Commit Graph

4426 Commits

Author SHA1 Message Date
Sharan Roongta fa974574af Merge branch 'consistent-damage-parameter-names' into 'development'
Consistent damage parameter names

See merge request damask/DAMASK!593
2022-06-04 11:27:28 +00:00
Martin Diehl 1b5194288b polished tests, consistent variable name 2022-06-04 07:26:45 +02:00
Franz Roters 5155a3f958 Merge branch 'MPI-FFTW-fix' into 'development'
bugfix: prevent segmentation fault

Closes #192

See merge request damask/DAMASK!595
2022-06-02 16:13:24 +00:00
Martin Diehl b8530bbcc9 Merge branch 'implicit-none-in-interface' into 'development'
implicit none does not propagate

See merge request damask/DAMASK!596
2022-05-31 12:06:03 +00:00
Philip Eisenlohr 3d10d29254 Merge branch 'warning-error-cleanup' into 'development'
more flexible error/warning handling

See merge request damask/DAMASK!594
2022-05-28 16:56:53 +00:00
Martin Diehl d858d753e3 implicit none does not propagate
each subroutine/function is independent and not part of a scope
2022-05-27 22:38:47 +02:00
Martin Diehl 6fa87b4038 not needed anymore 2022-05-27 20:58:18 +02:00
Martin Diehl 2bfcc4d0b8 standardized reporting 2022-05-27 20:53:16 +02:00
Martin Diehl b56ded62d9 encapsulate data, prefix only public functions 2022-05-27 18:01:40 +02:00
Philip Eisenlohr f85e8a3ff4 more precise warning for fftw_plan_mode 2022-05-27 12:00:53 -04:00
Philip Eisenlohr 75bd643258 more precise warning for fftw_plan_mode 2022-05-27 11:58:10 -04:00
Philip Eisenlohr d66c777d6e introduced "IO_panel" to write statements to STDERR 2022-05-27 11:47:14 -04:00
Martin Diehl dc59beb59c following advince from FFTW3 manual
silences a lot of gfortran warnings about unused variables
2022-05-27 09:54:00 +02:00
Martin Diehl 064b9c5cc4 bugfix: prevent segmentation fault
need to rely on FFTWs calculation for amount of storage needed
2022-05-27 09:53:04 +02:00
Martin Diehl 6c6b3e64b3 more flexible reporting 2022-05-27 09:38:47 +02:00
Martin Diehl cfa2be44e9 internal errors should trigger error stop 2022-05-27 08:48:42 +02:00
Martin Diehl f7d23ad151 internal variable, no prefix needed 2022-05-27 08:40:01 +02:00
Martin Diehl f31dd8cbe8 unknown error/warning numbers are internal errors 2022-05-27 08:34:52 +02:00
Martin Diehl f90523eb07 more flexible reporting 2022-05-27 08:33:50 +02:00
Martin Diehl 5f0ae4fe7a user errors are catched earlier
internal errors (essentially programming faults) should stop with 'error
stop'
2022-05-27 06:41:55 +02:00
Martin Diehl 79fad5ca19 IP was only used for an argument that is NOT the IP ID 2022-05-27 06:41:15 +02:00
Martin Diehl 1d4c476cd4 collect errors for more helpful message 2022-05-27 06:30:25 +02:00
Martin Diehl d5db083fec more convenient to see all invalid parameters 2022-05-27 00:25:25 +02:00
Martin Diehl 7a23699aa4 not needed
error occurs in <phase ... init>
2022-05-27 00:24:40 +02:00
Martin Diehl 724c4fcf5d not used
ip,el,g is anyways not flexible enough. Remove unused arguments before
developing a more flexible approach
2022-05-27 00:23:51 +02:00
Martin Diehl 19b78cb647 easier to read 2022-05-26 23:29:06 +02:00
Martin Diehl 82cad00f81 simplified interface 2022-05-26 19:07:23 +02:00
Martin Diehl a514d30060 standardized names 2022-05-26 19:03:29 +02:00
Martin Diehl 77f586b1d7 Merge remote-tracking branch 'origin/development' into consistent-damage-parameter-names 2022-05-22 23:35:42 +02:00
Martin Diehl f42bb5d175 mitigate rounding errors 2022-05-20 06:43:32 +02:00
Martin Diehl 491e2ec0b2 avoid negative zero when not needed 2022-05-20 06:30:07 +02:00
Martin Diehl 6b6c2548c5 Merge branch 'development' into initial-V_e 2022-05-19 17:03:09 +02:00
Martin Diehl 41a732f62d assumed rank solution gives wrong results #194 2022-05-19 16:38:37 +02:00
Martin Diehl 556d9d840e specifying V_e is more natural than F_i 2022-05-19 01:15:17 +02:00
Daniel Otto de Mentock 85fa92ee1f Merge branch 'use-openMP-module' into 'development'
using the module is the recommended approach

See merge request damask/DAMASK!582
2022-05-17 12:57:15 +00:00
Daniel Otto de Mentock 5f3203cecf Merge branch 'element-cleanup' into 'development'
only relevant for Marc

See merge request damask/DAMASK!583
2022-05-17 12:56:50 +00:00
Martin Diehl c0bb576117 Merge remote-tracking branch 'origin/development' into sigint-to-quit 2022-05-13 11:58:18 +02:00
Martin Diehl 3f85027b9e better use SIGINT for actions triggered by the user
SIGTERM is send by MPI if one process fails, catching it results in
deadlocks
2022-05-13 11:58:03 +02:00
Martin Diehl 6b17787be1 only relevant for Marc 2022-05-13 10:02:02 +02:00
Martin Diehl 05d3a92944 using the module is the recommended approach 2022-05-13 08:43:24 +02:00
Martin Diehl 6f7e8491bc Merge branch 'Marc2022.1' into 'development'
Marc2022.1

See merge request damask/DAMASK!579
2022-05-12 20:38:25 +00:00
Martin Diehl 0b6ead1a48 avoid console-spam during initialization 2022-05-12 15:36:12 +02:00
Martin Diehl 53796fce7a trustworthy reference 2022-05-08 20:10:12 +02:00
Martin Diehl b376b10b7a classic Intel gives FPE 2022-05-08 17:47:20 +02:00
Martin Diehl 10d8a63cb6 not used
and current code is not good (not using Horner scheme)
2022-05-08 17:18:15 +02:00
Martin Diehl d713026f7e fast evaluation for (most common) case of constant 2022-05-08 16:49:30 +02:00
Martin Diehl b2052cb3c7 less strict tolerances
values can be in the order of 1e5, so 1e-6 precision is not too bad
2022-05-08 16:41:19 +02:00
Martin Diehl 8f9fbb30e5 use fused multiply-add where possible
only possible for Intel compiler
2022-05-08 15:48:13 +02:00
Martin Diehl 5b8e5591ed Merge remote-tracking branch 'origin/development' into order4-polynomial 2022-05-08 14:35:48 +02:00
Martin Diehl 240426402c using (faster) Horner evaluation
https://rosettacode.org/wiki/Horner%27s_rule_for_polynomial_evaluation
2022-05-08 13:32:14 +02:00