Users

users@lists.einsteintoolkit.org

3 participants
3984 discussions

FFTW library
by Yosef Zlochower 30 Jun '11

30 Jun '11

Hi, One the CCE thorns that I coded up uses FFTW for filtering and calculating time derivatives. Is there a thorn that provides FFTW version 3 already in ET? I used the GSL thorn as a template to make such a thorn. I can commit it under PITTNullCode, but this would be redundant if the thorn already exists. -- Dr. Yosef Zlochower Center for Computational Relativity and Gravitation Assistant Professor School of Mathematical Sciences Rochester Institute of Technology 85 Lomb Memorial Drive Rochester, NY 14623 Office:74-2067 Phone: +1 585-475-6103 yosef(a)astro.rit.edu CONFIDENTIALITY NOTE: The information transmitted, including attachments, is intended only for the person(s) or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and destroy any copies of this information.

2 2

Carpet::baseextents, failed assertion
by Scott Hawley 30 Jun '11

30 Jun '11

Hey. I updated the section code which was using basextent.at() in an earlier version of Carpet, to the new syntax of baseextents.at(0).at() (Marked with "HERE" in below snippet of code.) The code compiles, but when it runs, there is a failed assertion. It seems the only value which changes in the entire run is "istr". Is there some other part of the code that also needs to change? Thanks, Scott I'll attach the par file too...(kind of a misnomer at this point) char const * coordinates = 0; string coordinates_str; { CCTK_INFO ("Getting coordinate data from Carpet's bounding boxes"); assert (maps == 1); int const m = 0; // block index assert (mglevels == 1); int const ml = 0; // convergence level vector<vector<rbbox> > gh (vhh.at(m)->reflevels()); // Loop over all refinement levels CCTK_INFO ("About to loop over all refinement levels"); for (int rl=0; rl<vhh.at(m)->reflevels(); ++rl) { // refinement level // Find the union of all regions on this level ibset level; for (int c=0; c<vhh.at(m)->components(rl); ++c) { // component ibbox const & ext = vdd.at(m)->boxes.at(ml).at(rl).at(c).exterior; level |= ext; } // for c level.normalize(); // Convert the set of regions to coordinates CCTK_INFO ("Converting the set of regions to coordinates"); printf("rl = %d\n",rl); for (ibset::const_iterator li = level.begin(); li != level.end(); ++ li) { printf("li = %d\n",li); ibbox const & ext = * li; rvect rlo, rup, rstr; ivect ilo, iup, istr; rvect lo, up, str; for (int d=0; d<3; ++d) { rlo[d] = domainspecs.at(m).exterior_min[d]; rup[d] = domainspecs.at(m).exterior_max[d]; rstr[d] = (rup[d] - rlo[d]) / (domainspecs.at(m).npoints[d] - 1); //------------------------- HERE ---------------------------- ilo[d] = vhh.at(m)->baseextents.at(m).at(rl).lower()[d]; iup[d] = vhh.at(m)->baseextents.at(m).at(rl).upper()[d]; istr[d] = vhh.at(m)->baseextents.at(m).at(rl).stride()[d]; //-------------------------------------------------------------- lo[d] = rlo[d] + (ext.lower()[d] - ilo[d]) * ((rup[d] - rlo[d]) / (iup[d] - ilo[d])); up[d] = rlo[d] + (ext.upper()[d] - ilo[d]) * ((rup[d] - rlo[d]) / (iup[d] - ilo[d])); str[d] = ext.stride()[d] * (rstr[d] / istr[d]); printf("d = %d, npoints = %d\n",d, domainspecs.at(m).npoints[d] ); printf(" rlo=%g, rup=%g, rstr=%g\n", rlo[d],rup[d],rstr[d]); printf(" ilo=%d, iup=%d, istr=%d\n", ilo[d],iup[d],istr[d]); } CCTK_INFO ("Just finished loop over d, about to rbbox"); rbbox rext (lo, up, str); Trace of the run of this part of the code looks like.... INFO (TexMEx): Getting coordinate data from Carpet's bounding boxes INFO (TexMEx): About to loop over all refinement levels INFO (TexMEx): Converting the set of regions to coordinates rl = 0 li = 368448976 d = 0, npoints = 49 rlo=-36, rup=36, rstr=1.5 ilo=0, iup=24576, istr=512 d = 1, npoints = 49 rlo=-36, rup=36, rstr=1.5 ilo=0, iup=24576, istr=512 d = 2, npoints = 49 rlo=-36, rup=36, rstr=1.5 ilo=0, iup=24576, istr=512 INFO (TexMEx): Just finished loop over d, about to rbbox INFO (TexMEx): About to gh.at(rl)... INFO (TexMEx): Converting the set of regions to coordinates rl = 1 li = 368448880 d = 0, npoints = 49 rlo=-36, rup=36, rstr=1.5 ilo=0, iup=24576, istr=256 d = 1, npoints = 49 rlo=-36, rup=36, rstr=1.5 ilo=0, iup=24576, istr=256 d = 2, npoints = 49 rlo=-36, rup=36, rstr=1.5 ilo=0, iup=24576, istr=256 INFO (TexMEx): Just finished loop over d, about to rbbox INFO (TexMEx): About to gh.at(rl)... INFO (TexMEx): Converting the set of regions to coordinates rl = 2 li = 368448816 d = 0, npoints = 49 rlo=-36, rup=36, rstr=1.5 ilo=0, iup=24576, istr=128 d = 1, npoints = 49 rlo=-36, rup=36, rstr=1.5 ilo=0, iup=24576, istr=128 d = 2, npoints = 49 rlo=-36, rup=36, rstr=1.5 ilo=0, iup=24576, istr=128 INFO (TexMEx): Just finished loop over d, about to rbbox INFO (TexMEx): About to gh.at(rl)... INFO (TexMEx): Converting the set of regions to coordinates rl = 3 li = 368443536 d = 0, npoints = 49 rlo=-36, rup=36, rstr=1.5 ilo=0, iup=24576, istr=64 d = 1, npoints = 49 rlo=-36, rup=36, rstr=1.5 ilo=0, iup=24576, istr=64 d = 2, npoints = 49 rlo=-36, rup=36, rstr=1.5 ilo=0, iup=24576, istr=64 INFO (TexMEx): Just finished loop over d, about to rbbox cactus_texmex: /home/shawley/Cactus/arrangements/Carpet/CarpetLib/src/bbox.cc:25: void bbox<T, D>::assert_bbox_limits() const [with T = double, D = 3]: Assertion `all((_upper-_lower)%_stride == T(0))' failed. Abort

2 3

OpenMP problem in Hydro_Analysis
by Erik Schnetter 29 Jun '11

29 Jun '11

In the course of a private email discussion I found the following problem. Since this may be of general interest I repeat my findings in public. The discussion concerned possible OpenMP problems in GRHydro: "The routine Hydro_Analysis_LocationSearch may return different results with OpenMP. If one component contains multiple maxima, then the last one is returned (without OpenMP), or a random one is returned (with OpenMP). However, if several components contain the multiple maxima, then the location of the maximum remains unset, i.e. at the value it had at the last timestep. "I suggest to switch off OpenMP in this routine, to insert a check whether a single component contains multiple maxima and treat this the same as when several components contain multiple maxima, and to set the location of the maximum to a special error value (or to introduce an error flag) so that it is obvious when the routine fails." -erik -- Erik Schnetter <schnetter(a)cct.lsu.edu> http://www.cct.lsu.edu/~eschnett/

2 1

change default value of CarpetRegrid2's min_distance
by Roland Haas 29 Jun '11

29 Jun '11

Hello all, I would like to propose change the default value of CarpetRegrid2::min_distance parameter from 4 (grid points) to 0. Currently: --8<-- CCTK_INT min_distance "Minimum distance (in grid points) between coarse and fine grid boundaries" STEERABLE=always { 0:* :: "" } 4 --8<-- and min_distance is used in enlarged_fine_grid to enforce a minimum distance between grid levels. A distance of 0 means that the end of the buffer+ghost zone region of the fine grid touches the end of the active (ie. non-buffer) region of the coarse grid. This can have unexpected effects when specifying "tight" boxes via the radius_n[] parameters when trying to translate old CarpetRegrid parameter files, since the radius_n[] parameters are no longer what actually specifies the final grid size. It also (I think) makes CarpetRegrid2 with boxes create a different grid than the equivalent manual-coordinates description in CarpetRegrid. Making min_distance's default value zero reduces the number of parameters the affect the grid structure in the default setting and makes it the user's responsibility to choose a decent (maybe 4) value for the emergency distance parameter min_distance. For the amr implementation in CarpetRegrid2 min_distance is also used and might be much more useful then for the nested boxes currently used. There a default of 4 is probably more sensible. Any comments on this? Yours, Roland -- My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from http://keys.gnupg.net.

1 0

Einstein Toolkit Meeting Reminder
by Frank Loeffler 27 Jun '11

27 Jun '11

Hi Please consider joining the weekly Einstein Toolkit phone call at 10 am US central time on Mondays. As usual, you can find instructions how to join on the following web site: http://einsteintoolkit.org/community/support/ I short: the number is (+1) 225-578-4942 or (+1) 866-573-0359 and the conference id is 118682#. Several regular members will not be able to make the call this time, and as far as I can tell there is no progress to report on either the paper nor concerning tests of the CCE code. Therefore I suggest to only have a short phone meeting to give anyone the opportunity to report/ask something, and to use the remaining usual meeting time to work on either the paper or CCE testing within the ET. To coordinate this, I booked an evo meeting/chat: http://evo.caltech.edu/evoNext/koala.jnlp?meeting=M9MIMi2a2uD2Dn9a9uDD99 Frank

1 0

ET linking problem related to regex
by Wolfgang Kastaun 23 Jun '11

23 Jun '11

Hi, I just tried to compile the ET (ET_RELEASE = ET_2010_11) and got many errors in the linking stage of the sort /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0x36): undefined reference to `regfreeA' For the full output see below. I was using the intel compiler version 10.1. The same source tree compiled fine before on other systems using gnu and xlc compilers. I don't know however if the problem is really related to the compiler, I'll try with gnu as well. During the configure stage, regexp.h was found on the system. I remember that Cactus has its own version of regexp, maybe there is a conflict. Any ideas ? Regards, Wolfgang. -------------------- Creating cactus_pizza_dev in /u/ap/kastaun/sim_build/pizza_dev/exe from EinsteinBase/ADMBase AEIThorns/AEILocalInterp CactusBase/Boundary Carpet/Carpet Carpet/CarpetEvolutionMask Carpet/CarpetIOASCII Carpet/CarpetIOBasic Carpet/CarpetIOHDF5 Carpet/CarpetIOScalar Carpet/CarpetInterp Carpet/CarpetInterp2 Carpet/CarpetLib Carpet/CarpetMask Carpet/CarpetReduce Carpet/CarpetRegrid Carpet/CarpetRegrid2 Carpet/CarpetSlab Carpet/CarpetTracker CactusBase/CartGrid3D CactusBase/CoordBase CactusBase/Fortran ExternalLibraries/GSL ExternalLibraries/HDF5 EinsteinBase/HydroBase PizzaDev/HydroBase_RNSID CactusBase/IOASCII CactusBase/IOBasic CactusBase/IOUtil CactusBase/InitBase CactusBase/LocalInterp CactusBase/LocalReduce Carpet/LoopControl CactusNumerical/MoL EinsteinAnalysis/Multipole CactusUtils/Nice CactusNumerical/Norms CactusNumerical/Periodic PizzaDev/Pizza PizzaDev/PizzaBase PizzaDev/PizzaIDBase PizzaDev/PizzaIDFile PizzaDev/PizzaMultipoles PizzaDev/PizzaMultipoles2 PizzaDev/PizzaPertEFF PizzaDev/PizzaPerturb PizzaDev/PizzaRNSA PizzaDev/PizzaRecycle PizzaDev/PizzaSample PizzaDev/PizzaStar PizzaDev/PizzaTOV CactusNumerical/ReflectionSymmetry CactusNumerical/RotatingSymmetry180 CactusNumerical/RotatingSymmetry90 CactusNumerical/Slab CactusNumerical/SlabTest CactusNumerical/SpaceMask CactusNumerical/SphericalSurface CactusBase/SymBase CactusUtils/TerminationTrigger CactusBase/Time CactusUtils/TimerReport /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/lib/libthorn_Cactus.a(Parameters.c.o): In function `CCTK_ParameterSetNotifyUnregister': /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0x36): undefined reference to `regfreeA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0x4a): undefined reference to `regfreeA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/lib/libthorn_Cactus.a(Parameters.c.o): In function `cctk_parametersetnotifyunregister_': /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0xad): undefined reference to `regfreeA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0xc1): undefined reference to `regfreeA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/lib/libthorn_Cactus.a(Parameters.c.o): In function `CCTK_ParameterSetNotifyRegister': /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0x1a2): undefined reference to `regcompA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0x1be): undefined reference to `regcompA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/lib/libthorn_Cactus.a(Parameters.c.o): In function `cctk_parametersetnotifyregister_': /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0x317): undefined reference to `regcompA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0x33b): undefined reference to `regcompA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/lib/libthorn_Cactus.a(Parameters.c.o): In function `CCTK_ParameterSet': /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0x1122): undefined reference to `regexecA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0x1141): undefined reference to `regexecA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/lib/libthorn_Cactus.a(Parameters.c.o): In function `cctk_parameterset_': /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0x1aba): undefined reference to `regexecA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/main/Parameters.c:(.text+0x1ad9): undefined reference to `regexecA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/lib/libthorn_Cactus.a(Misc.c.o): In function `CCTK_RegexMatch': /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/util/Misc.c:(.text+0x1ac): undefined reference to `regcompA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/util/Misc.c:(.text+0x1d1): undefined reference to `regexecA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/util/Misc.c:(.text+0x1dc): undefined reference to `regfreeA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/lib/libthorn_Cactus.a(Misc.c.o): In function `CCTK_SetStringInRegexList': /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/util/Misc.c:(.text+0x6bc): undefined reference to `regcompA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/util/Misc.c:(.text+0x6d7): undefined reference to `regexecA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/util/Misc.c:(.text+0x6e7): undefined reference to `regfreeA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/lib/libthorn_Cactus.a(Misc.c.o): In function `Util_IntInRange': /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/util/Misc.c:(.text+0xbca): undefined reference to `regcompA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/util/Misc.c:(.text+0xc0b): undefined reference to `regexecA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/util/Misc.c:(.text+0xc17): undefined reference to `regfreeA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/lib/libthorn_Cactus.a(Misc.c.o): In function `Util_DoubleInRange': /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/util/Misc.c:(.text+0x111d): undefined reference to `regcompA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/util/Misc.c:(.text+0x115e): undefined reference to `regexecA' /u/ap/kastaun/sim_build/pizza_dev/configs/pizza_dev/build/Cactus/util/Misc.c:(.text+0x116a): undefined reference to `regfreeA' make[1]: *** [/u/ap/kastaun/sim_build/pizza_dev/exe/cactus_pizza_dev] Error 1 make: *** [pizza_dev] Error 2

4 10

Re: [Users] [ET Trac] [Einstein Toolkit] #453: Formaline deletes files if lib and src are symlinks
by Roland Haas 21 Jun '11

21 Jun '11

Hello all, > Yes, that is the idea. The patch as stop-gap measure, and then trying to > understand the real cause of the problem and correcting git-add-thorn.pl. > > I do not think that Mac OS allows hard links to directories for regular > users. With these, one can create directory structures that are not trees, > which would break many Unix expectations; in particular, one can create > directory structures disconnected from the root directory and hence > unreachable. Root can call the "link" syscall (maybe even from the "ln" > command, possibly with a special option) to do this, but it's highly > discouraged. If this issue is just that Mac OS allows hard links for directories then simply changing link() to symlink() for directories: http://perldoc.perl.org/functions/symlink.html should fix the issue, shouldn't it? Something like: if(-d $object) { symlink(...) } else { link(...) } Yours, Roland -- My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from http://keys.gnupg.net.

3 3

ExternalLibraries/OpenSSL config on Mac 64
by Scott Hawley 21 Jun '11

21 Jun '11

Hi. I checked out the einstein toolkit new and installed from scratch an hour ago. Im getting.... + echo 'OpenSSL: Configuring...' OpenSSL: Configuring... + cd openssl-1.0.0d + ./config --prefix=/Users/shawley/Cactus/configs/etk/scratch/external/OpenSSL Operating system: i686-apple-darwinDarwin Kernel Version 10.7.4: Mon Apr 18 21:24:17 PDT 2011; root:xnu-1504.14.12~3/RELEASE_X86_64 WARNING! If you wish to build 64-bit library, then you have to invoke './Configure darwin64-x86_64-cc' *manually*. You have about 5 seconds to press Ctrl-C to abort. Configuring for darwin-i386-cc Configuring for darwin-i386-cc I want 64 bit. I read the ticket https://trac.einsteintoolkit.org/ticket/339 So I tried setting OPENSSL_DIR = /usr in my options file, but it seems to have no effect. My options file is attached. I then read ticket https://trac.einsteintoolkit.org/ticket/342 but I'm not sure how to apply the patch. Plus I imagine the patch has already been applied in today's version of ETK. What is the current way of dealing with this? Thanks, Scott -- Scott H. Hawley, Ph.D. Asst. Prof. of Physics Chemistry & Physics Dept Office: Hitch 100D Belmont University Tel: +1-615-460-6206 Nashville, TN 37212 USA Fax: +1-615-460-5458 PGP Key at http://sks-keyservers.net

3 2

What/where is thorn 'DGEMM'?
by Scott Hawley 20 Jun '11

20 Jun '11

I'm trying to run...any.. par file that came with the ETK checkout. Some of them require thorns not included in the distribution, e.g thorn PsiKadelia needed by many par files in EinsteinInitialData. The files in CarpetRegrid2/par/ require a thorn called DGEMM, which isnt part of the ETK list and doesn't even appear in Google: "Sorry, Cactus dgemm thorn did not return any results." How does one run, for example, staticbh.par? Thanks %Sent from my iPhone

3 3

Meeting minutes
by Roland Haas 20 Jun '11

20 Jun '11

Meeting minutes: Present were: Tanja, Roland, Frank, Josh, Yosef, Peter CCE code: * yosef tested the RIT SphericalHarmonicDecomp and SphericalHarmonicRecon thorns with the ET qc0.par example and confirms that they work fine with the ET new infrastructure thorns: * for SphericalHarmonicDecomp: create FFTW (3) thorn in ExternalLibraries * status of examples in ET paper: ** high-res TOV run (again) in queue on QueenBee, found fundamental mode with sourced by numerical noise, explicit perturbation was too strong, exciting non-linear higher modes ** high res single distorted black hole run is missing 16M due to Kraken crash, analysis on been started, will recover from checkpoint once the run is through the Kraken queue again fall ET workshop: * any help for writing the skeleton proposal is appreciated * so fat only Josh has send an email to Frank announcing interest Yours, Roland -- My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from http://keys.gnupg.net.

1 0

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

Users