much for your kind help! Regards Hongmei > Message: 2 > Date: Sun, 16 Feb 2003 23:36:26 -0500 > From: Nicola Marzari > To: pw_forum at pwscf.org > Subject: Re: [Pw_forum] install error > Reply-To: pw_forum at pwscf.org > > Dear Hongmei, > > as far as I know the blas are written in f77. Once they are > compiled onto a library, they can be linked to any f90-compiled > object. One caveat is that some compilers can be case-sensitive, > or can add a underscore to the calling sequence (so that a > subroutine xxxx becomes actually xxxx_ ) > > It would help if you'd let us know (offline - we can resort to > the list if we get stuck) the machine onto which you are > compiling PWSCF, and the error messages you get. > > nicola > > --------------------------------------------------------------------- > Prof Nicola Marzari Department of Materials Science and Engineering > 13-5066 MIT 77 Massachusetts Avenue Cambridge MA 02139-4307 USA > tel 617.4522758 fax 617.2586534 marzari at mit.edu http://nnn.mit.edu > > --__--__-- > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > > End of Pw_forum Digest -------------- next part -------------- A non-text attachment was scrubbed... Name: error.log Type: application/x-unknown-content-type-txtfile Size: 16286 bytes Desc: not available Url : /pipermail/attachments/20030217/11b9eb22/attachment.bin -------------- next part -------------- A non-text attachment was scrubbed... Name: make.sys Type: application/x-unknown-content-type-sysfile Size: 1556 bytes Desc: not available Url : /pipermail/attachments/20030217/11b9eb22/attachment-0001.bin From giannozz at nest.sns.it Mon Feb 17 09:48:53 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 17 Feb 2003 09:48:53 +0100 Subject: [Pw_forum] install error In-Reply-To: <3E509D45.68AB44CF@ihpc.a-star.edu.sg> References: <20030217063506.2137.70450.Mailman@democritos.sissa.it> <3E509D45.68AB44CF@ihpc.a-star.edu.sg> Message-ID: <200302170948.53804.giannozz@nest.sns.it> Hi > We are using Absoft fortran. We tried with UpperCase or > lowerCase with or without _ all the combination, but none > seem works. From your make.sys: LIBS = -L/usr/lib/lapack_LINUX.a -L/usr/lib/blas_linux.a \ -L/usr/lib/libfftw.a -L/usr/lib/libU77.a -L/usr/lib/libm.a This does not look like the right syntax to me. Option -L specifies a directory to be added to to the search path for libraries. Libraries must be specified with option -l, or with their explicit name. So "-Lmydir -lmylib" will load "mydir/libmylib.a" . "mydir/libmylib.a" will have the same effect. In any case you have to specify something. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050509412 Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 From pushpa at jncasr.ac.in Tue Feb 18 11:41:07 2003 From: pushpa at jncasr.ac.in (pushpa) Date: Tue, 18 Feb 2003 16:11:07 +0530 Subject: [Pw_forum] (no subject) Message-ID: <3E520DC3.7A4CA89F@jncasr.ac.in> Dear all, I am using pwscf to do some bulk calculations. I have compiled the code on alpha (single processor) machine. when I run it, I get the error as following. forrtl: error (75): floating point exception 0: __FINI_00_remove_gp_range [0x3ff81a6c374] 1: __FINI_00_remove_gp_range [0x3ff81a7422c] 2: __FINI_00_remove_gp_range [0x3ff800d0b9c] 3: __FINI_00_remove_gp_range [0x3ffbf764298] 4: __FINI_00_remove_gp_range [0x3ffbf91acec] 5: __FINI_00_remove_gp_range [0x3ffbf8be014] 6: __FINI_00_remove_gp_range [0x3ffbfb07754] 7: __FINI_00_remove_gp_range [0x3ffbfb2ce3c] 8: __FINI_00_remove_gp_range [0x3ffbfa354d0] 9: __FINI_00_remove_gp_range [0x3ffbfa4719c] 10: cdiaghg_ [cdiaghg.f90: 114, 0x12004cb88] 11: cegterg_ [cegterg.f90: 226, 0x120059a40] 12: c_bands_ [c_bands.f90: 193, 0x12003aa94] 13: electrons_ [electrons.f90: 125, 0x120085534] 14: pwscf_ [pwscf.f90: 41, 0x12010fcb0] 15: main [for_main.c: 203, 0x1201d1b9c] 16: __start [0x120016de8] and the program gets aborted. This does not happen for all calculations i.e. I am able to do the slab calculations. For some bulk calculations only it happens. But same kind of bulk calculations I am able to do on a PC. Has somebody encountered this kind of a problem, if so please let me know? Thanking you, Pushpa From francesco.antoniella at aquila.infn.it Mon Feb 17 13:09:11 2003 From: francesco.antoniella at aquila.infn.it (Francesco Antoniella) Date: Mon, 17 Feb 2003 13:09:11 +0100 Subject: [Pw_forum] (no subject) In-Reply-To: <3E520DC3.7A4CA89F@jncasr.ac.in> References: <3E520DC3.7A4CA89F@jncasr.ac.in> Message-ID: <200302171210.h1HCACU02557@antares> Il 11:41, marted? 18 febbraio 2003, hai scritto: > Dear all, > I am using pwscf to do some bulk calculations. I have compiled the code > on alpha (single processor) machine. when I run it, I get the error as > following. > forrtl: error (75): floating point exception > 0: __FINI_00_remove_gp_range [0x3ff81a6c374] > 1: __FINI_00_remove_gp_range [0x3ff81a7422c] > 2: __FINI_00_remove_gp_range [0x3ff800d0b9c] > 3: __FINI_00_remove_gp_range [0x3ffbf764298] > 4: __FINI_00_remove_gp_range [0x3ffbf91acec] > 5: __FINI_00_remove_gp_range [0x3ffbf8be014] > 6: __FINI_00_remove_gp_range [0x3ffbfb07754] > 7: __FINI_00_remove_gp_range [0x3ffbfb2ce3c] > 8: __FINI_00_remove_gp_range [0x3ffbfa354d0] > 9: __FINI_00_remove_gp_range [0x3ffbfa4719c] > 10: cdiaghg_ [cdiaghg.f90: 114, 0x12004cb88] > 11: cegterg_ [cegterg.f90: 226, 0x120059a40] > 12: c_bands_ [c_bands.f90: 193, 0x12003aa94] > 13: electrons_ [electrons.f90: 125, 0x120085534] > 14: pwscf_ [pwscf.f90: 41, 0x12010fcb0] > 15: main [for_main.c: 203, 0x1201d1b9c] > 16: __start [0x120016de8] > > and the program gets aborted. This does not happen for all calculations > i.e. I am able to do the slab calculations. For some bulk calculations > only it happens. But same kind of bulk calculations I am able to do on a > PC. Has somebody encountered this kind of a problem, if so please let me > know? I encountered the same problem. Are you using the dxml or cxml library from DEC/Compaq ? I worked around th problem linking the lapack and blas before dxml letting the dxml do only the FFT Regards Francesco Antoniella > > Thanking you, > Pushpa > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From verissim at df.ufscar.br Wed Feb 19 17:27:53 2003 From: verissim at df.ufscar.br (Marcos Verissimo Alves) Date: 19 Feb 2003 13:27:53 -0300 Subject: [Pw_forum] Problems compiling pwscf 1.2.0 Message-ID: <1045672073.3721.31.camel@rocknroll> Hi all, My name is Marcos and I am new to this list, so hello :). I have downloaded the latest version today and I am trying to compile it with lahey lf95. I am having some trouble with the compilation of the module io_base. The error message is the following: /lib/cpp -P -traditional -I/home/verissim/pwscf-1.2 -DPC -DFFTW -D"FFTWND_F77_ONE=fftwnd_f77_one_" -D"FFTW3D_F77_CREATE_PLAN=fftw3d_f77_create_plan_" io_base.f90 io_base.F90 lf95 --staticlink --dbl -O -I/home/verissim/pwscf-1.2e -I/home/verissim/pwscf-1.2/Modules -I/home/verissim/pwscf-1.2/PW -I/home/verissim/pwscf-1.2/PH -c io_base.F90 -o io_base.o Module subprogram name(write_restart_header2) 2005-W: "io_base.F90", line 274: idum is used but never set. Module subprogram name(read_restart_header1) 2005-W: "io_base.F90", line 554: idum is used but never set. Module subprogram name(write_restart_xdim2) 2005-W: "io_base.F90", line 638: idum is used but never set. Module subprogram name(write_restart_symmetry2) 2005-W: "io_base.F90", line 817: idum is used but never set. Module subprogram name(write_restart_pseudo3) 2005-W: "io_base.F90", line 1185: idum is used but never set. Module subprogram name(write_restart_pseudo2) 2005-W: "io_base.F90", line 1211: idum is used but never set. Module subprogram name(write_restart_gvec1) 2005-W: "io_base.F90", line 1873: idum is used but never set. Module subprogram name(write_restart_gvec2) 2005-W: "io_base.F90", line 1893: idum is used but never set. Module subprogram name(write_restart_gkvec1) 2005-W: "io_base.F90", line 2095: idum is used but never set. Module subprogram name(write_restart_gkvec2) 2005-W: "io_base.F90", line 2115: idum is used but never set. Module subprogram name(write_restart_cell1) 2043-S: "io_base.F90", line 2281: Value must be assigned for INTENT(OUT) variable htvel. 2043-S: "io_base.F90", line 2282: Value must be assigned for INTENT(OUT) variable xnosp. 2043-S: "io_base.F90", line 2283: Value must be assigned for INTENT(OUT) variable xnos0. 2043-S: "io_base.F90", line 2284: Value must be assigned for INTENT(OUT) variable xnosm. 2043-S: "io_base.F90", line 2285: Value must be assigned for INTENT(OUT) variable xnosm2. Module subprogram name(write_restart_cell2) 2005-W: "io_base.F90", line 2315: idum is used but never set. Module subprogram name(write_restart_ions2) 2005-W: "io_base.F90", line 2561: idum is used but never set. Module subprogram name(write_restart_electrons1) 2005-W: "io_base.F90", line 2874: idum is used but never set. Module subprogram name(write_restart_electrons2) 2005-W: "io_base.F90", line 2941: idum is used but never set. Module subprogram name(write_restart_wfc1) 2005-W: "io_base.F90", line 3360: idum is used but never set. Module subprogram name(write_restart_charge1) 2005-W: "io_base.F90", line 3775: idum is used but never set. Encountered 5 errors, 16 warnings in file io_base.F90. The lines containing "xxxx-S:" are the actual errors, the others are simply warnings. Although the compiler is not supported officially by the PWSCF team, I guess this error can be a rather common thing. This is the only thing that's missing to compile the new version with lahey f95. Does anyone have any guess about it ? By the way, if anyone wants the tips for compiling pwscf 1.1.x with lahey, I can send them. Thanks, Marcos From francesco.antoniella at aquila.infn.it Wed Feb 19 23:58:19 2003 From: francesco.antoniella at aquila.infn.it (antonief) Date: Wed, 19 Feb 2003 23:58:19 +0100 (CET) Subject: [Pw_forum] Problems compiling pwscf 1.2.0 In-Reply-To: <1045672073.3721.31.camel@rocknroll> Message-ID: On 19 Feb 2003, Marcos Verissimo Alves wrote: Hi Marcos, I'm Francesco Antoniella, I think whe know in each other. > Hi all, > > My name is Marcos and I am new to this list, so hello :). I have > downloaded the latest version today and I am trying to compile it with > lahey lf95. I am having some trouble with the compilation of the module > io_base. The error message is the following: > > > /lib/cpp -P -traditional -I/home/verissim/pwscf-1.2 -DPC -DFFTW > -D"FFTWND_F77_ONE=fftwnd_f77_one_" > -D"FFTW3D_F77_CREATE_PLAN=fftw3d_f77_create_plan_" io_base.f90 > io_base.F90~ > lf95 --staticlink --dbl -O -I/home/verissim/pwscf-1.2e > -I/home/verissim/pwscf-1.2/Modules -I/home/verissim/pwscf-1.2/PW > -I/home/verissim/pwscf-1.2/PH -c io_base.F90 -o io_base.o > Module subprogram name(write_restart_header2) > 2005-W: "io_base.F90", line 274: idum is used but never set. > Module subprogram name(read_restart_header1) > 2005-W: "io_base.F90", line 554: idum is used but never set. > Module subprogram name(write_restart_xdim2) > 2005-W: "io_base.F90", line 638: idum is used but never set. > Module subprogram name(write_restart_symmetry2) > 2005-W: "io_base.F90", line 817: idum is used but never set. > Module subprogram name(write_restart_pseudo3) > 2005-W: "io_base.F90", line 1185: idum is used but never set. > Module subprogram name(write_restart_pseudo2) > 2005-W: "io_base.F90", line 1211: idum is used but never set. > Module subprogram name(write_restart_gvec1) > 2005-W: "io_base.F90", line 1873: idum is used but never set. > Module subprogram name(write_restart_gvec2) > 2005-W: "io_base.F90", line 1893: idum is used but never set. > Module subprogram name(write_restart_gkvec1) > 2005-W: "io_base.F90", line 2095: idum is used but never set. > Module subprogram name(write_restart_gkvec2) > 2005-W: "io_base.F90", line 2115: idum is used but never set. > Module subprogram name(write_restart_cell1) > 2043-S: "io_base.F90", line 2281: Value must be assigned for > INTENT(OUT) variable htvel. This error is not common with the other compilers, but it seems that the intent OUT must be IN instead from the use made in the routine under examination Greetings francesco From mutombo at fzu.cz Thu Feb 20 15:39:49 2003 From: mutombo at fzu.cz (mutombo) Date: Thu, 20 Feb 2003 15:39:49 +0100 Subject: [Pw_forum] computation of STM maps References: Message-ID: <3E54E8B5.B89F6825@fzu.cz> Dear Users, I am trying to run pp.x in order to compute a STM map. Unfortunately I am getting the following error message: nbndx = 50 nbnd = 24 natomwfc = 50 npwx = 11066 nelec = 40.00 nkb = 50 ngl = 2156 Input/Output Error 153: Input file ended In Procedure: sum_band At Line: 81 Statement: Unformatted READ Unit: 7 Connected To: /home/crystal/tmp/pwscf.igk Form: Unformatted Access: Sequential Records Read : 0 Records Written: 0 End of diagnostics It seems like that the "pwscf.igk " file is needed. But when I checked the temporary directory, there is no such file. It appears as it is deleted at the end of the "scf "cycles. How can I tell the machine to keep this file in the temporary directory? Best regards, Pingo From pushpa at jncasr.ac.in Thu Feb 20 21:04:06 2003 From: pushpa at jncasr.ac.in (Raghani Pushpa) Date: Fri, 21 Feb 2003 01:34:06 +0530 (IST) Subject: [Pw_forum] (no subject) Message-ID: Dear All, I want to calculate the phonon density of states from pwscf but I don't know how to calculate it. Does the program calculate it directly or one has to modify it? Examples are given for calculating the electronic density of states. Thanks, pushpa -- From giannozz at nest.sns.it Thu Feb 20 22:25:25 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 20 Feb 2003 22:25:25 +0100 Subject: [Pw_forum] (no subject) In-Reply-To: References: Message-ID: <200302202225.25923.giannozz@nest.sns.it> Hi > I want to calculate the phonon density of states from pwscf but I don't > know how to calculate it. Does the program calculate it directly or one > has to modify it? in the new version, you can, using program "matdyn.x" (after you have calculated force constants in reals space: see example 6). See comments at the beginning of pwtools/matdyn.f90 Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050509412 Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 From fornari at phy.cmich.edu Fri Feb 21 00:57:51 2003 From: fornari at phy.cmich.edu (Marco Fornari) Date: Thu, 20 Feb 2003 18:57:51 -0500 Subject: [Pw_forum] look at that References: <200302202225.25923.giannozz@nest.sns.it> Message-ID: <3E556B7F.8090107@phy.cmich.edu> Ciao Paolo, spero tutto vada bene. Noi siamo finalmente in tre: http://www.phy.cmich.edu/people/fornari/Luca Abbracci, Marco From fornari at phy.cmich.edu Fri Feb 21 02:00:36 2003 From: fornari at phy.cmich.edu (Marco Fornari) Date: Thu, 20 Feb 2003 20:00:36 -0500 Subject: [Pw_forum] (no subject) References: <200302202225.25923.giannozz@nest.sns.it> Message-ID: <3E557A34.9060507@phy.cmich.edu> Please, disregard my previous message. I sent it by mystake to the list. THanks. MF From ahn22 at hermes.cam.ac.uk Sun Feb 23 23:53:54 2003 From: ahn22 at hermes.cam.ac.uk (Andriy H. Nevidomskyy) Date: Sun, 23 Feb 2003 22:53:54 +0000 (GMT) Subject: [Pw_forum] problems running test-system on Alpha. Any ideas? Message-ID: Dear PWSCF users, I have trouble running the test example with pw.x (version 1.2.0). This is part of the example No.1, which I downloaded from the PWSCF web-site, and it's about self-consistent calculation of Cu-metal. The run on DEC-Alpha fails with the following error message: --------------------------------------------------------- forrtl: error (65): floating invalid IOT trap (core dumped) --------------------------------------------------------- OR, on slightly different machine (also Alpha): ---------------------------------------------------------- forrtl: error (75): floating point exception IOT trap (core dumped) ---------------------------------------------------------- The analysis of the output file shows that the execution fails even before the first SCF step was taken. The input file that I used reads as follows: ---------------------------------------------------------------- &control calculation='scf' restart_mode='from_scratch', pseudo_dir = '/home/giannozz/O-sesame/pseudo//', outdir='/home/giannozz/tmp//' prefix='cu' tstress = .true. tprnfor = .true. / &system ibrav = 2, celldm(1) =6.73, nat= 1, ntyp= 1, ecutwfc = 25.0, ecutrho = 300.0 occupations='smearing', smearing='methfessel-paxton', degauss=0.02 / &electrons conv_thr = 1.0e-8 mixing_beta = 0.7 / ATOMIC_SPECIES Cu 63.55 CuUS.RRKJ3.UPF ATOMIC_POSITIONS Cu 0.0 0.0 0.0 K_POINTS (automatic) 8 8 8 0 0 0 -------------------------------------------------------------------- I played with different values of the input params: tried increasing _nbnd_, changing the _smearing_ scheme, increasing _degauss_ value - all in vain, the error is steadily reproducible! And of course, this error doesn't occur in the output file given by the code developers as a reference to the test [reference says that test was run on PC Pentium-4 machine]. Shall I thus assume, that this is a platform-dependent issue, specific for Alphas? However, the calculation for another metal, Al, doesn't produce this error! Why does it happen for Cu then? I would be very grateful for any insight that you could share with me on this problem. P.S> *It should be noted that presicely the same *error occurs in the SCF-calculation, as well as in the *band-structure calculations for another test example - Ni. Thanks! Andriy Nevidomskyy .------------------------------------------------------------------. | Mr Andriy H. Nevidomskyy | | | | TCM Group, Cavendish Laboratory, University of Cambridge | | Madingley Road, Cambridge CB3 0HE United Kingdom | | Tel:+44-1223-337433,Fax:+44-1223-337356, E-mail: ahn22 at cam.ac.uk | `------------------------------------------------------------------' From ahn22 at hermes.cam.ac.uk Mon Feb 24 00:09:04 2003 From: ahn22 at hermes.cam.ac.uk (Andriy H. Nevidomskyy) Date: Sun, 23 Feb 2003 23:09:04 +0000 (GMT) Subject: [Pw_forum] compilation problem on PC with Intel compiler Message-ID: Dear users of PWSCF, The compilation of PW-code fails on the PC-mashine when using Intel ifc compiler and Intel MKL library. It fails at the linking stage and produces the following log: (sorry for this long e-mail) cft_3.o: In function `cft_3_': cft_3.o(.text+0xe4): undefined reference to `fftw3d_f77_create_plan_' cft_3.o(.text+0x10c): undefined reference to `fftwnd_f77_one_' cft_fftw.o: In function `cft_1_': cft_fftw.o(.text+0x5a): undefined reference to `fftw_f77_create_plan_' cft_fftw.o(.text+0x7d): undefined reference to `fftw_f77_' cft_fftw.o: In function `cft_1s_': cft_fftw.o(.text+0x163): undefined reference to `fftw_f77_create_plan_' cft_fftw.o(.text+0x182): undefined reference to `fftw_f77_' cft_fftw.o: In function `cft_2_': cft_fftw.o(.text+0x2f3): undefined reference to `fftw_f77_create_plan_' cft_fftw.o(.text+0x32b): undefined reference to `fftw_f77_' cft_fftw.o(.text+0x354): undefined reference to `fftw_f77_create_plan_' cft_fftw.o(.text+0x3bd): undefined reference to `fftw_f77_' cft_fftw.o(.text+0x45a): undefined reference to `fftw_f77_create_plan_' cft_fftw.o(.text+0x4c2): undefined reference to `fftw_f77_' cft_fftw.o(.text+0x4ff): undefined reference to `fftw_f77_create_plan_' cft_fftw.o(.text+0x537): undefined reference to `fftw_f77_' cft_fftw.o: In function `cft_2s_': cft_fftw.o(.text+0x6cb): undefined reference to `fftw_f77_create_plan_' cft_fftw.o(.text+0x702): undefined reference to `fftw_f77_' cft_fftw.o(.text+0x72b): undefined reference to `fftw_f77_create_plan_' cft_fftw.o(.text+0x795): undefined reference to `fftw_f77_' cft_fftw.o(.text+0x82d): undefined reference to `fftw_f77_create_plan_' cft_fftw.o(.text+0x8a3): undefined reference to `fftw_f77_' cft_fftw.o(.text+0x8e0): undefined reference to `fftw_f77_create_plan_' cft_fftw.o(.text+0x917): undefined reference to `fftw_f77_' make[1]: *** [all] Error 1 make[1]: Leaving directory `/temp/ahn22/PWSCF/PW' make: *** [pw] Error 2 Apparantly this has something to do with the -D option of the compiler, where macroses are defined. However I didn't succeed to compile it successfully. Could anyone help me to resolve this problem, please? I attached to this e-mail the make.sys file that I was using for compilation. Best regards, Andriy .------------------------------------------------------------------. | Mr Andriy H. Nevidomskyy | | | | TCM Group, Cavendish Laboratory, University of Cambridge | | Madingley Road, Cambridge CB3 0HE United Kingdom | | Tel:+44-1223-337433,Fax:+44-1223-337356, E-mail: ahn22 at cam.ac.uk | `------------------------------------------------------------------' -------------- next part -------------- OSHOME=/temp/ahn22/PWSCF # # System-dependent Make definitions for Linux, Intel compiler (v.5-6) # Edit according to your needs # # add -DMKL if using the Intel Mathematical Kernel Library # Add -DADD_BLAS_ONE_UNDERSCORE if your blas/lapack library names contain # two underscores at the end # Define FFTW library names with one underscore less than they have in # the library (one underscore is added by the compiler) # CPPFLAGS = -I$(OSHOME)/include/ -DPC -DMKL -DFFTW \ -D"FFTWND_F77_ONE=fftwnd_f77_one" \ -D"FFTW3D_F77_CREATE_PLAN=fftw3d_f77_create_plan" \ -D"FFTW_F77=fftw_f77" \ -D"FFTW_F77_CREATE_PLAN=fftw_f77_create_plan" # Fortran compiler: # F90 = ifc # # fortran options: # -Vaxlib enables the use of portability libraries (getarg # and maybe other library calls) # -W0 disable warnings # -O3 more aggressive optimization # -tpp5 Pentium optimization # -tpp6 Pentium Pro, 2, 3 optimization # -tpp7 Pentium 4 optimization # FFLAGS =-Vaxlib -O2 -tpp6 F90FLAGS= $(FFLAGS) -fpp $(CPPFLAGS) # # This is needed to tell the compiler where modules are # use this for version < 7 # MODULEFLAG= -cl,./intel.pcl # use this for version = 7 #MODULEFLAG= -nomodule -I$(OSHOME)/Modules -I$(OSHOME)/PW -I$(OSHOME)/PH #MODULEFLAG= -nomodule -I$(OSHOME)/Modules -I$(OSHOME)/PW -I$(OSHOME)/PH MODULEFLAG= -cl,./intel.pcl # # Loader: # # The following is for Intel MKL libraries v. 5.1 on Pentium-4 # (remember -DMKL in CPPFLAGS !) # For MKL v. 5.2, add "-lguide" # Use libmkl_p3 for Pentium-3, libmkl_p4.a for Pentium-4 machines # (MKL will likely not work for other CPUs like Athlon) # #LIBS= -lfftw /opt/intel/mkl/lib/32/libmkl_lapack.a \ # /opt/intel/mkl/lib/32/libmkl_p4.a -lpthread LIBS= -lmkl_p3 -lmkl_lapack -lmkl_p3 -lguide -lpthread # # The following is for Atlas optimized blas/lapack libraries # You may not actually need -lcblas . See the manual for details # on how to fix I/O incompatibility problems with ifc # # LIBS = -lfftw -L$(HOME)/Linux_ATHLON256/ -llapack -lf77blas -lcblas -latlas # LD=$(F90) LFLAGS = -Vaxlib $(LIBS) # # ar: # AR = ar ARFLAGS = ruv From xhongjun at mail.ustc.edu.cn Mon Feb 24 02:49:46 2003 From: xhongjun at mail.ustc.edu.cn (xhongjun) Date: Mon, 24 Feb 2003 09:49:46 +0800 Subject: [Pw_forum] compilation problem on PC with Intel compiler In-Reply-To: References: Message-ID: <3E597A3A.3040909@mail.ustc.edu.cn> Andriy H. Nevidomskyy wrote: > Dear users of PWSCF, > > The compilation of PW-code fails on the PC-mashine when using Intel ifc > compiler and Intel MKL library. > It fails at the linking stage and produces the following log: > (sorry for this long e-mail) > > cft_3.o: In function `cft_3_': > cft_3.o(.text+0xe4): undefined reference to `fftw3d_f77_create_plan_' > cft_3.o(.text+0x10c): undefined reference to `fftwnd_f77_one_' > cft_fftw.o: In function `cft_1_': > cft_fftw.o(.text+0x5a): undefined reference to `fftw_f77_create_plan_' > cft_fftw.o(.text+0x7d): undefined reference to `fftw_f77_' > cft_fftw.o: In function `cft_1s_': > cft_fftw.o(.text+0x163): undefined reference to `fftw_f77_create_plan_' > cft_fftw.o(.text+0x182): undefined reference to `fftw_f77_' > cft_fftw.o: In function `cft_2_': > cft_fftw.o(.text+0x2f3): undefined reference to `fftw_f77_create_plan_' > cft_fftw.o(.text+0x32b): undefined reference to `fftw_f77_' > cft_fftw.o(.text+0x354): undefined reference to `fftw_f77_create_plan_' > cft_fftw.o(.text+0x3bd): undefined reference to `fftw_f77_' > cft_fftw.o(.text+0x45a): undefined reference to `fftw_f77_create_plan_' > cft_fftw.o(.text+0x4c2): undefined reference to `fftw_f77_' > cft_fftw.o(.text+0x4ff): undefined reference to `fftw_f77_create_plan_' > cft_fftw.o(.text+0x537): undefined reference to `fftw_f77_' > cft_fftw.o: In function `cft_2s_': > cft_fftw.o(.text+0x6cb): undefined reference to `fftw_f77_create_plan_' > cft_fftw.o(.text+0x702): undefined reference to `fftw_f77_' > cft_fftw.o(.text+0x72b): undefined reference to `fftw_f77_create_plan_' > cft_fftw.o(.text+0x795): undefined reference to `fftw_f77_' > cft_fftw.o(.text+0x82d): undefined reference to `fftw_f77_create_plan_' > cft_fftw.o(.text+0x8a3): undefined reference to `fftw_f77_' > cft_fftw.o(.text+0x8e0): undefined reference to `fftw_f77_create_plan_' > cft_fftw.o(.text+0x917): undefined reference to `fftw_f77_' > make[1]: *** [all] Error 1 > make[1]: Leaving directory `/temp/ahn22/PWSCF/PW' > make: *** [pw] Error 2 > > Apparantly this has something to do with the -D option of the compiler, > where macroses are defined. However I didn't succeed to compile it > successfully. > Could anyone help me to resolve this problem, please? > > I attached to this e-mail the make.sys file that I was using for > compilation. > > Best regards, > Andriy > .------------------------------------------------------------------. > | Mr Andriy H. Nevidomskyy | > | | > | TCM Group, Cavendish Laboratory, University of Cambridge | > | Madingley Road, Cambridge CB3 0HE United Kingdom | > | Tel:+44-1223-337433,Fax:+44-1223-337356, E-mail: ahn22 at cam.ac.uk | > `------------------------------------------------------------------' > > > > > ------------------------------------------------------------------------ > > OSHOME=/temp/ahn22/PWSCF > # > # System-dependent Make definitions for Linux, Intel compiler (v.5-6) > # Edit according to your needs > # > # add -DMKL if using the Intel Mathematical Kernel Library > # Add -DADD_BLAS_ONE_UNDERSCORE if your blas/lapack library names contain > # two underscores at the end > # Define FFTW library names with one underscore less than they have in > # the library (one underscore is added by the compiler) > # > CPPFLAGS = -I$(OSHOME)/include/ -DPC -DMKL -DFFTW \ > -D"FFTWND_F77_ONE=fftwnd_f77_one" \ > -D"FFTW3D_F77_CREATE_PLAN=fftw3d_f77_create_plan" \ > -D"FFTW_F77=fftw_f77" \ > -D"FFTW_F77_CREATE_PLAN=fftw_f77_create_plan" > > # Fortran compiler: > # > F90 = ifc > # > # fortran options: > # -Vaxlib enables the use of portability libraries (getarg > # and maybe other library calls) > # -W0 disable warnings > # -O3 more aggressive optimization > # -tpp5 Pentium optimization > # -tpp6 Pentium Pro, 2, 3 optimization > # -tpp7 Pentium 4 optimization > # > FFLAGS =-Vaxlib -O2 -tpp6 > > F90FLAGS= $(FFLAGS) -fpp $(CPPFLAGS) > # > # This is needed to tell the compiler where modules are > # use this for version < 7 > # MODULEFLAG= -cl,./intel.pcl > # use this for version = 7 > #MODULEFLAG= -nomodule -I$(OSHOME)/Modules -I$(OSHOME)/PW -I$(OSHOME)/PH > #MODULEFLAG= -nomodule -I$(OSHOME)/Modules -I$(OSHOME)/PW -I$(OSHOME)/PH > MODULEFLAG= -cl,./intel.pcl > > # > # Loader: > # > # The following is for Intel MKL libraries v. 5.1 on Pentium-4 > # (remember -DMKL in CPPFLAGS !) > # For MKL v. 5.2, add "-lguide" > # Use libmkl_p3 for Pentium-3, libmkl_p4.a for Pentium-4 machines > # (MKL will likely not work for other CPUs like Athlon) > # > > #LIBS= -lfftw /opt/intel/mkl/lib/32/libmkl_lapack.a \ > # /opt/intel/mkl/lib/32/libmkl_p4.a -lpthread > > LIBS= -lmkl_p3 -lmkl_lapack -lmkl_p3 -lguide -lpthread If you have libfftw.a , please add -lfftw to LIBS. Good luck. > # > # The following is for Atlas optimized blas/lapack libraries > # You may not actually need -lcblas . See the manual for details > # on how to fix I/O incompatibility problems with ifc > # > # LIBS = -lfftw -L$(HOME)/Linux_ATHLON256/ -llapack -lf77blas -lcblas -latlas > # > LD=$(F90) > LFLAGS = -Vaxlib $(LIBS) > > # > # ar: > # > AR = ar > ARFLAGS = ruv From giannozz at nest.sns.it Mon Feb 24 10:14:09 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 24 Feb 2003 10:14:09 +0100 Subject: [Pw_forum] problems on Alpha and PC In-Reply-To: References: Message-ID: <200302241014.09505.giannozz@nest.sns.it> Hi > The run on DEC-Alpha fails with the following error message: I know nothing - and don't want to know anything - about Alpha Dec/Compaq/HP/whatever thay are called now/ machines. I know that it was mentioned in this mailing list that some versions of the dxml/cxml/whatever they are called now/ libraries are buggy. Try to load compiled blas and lapack instead of or before cxml (you may still use FFT's from cxml). > pseudo_dir = '/home/giannozz/O-sesame/pseudo//', > outdir='/home/giannozz/tmp//' you shouldn't use the sample input file as is. Use the script "run_example". I might consider letting you read from my pseudopotentials, but I definitely won't let you write to my home directory :-) > cft_3.o(.text+0xe4): undefined reference to `fftw3d_f77_create_plan_' > cft_3.o(.text+0x10c): undefined reference to `fftwnd_f77_one_' > cft_fftw.o(.text+0x5a): undefined reference to `fftw_f77_create_plan_' > cft_fftw.o(.text+0x7d): undefined reference to `fftw_f77_' There are detailed explanations in the manual on how to choose the number of underscores for FFTW symbols. If you are compiling FFTW from scratch, verify that f77 symbols are created (you need a fortran compiler that is recognized by the FFTW installation mechanism: ifc is not, I think). In any event, you have to load the FFTW library, as noted by Xhongjun. Any idea on how to solve this issue once and for all is welcome. Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050509412 Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 From giannozz at nest.sns.it Mon Feb 24 13:47:36 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Mon, 24 Feb 2003 13:47:36 +0100 Subject: [Pw_forum] computation of STM maps In-Reply-To: <3E54E8B5.B89F6825@fzu.cz> References: <3E54E8B5.B89F6825@fzu.cz> Message-ID: <200302241347.36536.giannozz@nest.sns.it> Hi > I am trying to run pp.x in order to compute a STM map. Unfortunately I am > getting the following error message [...] > > Input/Output Error 153: Input file ended > > In Procedure: sum_band > At Line: 81 > > Statement: Unformatted READ > Unit: 7 > Connected To: /home/crystal/tmp/pwscf.igk this is a bug (actually, two bugs). Comment the line if (.not.lscf) call sum_band in PP/work_function.f90 . Don't use the output of a non-selfconsistent calculation to calculate stm maps: it will not work Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050509412 Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 From xhongjun at mail.ustc.edu.cn Mon Feb 24 14:24:12 2003 From: xhongjun at mail.ustc.edu.cn (xhongjun) Date: Mon, 24 Feb 2003 21:24:12 +0800 Subject: [Pw_forum] error output in phcg.x Message-ID: <3E5A1CFC.7080309@mail.ustc.edu.cn> Dear users of PWSCF, I use PWSCF 1.2.0. When I modify the run_example in example9 to the following: ... cat > sih4.nm.in << EOF normal modes for sih4 &inputph tr2_ph=1.0d-14, prefix='sih4', amass(1)=28.086, amass(2)=1.008, tmp_dir='$TMP_DIR', epsil=.true., trans=.true., asr=.true. deltatau=7.0d-2 , first=12 , last=15 , raman=.true. / 0.0 0.0 0.0 EOF ... Part of the output reads: Raman cross sections are in A^4/amu units # mode [cm-1] [THz] IR Raman 1 -511.77 -15.3425 0.0000 0.3570 2 -511.77 -15.3425 0.0000 0.3570 3 -511.77 -15.3425 0.0000 0.3572 4 0.00 0.0000 9.8026 1.4523 5 0.00 0.0000 9.8026 0.0000 6 0.00 0.0000 9.8026********** 7 740.87 22.2106 0.0000 0.0000 8 740.87 22.2106 0.0000 0.0000 9 903.88 27.0976 5.5553********** 10 903.88 27.0976 5.5553 0.0000 11 903.88 27.0976 5.5553********** 12 2519.42 75.5297 83.0350 0.0000 13 2519.42 75.5297 83.0350********** 14 2519.42 75.5297 83.0350 0.0000 15 2751.06 82.4742 0.0000 0.0000 This error was due to the bug in file raman.f90 from line 694 to line 700. It can be modified to : ... integer nu_temp ... if(nu>=first.and.nu<=last)then nu_temp=nu-first+1 ! alpha = (raman_activity(1,1,nu) + raman_activity(2,2,nu) + & ! raman_activity(3,3,nu))/3.d0 ! beta2 = ( (raman_activity(1,1,nu) - raman_activity(2,2,nu))**2 + & ! (raman_activity(1,1,nu) - raman_activity(3,3,nu))**2 + & ! (raman_activity(2,2,nu) - raman_activity(3,3,nu))**2 + 6.d0 * & ! (raman_activity(1,2,nu)**2 + raman_activity(1,3,nu)**2 + & ! raman_activity(2,3,nu)**2) )/2.d0 alpha = (raman_activity(1,1,nu_temp) + raman_activity(2,2,nu_temp) + & raman_activity(3,3,nu_temp))/3.d0 beta2 = ( (raman_activity(1,1,nu_temp) - & raman_activity(2,2,nu_temp))**2 + & (raman_activity(1,1,nu_temp) - & raman_activity(3,3,nu_temp))**2 + & (raman_activity(2,2,nu_temp) - & raman_activity(3,3,nu_temp))**2 + 6.d0 * & (raman_activity(1,2,nu_temp)**2 + & raman_activity(1,3,nu_temp)**2 + & raman_activity(2,3,nu_temp)**2) )/2.d0 else alpha = 0 beta2 = 0 endif ... From roma at cea.fr Mon Feb 24 14:28:43 2003 From: roma at cea.fr (Guido Roma) Date: Mon, 24 Feb 2003 14:28:43 +0100 Subject: [Pw_forum] problems on Alpha and PC References: <200302241014.09505.giannozz@nest.sns.it> Message-ID: <3E5A1E0B.98B5678E@cea.fr> Paolo Giannozzi wrote: > > Hi > > > The run on DEC-Alpha fails with the following error message: > > I know nothing - and don't want to know anything - about Alpha > Dec/Compaq/HP/whatever thay are called now/ machines. > I know that it was mentioned in this mailing list that some versions > of the dxml/cxml/whatever they are called now/ libraries are buggy. > Try to load compiled blas and lapack instead of or before cxml > (you may still use FFT's from cxml). > Hello, concerning the problem on DEC/Alpha machines**, I'm not sure that it is a problem of CXML libraries, or at least not only that; it could be due to the compiler version, perhaps coupled to the operating system version. I have the copper example running with correct results on a machine (let's call it machine1) running CXML version 5.2 (XMDLIB6520) and f90 version 5.5 (DFABASE551). On another machine (machine2), were I have an old version, f90 5.3 and CXML 3.6, it compiles but the copper example stops with floating invalid. If I link to the (static) library file of CXML 5.2 (which I copied from machine1) the problem is still there. The executable produced on machine1 doesn't work on machine2 (unresolved symbol: for_check_mult_overflow64). The executable produced on machine2 runs on machine1, but still stops for floating invalid. Both processors are of the ev6 family. The operating system version is not the same (machine1 runs Tru64 V5.1 732, machine2 runs Tru64 V4.0 1229). Hope this will give some hints, Guido ** I'm talking of machines running Tru64, I don't know anything about those running linux. -- Guido Roma -- CEA-Saclay - DEN/DMN/SRMP Bat.520/130 Phone: [+33]-1-69085738 -- Fax: ...6867 -- Mobile: [+33]-6-20069085 From pushpa at jncasr.ac.in Wed Feb 26 05:42:12 2003 From: pushpa at jncasr.ac.in (Raghani Pushpa) Date: Wed, 26 Feb 2003 10:12:12 +0530 (IST) Subject: [Pw_forum] (no subject) Message-ID: Dear All, Can I find the point group of a molecule or a cluster from pwscf. if so, could somebody tell me that how can I do it. I know the positions of atoms in the cluster. Thanks, Pushpa -- From degironc at sissa.it Wed Feb 26 09:49:03 2003 From: degironc at sissa.it (Stefano de Gironcoli) Date: Wed, 26 Feb 2003 09:49:03 +0100 Subject: [Pw_forum] (no subject) References: Message-ID: <3E5C7F7F.C1F5CC12@sissa.it> If you specify verbosity='high' in the &control namelist the program output more details about the symmetry operations that are found; not only their total number but also a "human readable" description of each, like "rotation by pi/2 around x" or similar. With a group theory booklet at hand it should be possible to identify the space group. Beware: pwscf only tries to apply symmetry operations that belong to the cubic or hexagonal groups with respect to the assumed cartesian axes. If your molecule is not properly oriented with respect to these axes no symmetry operation or only a subset of them could be found. All the best , stefano Raghani Pushpa wrote: > Dear All, > Can I find the point group of a molecule or a cluster from pwscf. if so, > could somebody tell me that how can I do it. I know the positions of atoms > in the cluster. > > Thanks, > Pushpa > > -- > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From giovanni.cangiani at epfl.ch Wed Feb 26 09:21:30 2003 From: giovanni.cangiani at epfl.ch (Giovanni Cangiani) Date: 26 Feb 2003 09:21:30 +0100 Subject: [Pw_forum] (no subject) In-Reply-To: References: Message-ID: <1046247690.2603.348.camel@irrmalin23.epfl.ch> If you increase the level of verbosity you get not only the number of symmetry operations, but also the rotation matrices. ASAIK, you do not get the name of the spacegroup. For this you can use the sginfo library: http://www.kristall.ethz.ch/LFK/software/sginfo/ or maybe also the crystallography toolbox: http://cctbx.sourceforge.net/index.html giovanni On Wed, 2003-02-26 at 05:42, Raghani Pushpa wrote: > Dear All, > Can I find the point group of a molecule or a cluster from pwscf. if so, > could somebody tell me that how can I do it. I know the positions of atoms > in the cluster. > > Thanks, > Pushpa From yan at im.imag.kiev.ua Wed Feb 26 17:01:39 2003 From: yan at im.imag.kiev.ua (B. Yanchitsky) Date: Wed, 26 Feb 2003 18:01:39 +0200 Subject: [Pw_forum] Changed ecut values Message-ID: <3E5CE4E3.C1BB5107@im.imag.kiev.ua> Dear users, I've a problem with pwscf, actually the problem is that i have got a converged system with some ecut value, and now i'd like to increase ecut, but using previously converged result, i.e. ecut(1)=25, ecut(2)=30, output_pot='pot', input_pot='pot', here 'pot' is the potential from a previous run. Such scheme does not work as far as i can see, ('pot' file exists, of course). A similar problem arises when trying to use the potential file from prev. calculation for slightly distorted unit cell, the program can't read the potential file correctly. So, is any way to use already converged potential (density?) from previous calculation for changed ecut value or distorted unit cell? Much thanks for your help, Bogdan Yanchitsky Institute of Magnetism Vernadsky Blvd., 36-b 03142 Kiev UKRAINE Tel. (+380-44) 444 34 20 Fax. (+380-44) 444 10 20 From eric at lambic.mit.edu Wed Feb 26 17:25:50 2003 From: eric at lambic.mit.edu (Eric Wu) Date: Wed, 26 Feb 2003 11:25:50 -0500 (EST) Subject: [Pw_forum] space group programs In-Reply-To: <1046247690.2603.348.camel@irrmalin23.epfl.ch> Message-ID: Hi Pushpa Another (free) crystallography program you can use is platon. http://www.cryst.chem.uu.nl/platon/ Platon will give space group, space group operations, and even suggest origin shifts. It has a visualizer, and can be coupled with ray tracing programs. It is a standard program used by crystallographers. Eric On 26 Feb 2003, Giovanni Cangiani wrote: > If you increase the level of verbosity you get not only the number of > symmetry operations, but also the rotation matrices. ASAIK, you do not > get the name of the spacegroup. For this you can use the sginfo library: > http://www.kristall.ethz.ch/LFK/software/sginfo/ or maybe also the > crystallography toolbox: http://cctbx.sourceforge.net/index.html > > > giovanni > > > On Wed, 2003-02-26 at 05:42, Raghani Pushpa wrote: > > Dear All, > > Can I find the point group of a molecule or a cluster from pwscf. if so, > > could somebody tell me that how can I do it. I know the positions of atoms > > in the cluster. > > > > Thanks, > > Pushpa > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum > From giuseppe at titus.phy.qub.ac.uk Thu Feb 27 10:51:40 2003 From: giuseppe at titus.phy.qub.ac.uk (Giuseppe Colizzi) Date: Thu, 27 Feb 2003 09:51:40 +0000 (GMT) Subject: [Pw_forum] (no subject) Message-ID: Hi all, I have some problems compiling the last release of pw on T3 Cray machine. Instead the very first release F77 compiles perfectly. I am using the same libraries for both the programs. Compiling the last release during the precompilation I receive several messages of WARNING of variables used before being defined and during the link of the objects I receive WARNING messages of undefined variables as : DSCAL, IDUM, TRANSPOSE and some others. Does anybody have some idea about how to solve this problem? Many Thanks for you attention and help . Regards Giuseppe From degironc at sissa.it Thu Feb 27 12:01:24 2003 From: degironc at sissa.it (Stefano de Gironcoli) Date: Thu, 27 Feb 2003 12:01:24 +0100 Subject: [Pw_forum] BUG in the LDA+U part of pwscf (Lixin He,Rutgers) Message-ID: <3E5DF004.1A57A6F9@sissa.it> Dear All, A bug in the LDA+U part of the PWSCF code has been reported by Lixin He in Vanderbilt's group at Rutgers university. The symmetrization of the occupation matrix used in LDA+U was done incorrectly using the d2 matrices instead of their transpose. In subroutine new_ns.F (or new_ns.f90 in the f90 version) the symmetrization step should be performed with the following bit of fortran nsnew(na,is,m1,m2) = nsnew(na,is,m1,m2) + + d2(m0,m1,isym) * nr(nb,is,m0,m00) * + d2(m00,m2,isym) / nsym instead of the original one nsnew(na,is,m1,m2) = nsnew(na,is,m1,m2) + + d2(m1,m0,isym) * nr(nb,is,m0,m00) * + d2(m2,m00,isym) / nsym i.e. using d2^{T} instead of d2, more precisely: change d2(m1,m0,isym)--> d2(m0,m1,isym) d2(m2,m00,isym)-->d2(m00,m2,isym). Please correct your version accordingly if you plan to use LDA+U. The bug is present in all versions containing LDA+U stuff, including (sob) the newly released 1.2.0 version. It has been fixed in the version under development. A similar error was also present in the PP/projwave.f90 file where atomic-state projected densities of states are computed (and symmetrized). The original code was all right for crystals that only have mirror symmetries or for which symmetries do not exchanges equivalent atoms in the unit cell. But it could fail for crystals that have rotational symmetries that exchange atoms. Thanks to Lixin He for reporting the problem (and the solution !!!) Stefano de Gironcoli From degironc at sissa.it Thu Feb 27 12:31:44 2003 From: degironc at sissa.it (Stefano de Gironcoli) Date: Thu, 27 Feb 2003 12:31:44 +0100 Subject: [Pw_forum] Changed ecut values References: <3E5CE4E3.C1BB5107@im.imag.kiev.ua> Message-ID: <3E5DF720.19181B20@sissa.it> At present it is not possible to restart a calculation with a potential generated with lower (or higher) cutoff. Stefano de Gironcoli "B. Yanchitsky" wrote: > Dear users, > > I've a problem with pwscf, actually the problem is that i have got a converged > system with some ecut value, and now i'd like to increase ecut, but using > previously converged result, i.e. > > ecut(1)=25, ecut(2)=30, > output_pot='pot', input_pot='pot', > > here 'pot' is the potential from a previous run. > Such scheme does not work as far as i can see, ('pot' file exists, of course). > > A similar problem arises when trying to use the potential file from > prev. calculation for slightly distorted unit cell, the program can't read the > potential file correctly. So, is any way to use already converged potential > (density?) from previous calculation for changed ecut value or distorted unit > cell? > > Much thanks for your help, > > Bogdan Yanchitsky > > Institute of Magnetism > Vernadsky Blvd., 36-b > 03142 Kiev > UKRAINE > > Tel. (+380-44) 444 34 20 > Fax. (+380-44) 444 10 20 > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum From giannozz at nest.sns.it Thu Feb 27 11:58:35 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 27 Feb 2003 11:58:35 +0100 Subject: [Pw_forum] Changed ecut values In-Reply-To: <3E5DF720.19181B20@sissa.it> References: <3E5CE4E3.C1BB5107@im.imag.kiev.ua> <3E5DF720.19181B20@sissa.it> Message-ID: <200302271158.35136.giannozz@nest.sns.it> On Thursday 27 February 2003 12:31, Stefano de Gironcoli wrote: > At present it is not possible to restart a calculation with a potential > generated with lower (or higher) cutoff [...] it should be possible soon. Right now the potential (or the charge density) is written in real space, on the FFT grid. It's easy and straightforward, but If the FFT grid changes, the potential cannot be (easily) read. We need to store the potential (or charge density) in G-space, and this is already implemented, it just needs to be put in place. > > I've a problem with pwscf, actually the problem is that i have got a > > converged system with some ecut value, and now i'd like to increase ecut, > > but using previously converged result [...] Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050509412 Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 From giannozz at nest.sns.it Thu Feb 27 12:20:19 2003 From: giannozz at nest.sns.it (Paolo Giannozzi) Date: Thu, 27 Feb 2003 12:20:19 +0100 Subject: [Pw_forum] T3 problems In-Reply-To: References: Message-ID: <200302271220.19735.giannozz@nest.sns.it> Hi > I have some problems compiling the last release of pw on T3 Cray > machine [...] I receive several messages of WARNING of variables > used before being defined this may or may not be harmless. In any case, it shouldn't happen. Can you please send the messages to any of the developers ? > during the link of the objects I receive WARNING messages of > undefined variables as : > DSCAL, IDUM, TRANSPOSE and some others. First, look at which routines are missing such objects. DSCAL is a blas routine. It should be preprocessed to "sscal". If it is not , either -DT3D is missing in preprocessing flags, or the line 'include "machine.h" ' is missing. TRANSPOSE: don't use -DT3D_BENCHLIB and see if the message disappears. It looks like there are two routines, both named 'transpose' ... Paolo -- Paolo Giannozzi e-mail: giannozz at nest.sns.it Scuola Normale Superiore Phone: +39/050509412 Piazza dei Cavalieri 7 Fax: +39/050509417, 050563513 I-56126 Pisa, Italy Office: Lab. NEST, Via della Faggiola 19 From yan at im.imag.kiev.ua Thu Feb 27 13:54:32 2003 From: yan at im.imag.kiev.ua (B. Yanchitsky) Date: Thu, 27 Feb 2003 14:54:32 +0200 Subject: [Pw_forum] Changed ecut values References: <3E5CE4E3.C1BB5107@im.imag.kiev.ua> <3E5DF720.19181B20@sissa.it> Message-ID: <3E5E0A88.3A46590B@im.imag.kiev.ua> Dear Paolo and Stefano, Thanks for your replies and clarification of the problem. Best regards, Bogdan Yanchitsky From yanming_ma at hotmail.com Thu Feb 27 17:54:13 2003 From: yanming_ma at hotmail.com (ma Yanming) Date: Fri, 28 Feb 2003 00:54:13 +0800 Subject: [Pw_forum] Electron-phonon coupling Message-ID: Dear PWSCF Users, I am trying to calculate the electron-phonon coupling coefficient. But I do not know how can I properly choose the k points in the first BZ to calculate, i.e., I don't know how many k points, what kind of k points (high symmetry or not) I can choose. What is the rule? Can anyone help me or provide me some references? Yanming Ma PhD National Research Councils of Canada. _________________________________________________________________ ??????????????? MSN Hotmail? http://www.hotmail.com