From agent-partner at itsavvinner.net Thu Aug 7 12:33:22 2003 From: agent-partner at itsavvinner.net (Wilmer Chandler) Date: Thu, 07 Aug 03 17:33:22 GMT Subject: [FLASH-BUGS] FW: Teachers Affordable Property Insurance Message-ID: <3cv0-76y$9$kmk-14$0$ixc@0620ws65y15e0a> An HTML attachment was scrubbed... URL: http://flash.uchicago.edu/pipermail/flash-bugs/attachments/20030807/714d928e/attachment.html From gawrysz at camk.edu.pl Thu Aug 7 15:57:38 2003 From: gawrysz at camk.edu.pl (Artur Gawryszczak) Date: Thu, 7 Aug 2003 22:57:38 +0200 Subject: [FLASH-BUGS] possible bugs in Poisson routines Message-ID: <200308072257.38596@phoenix.camk.edu.pl> Hi, I've attached two setups derived from dust_coll, which shows quite strange results. In both cases I'm using "natural" units, so G*density = 1, and box walls are set at +/-1 dust_coll.2d_cart_mgrd - sets up a collapse on 2-D cartesian grid using multigrid submodule. Everything works fine, except that free-fall time is longer than expected for given initial density. It takes 0.5 unit of time to collapse, while in my opinion it should take 1/sqrt(2*pi) (this is a collapse of an infinite cylinder). I've tried to run this problem in 3D, changed init_block.F90 to fill a cylinder instead of sphere and put periodic BC in z-direction. The free-fall time did not change. dust_coll.3d_cart_mpol - sets up a collapse on 3-D cartesian grid using multipole submodule. Strange bubble forms at the center of the sphere when "mpole_lmax" parameter is set high enough, i.e. 10 or more. -- Best Regards Artur -------------- next part -------------- A non-text attachment was scrubbed... Name: dust_coll_tests.tgz Type: application/x-tgz Size: 7140 bytes Desc: not available Url : http://flash.uchicago.edu/pipermail/flash-bugs/attachments/20030807/b195e5c1/attachment.bin From pmricker at uiuc.edu Thu Aug 7 16:12:35 2003 From: pmricker at uiuc.edu (Paul Ricker) Date: Thu, 07 Aug 2003 16:12:35 -0500 Subject: [FLASH-BUGS] possible bugs in Poisson routines In-Reply-To: <200308072257.38596@phoenix.camk.edu.pl> References: <200308072257.38596@phoenix.camk.edu.pl> Message-ID: <1060290755.10504.164.camel@localhost> Artur, The dust collapse problem requires isolated boundary conditions, which are not supported for multigrid in 2D Cartesian coordinates. What boundary conditions did you use for the 2D/3D multigrid case? If the boundary conditions are periodic or Dirichlet, instead of isolated, you will get a different potential, and the collapse time will be different. For multipole, the problem is spherically symmetric, so you shouldn't need to use such high mpole_lmax values. I suspect the problem in this case is that, because of the high power of r for high multipole order, errors in the computation of moments near the center are leading to force errors there. Paul -- --------------------------------------------------------------------- Paul M. Ricker Department of Astronomy Assistant Professor National Center for Supercomputing Applications pmricker at uiuc.edu University of Illinois at Urbana-Champaign http://www.astro.uiuc.edu/~pmricker Urbana IL 61801-3074 --------------------------------------------------------------------- On Thu, 2003-08-07 at 15:57, Artur Gawryszczak wrote: > Hi, > > I've attached two setups derived from dust_coll, which shows quite strange > results. > > In both cases I'm using "natural" units, so G*density = 1, and box walls are > set at +/-1 > > dust_coll.2d_cart_mgrd - sets up a collapse on 2-D cartesian grid using > multigrid submodule. Everything works fine, except that free-fall time is > longer than expected for given initial density. It takes 0.5 unit of time to > collapse, while in my opinion it should take 1/sqrt(2*pi) (this is a collapse > of an infinite cylinder). I've tried to run this problem in 3D, changed > init_block.F90 to fill a cylinder instead of sphere and put periodic BC in > z-direction. The free-fall time did not change. > > dust_coll.3d_cart_mpol - sets up a collapse on 3-D cartesian grid using > multipole submodule. Strange bubble forms at the center of the sphere when > "mpole_lmax" parameter is set high enough, i.e. 10 or more. From gawrysz at camk.edu.pl Thu Aug 7 18:54:23 2003 From: gawrysz at camk.edu.pl (Artur Gawryszczak) Date: Fri, 8 Aug 2003 01:54:23 +0200 Subject: [FLASH-BUGS] possible bugs in Poisson routines In-Reply-To: <1060290755.10504.164.camel@localhost> References: <200308072257.38596@phoenix.camk.edu.pl> <1060290755.10504.164.camel@localhost> Message-ID: <200308080154.23396@phoenix.camk.edu.pl> Hello, On Thursday 07 August 2003 23:12, Paul Ricker wrote: > The dust collapse problem requires isolated boundary conditions, which > are not supported for multigrid in 2D Cartesian coordinates. What > boundary conditions did you use for the 2D/3D multigrid case? I'm using `grav_boundary_type = "isolated"' in every run. Maybe this isn't supported, but both 2D and 3D setups does not complain about that. I made integrations once more (third or fourth time) and found the missing factor. I'm sorry about the mess. > For multipole, the problem is spherically symmetric, so you shouldn't > need to use such high mpole_lmax values. I suspect the problem in > this case is that, because of the high power of r for high multipole > order, errors in the computation of moments near the center are > leading to force errors there. Well, this was just an experiment. If it produces strange results in simple case, I'd rather not trust it and use multigrid submodule. -- Best Regards, Artur From m.s.gross at hw.ac.uk Thu Aug 21 09:14:25 2003 From: m.s.gross at hw.ac.uk (Markus Gross) Date: Thu, 21 Aug 2003 15:14:25 +0100 Subject: [FLASH-BUGS] FLASH USERS MAILINGLIST runns rampert Message-ID: <03082115142500.00707@lasersim> Hi! I am currently flooded with old old postings to the flash-users list What is going on? Markus -- _______________________________________________________________ Markus Gross AMIMechE BEng (Hons.) Mechanical Engineering Heriot Watt University Edinburgh School of Engineering and Physical Sciences James Nasmyth Building Edinburgh EH14 4AS UK Member of IMechE, SPIE, CSME and VDI _______________________________________________________________ further contact: Phone : +44 (0) 131 451 4737 UNiX talk: talk markus at lasersim.mce.hw.ac.uk _______________________________________________________________ "Plans are a place to begin," Grove said. "They rarely deliver you to where you expect. Make your plans knowing you are going to throw them away." _______________________________________________________________ From cvpdx6z at prodigy.com Thu Aug 21 18:56:08 2003 From: cvpdx6z at prodigy.com (Morgan Elmore) Date: Thu, 21 Aug 03 23:56:08 GMT Subject: [FLASH-BUGS] Best Business Opportunity of the Decade! mbwerp vhpkw xzddy Message-ID: An HTML attachment was scrubbed... URL: http://flash.uchicago.edu/pipermail/flash-bugs/attachments/20030821/5cd0b5f2/attachment.html From gardini at astro.uiuc.edu Sun Aug 24 14:47:45 2003 From: gardini at astro.uiuc.edu (Alessandro Gardini) Date: Sun, 24 Aug 2003 15:47:45 -0400 Subject: [FLASH-BUGS] configuration MPICH Message-ID: <200308241547.45153.gardini@astro.uiuc.edu> Dear Sir/Madam, I tried to install flash on my workstation and got troubles. The OS is Linux RH 9.0, the c++ and f90 compilers are the Intel free evaluation ones. I downloaded the MPI from MPICH site and HDF5 from ncsa. Testing the sedov test, the command make returns various errors, according to the way I configured the MPICH. Last time I used F90=ifc ./configure --prefix=/usr/local/mpich-1.2.5 c++=ifc cc=ifc --with-device=ch_shmem and it was returned make[1]: Entering directory `/home/gardini/work/FLASH2.3/object' mpif90 -c -r8 -i4 -O2 -tpp7 -u -DN_DIM=2 -DMAXBLOCKS=1000 -DNXB=8 -DNYB=8 -DNZB=1 CosmologicalFunctions.F90 /usr/local/mpich/bin/mpif90: line 332: eval: -c: invalid option eval: usage: eval [arg ...] make[1]: *** [CosmologicalFunctions.o] Error 2 make[1]: Leaving directory `/home/gardini/work/FLASH2.3/object' make: *** [default] Error 2 I wish to ask you if you know how to configure mpich for my software. Thank you for your attention and best regards, Alessandro Gardini ----------------------------------------------------------------------------------------------------------- Alessandro Gardini phone: 217 333 9378 Dept. of Astronomy e-mail: gardini at astro.uiuc.edu 1002 W. Green St 61801 Urbana USA From kmriley at zingiber.uchicago.edu Mon Aug 25 12:32:14 2003 From: kmriley at zingiber.uchicago.edu (Katherine M Riley) Date: Mon, 25 Aug 2003 12:32:14 -0500 Subject: [FLASH-BUGS] configuration MPICH In-Reply-To: Message from Alessandro Gardini of "Sun, 24 Aug 2003 15:47:45 EDT." <200308241547.45153.gardini@astro.uiuc.edu> Message-ID: <200308251732.h7PHWEIg027978@zingiber.uchicago.edu> Have you tested mpich to make sure it works properly? You can try the example in the examples/ directory of the source tree. Also - make sure the intel compiler works alright. I have had issues with the compiler on rh9. It sounds like either your compiler is installed incorrectly or mpich is incorrectly installed because '-c' is one of the most standard options. It tells to compiler just to create an object file. -katherine > Dear Sir/Madam, > > I tried to install flash on my workstation > and got troubles. > The OS is Linux RH 9.0, the c++ and f90 > compilers are the Intel free evaluation ones. > I downloaded the MPI from MPICH site and > HDF5 from ncsa. > > Testing the sedov test, the command make > returns various errors, according to the way I configured > the MPICH. > > Last time I used > > F90=ifc ./configure --prefix=/usr/local/mpich-1.2.5 c++=ifc cc=ifc > --with-device=ch_shmem > > and it was returned > > make[1]: Entering directory `/home/gardini/work/FLASH2.3/object' > mpif90 -c -r8 -i4 -O2 -tpp7 -u -DN_DIM=2 -DMAXBLOCKS=1000 -DNXB=8 -DNYB=8 > -DNZB=1 CosmologicalFunctions.F90 > /usr/local/mpich/bin/mpif90: line 332: eval: -c: invalid option > eval: usage: eval [arg ...] > make[1]: *** [CosmologicalFunctions.o] Error 2 > make[1]: Leaving directory `/home/gardini/work/FLASH2.3/object' > make: *** [default] Error 2 > > I wish to ask you if you know how to configure mpich for my software. > > Thank you for your attention and best regards, > > Alessandro Gardini > > ----------------------------------------------------------------------------- > ------------------------------ > Alessandro Gardini phone: 217 333 9378 > Dept. of Astronomy e-mail: gardini at astro.uiuc.edu > 1002 W. Green St > 61801 Urbana USA > > From weirs at flash.uchicago.edu Mon Aug 25 12:43:09 2003 From: weirs at flash.uchicago.edu (Greg Weirs) Date: Mon, 25 Aug 2003 12:43:09 -0500 Subject: [FLASH-BUGS] configuration MPICH In-Reply-To: <200308241547.45153.gardini@astro.uiuc.edu> Message-ID: At the very least, you should have c and c++ compilers instead of c++=ifc cc=ifc Greg __________________ Greg Weirs weirs at flash.uchicago.edu 773 834 3228 __________________ On Sun, 24 Aug 2003, Alessandro Gardini wrote: > Dear Sir/Madam, > > I tried to install flash on my workstation > and got troubles. > The OS is Linux RH 9.0, the c++ and f90 > compilers are the Intel free evaluation ones. > I downloaded the MPI from MPICH site and > HDF5 from ncsa. > > Testing the sedov test, the command make > returns various errors, according to the way I configured > the MPICH. > > Last time I used > > F90=ifc ./configure --prefix=/usr/local/mpich-1.2.5 c++=ifc cc=ifc > --with-device=ch_shmem > > and it was returned > > make[1]: Entering directory `/home/gardini/work/FLASH2.3/object' > mpif90 -c -r8 -i4 -O2 -tpp7 -u -DN_DIM=2 -DMAXBLOCKS=1000 -DNXB=8 -DNYB=8 > -DNZB=1 CosmologicalFunctions.F90 > /usr/local/mpich/bin/mpif90: line 332: eval: -c: invalid option > eval: usage: eval [arg ...] > make[1]: *** [CosmologicalFunctions.o] Error 2 > make[1]: Leaving directory `/home/gardini/work/FLASH2.3/object' > make: *** [default] Error 2 > > I wish to ask you if you know how to configure mpich for my software. > > Thank you for your attention and best regards, > > Alessandro Gardini > > ----------------------------------------------------------------------------------------------------------- > Alessandro Gardini phone: 217 333 9378 > Dept. of Astronomy e-mail: gardini at astro.uiuc.edu > 1002 W. Green St > 61801 Urbana USA > > From tomek at flash.uchicago.edu Mon Aug 25 12:53:38 2003 From: tomek at flash.uchicago.edu (Tomasz Plewa) Date: Mon, 25 Aug 2003 12:53:38 -0500 Subject: [FLASH-BUGS] configuration MPICH In-Reply-To: References: <200308241547.45153.gardini@astro.uiuc.edu> Message-ID: <20030825175338.GB10259389@flash.uchicago.edu> Also, since mpich is just a shell script, you may want to look into the source and see wheter -c is there at all. Tomek -- On Mon, Aug 25, 2003 at 12:43:09PM -0500, Greg Weirs wrote: > > > At the very least, you should have c and c++ compilers instead of > > c++=ifc cc=ifc > > Greg > > __________________ > > Greg Weirs > weirs at flash.uchicago.edu > 773 834 3228 > __________________ > > On Sun, 24 Aug 2003, Alessandro Gardini wrote: > > > Dear Sir/Madam, > > > > I tried to install flash on my workstation > > and got troubles. > > The OS is Linux RH 9.0, the c++ and f90 > > compilers are the Intel free evaluation ones. > > I downloaded the MPI from MPICH site and > > HDF5 from ncsa. > > > > Testing the sedov test, the command make > > returns various errors, according to the way I configured > > the MPICH. > > > > Last time I used > > > > F90=ifc ./configure --prefix=/usr/local/mpich-1.2.5 c++=ifc cc=ifc > > --with-device=ch_shmem > > > > and it was returned > > > > make[1]: Entering directory `/home/gardini/work/FLASH2.3/object' > > mpif90 -c -r8 -i4 -O2 -tpp7 -u -DN_DIM=2 -DMAXBLOCKS=1000 -DNXB=8 -DNYB=8 > > -DNZB=1 CosmologicalFunctions.F90 > > /usr/local/mpich/bin/mpif90: line 332: eval: -c: invalid option > > eval: usage: eval [arg ...] > > make[1]: *** [CosmologicalFunctions.o] Error 2 > > make[1]: Leaving directory `/home/gardini/work/FLASH2.3/object' > > make: *** [default] Error 2 > > > > I wish to ask you if you know how to configure mpich for my software. > > > > Thank you for your attention and best regards, > > > > Alessandro Gardini > > > > ----------------------------------------------------------------------------------------------------------- > > Alessandro Gardini phone: 217 333 9378 > > Dept. of Astronomy e-mail: gardini at astro.uiuc.edu > > 1002 W. Green St > > 61801 Urbana USA > > > > -- From jbgallag at flash.uchicago.edu Mon Aug 25 13:10:10 2003 From: jbgallag at flash.uchicago.edu (JB Gallagher) Date: Mon, 25 Aug 2003 13:10:10 -0500 Subject: [FLASH-BUGS] configuration MPICH References: <200308241547.45153.gardini@astro.uiuc.edu> Message-ID: <3F4A5102.737EAADC@flash.uchicago.edu> Alessandro, Try the following: setenv FC ifc setenv F90 ifc setenv FFLAGS "-Vaxlib" setenv RSHCOMMAND ssh (if using bource like shell, change the above to: export FC=ifc etc..) ./configure --prefix=/usr/local/mpich-1.2.5 --with-romio gmake ; gmake install This should get mpich built. Once it is installed edit /usr/local/mpich-1.2.5/bin/mpif90 and mpif77 and make sure "-Vaxlib" is defined in the variable "BASE_FLAGS". One thing I will point out is that you need to use ifc 6.0 for flash, there are known problems in building flash with ifc 7.0, if you can't get access to ifc 6.0 let me know. Thanks, --Brad Alessandro Gardini wrote: > Dear Sir/Madam, > > I tried to install flash on my workstation > and got troubles. > The OS is Linux RH 9.0, the c++ and f90 > compilers are the Intel free evaluation ones. > I downloaded the MPI from MPICH site and > HDF5 from ncsa. > > Testing the sedov test, the command make > returns various errors, according to the way I configured > the MPICH. > > Last time I used > > F90=ifc ./configure --prefix=/usr/local/mpich-1.2.5 c++=ifc cc=ifc > --with-device=ch_shmem > > and it was returned > > make[1]: Entering directory `/home/gardini/work/FLASH2.3/object' > mpif90 -c -r8 -i4 -O2 -tpp7 -u -DN_DIM=2 -DMAXBLOCKS=1000 -DNXB=8 -DNYB=8 > -DNZB=1 CosmologicalFunctions.F90 > /usr/local/mpich/bin/mpif90: line 332: eval: -c: invalid option > eval: usage: eval [arg ...] > make[1]: *** [CosmologicalFunctions.o] Error 2 > make[1]: Leaving directory `/home/gardini/work/FLASH2.3/object' > make: *** [default] Error 2 > > I wish to ask you if you know how to configure mpich for my software. > > Thank you for your attention and best regards, > > Alessandro Gardini > > ----------------------------------------------------------------------------------------------------------- > Alessandro Gardini phone: 217 333 9378 > Dept. of Astronomy e-mail: gardini at astro.uiuc.edu > 1002 W. Green St > 61801 Urbana USA -- Brad Gallagher ASCI Flash Center jbgallag at flash.uchicago.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: http://flash.uchicago.edu/pipermail/flash-bugs/attachments/20030825/12d2ef08/attachment.html From accessinsurance at itsavvinner.com Wed Aug 27 00:09:44 2003 From: accessinsurance at itsavvinner.com (Twila Mcnair) Date: Wed, 27 Aug 03 05:09:44 GMT Subject: [FLASH-BUGS] RE: Teachers Get Affordable Insurance For Your Assets Message-ID: <4tz5skdt3cs6$mb56-7-$d76-l0@chk4p> An HTML attachment was scrubbed... URL: http://flash.uchicago.edu/pipermail/flash-bugs/attachments/20030827/51e24b04/attachment.html From andrew at tapir.caltech.edu Wed Aug 27 03:51:45 2003 From: andrew at tapir.caltech.edu (Andrew MacFadyen) Date: Wed, 27 Aug 2003 01:51:45 -0700 Subject: [FLASH-BUGS] avisco bug for cylindrical coords Message-ID: <3F4C7121.7040902@tapir.caltech.edu> In cylindrical coordinates with a reflecting boundary for the radial coordinate and xmin = 0.0, x(i-1) = -x(i) for i = 5 so x(5)**2 = x(4)**2 and there is a division by zero for i = 5 in the computation of avis(i). avis(5) is set to zero anyway in hydro_1d for reflecting b.c. but the following change avoids the divide by zero. --- ../../source/hydro/explicit/split/ppm/avisco.F90 2003-06-10 12:31:23.000000000 -0700 +++ avisco.F90 2003-08-26 18:09:09.000000000 -0700 @@ -58,8 +58,12 @@ if (igeomx == 1) then do i = 5, nzn4+1 - avis(i) = (x(i) * u(i) - x(i-1) * u(i-1)) * 2.0 / & - (x(i)**2 - x(i-1)**2) + IF (x(i)**2 == x(i-1)**2) THEN + avis(i) = 0.0 + ELSE + avis(i) = (x(i) * u(i) - x(i-1) * u(i-1)) * 2.0 / & + (x(i)**2 - x(i-1)**2) + ENDIF avis(i) = - cvisc * avis(i) * (x(i) - x(i-1)) enddo end if -- Andrew MacFadyen http://www.its.caltech.edu/~aim Theoretical Astrophysics, Caltech andrew at tapir.caltech.edu MC 130-33, Pasadena, CA 91125 (626): 796-5675(fax); 395-8413(w); 396-8247(h) From tomek at flash.uchicago.edu Wed Aug 27 08:53:37 2003 From: tomek at flash.uchicago.edu (Tomasz Plewa) Date: Wed, 27 Aug 2003 08:53:37 -0500 Subject: [FLASH-BUGS] avisco bug for cylindrical coords In-Reply-To: <3F4C7121.7040902@tapir.caltech.edu> References: <3F4C7121.7040902@tapir.caltech.edu> Message-ID: <20030827135337.GA11204242@flash.uchicago.edu> Thanks, Andrew. We will immediately include safeguards to prevent it from happening. Tomek -- On Wed, Aug 27, 2003 at 01:51:45AM -0700, Andrew MacFadyen wrote: > > In cylindrical coordinates with a reflecting boundary for the radial > coordinate and xmin = 0.0, x(i-1) = -x(i) for i = 5 so x(5)**2 = x(4)**2 > and there is a division by zero for i = 5 in the computation of avis(i). > avis(5) is set to zero anyway in hydro_1d for reflecting b.c. but the > following change avoids the divide by zero. > > > --- ../../source/hydro/explicit/split/ppm/avisco.F90 2003-06-10 > 12:31:23.000000000 -0700 > +++ avisco.F90 2003-08-26 18:09:09.000000000 -0700 > @@ -58,8 +58,12 @@ > > if (igeomx == 1) then > do i = 5, nzn4+1 > - avis(i) = (x(i) * u(i) - x(i-1) * u(i-1)) * 2.0 / & > - (x(i)**2 - x(i-1)**2) > + IF (x(i)**2 == x(i-1)**2) THEN > + avis(i) = 0.0 > + ELSE > + avis(i) = (x(i) * u(i) - x(i-1) * u(i-1)) * 2.0 / & > + (x(i)**2 - x(i-1)**2) > + ENDIF > avis(i) = - cvisc * avis(i) * (x(i) - x(i-1)) > enddo > end if > > > -- > Andrew MacFadyen > http://www.its.caltech.edu/~aim > Theoretical Astrophysics, Caltech > andrew at tapir.caltech.edu > MC 130-33, Pasadena, CA 91125 (626): 796-5675(fax); 395-8413(w); > 396-8247(h) > > -- From 901papf at aol.com Wed Aug 27 11:10:23 2003 From: 901papf at aol.com (Steve Flowers) Date: Wed, 27 Aug 03 16:10:23 GMT Subject: [FLASH-BUGS] Quality conferencing only 15 cents per min. ibly r nausmnjqko Message-ID: <75-20t-$1xg$xxv@07iqg9.7.vsxk> An HTML attachment was scrubbed... URL: http://flash.uchicago.edu/pipermail/flash-bugs/attachments/20030827/2d81b2c2/attachment.html