mesa-users — open list for mesa comments, questions, & answers

[mesa-users] derived types, dynamic allocation, and pointers in mesa -- not your father's fortran

[Bill P. <pa...@ki...>]

Hi Ehsan,

On Jan 22, 2011, at 7:36 AM, Ehsan Moravveji wrote:
> What is the role of s%? How shall I interpret it, since it appears in all modules. What values does it take/return?
> 


That's such a good question that I'm cc'ing my reply to mesa-users.
It should be useful for anyone who will be looking at the code and doesn't 
already understand derived types, dynamic allocation, and pointers in fortran95.


In "old" fortran, we used common blocks.  Now we take all of the things that would have been in common blocks and call them a "derived type".  We then can dynamically allocate (i.e. during the run) as many copies as we need.  To tell the code which copy of the data it is using, we pass along a pointer to the copy.     The derived type for star is called 'star_info'.  The type declaration can be found in public/star_def.   The routines that access information from star_info will have a declaraction like this:
         type (star_info), pointer :: s
That says s is a pointer to an instance of the star_info derived type.   All of the stuff in star_data.dek is in star_info.  For example, to reference the variable 'nvar' in the instance of star_info data pointed to by s, we write
	s% nvar
The "%" is just the fortran syntax meaning "in the structure specified on the left of the %, access the data named by the stuff on the right of the %".   So s% nvar can act like any other variable -- it can appear as a value in expressions or it can appear on the left of an assignment statement when you want to change its value.

The items in the derived type can be arrays or other types, so, for example, the radius at the outer edge of cell k is
	s% r(k)

Cheers,
Bill

[mesa-users] mesa star build fails for linux gfortran 4.5.1

[Kent G. B. <kg...@kg...>]

Same error message as for gfortran 4.5.0.

Re: [mesa-users] [mesa-news] mesa release 2941

[Kent B. <kg...@la...>]

Bill,

I use 4.5.1 at home. I'll try it out there and see how it goes.

And then I'll try to get the sysadmin at work to upgrade my office
computer to 4.5.1. :(

On Fri, 2011-01-21 at 10:06 -0800, Bill Paxton wrote:
> Hi Kent,
> 
> Bummer.    gfortran 4.5.1 works fine on my mac of course.    sigh.....
> And I agree that asking for "tech support" for gfortran 4.5.0 is not an option.
> 
> Well, guys, I think the time has come.  Just as we need ifort 11.1 rather than 11.0,
> it seems we now need gfortran 4.5.1 rather than 4.5.0
> 
> But first -- let's see if  2941 actually works with 4.5.1 on linux.  
> Someone out there, please let us know.
> 
> Thanks,
> Bill
> 
> 
> 
> 
> On Jan 21, 2011, at 9:47 AM, Kent Budge wrote:
> 
> > Bill,
> > 
> > I'm getting a build error with this latest and gfortran 4.5.0 when
> > building the star module:
> > 
> > gfortran -fno-range-check  -fopenmp  -I../public -I../private
> > -I../../include -Wunused-value -Werror -W -fimplicit-none    -c
> > -ffree-form ../private/opacities.f
> > ../private/opacities.f:374:0: internal compiler error: Segmentation
> > fault
> > Please submit a full bug report,
> > with preprocessed source if appropriate.
> > See <http://gcc.gnu.org/bugs.html> for instructions.
> > make: *** [opacities.o] Error 1
> > 
> > Apparently the compiler is admitting it's the compiler and not the code.
> > Not sure how eager I am to throw a bottle in the ocean, if you know what
> > I mean.
> > 
> > --Kent G. Budge
> > 
> > On Fri, 2011-01-21 at 09:08 -0800, Bill Paxton wrote:
> >> Hello,
> >> 
> >> This is a minor release to fix a few things I screwed up last time.
> >> 
> >> The installation of pulse is now skipped by default.  but I think I've
> >> fixed the problem, so if you'd like to check it, please do so by
> >> setting INSTALL_PULSE = YES in your makefile header.
> >> 
> >> Cheers,
> >> Bill
> >> 
> >> 
> >> ------------------------------------------------------------------------------
> >> Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
> >> Finally, a world-class log management solution at an even better price-free!
> >> Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
> >> February 28th, so secure your free ArcSight Logger TODAY! 
> >> http://p.sf.net/sfu/arcsight-sfd2d
> >> _______________________________________________
> >> mesa-news mailing list
> >> mes...@li...
> >> https://lists.sourceforge.net/lists/listinfo/mesa-news
> > -- 
> > 
> 
-- 

Re: [mesa-users] [mesa-news] mesa release 2941

[Bill P. <pa...@ki...>]

Hi Kent,

Bummer.    gfortran 4.5.1 works fine on my mac of course.    sigh.....
And I agree that asking for "tech support" for gfortran 4.5.0 is not an option.

Well, guys, I think the time has come.  Just as we need ifort 11.1 rather than 11.0,
it seems we now need gfortran 4.5.1 rather than 4.5.0

But first -- let's see if  2941 actually works with 4.5.1 on linux.  
Someone out there, please let us know.

Thanks,
Bill




On Jan 21, 2011, at 9:47 AM, Kent Budge wrote:

> Bill,
> 
> I'm getting a build error with this latest and gfortran 4.5.0 when
> building the star module:
> 
> gfortran -fno-range-check  -fopenmp  -I../public -I../private
> -I../../include -Wunused-value -Werror -W -fimplicit-none    -c
> -ffree-form ../private/opacities.f
> ../private/opacities.f:374:0: internal compiler error: Segmentation
> fault
> Please submit a full bug report,
> with preprocessed source if appropriate.
> See <http://gcc.gnu.org/bugs.html> for instructions.
> make: *** [opacities.o] Error 1
> 
> Apparently the compiler is admitting it's the compiler and not the code.
> Not sure how eager I am to throw a bottle in the ocean, if you know what
> I mean.
> 
> --Kent G. Budge
> 
> On Fri, 2011-01-21 at 09:08 -0800, Bill Paxton wrote:
>> Hello,
>> 
>> This is a minor release to fix a few things I screwed up last time.
>> 
>> The installation of pulse is now skipped by default.  but I think I've
>> fixed the problem, so if you'd like to check it, please do so by
>> setting INSTALL_PULSE = YES in your makefile header.
>> 
>> Cheers,
>> Bill
>> 
>> 
>> ------------------------------------------------------------------------------
>> Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
>> Finally, a world-class log management solution at an even better price-free!
>> Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
>> February 28th, so secure your free ArcSight Logger TODAY! 
>> http://p.sf.net/sfu/arcsight-sfd2d
>> _______________________________________________
>> mesa-news mailing list
>> mes...@li...
>> https://lists.sourceforge.net/lists/listinfo/mesa-news
> -- 
> 

Re: [mesa-users] [mesa-news] mesa release 2941

[Kent B. <kg...@la...>]

Bill,

I'm getting a build error with this latest and gfortran 4.5.0 when
building the star module:

gfortran -fno-range-check  -fopenmp  -I../public -I../private
-I../../include -Wunused-value -Werror -W -fimplicit-none    -c
-ffree-form ../private/opacities.f
../private/opacities.f:374:0: internal compiler error: Segmentation
fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
make: *** [opacities.o] Error 1

Apparently the compiler is admitting it's the compiler and not the code.
Not sure how eager I am to throw a bottle in the ocean, if you know what
I mean.

--Kent G. Budge

On Fri, 2011-01-21 at 09:08 -0800, Bill Paxton wrote:
> Hello,
> 
> This is a minor release to fix a few things I screwed up last time.
> 
> The installation of pulse is now skipped by default.  but I think I've
> fixed the problem, so if you'd like to check it, please do so by
> setting INSTALL_PULSE = YES in your makefile header.
> 
> Cheers,
> Bill
> 
> 
> ------------------------------------------------------------------------------
> Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)!
> Finally, a world-class log management solution at an even better price-free!
> Download using promo code Free_Logger_4_Dev2Dev. Offer expires 
> February 28th, so secure your free ArcSight Logger TODAY! 
> http://p.sf.net/sfu/arcsight-sfd2d
> _______________________________________________
> mesa-news mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-news
-- 

[mesa-users] mesa release 2941

[Bill P. <pa...@ki...>]

Hello,

This is a minor release to fix a few things I screwed up last time.

The installation of pulse is now skipped by default.  but I think I've
fixed the problem, so if you'd like to check it, please do so by
setting INSTALL_PULSE = YES in your makefile header.

Cheers,
Bill

[mesa-users] Re-compiling EOS tables

[Dr. E. O. <ol...@sa...>]

Hi I tried to recompile the EOS tables but I got the following message when executing "./replace_DTdata" following the steps in the README file.

create eosDT_data
bzip2 -d eosDT_data.tar.bz2
eosDT_data/
mv: cannot stat `data/eosDT_data/*.data': No such file or directory

I then tried to get things working with a sledgehammer approach by recompiling all of MESA again with a  "./clean" and then a "./install". Things compiled fine untill the following error appeared:

 read  ../../data/eosDT_data/helm_table.dat
ERROR: failed to open ../../data/eosDT_data/helm_table.dat : ios      2
ERROR: read_helm_table failed
 failed in eos_init
 ERROR: read_helm_table failed
STOP 1

Can someone point me a way around this? I am trying to use the EOS module without radiation included, by trying to set crad=0 and then recompiling the tables. I am doing this "absurd" thing since I am wanting to use the EOS module in an application where I treat the radiation and gas separately and don't assume the gas and radiation is necessarily in equilibrium. 

Can anyone help? 

Thank you,
Enrico

[mesa-users] EOS without radiation pressure/energy density

[Dr. E. O. <ol...@sa...>]

Hi,

I would like to use the EOS module of MESA without radiation included. Will setting "crad=0" in const_def.f work? Or will it screw up something else in MESA as well?

Regards,

Enrico

Re: [mesa-users] MESA: doing something more than reporting...

[Bill P. <pa...@ki...>]

Hi,

Turns out that some fool (me) left a debugging trap enabled.  ; - (
I'll put out a new release asap with that fixed.  
If you'd like to make a tiny patch, replace this file in star/private.
Then in star/test do
./mk
./export

Sorry for the inconvenience. 

[mesa-users] MESA: doing something more than reporting...

[Evan O'C. <ev...@ca...>]

Hello all,

I just ran into something weird.  For the situation I am looking at, when I switch on 

report_all_dt_limits = .true.

the code crashes, if it is false then it is fine.  some of the error message is below.  

>>>>>>>>>>>>>>>
       1510   8.180391  1.244E+04   4.636411   4.755512  14.674456  11.547502   0.000000   0.012976   0.280000   0.477836  -3.242907    705      1
   2.248222   3.082165   1.683802   3.145216   3.469588  -6.929051   3.126954   0.979459   0.000502   0.020000   0.502238   0.041347      2      1
 1.1563E+07  19.104688   4.700862   4.088597   2.141492 -10.288952   0.000000   0.000728   0.004313   0.000396  1.993E-02  0.260E-04    varcontrol
 
                        dX n14  623    0.2006471728E-02    0.2000000000E-02    0.8955172923E+00    0.1445299224E-02
                             s% D_mix(k)         623    0.0000000000000000E+00
                                   X_old         623    1.4982242077306349E-02
                              s% xa(j,k)         623    1.2975770349680954E-02
                                      dX         623   -2.0064717276253950E-03
                                   D_mix         618    0.0000000000000000E+00    4.7017707893630550E-03
                                   D_mix         619    0.0000000000000000E+00    3.9011863067364982E-03
                                   D_mix         620    0.0000000000000000E+00    3.1006018241099415E-03
                                   D_mix         621    0.0000000000000000E+00    1.4994328588572581E-03
                                   D_mix         622    0.0000000000000000E+00    7.4971642942862904E-04
                                   D_mix         623    0.0000000000000000E+00    0.0000000000000000E+00
                                   D_mix         624    1.2954886313713439E+13   -1.4994328588571471E-03
                                   D_mix         625    1.6061308510682049E+13   -2.6642224508062681E-03
                                   D_mix         626    1.8407963009267238E+13   -3.8290120427557223E-03
                                   D_mix         627    2.0443522267985812E+13   -4.9938016347049544E-03
                                   D_mix         628    2.2259202775794098E+13   -6.1585912266542975E-03
 
check_dX
>>>>>>>>>>>>>>>>>

For me this is occurring right after helium burning in a 15Msun star (when the Ne20 and N14 levels are changing).  For reference, the inlist_15_z2m2 is identical to mesa/start/test_suite/15_si_burn/inlist_15_z2m2 except this flag is changed and I do not load the test_suite model but rather start from pre-main sequence).  I've also seen this occur in the same place in stars up to 19Msun.

It seems like the extra reporting causes mesa to fail.  I'm not sure why, thoughts?

Evan

Re: [mesa-users] Error during installation

[Kent B. <kg...@la...>]

Another data point for you, Bill. I can build pulse with gfortran 4.5.1
on my home computer, but not with 4.5.0 here at work.

--Kent G. Budge

On Sun, 2011-01-16 at 09:59 -0800, Bill Paxton wrote:
> Hi Evan,
> 
> 
> Excellent!  Thanks for that info.   
> I've now found that I get a crash at that same line when I run
> Red Hat 4.4.5-2
> gcc version 4.6.0 20100609
> If I comment out that deallocate, it finishes the test okay.
> There must be enough variation in the runtime dynamic allocation
> that we are breaking things for some systems but not for others.
> 
> 
> Until the problem is found and fixed, I'll change pulse installation
> to be optional.
> 
> 
> As a workaround, it is okay to comment out the "do_one pulse" line in
> the install script.
> 
> 
> Bill
> 
> 
> 
> 
> On Jan 16, 2011, at 9:26 AM, Evan O'Connor wrote:
> 
> > Hello,
> > 
> > 
> > I've also had issues with installation and pulse. I have ifort 12
> > (Mac) which works fine but with ifort 11.1(Linux) on another
> > computer I use I have an issue. The test itself go smoothly (the
> > correct output is in the tmp.txt file) but then gets hung up
> > indefinitely on the following line in pulse.f:do_free_pulse_info
> > 
> > 
> > deallocate(p% x, p% aa, p% y, p% yri)
> > 
> > 
> > This doesn't seem to be the same error as Naveen but maybe is
> > somehow related. 
> > 
> > 
> > Evan
> > 
> > On Jan 16, 2011, at 8:54, Bill Paxton <pa...@ki...> wrote:
> > 
> > 
> > 
> > > Hi Naveen,
> > > 
> > > 
> > > Thanks for the detailed report.   
> > > The installation goes okay with ifort and gfortran on my mac --
> > > that's using gcc version 4.5.1 20100506.
> > > Linux installation issues are out of my area, so we'll need to see
> > > if someone else can help there.
> > > 
> > > 
> > > If it turns into a show stopper, I'll add installation switches to
> > > turn off pulse in the same way we have
> > > the option to run without pgplot.   Let me know.
> > > 
> > > 
> > > In the meantime, edit the install script to comment out the line
> > > 
> > > 
> > > do_one pulse
> > > 
> > > 
> > > and try again.
> > > 
> > > 
> > > -Bill
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > > On Jan 16, 2011, at 5:34 AM, naveen yadav wrote:
> > > 
> > > >                                                 
> > > >                                                 
> > > >                                           Hi all,
> > > >                                      I am trying to
> > > >                                 install MESA and I am
> > > >                                    getting an error.
> > > >                                       The error is 
> > > > ******************************************************************************************
> > > >                              /home/astroboy/mesa/pulse
> > > >                                      building pulse
> > > >                                           package.
> > > >                                                 
> > > >                                                 
> > > >                                 make: `libpulse.a' is
> > > >                                        up to date.
> > > >                                    gfortran -fopenmp
> > > >                                        -o ../tester
> > > >                                        test_pulse.o
> > > >                                  -L../../make -lpulse
> > > >                                  -L../../../lib -leos
> > > >                                  -lkap -ljina -lweak
> > > >                                      -lnet -lscreen
> > > >                                  -lrates -lneu -lchem
> > > >                                        -linterp_2d
> > > >                                    -linterp_1d -lnum
> > > >                                     -lutils -lalert
> > > >                                       -lconst -lmtx
> > > >                                        -lmesalapack
> > > >                                         -lmesablas
> > > >                                                 
> > > >                                                 
> > > >                                                 
> > > >                                                 
> > > >                                            FAILED
> > > >                                                 
> > > >                                                 
> > > >                                                 
> > > >                                                 
> > > >                          /home/astroboy/mesa/pulse/test
> > > >                                      TEST FAILED --
> > > >                                  compare test_output
> > > >                                         to tmp.txt
> > > >                                                 
> > > >                                                 
> > > > ********************************************************************************************
> > > >                                      I am using GNU
> > > >                                  Fortran (GCC) 4.5.1
> > > >                                    20100924 (Red Hat
> > > >                                           4.5.1-4)
> > > >                                    I am using Pgplot
> > > >                                     without lpng and
> > > >                                           fortran
> > > >                   option ::SPECIAL_FC_FLAGS =  -fopenmpi.
> > > >                                  I am using the case
> > > >                                     (3) for BLAS and
> > > >                                     LAPACK libraries
> > > >                                      I am attaching
> > > >                                    the makefile ,the
> > > >                                    output of install
> > > >                                        step and the
> > > >                                          temp.txt.
> > > >                                                 
> > > >                                                 
> > > >                                                 
> > > >                                                 
> > > >                                           regards
> > > >                                           naveen 
> > > >                                                 
> > > >                                                 
> > > >                                                 
> > > > 
> > > > <install_output.txt.tar.gz>------------------------------------------------------------------------------
> > > > Protect Your Site and Customers from Malware Attacks
> > > > Learn about various malware tactics and how to avoid them.
> > > > Understand 
> > > > malware threats, the impact they can have on your business, and
> > > > how you 
> > > > can protect your company and customers by using code signing.
> > > > http://p.sf.net/sfu/oracle-sfdevnl_______________________________________________
> > > > mesa-users mailing list
> > > > mes...@li...
> > > > https://lists.sourceforge.net/lists/listinfo/mesa-users
> > > > 
> > > 
> > > 
> > > ------------------------------------------------------------------------------
> > > Protect Your Site and Customers from Malware Attacks
> > > Learn about various malware tactics and how to avoid them.
> > > Understand 
> > > malware threats, the impact they can have on your business, and
> > > how you 
> > > can protect your company and customers by using code signing.
> > > http://p.sf.net/sfu/oracle-sfdevnl
> > > _______________________________________________
> > > mesa-users mailing list
> > > mes...@li...
> > > https://lists.sourceforge.net/lists/listinfo/mesa-users
> > > 
> 
> 
> ------------------------------------------------------------------------------
> Protect Your Site and Customers from Malware Attacks
> Learn about various malware tactics and how to avoid them. Understand 
> malware threats, the impact they can have on your business, and how you 
> can protect your company and customers by using code signing.
> http://p.sf.net/sfu/oracle-sfdevnl
> _______________________________________________ mesa-users mailing list mes...@li... https://lists.sourceforge.net/lists/listinfo/mesa-users
-- 

Re: [mesa-users] Error during installation

[Billy V. (R. Student) <bx...@ri...>]

Dr. Paxton,

I can confirm that the installation in Ubuntu ( Linux ),  using ifort
12.0.0.20101116 hangs while compiling the pulse package.  I comment out the
pulse package in the installation script and it completes.

- Billy

On Sun, Jan 16, 2011 at 11:54 AM, Bill Paxton <pa...@ki...> wrote:

> Hi Naveen,
>
> Thanks for the detailed report.
> The installation goes okay with ifort and gfortran on my mac -- that's
> using gcc version 4.5.1 20100506.
> Linux installation issues are out of my area, so we'll need to see if
> someone else can help there.
>
> If it turns into a show stopper, I'll add installation switches to turn off
> pulse in the same way we have
> the option to run without pgplot.   Let me know.
>
> In the meantime, edit the install script to comment out the line
>
> do_one pulse
>
> and try again.
>
> -Bill
>
>
>
>
>
> On Jan 16, 2011, at 5:34 AM, naveen yadav wrote:
>
> Hi all,
> I am trying to install MESA and I am getting an error.
> The error is
>
> ******************************************************************************************
> /home/astroboy/mesa/pulse
> building pulse package.
>
> make: `libpulse.a' is up to date.
> gfortran -fopenmp -o ../tester test_pulse.o -L../../make -lpulse
> -L../../../lib -leos -lkap -ljina -lweak -lnet -lscreen -lrates -lneu -lchem
> -linterp_2d -linterp_1d -lnum -lutils -lalert -lconst -lmtx -lmesalapack
> -lmesablas
>
>
> FAILED
>
>
> /home/astroboy/mesa/pulse/test
> TEST FAILED -- compare test_output to tmp.txt
>
>
> ********************************************************************************************
> I am using GNU Fortran (GCC) 4.5.1 20100924 (Red Hat 4.5.1-4)
> I am using Pgplot without lpng and fortran option ::SPECIAL_FC_FLAGS =
>  -fopenmpi.
> I am using the case (3) for BLAS and LAPACK libraries
> I am attaching the makefile ,the output of install step and the temp.txt.
>
>
> regards
> naveen
>
> <install_output.txt.tar.gz>
> ------------------------------------------------------------------------------
> Protect Your Site and Customers from Malware Attacks
> Learn about various malware tactics and how to avoid them. Understand
> malware threats, the impact they can have on your business, and how you
> can protect your company and customers by using code signing.
>
> http://p.sf.net/sfu/oracle-sfdevnl_______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users
>
>
>
>
> ------------------------------------------------------------------------------
> Protect Your Site and Customers from Malware Attacks
> Learn about various malware tactics and how to avoid them. Understand
> malware threats, the impact they can have on your business, and how you
> can protect your company and customers by using code signing.
> http://p.sf.net/sfu/oracle-sfdevnl
>
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users
>
>

Re: [mesa-users] Error during installation

[Bill P. <pa...@ki...>]

Hi Evan,

Excellent!  Thanks for that info.   
I've now found that I get a crash at that same line when I run
	Red Hat 4.4.5-2
	gcc version 4.6.0 20100609
If I comment out that deallocate, it finishes the test okay.
There must be enough variation in the runtime dynamic allocation
that we are breaking things for some systems but not for others.

Until the problem is found and fixed, I'll change pulse installation to be optional.

As a workaround, it is okay to comment out the "do_one pulse" line in the install script.

Bill


On Jan 16, 2011, at 9:26 AM, Evan O'Connor wrote:

> Hello,
> 
> I've also had issues with installation and pulse. I have ifort 12 (Mac) which works fine but with ifort 11.1(Linux) on another computer I use I have an issue. The test itself go smoothly (the correct output is in the tmp.txt file) but then gets hung up indefinitely on the following line in pulse.f:do_free_pulse_info
> 
> deallocate(p% x, p% aa, p% y, p% yri)
> 
> This doesn't seem to be the same error as Naveen but maybe is somehow related. 
> 
> Evan
> 
> On Jan 16, 2011, at 8:54, Bill Paxton <pa...@ki...> wrote:
> 
>> Hi Naveen,
>> 
>> Thanks for the detailed report.   
>> The installation goes okay with ifort and gfortran on my mac -- that's using gcc version 4.5.1 20100506.
>> Linux installation issues are out of my area, so we'll need to see if someone else can help there.
>> 
>> If it turns into a show stopper, I'll add installation switches to turn off pulse in the same way we have
>> the option to run without pgplot.   Let me know.
>> 
>> In the meantime, edit the install script to comment out the line
>> 
>> do_one pulse
>> 
>> and try again.
>> 
>> -Bill
>> 
>> 
>> 
>> 
>> 
>> On Jan 16, 2011, at 5:34 AM, naveen yadav wrote:
>> 
>>> 
>>> Hi all,
>>> I am trying to install MESA and I am getting an error.
>>> The error is 
>>> ******************************************************************************************
>>> /home/astroboy/mesa/pulse
>>> building pulse package.
>>> 
>>> make: `libpulse.a' is up to date.
>>> gfortran -fopenmp -o ../tester test_pulse.o -L../../make -lpulse -L../../../lib -leos -lkap -ljina -lweak -lnet -lscreen -lrates -lneu -lchem -linterp_2d -linterp_1d -lnum -lutils -lalert -lconst -lmtx -lmesalapack -lmesablas
>>> 
>>> 
>>> FAILED
>>> 
>>> 
>>> /home/astroboy/mesa/pulse/test
>>> TEST FAILED -- compare test_output to tmp.txt
>>> 
>>> ********************************************************************************************
>>> I am using GNU Fortran (GCC) 4.5.1 20100924 (Red Hat 4.5.1-4)
>>> I am using Pgplot without lpng and fortran option ::SPECIAL_FC_FLAGS =  -fopenmpi.
>>> I am using the case (3) for BLAS and LAPACK libraries
>>> I am attaching the makefile ,the output of install step and the temp.txt.
>>> 
>>> 
>>> regards
>>> naveen 
>>> 
>>> <install_output.txt.tar.gz>------------------------------------------------------------------------------
>>> Protect Your Site and Customers from Malware Attacks
>>> Learn about various malware tactics and how to avoid them. Understand 
>>> malware threats, the impact they can have on your business, and how you 
>>> can protect your company and customers by using code signing.
>>> http://p.sf.net/sfu/oracle-sfdevnl_______________________________________________
>>> mesa-users mailing list
>>> mes...@li...
>>> https://lists.sourceforge.net/lists/listinfo/mesa-users
>> 
>> ------------------------------------------------------------------------------
>> Protect Your Site and Customers from Malware Attacks
>> Learn about various malware tactics and how to avoid them. Understand 
>> malware threats, the impact they can have on your business, and how you 
>> can protect your company and customers by using code signing.
>> http://p.sf.net/sfu/oracle-sfdevnl
>> _______________________________________________
>> mesa-users mailing list
>> mes...@li...
>> https://lists.sourceforge.net/lists/listinfo/mesa-users

Re: [mesa-users] Error during installation

[Evan O'C. <ev...@ca...>]

Hello,

I've also had issues with installation and pulse. I have ifort 12 (Mac) which works fine but with ifort 11.1(Linux) on another computer I use I have an issue. The test itself go smoothly (the correct output is in the tmp.txt file) but then gets hung up indefinitely on the following line in pulse.f:do_free_pulse_info

deallocate(p% x, p% aa, p% y, p% yri)

This doesn't seem to be the same error as Naveen but maybe is somehow related. 

Evan

On Jan 16, 2011, at 8:54, Bill Paxton <pa...@ki...> wrote:

> Hi Naveen,
> 
> Thanks for the detailed report.   
> The installation goes okay with ifort and gfortran on my mac -- that's using gcc version 4.5.1 20100506.
> Linux installation issues are out of my area, so we'll need to see if someone else can help there.
> 
> If it turns into a show stopper, I'll add installation switches to turn off pulse in the same way we have
> the option to run without pgplot.   Let me know.
> 
> In the meantime, edit the install script to comment out the line
> 
> do_one pulse
> 
> and try again.
> 
> -Bill
> 
> 
> 
> 
> 
> On Jan 16, 2011, at 5:34 AM, naveen yadav wrote:
> 
>> 
>> Hi all,
>> I am trying to install MESA and I am getting an error.
>> The error is 
>> ******************************************************************************************
>> /home/astroboy/mesa/pulse
>> building pulse package.
>> 
>> make: `libpulse.a' is up to date.
>> gfortran -fopenmp -o ../tester test_pulse.o -L../../make -lpulse -L../../../lib -leos -lkap -ljina -lweak -lnet -lscreen -lrates -lneu -lchem -linterp_2d -linterp_1d -lnum -lutils -lalert -lconst -lmtx -lmesalapack -lmesablas
>> 
>> 
>> FAILED
>> 
>> 
>> /home/astroboy/mesa/pulse/test
>> TEST FAILED -- compare test_output to tmp.txt
>> 
>> ********************************************************************************************
>> I am using GNU Fortran (GCC) 4.5.1 20100924 (Red Hat 4.5.1-4)
>> I am using Pgplot without lpng and fortran option ::SPECIAL_FC_FLAGS =  -fopenmpi.
>> I am using the case (3) for BLAS and LAPACK libraries
>> I am attaching the makefile ,the output of install step and the temp.txt.
>> 
>> 
>> regards
>> naveen 
>> 
>> <install_output.txt.tar.gz>------------------------------------------------------------------------------
>> Protect Your Site and Customers from Malware Attacks
>> Learn about various malware tactics and how to avoid them. Understand 
>> malware threats, the impact they can have on your business, and how you 
>> can protect your company and customers by using code signing.
>> http://p.sf.net/sfu/oracle-sfdevnl_______________________________________________
>> mesa-users mailing list
>> mes...@li...
>> https://lists.sourceforge.net/lists/listinfo/mesa-users
> 
> ------------------------------------------------------------------------------
> Protect Your Site and Customers from Malware Attacks
> Learn about various malware tactics and how to avoid them. Understand 
> malware threats, the impact they can have on your business, and how you 
> can protect your company and customers by using code signing.
> http://p.sf.net/sfu/oracle-sfdevnl
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users

Re: [mesa-users] Error during installation

[Bill P. <pa...@ki...>]

Hi Naveen,

Thanks for the detailed report.   
The installation goes okay with ifort and gfortran on my mac -- that's using gcc version 4.5.1 20100506.
Linux installation issues are out of my area, so we'll need to see if someone else can help there.

If it turns into a show stopper, I'll add installation switches to turn off pulse in the same way we have
the option to run without pgplot.   Let me know.

In the meantime, edit the install script to comment out the line

do_one pulse

and try again.

-Bill





On Jan 16, 2011, at 5:34 AM, naveen yadav wrote:

> 
> Hi all,
> I am trying to install MESA and I am getting an error.
> The error is 
> ******************************************************************************************
> /home/astroboy/mesa/pulse
> building pulse package.
> 
> make: `libpulse.a' is up to date.
> gfortran -fopenmp -o ../tester test_pulse.o -L../../make -lpulse -L../../../lib -leos -lkap -ljina -lweak -lnet -lscreen -lrates -lneu -lchem -linterp_2d -linterp_1d -lnum -lutils -lalert -lconst -lmtx -lmesalapack -lmesablas
> 
> 
> FAILED
> 
> 
> /home/astroboy/mesa/pulse/test
> TEST FAILED -- compare test_output to tmp.txt
> 
> ********************************************************************************************
> I am using GNU Fortran (GCC) 4.5.1 20100924 (Red Hat 4.5.1-4)
> I am using Pgplot without lpng and fortran option ::SPECIAL_FC_FLAGS =  -fopenmpi.
> I am using the case (3) for BLAS and LAPACK libraries
> I am attaching the makefile ,the output of install step and the temp.txt.
> 
> 
> regards
> naveen 
> 
> <install_output.txt.tar.gz>------------------------------------------------------------------------------
> Protect Your Site and Customers from Malware Attacks
> Learn about various malware tactics and how to avoid them. Understand 
> malware threats, the impact they can have on your business, and how you 
> can protect your company and customers by using code signing.
> http://p.sf.net/sfu/oracle-sfdevnl_______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users

[mesa-users] mesa release 2920

[Bill P. <pa...@ki...>]

Hi,

Highlights of changes in the 2 weeks since the last release (2891):

In the atm module, we now have tau=1 and tau=10 tables to match the tau=100 tables

The pulse module is now part of the standard install.  It is derived from
Jørgen Christensen-Dalsgaard's ADIPLS.   In addition to running
it from text file input, you can call it directly from star.  You can
find an example in star/test_suite/pulse.

There are several new additions to star/exercises.
I recommend the exercises both for new users and as a
source of "how to" ideas for various topics in "stellar engineering"
such as "How do I model a M=0.001Msun gas giant with an inert core?" 

Frank Timmes has provided a much simplified setup for the
installation of pgplot.   It is in mesa/utils/pgplot.tbz

The ionization package now includes Francois Hebert's
results for the ionization state of fe56 in he4.   
The diffusion package supports the new ionization,
and star supports the changes in diffusion.
See 'diffusion_min_Y_for_fe_in_he' in star_defaults.dek

Users can now write their own routines to add new plots to pgstar.  
See star/public/other_pgstar_plots.f

Cheers,
Bill


p.s.   I'm sending this to both mesa-news and mesa-users.
My thinking on these 2 lists has changed.  I now think people 
should be able to be on just one or the other, not both.
mesa-news will be limited to messages like this one,
i.e., new releases and other news items concerning mesa.
my idea is that it will be for people interested in following
developments in mesa but who are not active users.

mesa-users messages will include the ones on mesa-news, but
will also have requests for help and bug reports and
replies.   this is the list for all active users of mesa.
if you are on mesa-users, you don't need to be on mesa-news.

does that sound okay to you?

Re: [mesa-users] Appetizer error

[<ric...@ms...>]

After talking with Bill on this one, I found out the error was the way  
my quotes were written. In case anyone else has this error, a simple  
copy and paste from the webpage will not do since there are right and  
left quotes, you need to place straight quotes in manually.

Quoting Chris Richardson <ric...@ms...>:

> Hello all,
>
> I installed MESA and everything went fine using gfortran 4.5.1 but I'm
> getting an error with the appetizer. I altered the inlist to the what
> described on the website, saved the file and ran ./rn to get this result:
>
> Revision: 2891
> Fri Jan 7 12:52:05 EST 2011
> Failed while trying to read control namelist file inlist
> The following runtime error message might help you find the problem
>
> At line 1384 of file ../public/run_star_support.f (unit = 9, file =
> 'inlist')
> Fortran runtime error: Cannot match namelist object name
> ’inlist_first_thing_to_try’
>
> The file 'inlist_first_thing_to_try' is located in the same directory as
> 'inlist'. I figured this should work right off the bat so I wanted see
> if anyone had any ideas before I go altering things. Many thanks.
>
>
> ------------------------------------------------------------------------------
> Gaining the trust of online customers is vital for the success of any company
> that requires sensitive data to be transmitted over the Web.   Learn how to
> best implement a security strategy that keeps consumers' information secure
> and instills the confidence they need to proceed with transactions.
> http://p.sf.net/sfu/oracle-sfdevnl
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users
>

Re: [mesa-users] MESA Module Sample Issue

[Bill P. <pa...@ki...>]

Hi,

Well, it turns out to have been a bug in the test rather than a bug in the sample code.
Somewhere along the line, the test_output got bogus values, and it wasn't
being checked.   Now, I've added sample to the list of modules for install, so it will
be built and tested as part of every install.   That will be part of the next release.

Cheers,
Bill







On Jan 12, 2011, at 3:31 PM, Chenchong Zhu wrote:

> Hi all,
> 
> I'm currently trying to run the EOS module sample included with MESA
> (mesa/sample), as instructed in the "Using MESA modules in your project"
> subsection of Getting Started.  I've gotten the sample to run properly,
> but unfortunately all my input logPgas, c_P, etc, match my output values,
> which isn't in accordance with test_output.  Is this indicating that MESA
> has installed incorrectly?
> 
> Thanks for your help!
> 
> Charles
> 
> 
> ------------------------------------------------------------------------------
> Protect Your Site and Customers from Malware Attacks
> Learn about various malware tactics and how to avoid them. Understand 
> malware threats, the impact they can have on your business, and how you 
> can protect your company and customers by using code signing.
> http://p.sf.net/sfu/oracle-sfdevnl
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users

Re: [mesa-users] MESA Module Sample Issue

[Bill P. <pa...@ki...>]

Hi Charles,

I think MESA has installed okay, but you've discovered a problem in the sample!
It looks like the (density,temperature) version of the test is okay, but the
(pressure,temperature) is failing.  Unless you are planning to use the (P,T) eos,
I think you can go ahead while I figure out what's gone wrong.

Thanks for pointing out the problem!

-Bill



On Jan 12, 2011, at 3:31 PM, Chenchong Zhu wrote:

> Hi all,
> 
> I'm currently trying to run the EOS module sample included with MESA
> (mesa/sample), as instructed in the "Using MESA modules in your project"
> subsection of Getting Started.  I've gotten the sample to run properly,
> but unfortunately all my input logPgas, c_P, etc, match my output values,
> which isn't in accordance with test_output.  Is this indicating that MESA
> has installed incorrectly?
> 
> Thanks for your help!
> 
> Charles
> 
> 
> ------------------------------------------------------------------------------
> Protect Your Site and Customers from Malware Attacks
> Learn about various malware tactics and how to avoid them. Understand 
> malware threats, the impact they can have on your business, and how you 
> can protect your company and customers by using code signing.
> http://p.sf.net/sfu/oracle-sfdevnl
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users

[mesa-users] MESA Module Sample Issue

[Chenchong Z. <cc...@le...>]

Hi all,

I'm currently trying to run the EOS module sample included with MESA
(mesa/sample), as instructed in the "Using MESA modules in your project"
subsection of Getting Started.  I've gotten the sample to run properly,
but unfortunately all my input logPgas, c_P, etc, match my output values,
which isn't in accordance with test_output.  Is this indicating that MESA
has installed incorrectly?

Thanks for your help!

Charles

Re: [mesa-users] How to stop running when reached the ZAMS point?

[Falk H. <fh...@uv...>]

O good way to define the ZAMS is when log g has a maximum.
Falk
On 3-Jan-11, at 7:10 AM, Aaron Dotter wrote:

> Hi Zahra,
>
> Sometimes the experts might argue about how to define the ZAMS.  It  
> depends somewhat on the mass you are interested in.  However, One  
> good test is the first time that the luminosity from nuclear burning  
> is > 99% of the total luminosity.  Bill has kindly provided a check  
> for this controls inlist:
>
>          abs_diff_lg_LH_lg_Ls_limit = -1
>             ! stop when abs(lg_LH - lg_Ls) <= abs_diff_LH_Lsurf_limit
>             ! can be useful for deciding when pre-main sequence star  
> has reached ZAMS
>             ! set to negative value to disable
>
> If you set this to 2, you will get the test I described above.
>
> Best wishes,
> Aaron
>
>
>
>
> On Mon, Jan 3, 2011 at 2:58 AM, zahra altaha motahar <zam...@ya... 
> > wrote:
> Hi,
> I want to find the ZAMS starting point in my results. How can I stop  
> running when pre-main sequence star has reached ZAMS?
>
> Regards,
> Zahra
>
>
> ------------------------------------------------------------------------------
> Learn how Oracle Real Application Clusters (RAC) One Node allows  
> customers
> to consolidate database storage, standardize their database  
> environment, and,
> should the need arise, upgrade to a full multi-node Oracle RAC  
> database
> without downtime or disruption
> http://p.sf.net/sfu/oracle-sfdevnl
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users
>
>
> ------------------------------------------------------------------------------
> Learn how Oracle Real Application Clusters (RAC) One Node allows  
> customers
> to consolidate database storage, standardize their database  
> environment, and,
> should the need arise, upgrade to a full multi-node Oracle RAC  
> database
> without downtime or disruption
> http://p.sf.net/sfu/oracle-sfdevnl_______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users

--
Falk Herwig
Dept of Physics & Astronomy, U of Victoria
fh...@uv..., tel: +1 (250) 721-7743

--------------------------------------------------
I support http://www.rainforestsos.org

[mesa-users] Appetizer error

[Chris R. <ric...@ms...>]

Hello all,

I installed MESA and everything went fine using gfortran 4.5.1 but I'm 
getting an error with the appetizer. I altered the inlist to the what 
described on the website, saved the file and ran ./rn to get this result:

Revision: 2891
Fri Jan 7 12:52:05 EST 2011
Failed while trying to read control namelist file inlist
The following runtime error message might help you find the problem

At line 1384 of file ../public/run_star_support.f (unit = 9, file = 
'inlist')
Fortran runtime error: Cannot match namelist object name 
’inlist_first_thing_to_try’

The file 'inlist_first_thing_to_try' is located in the same directory as 
'inlist'. I figured this should work right off the bat so I wanted see 
if anyone had any ideas before I go altering things. Many thanks.

Re: [mesa-users] error during install

[Aaron D. <aar...@gm...>]

Hi Marco,

It looks there is a numerical precision problem with the ndiff utility that
is used to check the output of MESA test codes to the expected results.  If
you look at the file

mesa/utils/test/ck12

it has the line

../../utils/ndiff/bin/ndiff -quiet -relerr 1.0e-12 tmp.txt test_output

You edit this fiel to use a larger value or relerr, or try compiling your
own version of ndiff on the Windows machine to see if that helps.  Let us
know if it does not.

Also, to send messsages to mesa-users, we ask that you subscribe to the
mesa-users list.  You can do that here:

https://lists.sourceforge.net/lists/listinfo/MESA-users


Aaron

[mesa-users] error during install

[<mar...@in...>]

Installing mesa code on windows xp with install script , with lastest version 
of MinGW, I have this
error, :

"RELERR: 1.00e-012 is below machine epsilon 0.00000e+000" (see attachment for 
screenshot error)

How to solve this problem ?
Many thanks.

Marco Marchese 

Re: [mesa-users] How to stop running when reached the ZAMS point?

[Aaron D. <aar...@gm...>]

Hi Zahra,

Sometimes the experts might argue about how to define the ZAMS.  It depends
somewhat on the mass you are interested in.  However, One good test is the
first time that the luminosity from nuclear burning is > 99% of the total
luminosity.  Bill has kindly provided a check for this controls inlist:

         abs_diff_lg_LH_lg_Ls_limit = -1
            ! stop when abs(lg_LH - lg_Ls) <= abs_diff_LH_Lsurf_limit
            ! can be useful for deciding when pre-main sequence star has
reached ZAMS
            ! set to negative value to disable

If you set this to 2, you will get the test I described above.

Best wishes,
Aaron




On Mon, Jan 3, 2011 at 2:58 AM, zahra altaha motahar <zam...@ya...>wrote:

> Hi,
> I want to find the ZAMS starting point in my results. How can I stop
> running when pre-main sequence star has reached ZAMS?
>
> Regards,
> Zahra
>
>
>
> ------------------------------------------------------------------------------
> Learn how Oracle Real Application Clusters (RAC) One Node allows customers
> to consolidate database storage, standardize their database environment,
> and,
> should the need arise, upgrade to a full multi-node Oracle RAC database
> without downtime or disruption
> http://p.sf.net/sfu/oracle-sfdevnl
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users
>
>