From schnetter at cct.lsu.edu  Tue Nov  1 18:49:55 2005
From: schnetter at cct.lsu.edu (Erik Schnetter)
Date: Tue, 1 Nov 2005 18:49:55 -0600
Subject: [Patches] SampleBoundary: Check whether there are enough local grid
	points
Message-ID: <A9FCBAC7-A1ED-4842-8883-11B453D0EE7F@cct.lsu.edu>

Thorn CactusExamples/SampleBoundary applies an extrapolation boundary  
conditions.  For this it needs at least two interior grid points, in  
addition to the actual boundary points and the ghost points.  Ghost  
points cannot be used because the grid function may not have been  
synchronised before the boundary condition is applied.

The current implementation contains a check which disables -- without  
warning -- the boundary condition if there are not enough interior  
grid points.  Furthermore, this check is wrong because it does not  
take the ghost points into account.  This leads to spurious errors as  
the test case is run on more processors, because the local grid size  
shrinks.

This patch applies the boundary condition unconditionally, i.e.,  
independent of the local grid size.  If there are not enough local  
grid points, an error is returned.  A new error code 64 is used for  
that.

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>

My email is as private as my paper mail.  I therefore support encrypting
and signing email messages.  Get my PGP key from www.keyserver.net.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: sampleboundary.diff
Type: application/octet-stream
Size: 2125 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/patches/attachments/20051101/38d21584/attachment.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://www.cactuscode.org/pipermail/patches/attachments/20051101/38d21584/attachment.bin 

From schnetter at cct.lsu.edu  Wed Nov  2 15:09:22 2005
From: schnetter at cct.lsu.edu (Erik Schnetter)
Date: Wed, 2 Nov 2005 15:09:22 -0600
Subject: [Patches] Make PUGH's memory statistics work again
Message-ID: <5AE852DD-C14D-40AD-AE98-67A65F49D2BB@cct.lsu.edu>

PUGH's memory statistics were broken since it switched from En-/ 
DisableGroupStorage to In-/DecreaseGroupStorage.  The statistics were  
taken only when the old mechanism was used.

This patch moves the statistic keeping into a lower level routine,  
namely the routine that actually calls the malloc and free routines.   
This ensures that the statistics are always correct.  A disadvantage  
is that the distinction between grid functions and grid arrays is  
lost at that time.

In particular, the patch does the following:

(1) Change the variables that contain the memory statistics.  They  
are now double instead of float, track bytes instead of megabytes (no  
fractional part has more accuracy), and they are renamed according to  
their new meaning.  They also track the number grid variables times  
number of time levels, because it is too difficult to track the  
number of grid variables alone.  Personally, I find the new value  
also more interesting.

(2) Remove the old statistics code.

(3) Add the new statistics code.  This routine takes as arguments  
some values which are also available as parameters.  Since the  
routine now needs a CCTK_DECLARE_PARAMETERS to find out whether it  
should print to stdout, I rename the routine arguments.

(4) Update the routine that prints the statistics.

(5) Fix a typo in a comment.

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>

My email is as private as my paper mail.  I therefore support encrypting
and signing email messages.  Get my PGP key from www.keyserver.net.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: pugh-memstats.diff
Type: application/octet-stream
Size: 8593 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/patches/attachments/20051102/330dc873/attachment.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://www.cactuscode.org/pipermail/patches/attachments/20051102/330dc873/attachment.bin 

From schnetter at cct.lsu.edu  Wed Nov  2 16:41:53 2005
From: schnetter at cct.lsu.edu (Erik Schnetter)
Date: Wed, 2 Nov 2005 16:41:53 -0600
Subject: [Patches] SampleBoundary: Check whether there are enough local grid
	points
Message-ID: <63C87321-66D4-47FF-804B-F46A57E7F215@cct.lsu.edu>

[I'm resending this patch without signature.]

Thorn CactusExamples/SampleBoundary applies an extrapolation boundary  
conditions.  For this it needs at least two interior grid points, in  
addition to the actual boundary points and the ghost points.  Ghost  
points cannot be used because the grid function may not have been  
synchronised before the boundary condition is applied.

The current implementation contains a check which disables -- without  
warning -- the boundary condition if there are not enough interior  
grid points.  Furthermore, this check is wrong because it does not  
take the ghost points into account.  This leads to spurious errors as  
the test case is run on more processors, because the local grid size  
shrinks.

This patch applies the boundary condition unconditionally, i.e.,  
independent of the local grid size.  If there are not enough local  
grid points, an error is returned.  A new error code 64 is used for  
that.

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>

My email is as private as my paper mail.  I therefore support encrypting
and signing email messages.  Get my PGP key from www.keyserver.net.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: sampleboundary.diff
Type: application/octet-stream
Size: 2125 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/patches/attachments/20051102/cf89042b/attachment.obj 

From schnetter at cct.lsu.edu  Wed Nov  2 16:42:14 2005
From: schnetter at cct.lsu.edu (Erik Schnetter)
Date: Wed, 2 Nov 2005 16:42:14 -0600
Subject: [Patches] Make PUGH's memory statistics work again
Message-ID: <43EF69E5-DEE1-4B7D-AF0E-DCA8B50BDAD0@cct.lsu.edu>

[I'm resending this patch without signature.]

PUGH's memory statistics were broken since it switched from En-/ 
DisableGroupStorage to In-/DecreaseGroupStorage.  The statistics were  
taken only when the old mechanism was used.

This patch moves the statistic keeping into a lower level routine,  
namely the routine that actually calls the malloc and free routines.   
This ensures that the statistics are always correct.  A disadvantage  
is that the distinction between grid functions and grid arrays is  
lost at that time.

In particular, the patch does the following:

(1) Change the variables that contain the memory statistics.  They  
are now double instead of float, track bytes instead of megabytes (no  
fractional part has more accuracy), and they are renamed according to  
their new meaning.  They also track the number grid variables times  
number of time levels, because it is too difficult to track the  
number of grid variables alone.  Personally, I find the new value  
also more interesting.

(2) Remove the old statistics code.

(3) Add the new statistics code.  This routine takes as arguments  
some values which are also available as parameters.  Since the  
routine now needs a CCTK_DECLARE_PARAMETERS to find out whether it  
should print to stdout, I rename the routine arguments.

(4) Update the routine that prints the statistics.

(5) Fix a typo in a comment.

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>

My email is as private as my paper mail.  I therefore support encrypting
and signing email messages.  Get my PGP key from www.keyserver.net.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: pugh-memstats.diff
Type: application/octet-stream
Size: 8593 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/patches/attachments/20051102/3db276b2/attachment.obj 

From schnetter at cct.lsu.edu  Fri Nov  4 13:19:33 2005
From: schnetter at cct.lsu.edu (Erik Schnetter)
Date: Fri, 4 Nov 2005 13:19:33 -0600
Subject: [Patches] Correct problems with IBM Fortran compiler and F77FLAGS
Message-ID: <691283C1-F60D-418A-8F76-0929AE586EC7@cct.lsu.edu>

When setting F77 to F90 while configuring, we recently also started  
setting F77FLAGS to F90FLAGS.  Since this change, compiling F77 code  
with the IBM Fortran compiler seems to be broken.  The compiler does  
not recognise any more that F77 files are in fixed format.  This  
problem has been extensively discussed on the developers' list in the  
last weeks.  This patch corrects that problem.

Some known-architecture files produce makefile fragments that add, at  
run time, additional flags to F90FLAGS to set the source code form  
correctly (fixed or free format).  These fragments need to add the  
same flags to F77FLAGS as well.  This concerns the architectures AIX,  
BGL, and Darwin.

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>

My email is as private as my paper mail.  I therefore support encrypting
and signing email messages.  Get my PGP key from www.keyserver.net.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: F77F90.diff
Type: application/octet-stream
Size: 2998 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/patches/attachments/20051104/7121478d/attachment.obj 

From schnetter at cct.lsu.edu  Mon Nov  7 08:57:15 2005
From: schnetter at cct.lsu.edu (Erik Schnetter)
Date: Mon, 7 Nov 2005 08:57:15 -0600
Subject: [Patches] Correct problems with IBM Fortran compiler and F77FLAGS
In-Reply-To: <691283C1-F60D-418A-8F76-0929AE586EC7@cct.lsu.edu>
References: <691283C1-F60D-418A-8F76-0929AE586EC7@cct.lsu.edu>
Message-ID: <7C749A60-017F-465F-A438-A06B7ECE0207@cct.lsu.edu>

On Nov 4, 2005, at 13:19:33, Erik Schnetter wrote:

> When setting F77 to F90 while configuring, we recently also started  
> setting F77FLAGS to F90FLAGS.  Since this change, compiling F77  
> code with the IBM Fortran compiler seems to be broken.  The  
> compiler does not recognise any more that F77 files are in fixed  
> format.  This problem has been extensively discussed on the  
> developers' list in the last weeks.  This patch corrects that problem.
>
> Some known-architecture files produce makefile fragments that add,  
> at run time, additional flags to F90FLAGS to set the source code  
> form correctly (fixed or free format).  These fragments need to add  
> the same flags to F77FLAGS as well.  This concerns the  
> architectures AIX, BGL, and Darwin.

The previous version was not working correctly.  Steve White  
confirmed that this patch is better and solves his problems.

-erik

-- 
Erik Schnetter <schnetter at cct.lsu.edu>

My email is as private as my paper mail.  I therefore support encrypting
and signing email messages.  Get my PGP key from www.keyserver.net.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: F77F90.diff
Type: application/octet-stream
Size: 3604 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/patches/attachments/20051107/b77701c3/attachment.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: PGP.sig
Type: application/pgp-signature
Size: 186 bytes
Desc: This is a digitally signed message part
Url : http://www.cactuscode.org/pipermail/patches/attachments/20051107/b77701c3/attachment.bin 

From jabadi2 at gmail.com  Tue Nov 22 13:00:18 2005
From: jabadi2 at gmail.com (Josh Abadie)
Date: Tue, 22 Nov 2005 13:00:18 -0600
Subject: [Patches] Re: [CactusMaint] Cactus/1993:
	CreateConfigurationBindings.pl: INCLUDE is never set
In-Reply-To: <200511220536.jAM5aqHK005365@cactus.cct.lsu.edu>
References: <200511220536.jAM5aqHK005365@cactus.cct.lsu.edu>
Message-ID: <fa9006080511221100r25a21672j89c638191b2162f7@mail.gmail.com>

Having discussed it with Yaakoub, it would seem as if the INCLUDE property
was removed from configuration.ccl at some point in the past. I have deleted
the the last remaining reference to this property. Please apply and test.

Thanks a bunch.

On 11/21/05, schnetter at cct.lsu.edu <schnetter at cct.lsu.edu> wrote:
>
> >Number: 1993
> >Notify-List:
> >Category: Cactus
> >Synopsis: CreateConfigurationBindings.pl: INCLUDE is never set
> >Confidential: no
> >Severity: non-critical
> >Priority: medium
> >Responsible: goodale
> >State: open
> >Class: sw-bug
> >Submitter-Id: user
> >Arrival-Date: Mon Nov 21 23:36:51 -0600 2005
> >Originator: Erik Schnetter
> >Release: Cactus 4.0.b16
> >Organization:
> >Environment:
> >Description:
> The CST file CreateConfigurationBindings.pl checks the internal CST
> database for INCLUDE specification, which should presumably come from the
> script that is run from configuratiotn.ccl. The CST file
> ConfigScriptParser.pl never sets such a specification. The code in
> CreateConfigurationBindings.pl is thus never reached.
>
> Either ConfigScriptParser.pl should accept INCLUDE specifications, or
> CreateConfigurationBindings.pl should not allow them.
> >How-To-Repeat:
> >Fix:
> Unknown
> >Unformatted:
> _______________________________________________
> CactusMaint site list
> CactusMaint at cactuscode.org
> http://www.cactuscode.org/mailman/listinfo/cactusmaint
>



--
Josh Abadie
Computer Science Senior, LSU
Research Programmer.. kinda, CCT

Two men look out a jail window, one sees stars, the other sees mud. Look up
and out. The view is better, and just as valid

----------------
USA, LSU, OTC, OCS, CCT, JSG, JCG, CSC
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.cactuscode.org/pipermail/patches/attachments/20051122/5e41f8fe/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: CreateConfigurationBindings.pl.mine.diff
Type: application/octet-stream
Size: 760 bytes
Desc: not available
Url : http://www.cactuscode.org/pipermail/patches/attachments/20051122/5e41f8fe/attachment.obj