Trac

trac@lists.einsteintoolkit.org

8708 discussions

#2550: support compiling in Raspberry Pi 4
by Roland Haas 08 Aug '21

08 Aug '21

#2550: support compiling in Raspberry Pi 4 Reporter: Roland Haas Status: resolved Milestone: Version: development version Type: enhancement Priority: minor Component: Changes (by Roland Haas): status: resolved (was open) -- Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2550/support-compiling…

1 0

#2550: support compiling in Raspberry Pi 4
by Roland Haas 08 Aug '21

08 Aug '21

#2550: support compiling in Raspberry Pi 4 Reporter: Roland Haas Status: open Milestone: Version: development version Type: enhancement Priority: minor Component: Comment (by Roland Haas): Applied as git hash [585ada59](https://bitbucket.org/cactuscode/cactus/commits/585ada594dd81c098… "Cactus: regenerate configure" of [cactus](https://bitbucket.org/cactuscode/cactus) -- Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2550/support-compiling…

1 0

#2550: support compiling in Raspberry Pi 4
by Erik Schnetter 08 Aug '21

08 Aug '21

#2550: support compiling in Raspberry Pi 4 Reporter: Roland Haas Status: open Milestone: Version: development version Type: enhancement Priority: minor Component: Comment (by Erik Schnetter): This looks good. Please apply. -- Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2550/support-compiling…

1 0

#2550: support compiling in Raspberry Pi 4
by Roland Haas 08 Aug '21

08 Aug '21

#2550: support compiling in Raspberry Pi 4 Reporter: Roland Haas Status: open Milestone: Version: development version Type: enhancement Priority: minor Component: Changes (by Roland Haas): status: open (was new) Comment (by Roland Haas): Please review. -- Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2550/support-compiling…

1 0

#2550: support compiling in Raspberry Pi 4
by Roland Haas 08 Aug '21

08 Aug '21

#2550: support compiling in Raspberry Pi 4 Reporter: Roland Haas Status: new Milestone: Version: development version Type: enhancement Priority: minor Component: The Pi4 \(and maybe 3 and 2 as well, I don’t own a 3 and compiling on my 2 is going to be painful\) use gnueabihf which the current known-architectures do not catch. Pull request [https://bitbucket.org/cactuscode/cactus/pull-requests/125/support-raspberry… adds code to recognize and and also fixes some reliance on GNU awk \(or so it seems\) rather than eg mawk \(in raspbian minimal\). With these I can compile ET\_2021\_05 and run both the 2 and 1 process testsuites without a single error. -- Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2550/support-compiling…

1 0

#2546: CarpetIOHDF5: Don't set "delta" attribute to zero for zero-width components
by Erik Schnetter 08 Aug '21

08 Aug '21

#2546: CarpetIOHDF5: Don't set "delta" attribute to zero for zero-width components Reporter: Erik Schnetter Status: closed Milestone: Version: development version Type: enhancement Priority: minor Component: Changes (by Erik Schnetter): status: closed (was resolved) -- Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2546/carpetiohdf5-dont…

1 0

#2546: CarpetIOHDF5: Don't set "delta" attribute to zero for zero-width components
by Erik Schnetter 08 Aug '21

08 Aug '21

#2546: CarpetIOHDF5: Don't set "delta" attribute to zero for zero-width components Reporter: Erik Schnetter Status: resolved Milestone: Version: development version Type: enhancement Priority: minor Component: Changes (by Erik Schnetter): status: resolved (was open) -- Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2546/carpetiohdf5-dont…

1 0

#2546: CarpetIOHDF5: Don't set "delta" attribute to zero for zero-width components
by Erik Schnetter 08 Aug '21

08 Aug '21

#2546: CarpetIOHDF5: Don't set "delta" attribute to zero for zero-width components Reporter: Erik Schnetter Status: open Milestone: Version: development version Type: enhancement Priority: minor Component: Comment (by Erik Schnetter): Applied in [https://bitbucket.org/eschnett/carpet/commits/73edd0639548fb158c5f196f647f2…. -- Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2546/carpetiohdf5-dont…

1 0

#2548: Compilation failure: "const" assignment in LoopControl/src/loopcontrol.cc
by Roland Haas 06 Aug '21

06 Aug '21

1 0

#2548: Compilation failure: "const" assignment in LoopControl/src/loopcontrol.cc
by Roland Haas 06 Aug '21

06 Aug '21

#2548: Compilation failure: "const" assignment in LoopControl/src/loopcontrol.cc Reporter: Bernard Kelly Status: open Milestone: ET_2021_05 Version: ET_2021_05 Type: bug Priority: major Component: EinsteinToolkit thorn Comment (by Roland Haas): I can compile fine with Intel 17 on Blue Waters if I use a modern enough C\+\+ library. However if I force it to use gcc 4.3 by using --gxx-name then I get the same error that you see: ``` COMPILING Carpet/LoopControl/src/loopcontrol.cc /mnt/a/u/staff/rhaas/ET_Next/arrangements/Carpet/LoopControl/src/loopcontrol.cc(70): error: function call must have a constant value in a constant expression lc_random.max() - lc_random.min() + 1; ^ /mnt/a/u/staff/rhaas/ET_Next/arrangements/Carpet/LoopControl/src/loopcontrol.cc(70): error: function call must have a constant value in a constant expression lc_random.max() - lc_random.min() + 1; ^ compilation aborted for /mnt/a/u/staff/rhaas/ET_Next/configs/sim-intel/build/LoopControl/loopcontrol.cc (code 2) ``` so both this and #2547 are due to a too old stdc\+\+ library. Trying to work around this by using `minstd_rand::max()` does not work and gives: ```plaintext COMPILING Carpet/LoopControl/src/loopcontrol.cc /mnt/a/u/staff/rhaas/ET_Next/arrangements/Carpet/LoopControl/src/loopcontrol.cc(70): error: a nonstatic member reference must be relative to a specific object minstd_rand::max() - minstd_rand::min() + 1; ^ /mnt/a/u/staff/rhaas/ET_Next/arrangements/Carpet/LoopControl/src/loopcontrol.cc(70): error: a nonstatic member reference must be relative to a specific object minstd_rand::max() - minstd_rand::min() + 1; ^ compilation aborted for /mnt/a/u/staff/rhaas/ET_Next/configs/sim-intel/build/LoopControl/loopcontrol.cc (code 2) ``` indicating that the library does not declare the function correctly as a static class function and indeed digging into the header files on BlueWaters shows it to be defined as: ```c++ result_type max() const { return __m - 1; } ``` A workaround would be to remove `constexpr` from this line of code but given that thorn `Vectors` also fails with due to the same old C\+\+ library, I am not sure if the workaround would help a lot \(would only help in situations where `LoopControl` but not `Vectors` is used\). -- Ticket URL: https://bitbucket.org/einsteintoolkit/tickets/issues/2548/compilation-failu…

1 0

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

Trac