Trac

trac@lists.einsteintoolkit.org

8708 discussions

[Einstein Toolkit] #2021: compiling Carpet fails with intel 17 and optimisation
by Einstein Toolkit 22 Jan '18

22 Jan '18

#2021: compiling Carpet fails with intel 17 and optimisation -----------------------------------------+---------------------------------- Reporter: wolfgang.kastaun@… | Owner: eschnett Type: defect | Status: new Priority: unset | Milestone: Component: Carpet | Version: Keywords: | -----------------------------------------+---------------------------------- Compiling carpet with intel 17.0.1 and -O2 or -O3 triggers an internal compiler error ("internal error: 0_76"). It only affects the file bbox.cc. Happened on two different clusters. ET version is Payne. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/2021> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 22

[Einstein Toolkit] #2072: Carpet: Disable OpenMP parallelization of transport operators
by Einstein Toolkit 22 Jan '18

22 Jan '18

#2072: Carpet: Disable OpenMP parallelization of transport operators -------------------------+-------------------------------------------------- Reporter: eschnett | Owner: eschnett Type: enhancement | Status: new Priority: major | Milestone: Component: Carpet | Version: development version Keywords: | -------------------------+-------------------------------------------------- This disables the OpenMP parallelization of Carpet's transport operators. I have observed that this leads to a significant speedup when many threads are used. The likely reason is that the regions which are parallelized are typically small. A typical reason would e.g. be the lower x-boundary of one component of one grid variable. The OpenMP thread startup overhead and the cache misses caused by parallelizing this are then larger than any benefit. See <https://bitbucket.org/eschnett/carpet/pull-requests/18/carpet- disable-openmp-parallelization-of/diff>. In a next step (not proposed here), we can parallelize transport operators again, but at a much higher level, e.g. at the level of the loop over all variables that need to be prolongated. However, Carpet currently (and quite unfortunately) uses static variables to hold pointers to timers, and these are not thread-safe. (Neither the static variables nor the timer implementation itself are.) This needs to be either corrected or disabled, which will be the topic of a further pull request. At this time, I ask some of those who are interested in performance in trying this pull request on a few iterations of a production simulation that uses many OpenMP threads and report back here. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/2072> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 10

[Einstein Toolkit] #1784: osx-homebrew.cfg does not work out of the box on Yosemite
by Einstein Toolkit 22 Jan '18

22 Jan '18

#1784: osx-homebrew.cfg does not work out of the box on Yosemite ------------------------+--------------------------------------------------- Reporter: anonymous | Owner: Type: task | Status: new Priority: optional | Milestone: Component: SimFactory | Version: ET_2015_05 Keywords: | ------------------------+--------------------------------------------------- When installing ET anew on Yosemite with the instructions in osx-homebrew.cfg, the build fails because homebrew gets the latest version of gcc (5.1), which is currently not the one then specified in the cactus compilation options (4.9). This problem was encountered by a student who approached ET for the first time. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/1784> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 3

[Einstein Toolkit] #1757: Stampede needs to be updated
by Einstein Toolkit 22 Jan '18

22 Jan '18

#1757: Stampede needs to be updated ----------------------+----------------------------------------------------- Reporter: hinder | Owner: Type: task | Status: new Priority: critical | Milestone: Component: Other | Version: development version Keywords: backport | ----------------------+----------------------------------------------------- The SimFactory definition for Stampede currently uses the intel/13.1.1.163 compiler. This compiler will be removed next Tuesday (see https://portal.tacc.utexas.edu/user-guides/stampede/intel15). According to TACC, the current default is the older intel/13.0.2.146, and this is the recommended and default compiler, and will remain so. Both compilers have the "restrict" keyword blacklisted in Cactus (#1276). They will install intel/15.0.1 at the end of April. There is also intel/14.0.1.106 available, but it is labeled as "limited software stack". It is not clear why they are removing a newer compiler, or why they chose to do this a month before installing an even newer one. Our options are: 1. Upgrade to intel/14.0.1.106, which is labeled as "limited software stack" 2. Downgrade to intel/13.0.2.146, which is the default and will remain so for a while This should be done both for the trunk and the release branch. I suggest that option 2 is the most conservative, and probably the easiest, as there might be libraries which are not compiled for intel/14.0.1.106. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/1757> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 8

[Einstein Toolkit] #2007: MoL fails to build with Intel compiler
by Einstein Toolkit 22 Jan '18

22 Jan '18

#2007: MoL fails to build with Intel compiler ---------------------+------------------------------------------------------ Reporter: hinder | Owner: eschnett Type: defect | Status: new Priority: blocker | Milestone: Component: Cactus | Version: development version Keywords: | ---------------------+------------------------------------------------------ Compiling MoL gives this error: {{{ /home/ianhin/Cactus/EinsteinToolkitGit/arrangements/CactusNumerical/MoL/src/RK4-RK2.c(60): error: unrecognized OpenMP #pragma #pragma omp /*parallel for*/ ^ compilation aborted for /home/ianhin/Cactus/EinsteinToolkitGit/configs/sim/build/MoL/RK4-RK2.c (code 2) }}} This code was introduced in https://bitbucket.org/cactuscode/cactusnumerical/commits/09a209266daa281a3e…. Removing the commented portion doesn't help. Replacing it with {{{ #pragma omp parallel }}} allows the code to compile. This is on Minerva, with icc (ICC) 16.0.1 20151021. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/2007> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 10

[Einstein Toolkit] #1458: intel 13 2013.1.117 miss-compiles asserts in TwoPunctures tp_utils.c
by Einstein Toolkit 22 Jan '18

22 Jan '18

#1458: intel 13 2013.1.117 miss-compiles asserts in TwoPunctures tp_utils.c -----------------------------------+---------------------------------------- Reporter: rhaas | Owner: Type: defect | Status: new Priority: major | Milestone: Component: EinsteinToolkit thorn | Version: development version Keywords: TwoPunctures | -----------------------------------+---------------------------------------- this continues the thread in #1429 in particular https://trac.einsteintoolkit.org/ticket/1429#comment:14. Currently we need to determine which versions are affected. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/1458> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 4

[Einstein Toolkit] #2095: Upgrade Trac from 0.12
by Einstein Toolkit 12 Jan '18

12 Jan '18

#2095: Upgrade Trac from 0.12 -----------------------------------+---------------------------------------- Reporter: hinder | Owner: sbrandt Type: task | Status: new Priority: minor | Milestone: Component: Server Infrastructure | Version: development version Keywords: | -----------------------------------+---------------------------------------- The current version of Trac is 0.12, which is very old. It should be upgraded to something more modern, even if only to avoid security problems. It's also likely that newer versions have more functionality. Documentation for performing the upgrade is at https://trac.edgewall.org/wiki/TracUpgrade. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/2095> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 1

[Einstein Toolkit] #2091: ET wiki is down
by Einstein Toolkit 08 Jan '18

08 Jan '18

#2091: ET wiki is down -----------------------------------+---------------------------------------- Reporter: hinder | Owner: sbrandt Type: defect | Status: new Priority: blocker | Milestone: Component: Server Infrastructure | Version: development version Keywords: | -----------------------------------+---------------------------------------- The ET wiki went down at 03:12 CET on 10-Dec-2017. -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/2091> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 3

[Einstein Toolkit] #2094: Regularly Send Doodle/WhenIsGood Polls for ETK Telecons
by Einstein Toolkit 04 Jan '18

04 Jan '18

#2094: Regularly Send Doodle/WhenIsGood Polls for ETK Telecons -----------------------+---------------------------------------------------- Reporter: anonymous | Owner: Type: defect | Status: new Priority: unset | Milestone: Component: Other | Version: development version Keywords: | -----------------------+---------------------------------------------------- To maximize participation in ETK telecons throughout the year, particularly among those of us who have to teach or have other commitments, a Doodle/WhenIsGood poll should be sent out regularly. Given that most of us exist at universities with semester schedules, I would suggest sending out polls 3 times per year: 1) at the start of the Fall semester (in Aug/Sept) 2) at the start of the Spring semester (in Jan) 3) at the start of the summer (in May). -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/2094> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 0

[Einstein Toolkit] #1599: CarpetInterp/waveinterp_2p tests constant in time initial data
by Einstein Toolkit 27 Dec '17

27 Dec '17

#1599: CarpetInterp/waveinterp_2p tests constant in time initial data --------------------------+------------------------------------------------- Reporter: rhaas | Owner: eschnett Type: defect | Status: new Priority: minor | Milestone: Component: Carpet | Version: development version Keywords: CarpetInterp | --------------------------+------------------------------------------------- the test waveinterp_2p sets up initial data using IDWaveMoL (fom CactusExamples) using kx=ky=kz = 0. and slopet = 0.1. Ie the initial data is constant in space and has frequency 0.1. It then evolves this using WaveMoL and interpolates in space and time with CarpetInterp. Since the function is constant in space (exactly so, see eg the output in InterpToArray::array1d_vars[0]) the interpolation in space just tests roundoff errors. The second time derivative (array1d_vars[2]) should be zero and all we measure in the test is truncation errors on the level of 1e-9. This makes the test very sensitive to -Ofast, -O0, gcc vs. intel etc. and not a very good test. I would rather either use a Gausian wave for initial data so that there is some actual data in the result and not just numerical noise, or leave out the 2nd time derivative from the test data. The test fails due to significant errors in a virtual machine using ubuntu.cfg (which usies -ffast-math and -O2). -- Ticket URL: <https://trac.einsteintoolkit.org/ticket/1599> Einstein Toolkit <http://einsteintoolkit.org> The Einstein Toolkit

1 11

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

Trac