| 5e482e8 | Riccardo De Maria | 24 March 2015, 14:18:50 UTC | Update from Kyrre et al. What is new: * DYNK block — documented on TWIKI: https://twiki.cern.ch/twiki/bin/view/LHCAtHome/SixTrackDoc#Dynamic_Kicks_DYNK_input_block * DUMP block — documented on TWIKI: https://twiki.cern.ch/twiki/bin/view/LHCAtHome/SixTrackDoc#Dumping_of_beam_population_DUMP * DBDCUM for computing s in every element (used for outputting, need to be always in place) (Alessio) * Removed RIPP - functionality replaced by DYNK & Documented on TWIKI * In collimat: - Make sure all elements are always in place, i.e. kz=53/54/57/58/59/60/61/62 (crabs), 55 (dipedge), 56 (solenoid) - Bugfixes for HDF5 (on top of what has already been released) * Fixes in LINOPT (Alessio): - De-duplicate writelin calls - Always initialize variable “k” - Use bez(jk) (single element inside block) instead of bez(ix) (block name) for output - Always write final bez after block - Thin tracking+block that is not pure drift -> treat properly - Fix trom06 - Correctly handle crabs in collimat - Element kz > 28: Dump linear optics - No jk outside BLOCK loop kz=11: Dump * Remove nonsense "if(abs(kp(ix).eq. 0 OR 1) -> GOTO 650" from thin6d * Move file closing code into subroutine “close” (Alessio) * In make_six: - Check code using beauty for missed astuce “+ei” - Check that cr and collimat is not enabled at the same time - This version / already on SVN * h5 in exec name if built with HDF5 * Bugfixes for linking etc. * In beauty: - Return error to make_six if astuce not closing * Astuce files sixve.ast & track.ast - Add support for decks dynkancil, dynktrack, cadcum, close Main tests that has been ran, all successfully: * Check DYNKs SET functionality for kz=+1, -1, +2, -2, +3, +4, (+/-1 to +/- 10 shares most of the code), 23 (voltage/amplitude/phase) * Check DYNKs FUN functionality / function generation. * Check Checkpoint/Restart functionality for DYNK and DUMP, including correct truncation and continuation of output files. * Check collimation examples * Check HDF5 * Check that RIPP is exactly reproduced * Check that SIXTRAC/DYNK/DUMP works with Intel fortran, both standard and checkpoint/restart version. * Sixtest tests (run_pro and run_kill): bnl exact crabamp javier thick4 lost tilt dipedge bb notilt lostnotilt lostevery s316 eric frs frs60 bbe51 bbe52 bbe571ib0 distance All in all, I think the test coverage was quite extensive, and in total we have probably removed more bugs than we have added. In total, the code grew by 4943 lines of code (~33 LOC/day), and we also deleted or modified many many many other lines. Different executables successfully produced and tested: ./make_six gfortran collimat ./make_six gfortran collimat hdf5 ./make_six gfortran crlibm ./make_six gfortran crlibm cr ./make_six ifort crlibm ./make_six ifort crlibm cr Most of the test notes can be found at (warning: messy and not really intended for general consumption): https://www.evernote.com/l/AHNpfE_-KixNka5uH8mK8deDXuSUQ3cm_CA https://www.evernote.com/l/AHNpfE_-KixNka5uH8mK8deDXuSUQ3cm_CA https://www.evernote.com/l/AHNxnoAJe15EOax8Z8A49WfhtBmU-OCDpa4 https://www.evernote.com/l/AHPvDfsLTjNF77wIqiFH0M5an9I1A6kHZgo A complete modified version of the SVN repo folder, comparable to trunk/Sixtrack, can be found at: https://github.com/KFubuki/time_dependant_kicks/tree/devel_upstreammerge2 -> subfolder sixtrack_SVN/sixtrack Note: This commit was originally two commits in the old SVN repository. There, the files where originally committed with DOS line endings, making the diff go crazy. This was undone in a second commit, here those two are squashed into one. Also, the second commit included count.awk, which was forgotten in the first commit. Affected files: M ast_mask/sixve.ast M ast_mask/track.ast M beauty/count.awk M make_six M sixtrack.s To see a proper diff (i.e. not all lines), use: svn diff -x "-w --ignore-eol-style" ast_mask/sixve.ast ast_mask/track.ast beauty/count.awk make_six sixtrack.s | 19 August 2015, 09:26:23 UTC |
| 43dcc92 | Riccardo De Maria | 11 March 2015, 14:30:04 UTC | fixing sse2 sse3 option in windows for new Makefile | 11 March 2015, 14:30:04 UTC |
| aa8cfcb | Alvaro Gonzalez Alvarez | 04 March 2015, 15:50:04 UTC | More test | 04 March 2015, 15:50:04 UTC |
| 1ee677d | Svn Administrator | 04 March 2015, 14:25:44 UTC | Test HTTPS with agonzale INC0744512 | 04 March 2015, 14:25:44 UTC |
| e38d635 | Svn Administrator | 04 March 2015, 14:13:26 UTC | Second test HTTPS | 04 March 2015, 14:13:26 UTC |
| a17bc41 | Alvaro Gonzalez Alvarez | 04 March 2015, 14:09:52 UTC | Test INC0744512 | 04 March 2015, 14:09:52 UTC |
| 1412737 | Kyrre Ness Sjobak | 04 March 2015, 14:02:23 UTC | HDF5 fixes from Regina Kwee-Hinzmann; This fixes bugs in the HDF5 module, which prevented compilation of HDF5, created runtime errors, and created incorrect physics results (counter was incremented in the wrong place in the code). This module is enabled by adding the "hdf5" flag to make_six (requiring collimat). It was tested by Regina and Kyrre, and found to produce correct output. When used, the CPU time and disk space requirements for tracks2.dat (replaced by tracks2.h5) go down by a significant factor. | 04 March 2015, 14:02:23 UTC |
| 27480a1 | Kyrre Ness Sjobak | 04 March 2015, 13:49:59 UTC | Reverting WIRE changes (back to r184) | 04 March 2015, 13:49:59 UTC |
| ad04f94 | Andrei Patapenka | 02 March 2015, 16:27:34 UTC | wire element added | 02 March 2015, 16:27:34 UTC |
| e18c877 | Eric Mcintosh | 18 February 2015, 16:04:33 UTC | SixTrack Version: 4.5.19 Eric for Riccardo -- Uncommented the BUG fix? go to 360 in LIMI block -- Fixed undefine variable time in sixda -- Added SixTrack_da link for Frank McIntosh , 14th February, 2015 | 18 February 2015, 16:04:33 UTC |
| 820761d | Riccardo De Maria | 26 January 2015, 13:57:05 UTC | small update on Makefile | 26 January 2015, 13:57:05 UTC |
| d74b7a5 | Riccardo De Maria | 26 January 2015, 09:57:58 UTC | reverting 180 make_six sixtrack.s | 26 January 2015, 09:57:58 UTC |
| 5022f26 | Riccardo De Maria | 26 January 2015, 09:54:58 UTC | adding new makefile | 26 January 2015, 09:54:58 UTC |
| 618a087 | Riccardo De Maria | 22 August 2014, 12:02:24 UTC | Improving sixsrc.py for html page and fortran generation | 22 August 2014, 12:02:24 UTC |
| 26e85ad | Eric Mcintosh | 14 May 2014, 08:24:28 UTC | SixTrack Version: 4.5.18 Adriana -- Collimation changes -- Line 29951: formatting in the output file 38 (tracks2.dat) -- Lines 57712 and 57715: input data for tracking (carbon density and free path length respectively) -- Lines 58583 and 58584 : correction of the calculation of momentum and energy in the subroutine CALC_ION_LOSS Adriana and McIntosh , 14th May, 2014 | 14 May 2014, 08:24:28 UTC |
| 25c8758 | Eric Mcintosh | 09 May 2014, 07:41:07 UTC | SixTrack Version: 4.5.17 Eric -- Removed boinc_zipitall calls -- added CLOSE 6 and 93 to routine abend McIntosh , 9th May, 2014 | 09 May 2014, 07:41:07 UTC |
| e3007ab | Eric Mcintosh | 08 May 2014, 15:24:45 UTC | SixTrack Version: 4.5.16 Eric -- Going back to Version 4513 (best so far) -- Added Riccardo's physics from SVN Version 176 From Javier: increase bb 6d max number of slices, change logic for lhc option (strong beam optics): lhc=0 symmetric optics; lhc=1 antysymmetric; lhc=2, like lhc=0 but sigma 11,33,55 overwriten McIntosh, Riccardo, Javier, 8th May, 2014 | 08 May 2014, 15:24:45 UTC |
| 04a57fa | Riccardo De Maria | 07 May 2014, 07:54:25 UTC | From Javier: increase bb 6d max number of slices, change logic for lhc option (strong beam optics): lhc=0 symmetric optics; lhc=1 antysymmetric; lhc=2, like lhc=0 but sigma 11,33,55 overwriten | 07 May 2014, 07:54:25 UTC |
| 8523149 | Eric Mcintosh | 04 May 2014, 14:41:08 UTC | SixTrack Version: 4.5.15 Eric -- Rebuilding with absolutely NO zip stuff -- Modified the Makefile.boinc.template and README -- boinc_api_fortran.cpp (sixtrack.s of course) McIntosh , 3rd May, 2014 | 04 May 2014, 14:41:08 UTC |
| 89b203b | Eric Mcintosh | 03 May 2014, 18:15:16 UTC | SixTrack Version: 4.5.14 Eric -- Rebuilding WITHOUT any zip calls but with the lib boinc_zip -- Just replaced the Makefile.boinc.template McIntosh , 3rd May, 2014 | 03 May 2014, 18:15:16 UTC |
| 28a9f17 | Eric Mcintosh | 03 May 2014, 17:38:06 UTC | SixTrack Version: 4.5.14 Eric -- Rebuilding WITHOUT any zip stuff. McIntosh , 3rd May, 2014 | 03 May 2014, 17:38:06 UTC |
| 25877eb | Eric Mcintosh | 30 April 2014, 15:01:31 UTC | SixTrack Version: 4.5.13 Adriana Modifications to Collimation Routines -- Add output file Coll_Scatter.dat (within the Jaw subroutine) with all interactions at collimators -- Add output file FLUKA_impacts_all.dat with all absorptions at collimators -- Add Single Diffractive events in FLUKA_impacts.dat -- Add Pencil Beam type 3 (R.Bruce): The new pencil beam model uses the existing routines for annular halo generation and samples a matched halo at the face or end of any given collimator, with all particles hitting the jaws at a variable depth. There is no unphysical tilt of the jaws with the beam envelope as in the previous pencil beam -- Replacement of the scattering routine with an upgraded version (C.Tambasco and B.Salvachua). At the moment the old routine is commented out, to be seen if to delete it or put it as a compilation option. The new routine includes: A different Carbon density that better fits the real collimator jaw material. The addition of the logarithmic term in the in the rms angle equation of the multiple Coulomb scattering has been added. Implementation of the Bethe-Bloch equation (in place of a fix value) to better estimate the ionization losses. Computation of new fits of experimental data (TOTEM collaboration) to better extrapolate the proton-proton elastic and total cross sections. Updating of the proton-proton single diffractive cross section considering a recent parametrization based on the renormalized pomeron flux exchange. Updating of the proton-nucleus inelastic and total cross sections considering the new available data from the Particle Data Group. -- Further modification to the computation of ionisation losses taking into account the tale of the Landau distribution, since Bethe-Bloch is not valid for beta.gamma>1000 (D. Mirarchi). McIntosh/Adriana, 30th April, 2014 SixTrack Version: 4.5.12 Eric -- Just for testing new build by Xavier (not committed) McIntosh , 30th April, 2014 | 30 April 2014, 15:01:31 UTC |
| b510fe5 | Eric Mcintosh | 25 April 2014, 13:17:46 UTC | SixTrack Version: 4.5.11 Eric -- Put back call to unzip and zipitall -- Waiting for Riccardo's changes -- Commit for a build anyway to use on boinctest McIntosh , 25th April, 2014 | 25 April 2014, 13:17:46 UTC |
| e8510b3 | Eric Mcintosh | 20 April 2014, 22:02:18 UTC | SixTrack Version: 4.5.10 Eric -- NO call boinc_zipitall (missed one in 4.5.09) -- Re-instated call boinc_time_to checkpoint -- Changed CR ABEND Message McIntosh , 14th April, 2014 | 20 April 2014, 22:02:18 UTC |
| 88e1eeb | Eric Mcintosh | 26 March 2014, 12:20:06 UTC | SixTrack Version: 4.5.09 Eric -- Based on 4508 but NO call boinc_unzip -- Based on 4508 but NO call boinc_zipitall -- ???Just to test ifort 2013_sp1.2.144??? McIntosh , 26th March, 2014 | 26 March 2014, 12:20:06 UTC |
| 9737a17 | Eric Mcintosh | 07 March 2014, 13:41:45 UTC | SixTrack Version: 4.5.08 Eric -- Added parentheses to FOXY EXACT code to match the actual computation (and removed redundant parentheses) McIntosh , 7th March, 2014 | 07 March 2014, 13:41:45 UTC |
| 466381c | Eric Mcintosh | 06 March 2014, 08:05:57 UTC | SixTrack Version: 4.5.06 Eric and Mattias -- Implemented "exact" drift from Mattias -- Eric changed to just use sqrt McIntosh 6th March, 2014 SixTrack Version: 4.5.05 Riccardo -- Added licence info Riccardo 24th February, 2014 | 06 March 2014, 08:05:57 UTC |
| bb6f113 | Riccardo De Maria | 27 February 2014, 14:25:47 UTC | fix a bug and read all fort.90 | 27 February 2014, 14:25:47 UTC |
| efae31c | Riccardo De Maria | 21 February 2014, 09:48:52 UTC | Adding python script to read binary files fort.90 fort.89 ... produced by Sixtrack | 21 February 2014, 09:48:52 UTC |
| 24d353d | Riccardo De Maria | 21 February 2014, 09:42:15 UTC | Adding Exact drift flag to track with a more accurate equation for drift spaces. An option iexact is added at the end of the tracking parameter block | 21 February 2014, 09:42:15 UTC |
| 1a0cc03 | Riccardo De Maria | 05 February 2014, 21:44:55 UTC | removed space in between + and cd | 05 February 2014, 21:44:55 UTC |
| 7be138a | Riccardo De Maria | 31 January 2014, 10:18:22 UTC | adding license informations | 31 January 2014, 10:18:22 UTC |
| 9b423d2 | Riccardo De Maria | 31 January 2014, 10:15:48 UTC | adding license informations | 31 January 2014, 10:15:48 UTC |
| def9db5 | Riccardo De Maria | 31 January 2014, 09:59:04 UTC | adding license informations | 31 January 2014, 09:59:04 UTC |
| af4ad0e | Eric Mcintosh | 24 January 2014, 16:45:51 UTC | SixTrack Version: 4.5.04 Eric -- Removed +if crlibm from the new vars posttime, pretime and tottime McIntosh 24th January, 2014 | 24 January 2014, 16:45:51 UTC |
| 0593148 | Riccardo De Maria | 20 January 2014, 13:27:08 UTC | added more detail on the README for 4.4.98 version | 20 January 2014, 13:27:08 UTC |
| 6386369 | Eric Mcintosh | 11 January 2014, 11:26:49 UTC | SixTrack Version: 4.5.03 Eric -- Fixed reported CPU times. Finally used new names, removed crtime0/crtime1 and used time3 in crpoint (as well as post) for crtime3 McIntosh 11th January, 2014 SixTrack Version: 4.5.02 Eric -- Now try and correct the turns reported in fort.6 -- Should also try and fix CPU time reports -- Leaving debug code until I check LOST particles McIntosh 6th January, 2014 SixTrack Version: 4.5.01 Eric -- Now try and correct sumda(22/23) and Total Turn(s) Tricky, because of nnuml and multiple restarts Recomputing napxto in maincr and sumda(22/23) in postpr Leaving debug code until I check LOST particles McIntosh 5th January, 2014 | 11 January 2014, 11:26:49 UTC |
| 90d090b | Eric Mcintosh | 05 January 2014, 16:16:04 UTC | SixTrack Version: 4.5.01 Eric -- Now try and correct sumda(22/23) and Total Turn(s) Tricky, because of nnuml and multiple restarts Recomputing napxto in maincr and sumda(22/23) in postpr Leaving debug code until I check LOST particles McIntosh 5th January, 2014 M sixtrack.s M README | 05 January 2014, 16:16:04 UTC |
| f35d3ed | Eric Mcintosh | 30 December 2013, 14:52:19 UTC | SixTrack Version: 4.4.99 Eric -- Removed the now hopefully unnecessary checkpoint of OIDPSV McIntosh 29th December, 2013 SixTrack Version: 4.4.98 Riccardo -- Fixes from Riccardo for the OIDPSV problem etc -- based on Version 4.4.97 McIntosh 28th December, 2013 | 30 December 2013, 14:52:19 UTC |
| 9309b39 | Eric Mcintosh | 26 December 2013, 16:18:24 UTC | SixTrack Version: 4.4.97 Eric -- Default numlcp to 1000 -- Reading numlcp and numlmax as 10th/11th items in fort.3 Tracking Parameters Line 1 (was done already) but remember to set 8th/9th items niu(1 and 2) to 0 and re-create Sixin.zip -- Commented out call system in crpoint to check timing of C/R McIntosh 26th December, 2013 SixTrack Version: 4.4.96 Eric -- Added a fort.93 log message to clarify start up -- Fixed missing +ei at 611 for =if boinc McIntosh 23rd December, 2013 SixTrack Version: 4.4.95 Eric -- Put back and added checks "unzip of Sixin.zip" McIntosh 23rd December, 2013 SixTrack Version: 4.4.94 Eric -- Commented out the "unzip of Sixin.zip" for tests with Igor McIntosh 23rd December, 2013 | 26 December 2013, 16:18:24 UTC |
| 0c886ac | Eric Mcintosh | 19 December 2013, 13:09:02 UTC | SixTrack Version: 4.4.93 Eric -- FIXED the mywwerf problem by using a CONSTANT P -- Found that unzip Sixin.zip was over-writing fort.93 etc -- Decided to just re-position fort.93 again and empty fort.92 -- WRITEBIN not to write message if +bnlelens -- Do NOT enable_xp in dabnew.s for Lahey lf95 -- Enable for other compilers if +crlibm, but I use the binary fort.111 -- In sixtrack.s it is done only for +fio which isn't supported by lf95 (should carefully check when testing fio) -- Commented out some +if debug commands -- Riccardo will add fix for OIDPSV and then I will test -- See README for all changes since Version 4462 McIntosh 19th December, 2013 | 19 December 2013, 13:09:02 UTC |
| db72ee8 | Laurent Deniau | 12 December 2013, 14:07:08 UTC | compilation of tools added using mad-x build system | 12 December 2013, 14:07:08 UTC |
| 99d4b57 | Eric Mcintosh | 13 November 2013, 16:05:45 UTC | SixTrack Version: 4.4.91 Eric -- Merged version of sixtrack.s.v4490 with Yngve's -- collimations mods based on v4463 McIntosh 13th November, 2013 SixTrack Version: 4.4.90 Eric -- Adding oidpsv(j) to C/R as a TEMPORARY??? fix for test javier. -- Still to run regression tests before testing new NUMLMAX NUMLCP -- Modified from Version 4.4.87. -- See README for full info. McIntosh 12th November, 2013 | 13 November 2013, 16:05:45 UTC |
| d7c230f | Yngve Levinsen | 01 October 2013, 07:08:11 UTC | not writing debugfile.txt | 01 October 2013, 07:08:11 UTC |
| cad6944 | Yngve Levinsen | 01 October 2013, 07:08:08 UTC | in beamgas module if a long list of dpmjet events is provided, skip the rest of the lines in the file | 01 October 2013, 07:08:08 UTC |
| 9c1d0e5 | Eric Mcintosh | 31 August 2013, 19:13:37 UTC | SixTrack Version: 4.4.63 Eric -- Just added a call to boincrf for fort.zip for Windows. McIntosh 31st August, 2013 | 31 August 2013, 19:13:37 UTC |
| e175fb2 | Eric Mcintosh | 31 August 2013, 19:12:25 UTC | This is the REAL sixtrack.s.v4462 | 31 August 2013, 19:12:25 UTC |
| 2e27c6f | Eric Mcintosh | 31 August 2013, 19:07:28 UTC | SixTrack Version: 4.4.62 Eric -- Fixed nasty bug in daten iclr.eq.2 concerning exz in tracking input -- Fixing all "formatted read" for fio and _xp stuff. -- dabnew.s and sixtrack.s. if fio (but NOT Lahey lf95) then enable/disable _xp before/after the READ "NEAREST" fio overrules crlibm for formatted input McIntosh 29th August, 2013 | 31 August 2013, 19:07:28 UTC |
| 1d3dc97 | Riccardo De Maria | 22 August 2013, 16:11:18 UTC | adding boinc address to Makefile.boinc.template for linux | 22 August 2013, 16:11:18 UTC |
| be6c397 | Eric Mcintosh | 21 August 2013, 13:40:18 UTC | SixTrack Version: 4.4.61 Eric -- Added code to daten for crlibm and read of xstr McIntosh 16th August, 2013 | 21 August 2013, 13:40:18 UTC |
| a89b4ba | Eric Mcintosh | 12 August 2013, 14:10:55 UTC | SixTrack Version: 4.4.52 eric and Igor and Laurent -- Just fixed the type definitions in myboinc.f which is for internal use only. McIntosh 12th August, 2013 | 12 August 2013, 14:10:55 UTC |
| d33db0d | Eric Mcintosh | 12 August 2013, 13:37:36 UTC | SixTrack Version: 4.4.52 eric and Igor and Laurent -- Deleted a redundant +ca cseeds (came in v4443!)??? McIntosh 4th July, 2013 | 12 August 2013, 13:37:36 UTC |
| e7fc706 | Eric Mcintosh | 01 July 2013, 15:11:03 UTC | SixTrack Version: 4.4.51 eric -- Really fixes checkpoint and adds log message to fort.93 for the last checkpoint -- Adds parameters 8,0 to boinc_zip call (string lengths) McIntosh 1st July, 2013 | 01 July 2013, 15:11:03 UTC |
| 20ca74c | Eric Mcintosh | 29 June 2013, 15:37:44 UTC | SixTrack Version: 4.4.50 eric -- Always checkpoint (do NOT call BOINC time to checkpoint) -- Do a checkpoint when all turns completed before post-processing (unless restarted from that checkpoint) in order to facilitate ten million turns etc -- Replaced g77 option by pgf90 (preliminary) McIntosh 27th June, 2013 | 29 June 2013, 15:37:44 UTC |
| 45e7eb4 | Riccardo De Maria | 04 December 2012, 11:32:30 UTC | fix sixsrc.py lxplus deps | 04 December 2012, 11:32:30 UTC |
| 949b7e3 | Riccardo De Maria | 04 December 2012, 11:29:43 UTC | make sixsrc.py executable | 04 December 2012, 11:29:43 UTC |
| b926be8 | Riccardo De Maria | 04 December 2012, 11:01:02 UTC | Adding python code browser | 04 December 2012, 11:01:02 UTC |
| 650c798 | Eric Mcintosh | 26 September 2012, 15:15:56 UTC | SixTrack Version: 4.4.46 eric and laurent -- Now call boinc_zip_ directly and use boinc_fraction_done instead of boinc_sixtrack_progress -- Added boinc_fraction_done etc to myboinc.f McIntosh 26th September, 2012 | 26 September 2012, 15:15:56 UTC |
| 8702c28 | Eric Mcintosh | 26 September 2012, 14:05:19 UTC | SixTrack Version: 4.4.45 eric -- Just made sure MMAC is zero (when NMS=0) for lost particles McIntosh 26th September, 2012 | 26 September 2012, 14:05:19 UTC |
| 91835d1 | Eric Mcintosh | 26 September 2012, 12:42:07 UTC | SixTrack Version: 4.4.44 eric and laurent -- Removed the integer parameter from call to boinc_init McIntosh 22nd September, 2012 | 26 September 2012, 12:42:07 UTC |
| 75b0db8 | Eric Mcintosh | 25 September 2012, 13:08:53 UTC | SixTrack Version: 4.4.43 eric and javier -- Javier's mods for new elements called allMULT -- Fix for sumpos for mmac and nms -- make_six new SSE defaults and checks and removed set -x McIntosh 22nd September, 2012 | 25 September 2012, 13:08:53 UTC |
| 7705b55 | Eric Mcintosh | 22 July 2012, 13:27:19 UTC | SixTrack Version: 4.4.42 eric -- Fixed SUMDA 59/60 for all particles lost early McIntosh 22nd July, 2012 | 22 July 2012, 13:27:19 UTC |
| 07037ba | Eric Mcintosh | 05 July 2012, 12:51:08 UTC | Added Makefile.boinc.mac.template and fixed make_six for MAC McIntosh 5th July, 2012 | 05 July 2012, 12:51:08 UTC |
| 2942e1b | Eric Mcintosh | 03 July 2012, 15:15:11 UTC | SixTrack Version: 4.4.41 eric -- Quickly changed to SSE3 (but still called SSE2) for Igor for BOINC -- Added total turns and CPU time to fort.10 words 59/60 -- uname -m instead of fs sysname -- sse3 instead of sse2 and nothing for Macs McIntosh 3rd July, 2012 | 03 July 2012, 15:15:11 UTC |
| a3140dc | Eric Mcintosh | 22 June 2012, 12:26:20 UTC | SixTrack Version: 4.4.40 eric -- Same SixTrack but new make_six for MacOS -- Defaults to ifort instead of lf95 -- uname -m instead of fs sysname -- sse3 instead of sse2 for Macs -- enable_xp.c/disable_xp.c dummied on ifdef APPLE -- Comma separated map options for FCLMAP -- NO static linking for Mac -- Fixed bug with make_six options and added lf95 McIntosh 22nd June, 2012 | 22 June 2012, 12:26:20 UTC |
| e4836d5 | Eric Mcintosh | 18 June 2012, 14:49:41 UTC | Just to try and clean up new/old entries. Eric. | 18 June 2012, 14:49:41 UTC |
| 10a19e5 | Eric Mcintosh | 18 June 2012, 14:40:49 UTC | SixTrack Version: 4.4.40 eric -- Put write 210 (fort.10 HEX) under +if debug -- Added the crlibm .f routines (acos_rn.f asin_rn.f atan2_rn.f) -- to sixtrack.s -- make_six (for MacOS), boinc_api_fortran[.windows] -- and new make_windows with new templates and -- flags for +/-SSE2 McIntosh 18th June, 2012 | 18 June 2012, 14:40:49 UTC |
| 82d914e | Eric Mcintosh | 20 May 2012, 11:55:02 UTC | SixTrack Version: 4.4.35 eric -- Just a commit to add dtoa_c.c, Makefile.bat, dafor and astuce. McIntosh 20th May, 2012 | 20 May 2012, 11:55:02 UTC |
| c987927 | Eric Mcintosh | 20 May 2012, 11:28:22 UTC | SixTrack Version: 4.4.35 eric and Harry -- based on 4.4.34 -- Javier's mods for crabamp with Harry's fixes labelled hr13 -- Removed a double assignment to ferrno in round_near.c McIntosh 18th May, 2012 SixTrack Version: 4.4.34 eric -- Fixed assignment of data to sigma0 and qw0 in daten. -- Fixed a few problems with g77 and concatenation in new daten subs splitfld, spliterr, rounderr. -- Unfixed initialisation of DA variables in dafor.f. -- Not committed McIntosh 28th April, 2012 | 20 May 2012, 11:28:22 UTC |
| 6513a24 | Frank Schmidt | 19 April 2012, 08:37:30 UTC | Missing routine dtostr Frank & Eric | 19 April 2012, 08:37:30 UTC |
| 5facb3c | Eric Mcintosh | 11 April 2012, 15:55:45 UTC | SixTrack Version: 4.4.33 eric -- based on 4.4.11 -- Uses strtod and sprintf and fround for all input variables and for formatted write of fort.10. This is the first 0 ULP version giving identical results on Linux with 4 different Fortran compilers. McIntosh 11th April, 2012 | 11 April 2012, 15:55:45 UTC |
| 3b0deec | Eric Mcintosh | 25 March 2012, 04:54:20 UTC | SixTrack Version: 4.4.11 eric McIntosh 8th February, 2012 -- based on 4.4.01 (never committed) -- Uses round_ulp from Laurent roundnulp( ,1024) only for the same variables, extaux(k), alignx|z and tilt McIntosh 25th March, 2012 | 25 March 2012, 04:54:20 UTC |
| 21f0190 | Frank Schmidt | 12 January 2012, 16:19:26 UTC | version 4.2.16 on 09.01.2012 | 12 January 2012, 16:19:26 UTC |
| 558c143 | Frank Schmidt | 09 January 2012, 15:19:22 UTC | 1) In the "iteration errors" input block and on the fourth input line the parameters aper(1) and aper(2) can be changed from their default value of 1m. 2) In the subroutine "umlauf" the aperture is now being checked against aper(1)/aper(2) which can now be modified to larger values to avoid fake crashes. | 09 January 2012, 15:19:22 UTC |
| e8146dc | Yngve Levinsen | 29 September 2011, 16:08:54 UTC | This is highly related to revision 115. Apparently the bug was present in two locations, not one. Someone should seriously consider to refactor some of this code... | 29 September 2011, 16:08:54 UTC |
| 1b7555f | Yngve Levinsen | 29 September 2011, 14:49:11 UTC | updated readme moved hdf5 libraries to my scratch0 where I have more space.. | 29 September 2011, 14:49:11 UTC |
| a391deb | Yngve Levinsen | 29 September 2011, 14:49:09 UTC | Fix for finding correct nsig_tct for the vertical collimators in ir2/8, nsig was not set correctly from what I can see. I thought this was already fixed. I now changes so that if the name is tct(h/v), the nsig is selected ONLY based on the IR number, not requiring the ".b" or ".B" afterwarrds | 29 September 2011, 14:49:09 UTC |
| d6e191c | Yngve Levinsen | 23 September 2011, 19:49:25 UTC | updated readme | 23 September 2011, 19:49:25 UTC |
| 76b4b25 | Yngve Levinsen | 23 September 2011, 19:49:22 UTC | [hdf5] some debug statements are now only printed if debug flag is used when running make_six | 23 September 2011, 19:49:22 UTC |
| 9bf4d58 | Yngve Levinsen | 23 September 2011, 19:49:20 UTC | makedis_ga now using mynex/myney to set cut cleaning up beamgas code somewhat | 23 September 2011, 19:49:20 UTC |
| b378dae | Yngve Levinsen | 20 September 2011, 09:11:49 UTC | For Frank: - astuce/dafor not statically linked, only flag used is -m32 -m32 probably not needed, but also probably doesn't hurt.. | 20 September 2011, 09:11:49 UTC |
| de33f0f | Frank Schmidt | 19 September 2011, 13:20:26 UTC | mising in previous version Eric&Frank | 19 September 2011, 13:20:26 UTC |
| c7ceb8f | Yngve Levinsen | 16 September 2011, 07:14:13 UTC | Fixed problem with collimat -crlibm. It's been a while, I had a space too much in my "+if .not.beamgas"... | 16 September 2011, 07:14:13 UTC |
| 5e738b0 | Yngve Levinsen | 15 September 2011, 14:27:29 UTC | Updated README | 15 September 2011, 14:27:29 UTC |
| f04a3c5 | Yngve Levinsen | 15 September 2011, 14:22:24 UTC | using make_six +option is not valid anymore. | 15 September 2011, 14:22:24 UTC |
| 29fe333 | Yngve Levinsen | 15 September 2011, 14:22:22 UTC | for beamgas module, increased number of blocks/elements again. I had forgotten about that a while back... | 15 September 2011, 14:22:22 UTC |
| ac548de | Yngve Levinsen | 15 September 2011, 14:22:19 UTC | added gcc 4.6 to list of acceptable versions for hdf5 | 15 September 2011, 14:22:19 UTC |
| 193b76a | Yngve Levinsen | 15 September 2011, 14:22:17 UTC | removed Makefile, only Makefile.template should be part of repository. Makefile is generated by make_six added some fixes to make_six for hdf5/beamgas | 15 September 2011, 14:22:17 UTC |
| 0c2cf14 | Eric Mcintosh | 19 July 2011, 10:39:28 UTC | SixTrack Version: 4.2.15 Eric -- Just added correction to make_six for checking that not both da and collimation are selected McIntosh 19th July, 2011 | 19 July 2011, 10:39:28 UTC |
| 9f8ec1f | Eric Mcintosh | 19 July 2011, 10:31:13 UTC | SixTrack Version: 4.2.15 Eric -- A few small changes for maintenance mainly. Extended Precision is disabled if crlibm is selected (was done by Lahey previously). Added the disable_xp routine to crlibm and prototypes for both enable/disable_xp. -- beauty/count.awk now writes to count.log as well as displaying report on the astuce directives in a .s file -- Added crlibm/Makefile.template to SVN. McIntosh 19th July, 2011 | 19 July 2011, 10:31:13 UTC |
| b6f8a96 | Eric Mcintosh | 03 July 2011, 14:21:19 UTC | SixTrack Version: 4.2.14 Eric -- Used SVN Rev 97 with all the Collimation mods to sixtrack.s, make_six and Makefile and added ylevinse SVN notes to this README (and fixed open(11, for BOINC) -- Updated NBLO from 400 to 600 (but NOT for collimation) -- Added/removed "&"s at lines 7398, 14818, and 30250 for consistency -- Added a hundred or so missing ampersands for f90 compatibility -- Temporarily removed the fix to real**int in the function errf to preserve numerical compatibility -- Re-arranged some of [my] C/R variables to align them for ifort -- Added ifort and fio options to make_six and fixed the ast files -- Fixed setting of make_six options selected with a + -- Added Makefile.template and made beamgas.f compilation conditional -- Modified the naming convention for the executables. | 03 July 2011, 14:21:19 UTC |
| 5f39609 | Eric Mcintosh | 15 June 2011, 10:58:22 UTC | Just restored previous version Rev 97 (replaced by accident). | 15 June 2011, 10:58:22 UTC |
| 3c7b7f5 | Eric Mcintosh | 13 June 2011, 13:53:07 UTC | Testing | 13 June 2011, 13:53:07 UTC |
| 3ad77b2 | Yngve Levinsen | 28 April 2011, 11:11:25 UTC | added changes to make_six. New (optional) flags hdf5 and beamgas. These are off by default added linking to libsz.a in make_six fixes to make_six for hdf5 - hdf5 mod files must be made from same compiler you are using (??) - tested for gfortran 4.5, new H5ROOT should point to correct base for hdf5 lib/include folders. | 28 April 2011, 11:11:25 UTC |
| 7d4df76 | Yngve Levinsen | 28 April 2011, 11:11:23 UTC | BEAMGAS source code This is the source code needed for the beamgas module. This patch is originating from many different commits on a different branch. Together with the previous commit the final code will be tested and checked before committed. | 28 April 2011, 11:11:23 UTC |
| 67fbcee | Yngve Levinsen | 28 April 2011, 11:11:20 UTC | HDF5 source code this commit contains the source code needed for HDF5 support. it is a copy of the diff from many other earlier commits on a different branch. There has been an attempt to clean it, so that the patches are easier to read. | 28 April 2011, 11:11:20 UTC |
| e59da94 | Yngve Levinsen | 28 April 2011, 11:11:17 UTC | Fixed sixtrack.s so it works with gfortran again - replaced "777 end do" with "777 continue - end do" - write(11) -> write(11,*) - also added the calls hdf5 module - changed fort.11 to formatted output (got runtime error with gfortran) | 28 April 2011, 11:11:17 UTC |
| d19b8a2 | Yngve Levinsen | 28 April 2011, 11:11:14 UTC | further changes to make_six: - hdf5 currently only works with gfortran - make astuce/dafor didn't work because of previous commit | 28 April 2011, 11:11:14 UTC |
| 68f05e4 | Yngve Levinsen | 28 April 2011, 11:11:12 UTC | added new code in sixtrack.s - HDF5 module/subroutines for tracks2.h5 - The modules for beamgas, subroutines are not present yet These modules just holds common variables for beamgas routines.. | 28 April 2011, 11:11:12 UTC |
| 1d4d483 | Yngve Levinsen | 28 April 2011, 11:11:09 UTC | updated astuce masks for new modules, added one (beamgas.ast) | 28 April 2011, 11:11:09 UTC |
