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

Re: [mesa-users] New release of the MESA SDK

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

On Jan 21, 2012, at 8:48 AM, Richard Townsend wrote:

> As usual, please let me know if you run into any difficulties.

Hi Rich,

Sorry, I can't report any difficulties!  ;-)
It all went smoothly on my Mac with  OS X 10.6.8
Good work, as usual.  Thanks again for supplying the mesa_sdk.

Cheers,
Bill

[mesa-users] Compiling with MESA SDK on Mac OS X 10.5.8

[Chris M. <cma...@um...>]

Hi all,

I've unpacked today's release of the SDK and I'm having issues trying to
install MESA.  I run into problems trying to build the mtx package.  I'm
not sure what's going on, so here are some of the interesting snippets:

gfortran -fno-range-check  -fopenmp  -I../public -I../private
-I../../include -Wunused-value -Werror -W -fimplicit-none   -O2 -c
-ffree-form ../public/mtx_lib.f
ar crs libmtx.a mtx_def.o mod_mkl_pardiso_stub.o mtx_support.o mod_klu.o
thomas_block.o thomas_block_sparse.o d_and_c_block.o mtx_lib.o
ranlib: file: libmtx.a(mtx_def.o) has no symbols

. . .

cc -O3 -fno-common -fexceptions -c -I../klu_src ../klu_src/klu_c_fortran.c
-o klu_c_fortran.o
../klu_src/klu_c_fortran.c: In function ‘c_fortran_klu_’:
../klu_src/klu_c_fortran.c:45: warning: cast to pointer from integer of
different size
../klu_src/klu_c_fortran.c:52: warning: cast from pointer to integer of
different size

. . .

gfortran -fopenmp -o ../tester test_pardiso_stub.o test_klu.o
test_mtx_support.o test_banded.o test_block_tridiagonal.o test_square.o
test_sparse.o test_mtx.o  \
-L../../make -lmtx -lmesaklu `mesasdk_lapack_link` `mesasdk_blas_link`
-L../../../lib -lutils -lalert
/bin/sh: mesasdk_lapack_link: command not found
/bin/sh: mesasdk_blas_link: command not found
Undefined symbols:
  "_dgbcon_", referenced from:
      ___mtx_lib_MOD_mtx_rcond_banded in libmtx.a(mtx_lib.o)
  "_dger_", referenced from:
      ___thomas_block_MOD_thomas_factor in libmtx.a(thomas_block.o)

(the output then goes on to list another dozen or so undefined symbols.)

Any ideas?


Cheers,
Chris

[mesa-users] New release of the MESA SDK

[Richard T. <tow...@as...>]

Hi folks --

I've just made a new release of the MESA SDK. Important changes:

*) The Linux versions (i686 and x86_64) are now built on virtual Fedora Core 6 distributions. This should eliminate the "version 'GLIBC_2.11' not found (required by gfortran)" errors some people were reporting.

*) The SDK is now installed in a fixed location, /opt/mesasdk. THIS REQUIRES SOME KIND OF ROOT USER PRIVILEGES! Read through the instructions on the webpage for the revised installation instructions, and some reasoning behind this change.

As usual, please let me know if you run into any difficulties.

cheers,

Rich

Webpage: http://www.astro.wisc.edu/~townsend/static.php?ref=mesasdk

Re: [mesa-users] 2012 MESA Summer School

[Francis T. <fx...@ma...>]

hi everyone,

a gentle reminder applications close february 1,
and there are a limited number of seats remaining.
apply soon!

fxt




> Dear MESA community,
> 
> The 2012 MESA Summer School is now accepting applications:
> 
> http://cococubed.asu.edu/mesa_school_website
> 
> This 2012 MESA Summer School is aimed at graduate students and
> postdoctoral researchers who are interested in using, developing,
> and sharing their research experiences with MESA. Interested faculty
> should contact one of the School Directors. Themes for the 2012
> MESA Summer School include pulsation, rotation, and astroseismology.
> 
> Enrollment will be limited to no more than 44 participants. 
> Applications close February 1, 2012, so apply now!
> 
> The School Directors

Re: [mesa-users] eos choice

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

Hi Jing,

There is no built-in option that I'm aware of to use a polytrope EoS in
MESA.  However, one can implement any EoS in MESA provided that it returns
all the necessary information.

See mesa/star/public/other_eos.f and subroutine extras_controls in
mesa/star/test/src/run_star_extras.f.

Aaron



On Fri, Jan 20, 2012 at 2:10 AM, Jing Luan <jin...@ca...> wrote:

> Hello,
>
> May I simply adopt a polytrope eos in mesa?
>
> Thank you :-)
>
> Sincerely,
> Jing
>
>
> ------------------------------------------------------------------------------
> Keep Your Developer Skills Current with LearnDevNow!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-d2d
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users
>

[mesa-users] eos choice

[Jing L. <jin...@ca...>]

Hello,

May I simply adopt a polytrope eos in mesa?

Thank you :-)

Sincerely,
Jing

Re: [mesa-users] MESA sdk interfering with Tioga installation?

[Kevin M. <km...@ph...>]

Ok - I'll be glad to test a new version once you've made it. 

I just successfully installed Tioga after removing the whole sdk, so it looks like it was something to do with it… somewhere.

On Jan 18, 2012, at 4:50 PM, Richard Townsend wrote:

> Hi Kevin --
> 
> OK, my bad -- your DYLD_LIBRARY_PATH looks fine.
> 
> I think this is one of those messy issues which arises when libraries are moved around from their original installation locations. On Linux, it can get tricky; on OS X, nightmarish. DYLD_LIBRARY_PATH is a workaround that sometimes works, and sometimes makes things worse.
> 
> Right now I'm building the fixed-location SDK, which won't have these problems. I plan to upload it later tonight -- I'll let you know.
> 
> cheers,
> 
> Rich
> 
> On Jan 18, 2012, at 6:44 PM, Kevin Moore wrote:
> 
>> Thanks for the response, Rich. All the required setup should be in my .profile. I've used the sdk several times on my mac and haven't had anything like this happen before.
>> 
>> echo $DYLD_LIBRARY_PATH gives
>> /Users/Kevin/mesa/utils/mesasdk/lib:/opt/intel/composerxe-2011.4.184/compiler/lib:/opt/intel/composerxe-2011.4.184/mkl/lib:/opt/intel/composerxe-2011.4.184/compiler/lib:/opt/intel/composerxe-2011.4.184/compiler/lib/intel64:/opt/intel/composerxe-2011.4.184/compiler/lib:/opt/intel/composerxe-2011.4.184/mkl/lib
>> 
>> I suppose I can try removing everything and see if that fixes it.
>> 
>> 
>> On Jan 18, 2012, at 3:53 PM, Richard Townsend wrote:
>> 
>>> Hi Kevin --
>>> 
>>> It looks like you've not sourced the mesasdk_init.sh (or mesasdk_init.csh) setup script. Your problem is symptomatic of the fact that the DYLD_LIBRARY_PATH environment has not been set (which the script takes care of for you).
>>> 
>>> This will soon become moot, because I'm migrating toward having the SDK installed in a fixed, non-varying location; this will clear up a number of issues people are experiencing. But for the moment, I advise you to check that the setup script is being properly sourced.
>>> 
>>> cheers,
>>> 
>>> Rich
>>> 
>>> On Jan 18, 2012, at 5:45 PM, Kevin Moore wrote:
>>> 
>>>> Hi all,
>>>> 
>>>> In my ongoing list of problems with MESA (this time on osx lion), I recently found that Tioga is no longer working and reinstalling it gives a weird error that seems to indicate that it has something to do with the mesa sdk:
>>>> 
>>>> gcc -I. -I/Users/Kevin/.rvm/rubies/ruby-1.8.7-p357/lib/ruby/1.8/i686-darwin11.2.0 -I/Users/Kevin/.rvm/rubies/ruby-1.8.7-p357/lib/ruby/1.8/i686-darwin11.2.0 -I/Users/Kevin/Desktop/tioga-1.14/ext/Dobjects/Dtable -I../../includes -I../Dvector/include -D_XOPEN_SOURCE -D_DARWIN_C_SOURCE -fno-common -g -O2  -fno-common -pipe -fno-common   -c dtable.c
>>>> dyld: Library not loaded: /Users/townsend/mesasdk_build/x86_64/darwin/current/lib/libmpc.2.dylib
>>>> Referenced from: /Users/Kevin/mesasdk/bin/../libexec/gcc/x86_64-apple-darwin10.8.0/4.7.0/cc1
>>>> Reason: image not found
>>>> gcc: internal compiler error: Trace/BPT trap: 5 (program cc1)
>>>> Please submit a full bug report,
>>>> with preprocessed source if appropriate.
>>>> See <http://gcc.gnu.org/bugs.html> for instructions.
>>>> make: *** [dtable.o] Error 4
>>>> 
>>>> For some reason it's looking for libmpc.2.dylib in /Users/townsend/mesasdk_build/x86_64/darwin/current/lib/, while it should be looking in /Users/Kevin/mesasdk/lib. I can't find anything in my environment variables with that path, and grepping around the sdk shows that there are a few files with this path in it:
>>>> 
>>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/cc1 matches
>>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/cc1plus matches
>>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/collect2 matches
>>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/f951 matches
>>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/install-tools/fixincl matches
>>>> ./x86_64-apple-darwin10.8.0/4.7.0/install-tools/mkheaders:  #prefix=/Users/townsend/mesasdk_build/x86_64/darwin/current
>>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/lto-wrapper matches
>>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/plugin/gengtype matches
>>>> 
>>>> I don't know how/when these are created, but is there a simple way to recreate these executables with the correct paths?
>>>> 
>>>> -Kevin
>>>> ------------------------------------------------------------------------------
>>>> Keep Your Developer Skills Current with LearnDevNow!
>>>> The most comprehensive online learning library for Microsoft developers
>>>> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
>>>> Metro Style Apps, more. Free future releases when you subscribe now!
>>>> http://p.sf.net/sfu/learndevnow-d2d_______________________________________________
>>>> mesa-users mailing list
>>>> mes...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/mesa-users
>>> 
>>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> Keep Your Developer Skills Current with LearnDevNow!
>> The most comprehensive online learning library for Microsoft developers
>> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
>> Metro Style Apps, more. Free future releases when you subscribe now!
>> http://p.sf.net/sfu/learndevnow-d2d
>> _______________________________________________
>> mesa-users mailing list
>> mes...@li...
>> https://lists.sourceforge.net/lists/listinfo/mesa-users
> 
> 

Re: [mesa-users] MESA sdk interfering with Tioga installation?

[Richard T. <tow...@as...>]

Hi Kevin --

OK, my bad -- your DYLD_LIBRARY_PATH looks fine.

I think this is one of those messy issues which arises when libraries are moved around from their original installation locations. On Linux, it can get tricky; on OS X, nightmarish. DYLD_LIBRARY_PATH is a workaround that sometimes works, and sometimes makes things worse.

Right now I'm building the fixed-location SDK, which won't have these problems. I plan to upload it later tonight -- I'll let you know.

cheers,

Rich

On Jan 18, 2012, at 6:44 PM, Kevin Moore wrote:

> Thanks for the response, Rich. All the required setup should be in my .profile. I've used the sdk several times on my mac and haven't had anything like this happen before.
> 
> echo $DYLD_LIBRARY_PATH gives
> /Users/Kevin/mesa/utils/mesasdk/lib:/opt/intel/composerxe-2011.4.184/compiler/lib:/opt/intel/composerxe-2011.4.184/mkl/lib:/opt/intel/composerxe-2011.4.184/compiler/lib:/opt/intel/composerxe-2011.4.184/compiler/lib/intel64:/opt/intel/composerxe-2011.4.184/compiler/lib:/opt/intel/composerxe-2011.4.184/mkl/lib
> 
> I suppose I can try removing everything and see if that fixes it.
> 
> 
> On Jan 18, 2012, at 3:53 PM, Richard Townsend wrote:
> 
>> Hi Kevin --
>> 
>> It looks like you've not sourced the mesasdk_init.sh (or mesasdk_init.csh) setup script. Your problem is symptomatic of the fact that the DYLD_LIBRARY_PATH environment has not been set (which the script takes care of for you).
>> 
>> This will soon become moot, because I'm migrating toward having the SDK installed in a fixed, non-varying location; this will clear up a number of issues people are experiencing. But for the moment, I advise you to check that the setup script is being properly sourced.
>> 
>> cheers,
>> 
>> Rich
>> 
>> On Jan 18, 2012, at 5:45 PM, Kevin Moore wrote:
>> 
>>> Hi all,
>>> 
>>> In my ongoing list of problems with MESA (this time on osx lion), I recently found that Tioga is no longer working and reinstalling it gives a weird error that seems to indicate that it has something to do with the mesa sdk:
>>> 
>>> gcc -I. -I/Users/Kevin/.rvm/rubies/ruby-1.8.7-p357/lib/ruby/1.8/i686-darwin11.2.0 -I/Users/Kevin/.rvm/rubies/ruby-1.8.7-p357/lib/ruby/1.8/i686-darwin11.2.0 -I/Users/Kevin/Desktop/tioga-1.14/ext/Dobjects/Dtable -I../../includes -I../Dvector/include -D_XOPEN_SOURCE -D_DARWIN_C_SOURCE -fno-common -g -O2  -fno-common -pipe -fno-common   -c dtable.c
>>> dyld: Library not loaded: /Users/townsend/mesasdk_build/x86_64/darwin/current/lib/libmpc.2.dylib
>>> Referenced from: /Users/Kevin/mesasdk/bin/../libexec/gcc/x86_64-apple-darwin10.8.0/4.7.0/cc1
>>> Reason: image not found
>>> gcc: internal compiler error: Trace/BPT trap: 5 (program cc1)
>>> Please submit a full bug report,
>>> with preprocessed source if appropriate.
>>> See <http://gcc.gnu.org/bugs.html> for instructions.
>>> make: *** [dtable.o] Error 4
>>> 
>>> For some reason it's looking for libmpc.2.dylib in /Users/townsend/mesasdk_build/x86_64/darwin/current/lib/, while it should be looking in /Users/Kevin/mesasdk/lib. I can't find anything in my environment variables with that path, and grepping around the sdk shows that there are a few files with this path in it:
>>> 
>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/cc1 matches
>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/cc1plus matches
>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/collect2 matches
>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/f951 matches
>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/install-tools/fixincl matches
>>> ./x86_64-apple-darwin10.8.0/4.7.0/install-tools/mkheaders:  #prefix=/Users/townsend/mesasdk_build/x86_64/darwin/current
>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/lto-wrapper matches
>>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/plugin/gengtype matches
>>> 
>>> I don't know how/when these are created, but is there a simple way to recreate these executables with the correct paths?
>>> 
>>> -Kevin
>>> ------------------------------------------------------------------------------
>>> Keep Your Developer Skills Current with LearnDevNow!
>>> The most comprehensive online learning library for Microsoft developers
>>> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
>>> Metro Style Apps, more. Free future releases when you subscribe now!
>>> http://p.sf.net/sfu/learndevnow-d2d_______________________________________________
>>> mesa-users mailing list
>>> mes...@li...
>>> https://lists.sourceforge.net/lists/listinfo/mesa-users
>> 
>> 
> 
> 
> ------------------------------------------------------------------------------
> Keep Your Developer Skills Current with LearnDevNow!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-d2d
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users

Re: [mesa-users] MESA sdk interfering with Tioga installation?

[Kevin M. <km...@ph...>]

Thanks for the response, Rich. All the required setup should be in my .profile. I've used the sdk several times on my mac and haven't had anything like this happen before.

echo $DYLD_LIBRARY_PATH gives
/Users/Kevin/mesa/utils/mesasdk/lib:/opt/intel/composerxe-2011.4.184/compiler/lib:/opt/intel/composerxe-2011.4.184/mkl/lib:/opt/intel/composerxe-2011.4.184/compiler/lib:/opt/intel/composerxe-2011.4.184/compiler/lib/intel64:/opt/intel/composerxe-2011.4.184/compiler/lib:/opt/intel/composerxe-2011.4.184/mkl/lib

I suppose I can try removing everything and see if that fixes it.


On Jan 18, 2012, at 3:53 PM, Richard Townsend wrote:

> Hi Kevin --
> 
> It looks like you've not sourced the mesasdk_init.sh (or mesasdk_init.csh) setup script. Your problem is symptomatic of the fact that the DYLD_LIBRARY_PATH environment has not been set (which the script takes care of for you).
> 
> This will soon become moot, because I'm migrating toward having the SDK installed in a fixed, non-varying location; this will clear up a number of issues people are experiencing. But for the moment, I advise you to check that the setup script is being properly sourced.
> 
> cheers,
> 
> Rich
> 
> On Jan 18, 2012, at 5:45 PM, Kevin Moore wrote:
> 
>> Hi all,
>> 
>> In my ongoing list of problems with MESA (this time on osx lion), I recently found that Tioga is no longer working and reinstalling it gives a weird error that seems to indicate that it has something to do with the mesa sdk:
>> 
>> gcc -I. -I/Users/Kevin/.rvm/rubies/ruby-1.8.7-p357/lib/ruby/1.8/i686-darwin11.2.0 -I/Users/Kevin/.rvm/rubies/ruby-1.8.7-p357/lib/ruby/1.8/i686-darwin11.2.0 -I/Users/Kevin/Desktop/tioga-1.14/ext/Dobjects/Dtable -I../../includes -I../Dvector/include -D_XOPEN_SOURCE -D_DARWIN_C_SOURCE -fno-common -g -O2  -fno-common -pipe -fno-common   -c dtable.c
>> dyld: Library not loaded: /Users/townsend/mesasdk_build/x86_64/darwin/current/lib/libmpc.2.dylib
>> Referenced from: /Users/Kevin/mesasdk/bin/../libexec/gcc/x86_64-apple-darwin10.8.0/4.7.0/cc1
>> Reason: image not found
>> gcc: internal compiler error: Trace/BPT trap: 5 (program cc1)
>> Please submit a full bug report,
>> with preprocessed source if appropriate.
>> See <http://gcc.gnu.org/bugs.html> for instructions.
>> make: *** [dtable.o] Error 4
>> 
>> For some reason it's looking for libmpc.2.dylib in /Users/townsend/mesasdk_build/x86_64/darwin/current/lib/, while it should be looking in /Users/Kevin/mesasdk/lib. I can't find anything in my environment variables with that path, and grepping around the sdk shows that there are a few files with this path in it:
>> 
>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/cc1 matches
>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/cc1plus matches
>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/collect2 matches
>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/f951 matches
>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/install-tools/fixincl matches
>> ./x86_64-apple-darwin10.8.0/4.7.0/install-tools/mkheaders:  #prefix=/Users/townsend/mesasdk_build/x86_64/darwin/current
>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/lto-wrapper matches
>> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/plugin/gengtype matches
>> 
>> I don't know how/when these are created, but is there a simple way to recreate these executables with the correct paths?
>> 
>> -Kevin
>> ------------------------------------------------------------------------------
>> Keep Your Developer Skills Current with LearnDevNow!
>> The most comprehensive online learning library for Microsoft developers
>> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
>> Metro Style Apps, more. Free future releases when you subscribe now!
>> http://p.sf.net/sfu/learndevnow-d2d_______________________________________________
>> mesa-users mailing list
>> mes...@li...
>> https://lists.sourceforge.net/lists/listinfo/mesa-users
> 
> 

Re: [mesa-users] MESA sdk interfering with Tioga installation?

[Richard T. <tow...@as...>]

Hi Kevin --

It looks like you've not sourced the mesasdk_init.sh (or mesasdk_init.csh) setup script. Your problem is symptomatic of the fact that the DYLD_LIBRARY_PATH environment has not been set (which the script takes care of for you).

This will soon become moot, because I'm migrating toward having the SDK installed in a fixed, non-varying location; this will clear up a number of issues people are experiencing. But for the moment, I advise you to check that the setup script is being properly sourced.

cheers,

Rich

On Jan 18, 2012, at 5:45 PM, Kevin Moore wrote:

> Hi all,
> 
> In my ongoing list of problems with MESA (this time on osx lion), I recently found that Tioga is no longer working and reinstalling it gives a weird error that seems to indicate that it has something to do with the mesa sdk:
> 
> gcc -I. -I/Users/Kevin/.rvm/rubies/ruby-1.8.7-p357/lib/ruby/1.8/i686-darwin11.2.0 -I/Users/Kevin/.rvm/rubies/ruby-1.8.7-p357/lib/ruby/1.8/i686-darwin11.2.0 -I/Users/Kevin/Desktop/tioga-1.14/ext/Dobjects/Dtable -I../../includes -I../Dvector/include -D_XOPEN_SOURCE -D_DARWIN_C_SOURCE -fno-common -g -O2  -fno-common -pipe -fno-common   -c dtable.c
> dyld: Library not loaded: /Users/townsend/mesasdk_build/x86_64/darwin/current/lib/libmpc.2.dylib
>  Referenced from: /Users/Kevin/mesasdk/bin/../libexec/gcc/x86_64-apple-darwin10.8.0/4.7.0/cc1
>  Reason: image not found
> gcc: internal compiler error: Trace/BPT trap: 5 (program cc1)
> Please submit a full bug report,
> with preprocessed source if appropriate.
> See <http://gcc.gnu.org/bugs.html> for instructions.
> make: *** [dtable.o] Error 4
> 
> For some reason it's looking for libmpc.2.dylib in /Users/townsend/mesasdk_build/x86_64/darwin/current/lib/, while it should be looking in /Users/Kevin/mesasdk/lib. I can't find anything in my environment variables with that path, and grepping around the sdk shows that there are a few files with this path in it:
> 
> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/cc1 matches
> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/cc1plus matches
> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/collect2 matches
> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/f951 matches
> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/install-tools/fixincl matches
> ./x86_64-apple-darwin10.8.0/4.7.0/install-tools/mkheaders:  #prefix=/Users/townsend/mesasdk_build/x86_64/darwin/current
> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/lto-wrapper matches
> Binary file ./x86_64-apple-darwin10.8.0/4.7.0/plugin/gengtype matches
> 
> I don't know how/when these are created, but is there a simple way to recreate these executables with the correct paths?
> 
> -Kevin
> ------------------------------------------------------------------------------
> Keep Your Developer Skills Current with LearnDevNow!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-d2d_______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users

[mesa-users] MESA sdk interfering with Tioga installation?

[Kevin M. <km...@ph...>]

Hi all,

In my ongoing list of problems with MESA (this time on osx lion), I recently found that Tioga is no longer working and reinstalling it gives a weird error that seems to indicate that it has something to do with the mesa sdk:

gcc -I. -I/Users/Kevin/.rvm/rubies/ruby-1.8.7-p357/lib/ruby/1.8/i686-darwin11.2.0 -I/Users/Kevin/.rvm/rubies/ruby-1.8.7-p357/lib/ruby/1.8/i686-darwin11.2.0 -I/Users/Kevin/Desktop/tioga-1.14/ext/Dobjects/Dtable -I../../includes -I../Dvector/include -D_XOPEN_SOURCE -D_DARWIN_C_SOURCE -fno-common -g -O2  -fno-common -pipe -fno-common   -c dtable.c
dyld: Library not loaded: /Users/townsend/mesasdk_build/x86_64/darwin/current/lib/libmpc.2.dylib
 Referenced from: /Users/Kevin/mesasdk/bin/../libexec/gcc/x86_64-apple-darwin10.8.0/4.7.0/cc1
 Reason: image not found
gcc: internal compiler error: Trace/BPT trap: 5 (program cc1)
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
make: *** [dtable.o] Error 4

For some reason it's looking for libmpc.2.dylib in /Users/townsend/mesasdk_build/x86_64/darwin/current/lib/, while it should be looking in /Users/Kevin/mesasdk/lib. I can't find anything in my environment variables with that path, and grepping around the sdk shows that there are a few files with this path in it:

Binary file ./x86_64-apple-darwin10.8.0/4.7.0/cc1 matches
Binary file ./x86_64-apple-darwin10.8.0/4.7.0/cc1plus matches
Binary file ./x86_64-apple-darwin10.8.0/4.7.0/collect2 matches
Binary file ./x86_64-apple-darwin10.8.0/4.7.0/f951 matches
Binary file ./x86_64-apple-darwin10.8.0/4.7.0/install-tools/fixincl matches
./x86_64-apple-darwin10.8.0/4.7.0/install-tools/mkheaders:  #prefix=/Users/townsend/mesasdk_build/x86_64/darwin/current
Binary file ./x86_64-apple-darwin10.8.0/4.7.0/lto-wrapper matches
Binary file ./x86_64-apple-darwin10.8.0/4.7.0/plugin/gengtype matches

I don't know how/when these are created, but is there a simple way to recreate these executables with the correct paths?

-Kevin

Re: [mesa-users] error in project compiling PROBLEM SOLVED

[<umb...@un...>]

Problem solved!!!!

Quoting umb...@un...:

> Hi all,
> I'm trying to compile simple project.. and actually is not the first
> time.. as usual I typed ./clean and ./mk in my work directory. After
> that the following message is displayed:
>
> gfortran -fno-range-check  -fopenmp -fbounds-check -g
> -I/home/umberto/mesa/include -I/home/umberto/mesasdk/include -c
> -ffree-form ../src/run_star_extras.f
> make: *** No rule to make target `run_star.o', needed by
> `star_solar_life'.  Stop.
>
> I don't know why it fails in making run_star.o, since it did not have
> problem few days ago in doing that. And I also think is something very
> simple, but despite that I'm still confused.. can you help me? Thanks
> very much!
>
>   Umberto Battino
>
> ----------------------------------------------------------------
> This message was sent using IMP, the Internet Messaging Program.
>
>
>
> ------------------------------------------------------------------------------
> Keep Your Developer Skills Current with LearnDevNow!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-d2d
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users
>



----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

[mesa-users] error in project compiling

[<umb...@un...>]

Hi all,
I'm trying to compile simple project.. and actually is not the first  
time.. as usual I typed ./clean and ./mk in my work directory. After  
that the following message is displayed:

gfortran -fno-range-check  -fopenmp -fbounds-check -g  
-I/home/umberto/mesa/include -I/home/umberto/mesasdk/include -c  
-ffree-form ../src/run_star_extras.f
make: *** No rule to make target `run_star.o', needed by  
`star_solar_life'.  Stop.

I don't know why it fails in making run_star.o, since it did not have  
problem few days ago in doing that. And I also think is something very  
simple, but despite that I'm still confused.. can you help me? Thanks  
very much!

  Umberto Battino

----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

[mesa-users] Seeking community input on MESA's web presence

[Alfred G. <al...@ga...>]

On one side, I personally find the current mailing-list system
informative and helpful; on the other hand, it is terrible to keep track
of a particular topic the mailing-list. At least to me, it is a pain in
the neck or even deterrent to try to follow a thread that develops over
a number of days or a series of back-and-forth exchanges between
parties. Things get particularly bad if all the previous levels of
discussions are attached to the messages and/or links to useful
informations are skipped or get lost. 

The best illustration of the above whining is how I have to answer to
the current thread, which really takes some will to contribute: I am
forced to write the mail in an external application, copy&paste the
thread title to the "concern" line, and send it to the mailing-list
rather than being able to just click on something like "Post Reply"
which opens a submission window (as it frequently/usually happens in a
forum) and lets me contribute to the topic at hand. 
[Somehow, I never get past the "mesa-users *(read-only)*" status in the
email archive; if somebody manages to interact in a more sophisticated
way, one of the elderly-ones (aka the writer) would appreciate to learn
about it.]  

Resorting to a forum (such as supported in sourceforge and ready but
essentially unused in the case of MESA) would at least eliminate the
problem of fogging-in the evolution and hopefully the resolution of a
particular thread on whatever topic concerning MESA. If each thread
would get a proper title, the temptation to repeatedly chew on a
particular topic, or open a knew thread each time, would be considerably
reduced. Furthermore, since forums also have the possibility of fulltext
searches it would be rather easy to find (or not) topics fitting user
search-words. I think that a forum could essentially serve most needs to
help users and to motivate users to contribute rather painlessly while
maintaining a decently manageable order in the whole collection.

As an example, to me one of the best expriences with forums is that
associated with the Yorick project: See e.g.

http://yorick.sourceforge.net/phpBB3/

It is simple enough to let me find my way with ease even when visiting
it only casually; it is comprehensive enough to have served essentially
all the needs I ever had.

A digest of the forum activity might still be mailed off to mailing-list
subscribers to satisfy those who want to be actively pushed towards MESA
activities; or at least being informed on topics of current concern. 


Running an elementary FAQ on the *MESA webpage* itself would certainly
be helpful, in particular when dealing with elementary installation and
start-up issues for those who consider earnestly to jump this
bandwagon. 


IMO, a MESA manual for educational purposes, which was discussed
recently, would probably be best served via a Wiki installation with a
diligent, critical, but supportive curator in the background. The wiki
could be open to all users who have applied to get access to the forum
(i.e. to registered users) - hence, one could probably get rid of much
potential misuse that way already. However, it appears to me that MESA
is still changing so rapidly, and these changes are not easily visible
to the common user since they appear to happen along multiple fronts and
sometimes deep down in the code, that it will be challenging to maintain
an up-to-date manual at the present stage of the development. 

Alfred

Re: [mesa-users] Seeking community input on MESA's web presence

[Conny A. <co...@st...>]

Hi MESAs,

while not an active daily user of MESA, I follow the email discussions 
with great interest (even the mails on the technicalities are much more 
interesting than SPAM or boring admin mails), something that would 
certainly not happen if I were to be obliged to go and search in 
documents or on a Twiki-type info channel. FAQ lists are always useful. 
But I am not the main customer you should aim at, so the intensive users 
should have the decisions in hand.

Best regards,
Conny

-- 
--------------------------------------------------------
Conny Aerts                         tel: +32/16/32 70 28
Director                            fax: +32/16/32 79 99
Instituut voor Sterrenkunde
K.U.Leuven
Celestijnenlaan 200D03.29
3001 Leuven                e-mail:co...@st...
Belgium                      http://fys.kuleuven.be/ster
--------------------------------------------------------
Professor of Asteroseismology
Afdeling Sterrenkunde, IMAPP        tel: +31/24/365 2804
Radboud University Nijmegen         fax: +31/24/365 2807
Heyendaalseweg 135              e-mail:ae...@as...
6525 AJ Nijmegen                  http://www.astro.ru.nl
The Netherlands



On 01/17/2012 02:51 AM, Aaron Dotter wrote:
> Dear MESA community,
>
> In an effort to improve the way in which we obtain help from, and 
> share information with, other MESA users, we are seeking your input.
>
> The current standard relies entirely on the webpage and the mesa-users 
> email list.  This arrangement is not ideal as many new users have the 
> same basic questions and the email list experiences many repeated 
> questions.  Much of the information currently provided via email needs 
> to be preserved in an organized, searchable format.
>
> We have access to a community forum through sourceforge.net 
> <http://sourceforge.net> and could use it for some things, such as 
> creating an FAQ for new users.  The forum structure does not meet all 
> the needs that we can envision, and something along the lines of a 
> wiki/plone could serve the community better.  However, to do so 
> requires moving out of the sourceforge domain.
>
> Please respond to the following queries:
>
> 1.  What sort of MESA web presence would be most useful to you?
>
>
> 2.  If you've had experience with this sort of thing in the past, what 
> would you recommend that we do (or not do)?
>
>
>
> Thank you for your input!
> Aaron / MESA Council
>
>
> ------------------------------------------------------------------------------
> Keep Your Developer Skills Current with LearnDevNow!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-d2d
>
>
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users

[mesa-users] Seeking community input on MESA's web presence

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

Dear MESA community,

In an effort to improve the way in which we obtain help from, and share
information with, other MESA users, we are seeking your input.

The current standard relies entirely on the webpage and the mesa-users
email list.  This arrangement is not ideal as many new users have the same
basic questions and the email list experiences many repeated questions.
 Much of the information currently provided via email needs to be preserved
in an organized, searchable format.

We have access to a community forum through sourceforge.net and could use
it for some things, such as creating an FAQ for new users.  The forum
structure does not meet all the needs that we can envision, and something
along the lines of a wiki/plone could serve the community better.  However,
to do so requires moving out of the sourceforge domain.

Please respond to the following queries:

1.  What sort of MESA web presence would be most useful to you?


2.  If you've had experience with this sort of thing in the past, what
would you recommend that we do (or not do)?



Thank you for your input!
Aaron / MESA Council

Re: [mesa-users] Help, error compiling mesa on 64-bit Ubuntu

[Matt C. <mco...@ph...>]

Hi Richard,

Thank you, running "export LIBRARY_PATH=/usr/lib/x86_64-linux-gnu" fixed
that problem and I was able to install MESA using the mesasdk.

On Sun, Jan 15, 2012 at 6:39 AM, Richard Townsend
<tow...@as...>wrote:

> Hi Matt --
>
> Before compiling with the SDK, could you try running this command:
>
> export LIBRARY_PATH=/usr/lib/x86_64-linux-gnu
>
> (for bash), or
>
> setenv LIBRARY_PATH /usr/lib/x86_64-linux-gnu
>
> (for tcsh). This should fix the issue with the missing crt1.o etc. If it
> works, you can add it to the appropriate login scripts. FYI, this is an
> ongoing bug with
>
> cheers,
>
> Rich
>
>
> On Jan 14, 2012, at 9:53 PM, Richard Townsend wrote:
>
> > Hi Matt --
> >
> > This is a known problem with the SDK, which I'm still trying to rectify.
> >
> > cheers,
> >
> > Rich
> >
> > On Jan 14, 2012, at 7:58 PM, Matt Coleman wrote:
> >
> >> Hi MESA users,
> >>
> >> I have been having problems installing MESA on 64-bit Ubuntu, both when
> I compile it using mesasdk and without. Any help would be greatly
> appreciated.
> >>
> >>
> >> With mesasdk I get the following:
> >>
> >> checking if ndiff is installed ...
> >> yes!
> >>
> >> /home/mcoleman/mesa/const
> >> building const package.
> >>
> >> make: `libconst.a' is up to date.
> >> make: `libconst.a' is up to date.
> >> gfortran -fopenmp -o ../tester test_const.o -L../../make -lconst
> >> /usr/bin/ld: cannot find crt1.o: No such file or directory
> >> /usr/bin/ld: cannot find crti.o: No such file or directory
> >> collect2: error: ld returned 1 exit status
> >> make: *** [tester] Error 1
> >>
> >>
> ------------------------------------------------------------------------------------------
> >>
> >> Compiling directly (without mesasdk) I get to module 'num' then it
> fails with the following output:
> >>
> >> /home/mcoleman/mesa/num
> >> building num package.
> >>
> >> gfortran -fno-range-check -fopenmp -I../public -I../private
> -I../../include -Wunused-value -Werror -W -fimplicit-none -O2 -c
> -ffixed-form -ffixed-line-length-132 ../private/mod_root.f
> >> ../private/mod_root.f:38.132:
> >>
> >>
> >> 1
> >> Fatal Error: Wrong module version '7' (expected '6') for file
> 'const_def.mod' opened at (1)
> >> make: *** [mod_root.o] Error 1
> >>
> >> /home/mcoleman/mesa/num/make
> >> FAILED
> >>
> >>
> >> Thanks,
> >>
> >> --
> >> - Matt
> >>
> ------------------------------------------------------------------------------
> >> RSA(R) Conference 2012
> >> Mar 27 - Feb 2
> >> Save $400 by Jan. 27
> >> Register now!
> >>
> http://p.sf.net/sfu/rsa-sfdev2dev2_______________________________________________
> >> mesa-users mailing list
> >> mes...@li...
> >> https://lists.sourceforge.net/lists/listinfo/mesa-users
> >
> >
> >
> ------------------------------------------------------------------------------
> > RSA(R) Conference 2012
> > Mar 27 - Feb 2
> > Save $400 by Jan. 27
> > Register now!
> > http://p.sf.net/sfu/rsa-sfdev2dev2
> > _______________________________________________
> > mesa-users mailing list
> > mes...@li...
> > https://lists.sourceforge.net/lists/listinfo/mesa-users
>
>
>


-- 
- Matt

Re: [mesa-users] Help, error compiling mesa on 64-bit Ubuntu

[Richard T. <tow...@as...>]

Hi Matt --

Before compiling with the SDK, could you try running this command:

export LIBRARY_PATH=/usr/lib/x86_64-linux-gnu

(for bash), or

setenv LIBRARY_PATH /usr/lib/x86_64-linux-gnu 

(for tcsh). This should fix the issue with the missing crt1.o etc. If it works, you can add it to the appropriate login scripts. FYI, this is an ongoing bug with 

cheers,

Rich


On Jan 14, 2012, at 9:53 PM, Richard Townsend wrote:

> Hi Matt --
> 
> This is a known problem with the SDK, which I'm still trying to rectify.
> 
> cheers,
> 
> Rich
> 
> On Jan 14, 2012, at 7:58 PM, Matt Coleman wrote:
> 
>> Hi MESA users,
>> 
>> I have been having problems installing MESA on 64-bit Ubuntu, both when I compile it using mesasdk and without. Any help would be greatly appreciated.
>> 
>> 
>> With mesasdk I get the following:
>> 
>> checking if ndiff is installed ...
>> yes!
>> 
>> /home/mcoleman/mesa/const
>> building const package.
>> 
>> make: `libconst.a' is up to date.
>> make: `libconst.a' is up to date.
>> gfortran -fopenmp -o ../tester test_const.o -L../../make -lconst
>> /usr/bin/ld: cannot find crt1.o: No such file or directory
>> /usr/bin/ld: cannot find crti.o: No such file or directory
>> collect2: error: ld returned 1 exit status
>> make: *** [tester] Error 1
>> 
>> ------------------------------------------------------------------------------------------
>> 
>> Compiling directly (without mesasdk) I get to module 'num' then it fails with the following output:
>> 
>> /home/mcoleman/mesa/num
>> building num package.
>> 
>> gfortran -fno-range-check -fopenmp -I../public -I../private -I../../include -Wunused-value -Werror -W -fimplicit-none -O2 -c -ffixed-form -ffixed-line-length-132 ../private/mod_root.f
>> ../private/mod_root.f:38.132:
>> 
>> 
>> 1 
>> Fatal Error: Wrong module version '7' (expected '6') for file 'const_def.mod' opened at (1)
>> make: *** [mod_root.o] Error 1
>> 
>> /home/mcoleman/mesa/num/make
>> FAILED
>> 
>> 
>> Thanks,
>> 
>> --
>> - Matt
>> ------------------------------------------------------------------------------
>> RSA(R) Conference 2012
>> Mar 27 - Feb 2
>> Save $400 by Jan. 27
>> Register now!
>> http://p.sf.net/sfu/rsa-sfdev2dev2_______________________________________________
>> mesa-users mailing list
>> mes...@li...
>> https://lists.sourceforge.net/lists/listinfo/mesa-users
> 
> 
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Mar 27 - Feb 2
> Save $400 by Jan. 27
> Register now!
> http://p.sf.net/sfu/rsa-sfdev2dev2
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users

Re: [mesa-users] bad CPU in executable

[Richard T. <tow...@as...>]

Hi Anders --

It was my understanding that 10.4 could still run 64-bit executables, even though it runs in 32-bit mode. However, that doesn't appear to be the case -- your computer doesn't understand how to run the code in the SDK, which is 64-bit.

I'm afraid you may have to upgrade to a more-recent version of OS X.

cheers,

Rich

On Jan 13, 2012, at 11:14 PM, andres suarez wrote:

> Sorry, I did mean Tiger...
> 
> uname -m
> i386
> 
> ____________________
> a.
> 
> 
> On Fri, Jan 13, 2012 at 9:10 PM, Richard Townsend <tow...@as...> wrote:
> Hi Andres -
> 
> Do you mean OS X Tiger? Because Lion is 10.7, not 10.4.
> 
> In any case, can you send me the output from 'uname -m'
> 
> Cheers,
> 
> Rich
> 
> 
> On Jan 13, 2012, at 10:55 PM, andres suarez <asu...@uc...> wrote:
> 
>>> Hi,
>>> I am trying to install mesa on a MacPro 1.1 running MAC OS 10.4.11
>>> When I try to use the mesasdk package for Darwin, the gfortran and gcc compilers provided seem to not be supported by my system, as I get the errors:
>>> 
>>>        -bash ./gfortran: Bad CPU type in executable
>>>        -bash: ./gcc: Bad CPU type in executable
>>> 
>>> I tried getting the gfortran appropriate to my system, but I cannot find anything beyond gfortran4.2 that runs with Lion Mac OS (10.4), and when trying to install without mesasdk, I noticed that mesa explicitly requires a gfortran equal or higher to 4.5.
>>> 
>>> so, anyone can tell me if there is a gfortran4.5 that runs in Mac OS 10.4 and where to find it?
>>> Is there a workaround?
>>> 
>>> Thank you.
>>> a.
>> ------------------------------------------------------------------------------
>> RSA(R) Conference 2012
>> Mar 27 - Feb 2
>> Save $400 by Jan. 27
>> Register now!
>> http://p.sf.net/sfu/rsa-sfdev2dev2
>> _______________________________________________
>> mesa-users mailing list
>> mes...@li...
>> https://lists.sourceforge.net/lists/listinfo/mesa-users
> 

Re: [mesa-users] Help, error compiling mesa on 64-bit Ubuntu

[Francis T. <fx...@ma...>]

for the non-sdk attempt:

this error message indicates that your module "mod_root" has been 
compiled with a different compiler version than the one you're using 
to compile const_def.mod. please make sure you execute the "./clean"
script at the top level of mesa. alternatively, put new versions of
mesa in new directories - this will ensure only one compiler does all the work.

fxt





On Jan 14, 2012, at 8:53 PM, Richard Townsend wrote:

> Hi Matt --
> 
> This is a known problem with the SDK, which I'm still trying to rectify.
> 
> cheers,
> 
> Rich
> 
> On Jan 14, 2012, at 7:58 PM, Matt Coleman wrote:
> 
>> Hi MESA users,
>> 
>> I have been having problems installing MESA on 64-bit Ubuntu, both when I compile it using mesasdk and without. Any help would be greatly appreciated.
>> 
>> 
>> With mesasdk I get the following:
>> 
>> checking if ndiff is installed ...
>> yes!
>> 
>> /home/mcoleman/mesa/const
>> building const package.
>> 
>> make: `libconst.a' is up to date.
>> make: `libconst.a' is up to date.
>> gfortran -fopenmp -o ../tester test_const.o -L../../make -lconst
>> /usr/bin/ld: cannot find crt1.o: No such file or directory
>> /usr/bin/ld: cannot find crti.o: No such file or directory
>> collect2: error: ld returned 1 exit status
>> make: *** [tester] Error 1
>> 
>> ------------------------------------------------------------------------------------------
>> 
>> Compiling directly (without mesasdk) I get to module 'num' then it fails with the following output:
>> 
>> /home/mcoleman/mesa/num
>> building num package.
>> 
>> gfortran -fno-range-check -fopenmp -I../public -I../private -I../../include -Wunused-value -Werror -W -fimplicit-none -O2 -c -ffixed-form -ffixed-line-length-132 ../private/mod_root.f
>> ../private/mod_root.f:38.132:
>> 
>> 
>> 1 
>> Fatal Error: Wrong module version '7' (expected '6') for file 'const_def.mod' opened at (1)
>> make: *** [mod_root.o] Error 1
>> 
>> /home/mcoleman/mesa/num/make
>> FAILED
>> 
>> 
>> Thanks,
>> 
>> --
>> - Matt
>> ------------------------------------------------------------------------------
>> RSA(R) Conference 2012
>> Mar 27 - Feb 2
>> Save $400 by Jan. 27
>> Register now!
>> http://p.sf.net/sfu/rsa-sfdev2dev2_______________________________________________
>> mesa-users mailing list
>> mes...@li...
>> https://lists.sourceforge.net/lists/listinfo/mesa-users
> 
> 
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Mar 27 - Feb 2
> Save $400 by Jan. 27
> Register now!
> http://p.sf.net/sfu/rsa-sfdev2dev2
> _______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users

Re: [mesa-users] error compiling MESA rev 3851: bad value for which_atm_option

[Francis T. <fx...@ma...>]

i suggest downloading the latest version of mesa into a new directory
and trying again. let me know what happens.

fxt



On Jan 14, 2012, at 8:57 PM, Jean-Claude Passy wrote:

> Hi all,
> 
> I am having trouble installing the latest revision. The error occurs when building the atm package:
> 
> ---------------------------------------------------------------------
> /Users/jean-claudepassy/Work/MESA/mesa/atm
> building atm package.
> 
> make: `libatm.a' is up to date.
> make: `libatm.a' is up to date.
> ifort -openmp -threads -o ../tester test_atm_support.o test_atm.o -L../../make -latm -L../../../lib -lmlt -leos -lkap -lnet -lscreen -lrates -lreaclib -lweak -lneu -lchem -linterp_2d -linterp_1d -lnum -lutils -lalert -lconst -lmtx -lmesaklu -lmesalapack -lmesablas
> ifort -openmp -threads -o ../test_quietly test_atm_support.o test_atm_quietly.o -L../../make -latm -L../../../lib -lmlt -leos -lkap -lnet -lscreen -lrates -lreaclib -lweak -lneu -lchem -linterp_2d -linterp_1d -lnum -lutils -lalert -lconst -lmtx -lmesaklu -lmesalapack -lmesablas
> ifort -openmp -threads -o ../plotter test_atm_support.o plot_atm.o -L../../make -latm -L../../../lib -lmlt -leos -lkap -lnet -lscreen -lrates -lreaclib -lweak -lneu -lchem -linterp_2d -linterp_1d -lnum -lutils -lalert -lconst -lmtx -lmesaklu -lmesalapack -lmesablas
>  bad value for which_atm_option          11
>  bad value for which_atm_option           7
> 
> 3c3
> <  bad value for which_atm_option          11
> --- field 5
> ...
> ---------------------------------------------------------------------
> 
> Any help much appreciated!
> Thanks a lot,
> 
> JC
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Mar 27 - Feb 2
> Save $400 by Jan. 27
> Register now!
> http://p.sf.net/sfu/rsa-sfdev2dev2_______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users

Re: [mesa-users] Help, error compiling mesa on 64-bit Ubuntu

[Richard T. <tow...@as...>]

Hi Matt --

This is a known problem with the SDK, which I'm still trying to rectify.

cheers,

Rich

On Jan 14, 2012, at 7:58 PM, Matt Coleman wrote:

> Hi MESA users,
> 
> I have been having problems installing MESA on 64-bit Ubuntu, both when I compile it using mesasdk and without. Any help would be greatly appreciated.
> 
> 
> With mesasdk I get the following:
> 
> checking if ndiff is installed ...
> yes!
> 
> /home/mcoleman/mesa/const
> building const package.
> 
> make: `libconst.a' is up to date.
> make: `libconst.a' is up to date.
> gfortran -fopenmp -o ../tester test_const.o -L../../make -lconst
> /usr/bin/ld: cannot find crt1.o: No such file or directory
> /usr/bin/ld: cannot find crti.o: No such file or directory
> collect2: error: ld returned 1 exit status
> make: *** [tester] Error 1
> 
> ------------------------------------------------------------------------------------------
> 
> Compiling directly (without mesasdk) I get to module 'num' then it fails with the following output:
> 
> /home/mcoleman/mesa/num
> building num package.
> 
> gfortran -fno-range-check -fopenmp -I../public -I../private -I../../include -Wunused-value -Werror -W -fimplicit-none -O2 -c -ffixed-form -ffixed-line-length-132 ../private/mod_root.f
> ../private/mod_root.f:38.132:
> 
> 
> 1 
> Fatal Error: Wrong module version '7' (expected '6') for file 'const_def.mod' opened at (1)
> make: *** [mod_root.o] Error 1
> 
> /home/mcoleman/mesa/num/make
> FAILED
> 
> 
> Thanks,
> 
> --
> - Matt
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Mar 27 - Feb 2
> Save $400 by Jan. 27
> Register now!
> http://p.sf.net/sfu/rsa-sfdev2dev2_______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users

[mesa-users] error compiling MESA rev 3851: bad value for which_atm_option

[Jean-Claude P. <jc...@gm...>]

Hi all,

I am having trouble installing the latest revision. The error occurs 
when building the atm package:

---------------------------------------------------------------------

/Users/jean-claudepassy/Work/MESA/mesa/atm

building atm package.


make: `libatm.a' is up to date.

make: `libatm.a' is up to date.

ifort -openmp -threads -o ../tester test_atm_support.o test_atm.o 
-L../../make -latm -L../../../lib -lmlt -leos -lkap -lnet -lscreen 
-lrates -lreaclib -lweak -lneu -lchem -linterp_2d -linterp_1d -lnum 
-lutils -lalert -lconst -lmtx -lmesaklu -lmesalapack -lmesablas

ifort -openmp -threads -o ../test_quietly test_atm_support.o 
test_atm_quietly.o -L../../make -latm -L../../../lib -lmlt -leos -lkap 
-lnet -lscreen -lrates -lreaclib -lweak -lneu -lchem -linterp_2d 
-linterp_1d -lnum -lutils -lalert -lconst -lmtx -lmesaklu -lmesalapack 
-lmesablas

ifort -openmp -threads -o ../plotter test_atm_support.o plot_atm.o 
-L../../make -latm -L../../../lib -lmlt -leos -lkap -lnet -lscreen 
-lrates -lreaclib -lweak -lneu -lchem -linterp_2d -linterp_1d -lnum 
-lutils -lalert -lconst -lmtx -lmesaklu -lmesalapack -lmesablas

  bad value for which_atm_option          11

  bad value for which_atm_option           7


3c3

<  bad value for which_atm_option          11

--- field 5

...

---------------------------------------------------------------------


Any help much appreciated!
Thanks a lot,

JC

[mesa-users] Help, error compiling mesa on 64-bit Ubuntu

[Matt C. <mco...@ph...>]

Hi MESA users,

I have been having problems installing MESA on 64-bit Ubuntu, both when I
compile it using mesasdk and without. Any help would be greatly appreciated.


With mesasdk I get the following:

checking if ndiff is installed ...
yes!

/home/mcoleman/mesa/const
building const package.

make: `libconst.a' is up to date.
make: `libconst.a' is up to date.
gfortran -fopenmp -o ../tester test_const.o -L../../make -lconst
/usr/bin/ld: cannot find crt1.o: No such file or directory
/usr/bin/ld: cannot find crti.o: No such file or directory
collect2: error: ld returned 1 exit status
make: *** [tester] Error 1

------------------------------------------------------------------------------------------

Compiling directly (without mesasdk) I get to module 'num' then it fails
with the following output:

/home/mcoleman/mesa/num
building num package.

gfortran -fno-range-check -fopenmp -I../public -I../private -I../../include
-Wunused-value -Werror -W -fimplicit-none -O2 -c -ffixed-form
-ffixed-line-length-132 ../private/mod_root.f
../private/mod_root.f:38.132:


1
Fatal Error: Wrong module version '7' (expected '6') for file
'const_def.mod' opened at (1)
make: *** [mod_root.o] Error 1

/home/mcoleman/mesa/num/make
FAILED


Thanks,

--
- Matt

Re: [mesa-users] MESA summer school and MESA education

[Francis T. <fx...@ma...>]

hi dave,

this is a good idea. time resources have been a limiting factor 
to producing a quality users manual, but maybe this could be overcome 
if said manual became partly or wholly a community driven project
that starts from a summer school generated seed. 

fxt



On Jan 14, 2012, at 12:10 PM, david arnett wrote:

> Hi Frank,
> Why not make a MESA "intro manual" as a spin-off of the exercises in the summer school? With such  a base it could become a community project. That might be feasible if the work was shared with new and eager MESAns.
> Dave
> 
> -- 
> David Arnett
> Regents Professor
> Steward Observatory
> University of Arizona
> ------------------------------------------------------------------------------
> RSA(R) Conference 2012
> Mar 27 - Feb 2
> Save $400 by Jan. 27
> Register now!
> http://p.sf.net/sfu/rsa-sfdev2dev2_______________________________________________
> mesa-users mailing list
> mes...@li...
> https://lists.sourceforge.net/lists/listinfo/mesa-users