Szilard Pall [Fri, 8 Apr 2011 22:23:34 +0000 (00:23 +0200)]
fixed g_tune_pme to pass correctly boolean arguments to mdrun
Justin A. Lemkul [Thu, 7 Apr 2011 01:20:26 +0000 (21:20 -0400)]
Revised wording in pdb2gmx.c, hopefully clearer now.
David van der Spoel [Wed, 6 Apr 2011 16:16:12 +0000 (18:16 +0200)]
Fixed header for gmx_conect_init
Justin A. Lemkul [Wed, 6 Apr 2011 12:02:38 +0000 (08:02 -0400)]
Added documentation about proper use of pdb2gmx -ter.
David van der Spoel [Mon, 4 Apr 2011 13:08:46 +0000 (15:08 +0200)]
Merge branch 'release-4-5-patches' of git@git.gromacs.org:gromacs into release-4-5-patches
David van der Spoel [Mon, 4 Apr 2011 13:08:04 +0000 (15:08 +0200)]
Removed superfluous file.
David van der Spoel [Mon, 4 Apr 2011 13:07:29 +0000 (15:07 +0200)]
Made new programs compile with gcc -pedantic.
David van der Spoel [Mon, 4 Apr 2011 12:31:56 +0000 (14:31 +0200)]
New entry
Justin Lemkul [Mon, 4 Apr 2011 12:05:31 +0000 (08:05 -0400)]
A few more additions to wman.c to account for symbol encoding.
David van der Spoel [Mon, 4 Apr 2011 11:13:54 +0000 (13:13 +0200)]
Added 1 analysis tool manual.
David van der Spoel [Mon, 4 Apr 2011 10:45:16 +0000 (12:45 +0200)]
Merge branch 'release-4-5-patches' into orderparm
David van der Spoel [Mon, 4 Apr 2011 10:42:15 +0000 (12:42 +0200)]
Latest fixes from Bjorn Steen Saethre
Sander Pronk [Fri, 1 Apr 2011 14:24:00 +0000 (16:24 +0200)]
Fixed g_pme_error to use MPI, but not threads.
This fixes issue #729.
Berk Hess [Thu, 31 Mar 2011 08:55:50 +0000 (10:55 +0200)]
fixed genbox -ci only filling part of triclinic boxes
This fixes #724
Mark Abraham [Wed, 30 Mar 2011 19:52:24 +0000 (06:52 +1100)]
Removed useless const qualifier
xdr_zero was declared as a pointer-to-const, but was only used as
an argument to a function taking pointer-to-non-const. This provokes
a compiler warning, so I removed the useless const.
Mark Abraham [Wed, 30 Mar 2011 19:47:33 +0000 (06:47 +1100)]
Suppressed Cygwin C-only compiler warnings
Cygwin uses newlib, which by default uses macros for things like
isupper() defined in ctype.h. Apparently, by design, these are
intended to provoke compiler warnings when called with signed char
arguments. I don't know why. Undefining these macros falls back to
function versions that accept signed int, and so produce no warning.
Justin Lemkul [Wed, 30 Mar 2011 17:07:41 +0000 (13:07 -0400)]
Updated g_energy help information.
Carsten Kutzner [Wed, 30 Mar 2011 14:17:16 +0000 (16:17 +0200)]
Checks for zero-mass atoms in ED, better ordering of options in make_edi
Mark Abraham [Sat, 26 Mar 2011 04:47:08 +0000 (15:47 +1100)]
Fixed g_cluster logic error
g_cluster was attempting to deallocate a gmx_rmpbc_t object when it
might not have been initialized. Since that object is only initialized
under a simple combination of conditions, it makes most sense to clean
up only under those conditions (as well as fix the real reason for the
segfault).
Mark Abraham [Sat, 26 Mar 2011 04:34:07 +0000 (15:34 +1100)]
Made gmx_rmpbc_t object functions not segfault
In C++, member functions can assume that they are called on an allocated
object. C pseudo-member-functions don't have the luxury of that
knowledge about the object.
Mark Abraham [Sat, 26 Mar 2011 04:32:19 +0000 (15:32 +1100)]
Fixed matio.c sfree() that made DvdS cry.
The error was merely in deallocating the modified pointer,
rather than the allocated one.
Sander Pronk [Fri, 25 Mar 2011 14:41:31 +0000 (15:41 +0100)]
Fixed deallocation bug in energy history checkpoint
Justin Lemkul [Thu, 24 Mar 2011 12:00:17 +0000 (08:00 -0400)]
Fixed incorrect reference in g_tcaf.
David van der Spoel [Thu, 24 Mar 2011 08:50:35 +0000 (09:50 +0100)]
Added Bjorn's code, implemented running average filtering.
Justin Lemkul [Wed, 23 Mar 2011 00:41:46 +0000 (20:41 -0400)]
Tons of small fixes to documentation.
These are a lot of editorial fixes (style, typos, etc) but also an
introduction of symbols in equations and normal text that I think (1)
makes the manual look prettier and (2) makes the terminology easier to
follow throughout the manual, back to earlier chapters, etc.
Justin Lemkul [Tue, 22 Mar 2011 14:55:40 +0000 (10:55 -0400)]
Lots of editorial fixes to descriptions, etc. to make the manual
a bit cleaner and more consistent.
Berk Hess [Tue, 22 Mar 2011 13:42:15 +0000 (14:42 +0100)]
fixed replica exchange quantity consistency check
Rossen Apostolov [Mon, 21 Mar 2011 15:18:18 +0000 (16:18 +0100)]
New maintenance release: 4.5.4
Sander Pronk [Mon, 21 Mar 2011 15:02:23 +0000 (16:02 +0100)]
The -version flag now prints more information.
mdrun -version now prints whether the binary is single/double
precision, the parallellization library used, and the FFT library.
Erik Marklund [Mon, 21 Mar 2011 12:46:33 +0000 (13:46 +0100)]
Fixed warning arising from gmx_hbond.c
Rossen Apostolov [Mon, 21 Mar 2011 11:26:12 +0000 (12:26 +0100)]
Fixed more real/double issues in gmx_density
Rossen Apostolov [Mon, 21 Mar 2011 10:52:22 +0000 (11:52 +0100)]
Removed duplicated EXTRA_DIST entry in src/tools/Makefiles.am
Rossen Apostolov [Mon, 21 Mar 2011 10:11:09 +0000 (11:11 +0100)]
Added back Template.mak.
Sander Pronk [Mon, 21 Mar 2011 09:58:28 +0000 (10:58 +0100)]
Fixed real/double issue in gmx_density
Rossen Apostolov [Mon, 21 Mar 2011 09:35:28 +0000 (10:35 +0100)]
Removed some template files from installation.
Sander Pronk [Thu, 17 Mar 2011 10:58:42 +0000 (11:58 +0100)]
Various fixes for Windows builds.
Justin Lemkul [Wed, 16 Mar 2011 15:17:41 +0000 (11:17 -0400)]
Made g_sgangle fatal error a bit more verbose.
Sander Pronk [Wed, 16 Mar 2011 11:54:12 +0000 (12:54 +0100)]
mdrun -multi, multidir and replica exchange now exchange large int step sizes.
When coordinating step numbers, multirun now sends large ints.
An MPI datatype for that is now introduced: GMX_MPI_LARGE_INT.
Sander Pronk [Tue, 15 Mar 2011 14:37:55 +0000 (15:37 +0100)]
Multisim simulations now all stop when the first one stops.
This fixes a bug where all MPI nodes would hang once the first
simulation in a multisim finishes.
Rossen Apostolov [Mon, 14 Mar 2011 13:30:36 +0000 (14:30 +0100)]
g_options source files are now properly packaged
The tool is installed by default by both CMake and autoconf
Berk Hess [Mon, 14 Mar 2011 07:25:29 +0000 (08:25 +0100)]
fixed bd with bd_fric>0 making all coordinates 0
Berk Hess [Mon, 14 Mar 2011 07:25:00 +0000 (08:25 +0100)]
fixed precision of g_traj -com for large groups
Sander Pronk [Fri, 11 Mar 2011 15:31:38 +0000 (16:31 +0100)]
Merge branch 'release-4-5-patches' of git.gromacs.org:gromacs into release-4-5-patches
Berk Hess [Fri, 11 Mar 2011 10:57:34 +0000 (11:57 +0100)]
add pdb2gmx check for multiple .ff directories
pdb2gmx will currently always read from the first .ff directory
in the library directory search list. Therefore selecting a second
force field with the same directory name would still lead to reading
the files in the first directory. A verbose fatal error now occurs
when not choosing the first of multiple entries.
This fixes #619
Berk Hess [Fri, 11 Mar 2011 10:12:18 +0000 (11:12 +0100)]
fixed memory error in element setting in pdbio
Atom types without element would cause a NULL pointer to be passed
to sprintf in get_pdb_atomnumber. This could cause the memory of
pdb2gmx to get corrupted and print incorrect atom types and charges.
This would not affect normal pdb files and force fields, since all
atom in there are "real" atoms. This fixes #618
Berk Hess [Fri, 11 Mar 2011 09:43:20 +0000 (10:43 +0100)]
added histidine names for gen_vsite, fixes 706
Sander Pronk [Thu, 10 Mar 2011 16:30:39 +0000 (17:30 +0100)]
More allocation fixes for NUMA allocator.
Sander Pronk [Thu, 10 Mar 2011 16:27:26 +0000 (17:27 +0100)]
Added documentation for multidir.
Sander Pronk [Thu, 10 Mar 2011 16:24:12 +0000 (17:24 +0100)]
Added more support for multidir.
Maarten Wolf [Thu, 10 Mar 2011 09:53:32 +0000 (10:53 +0100)]
g_membed bug #697 fix. Remove empty molblocks due to
removing these molecules
Mark Abraham [Tue, 8 Mar 2011 12:29:13 +0000 (23:29 +1100)]
Added grompp note about avoiding zero-step EM
People try to do zero-step EM to evaluate energies of single
configurations. grompp now writes a note that this will alter the
coordinates before computing the energy, and recommends zero-step MD,
or the use of mdrun -rerun.
Berk Hess [Mon, 7 Mar 2011 21:10:02 +0000 (22:10 +0100)]
removed global #atoms allocation for empty DD cells
Berk Hess [Mon, 7 Mar 2011 15:00:01 +0000 (16:00 +0100)]
removed a global #atoms alloc in mdrun initialization
David van der Spoel [Sat, 5 Mar 2011 19:07:30 +0000 (20:07 +0100)]
Merge branch 'release-4-5-patches' of git@git.gromacs.org:gromacs into release-4-5-patches
David van der Spoel [Sat, 5 Mar 2011 19:06:53 +0000 (20:06 +0100)]
Fixed SEGV in eneconv when using the -e flag.
Sander Pronk [Thu, 3 Mar 2011 17:17:24 +0000 (18:17 +0100)]
Replaced strdups with gmx_strdups so that they can be sfree()'d safely.
Patch providided by Scott Field from Microsoft as part of NUMA optimizations.
Berk Hess [Thu, 3 Mar 2011 16:08:38 +0000 (17:08 +0100)]
made the g_density accumulation double precision
Christoph Junghans [Thu, 3 Mar 2011 13:39:38 +0000 (14:39 +0100)]
fixed a typo in cmake pkg-config files
Carsten Kutzner [Thu, 3 Mar 2011 10:56:50 +0000 (11:56 +0100)]
Fixed wrong error message when using flooding code as harmonic
restraint. Also added comments.
Justin Lemkul [Tue, 1 Mar 2011 14:59:35 +0000 (09:59 -0500)]
Small fixes to g_dielectric description to improve readability.
Carsten Kutzner [Tue, 1 Mar 2011 12:06:31 +0000 (13:06 +0100)]
Fixed bug #714: deprecated structure in gettimeofday
Justin Lemkul [Sat, 26 Feb 2011 13:32:53 +0000 (08:32 -0500)]
Another typo.
Justin Lemkul [Sat, 26 Feb 2011 13:29:23 +0000 (08:29 -0500)]
Fixed some typos in mdp_opt.html
Justin Lemkul [Fri, 25 Feb 2011 23:44:47 +0000 (18:44 -0500)]
Fixed typo in citation.
Carsten Kutzner [Fri, 25 Feb 2011 18:37:58 +0000 (19:37 +0100)]
Corrected output of a warning message.
Carsten Kutzner [Fri, 25 Feb 2011 18:30:32 +0000 (19:30 +0100)]
Added the -fix option to g_tune_pme for tuning of rcoulomb and grid
at a fixed number of PME-only nodes.
Berk Hess [Thu, 24 Feb 2011 16:39:28 +0000 (17:39 +0100)]
also renamed options to g_options in Makefile.am
Berk Hess [Thu, 24 Feb 2011 14:37:26 +0000 (15:37 +0100)]
added missing density factor to cosaccel 1/viscosity output
Berk Hess [Thu, 24 Feb 2011 13:04:14 +0000 (14:04 +0100)]
pdb model numbers now start at 1
The general pdb writing code and trjconv could write pdb files
with 0 as the first model number. The minimum been changed to 1.
This fixes #709
Szilard Pall [Wed, 23 Feb 2011 12:59:32 +0000 (13:59 +0100)]
added cmake warning when the cache generated on another host is reused
Berk Hess [Wed, 23 Feb 2011 12:51:24 +0000 (13:51 +0100)]
fixed incorrect virial and pressure averages
Removed a second, incorrect setting of flags for storage of energy averages
in mdrun.
This would cause incorrect averages of the virial and pressure to be stored
in the energy file and printed at the end of the log file when the lowest
common denominator of nstcomm, nsttcouple and/or nstpcouple was less than
nstcalcenergy. Note that the instantaneous values and the actual simulation
results were all correct.
Additionally there was a potential issue with parallel checkpoint signalling
with -gcom larger than nstlist which could lead to hanging of mdrun.
This fixes #696
Berk Hess [Mon, 21 Feb 2011 14:26:10 +0000 (15:26 +0100)]
added support for harmonic potential (bond type 7) in OpenMM
Berk Hess [Mon, 21 Feb 2011 14:24:46 +0000 (15:24 +0100)]
made maximum load balancing tunable with env.var. GMX_DLB_MAX
Berk Hess [Sun, 20 Feb 2011 18:51:10 +0000 (19:51 +0100)]
fixed dynamic load balancing inactive with -gcom and pressure scaling with -gcom when nstpcouple<nstcalcenergy
Berk Hess [Sun, 20 Feb 2011 18:49:37 +0000 (19:49 +0100)]
added some debug prints to gmxcpp.c
Sander Pronk [Sun, 20 Feb 2011 09:54:14 +0000 (10:54 +0100)]
Minor fix for when write_dhdl_file=false and delta_lambda != 0
Carsten Kutzner [Wed, 16 Feb 2011 17:05:06 +0000 (18:05 +0100)]
Merge branch 'release-4-5-patches' of git.gromacs.org:gromacs into release-4-5-patches
Carsten Kutzner [Wed, 16 Feb 2011 16:37:06 +0000 (17:37 +0100)]
Fixed int overflow happening in genion when number of water molecules is large
Teemu Murtola [Wed, 16 Feb 2011 15:53:32 +0000 (17:53 +0200)]
Fixed a rare bug in selection position handling.
If the "whole" prefix was used on a position calculation that applied to
a static selection (e.g., "whole_res_com of name CA CB"), the block data
for the output positions was incorrect.
Justin Lemkul [Wed, 16 Feb 2011 15:38:39 +0000 (10:38 -0500)]
Add -tu option to g_hbond.
IssueID #580
Mark Abraham [Wed, 16 Feb 2011 06:14:00 +0000 (17:14 +1100)]
Clarified source of g_rmsf reference frame
Justin Lemkul [Tue, 15 Feb 2011 19:07:38 +0000 (14:07 -0500)]
Small fix to improve xpm2ps output
Mark Abraham [Sun, 13 Feb 2011 01:11:35 +0000 (12:11 +1100)]
Added workaround for F_GB13 duplicates
Under some conditions, identical reference bond lengths could lead to
different computed 1-3 lengths (using icc 11 and double precision, at
least). These then failed the memcmp() test that removes duplicate
parameter lists, which led to near-duplicates in the
function-parameters lists. In theory, we shouldn't test for
floating-point equality with memcmp(), but in theory the test should
work, too!
Instead of memcmp(), for F_GB13 only, tests with
gmx_within_tol(a,b,1e-6) are now used. That should be an acceptable
compromise between expected function, numerical accuracy and grompp
execution time.
IssueID #699
Mark Abraham [Sun, 13 Feb 2011 00:27:24 +0000 (11:27 +1100)]
Fixed erroneous floating-point comparison
IssueID #699
Mark Abraham [Sat, 12 Feb 2011 02:20:58 +0000 (13:20 +1100)]
Added note about doing repeat invocations of genbox
Szilard Pall [Fri, 11 Feb 2011 18:14:09 +0000 (19:14 +0100)]
fixed buggy detection of hostname and time of build in CMake
Mark Abraham [Fri, 11 Feb 2011 07:31:52 +0000 (18:31 +1100)]
Removed ancient CVS tag lines
Mark Abraham [Fri, 11 Feb 2011 07:11:18 +0000 (18:11 +1100)]
Used some spacing to make code readable
Mark Abraham [Fri, 11 Feb 2011 06:57:27 +0000 (17:57 +1100)]
Used symbolic constant more consistently
Mark Abraham [Fri, 11 Feb 2011 06:43:22 +0000 (17:43 +1100)]
Made the "multiple settles" fatal error more helpful
The error message now explains the origin of the issue, hints at a
common reason for erroneously provoking this error, and still
provides the previously-given tip for resolving the issue for users
who really need two molecule blocks of constrained solvent.
Sander Pronk [Thu, 10 Feb 2011 10:39:57 +0000 (11:39 +0100)]
Fixed a bug in g_energy -odh
Can now succesfully extract dhdl values with g_energy when
delta_lambda>0
Roland Schulz [Thu, 23 Sep 2010 05:34:02 +0000 (01:34 -0400)]
using set_mm_ps to make pgi happy
Asked about this commit 10Apr2010 and got no answer. Committing it now so that it deosn't get lost
Should have no effect on gcc and fixes it for PGI
Justin Lemkul [Thu, 3 Feb 2011 00:51:41 +0000 (19:51 -0500)]
Fixed a typo.
Mark Abraham [Wed, 2 Feb 2011 04:48:20 +0000 (15:48 +1100)]
Fixed minor issue with replica exchange
If the user chose mdrun -replex that was not a multiple of
nstcalcenergy, then replica exchange could occur when energies were
not current. Fixed by requiring repl_ex_nst to be a multiple of
nstcalcenergy, as with similar mechanisms for global
intra-simulation communication.
IssueID #695
Erik Lindahl [Sun, 30 Jan 2011 20:59:08 +0000 (21:59 +0100)]
Improved PBC handling for large molecules during cluster PBC
David van der Spoel [Sun, 30 Jan 2011 20:33:11 +0000 (21:33 +0100)]
Fixed more PBC and indexing stuff.
Erik Lindahl [Sun, 30 Jan 2011 20:15:41 +0000 (21:15 +0100)]
Fixed the pbc cluster option of trjconv by switching to a slower algorithm
David van der Spoel [Sun, 30 Jan 2011 19:24:39 +0000 (20:24 +0100)]
Fixed typo.
David van der Spoel [Sun, 30 Jan 2011 18:24:40 +0000 (19:24 +0100)]
Merge branch 'release-4-5-patches' into orderparm
Merged in some new code from Bjorn Steen Saethre.
Conflicts:
src/tools/Makefile.am
src/tools/expfit.c
src/tools/levenmar.c
David van der Spoel [Sat, 29 Jan 2011 21:19:45 +0000 (22:19 +0100)]
Merge branch 'release-4-5-patches' of git@git.gromacs.org:gromacs into release-4-5-patches