Rossen Apostolov [Wed, 21 Sep 2011 12:55:56 +0000 (14:55 +0200)]
Updated version: now 4.6
Change-Id: I309ddac1ba41c8c2bfe9c8880848c061c384720c
David van der Spoel [Wed, 21 Sep 2011 09:03:07 +0000 (11:03 +0200)]
Commented out part of this new program that was not yet validated, in order to protent users from incorrect results.
Change-Id: I02a0065205aae881438c15c9aec9dae722b23206
David van der Spoel [Wed, 21 Sep 2011 07:02:20 +0000 (09:02 +0200)]
Fixed documentation issue described in redmine issue 681. Added warning line to terminal output about interpretation of results.
Change-Id: I494596d5259038c96d9aad395089434520cad5c0
David van der Spoel [Wed, 21 Sep 2011 06:56:46 +0000 (08:56 +0200)]
Fixed documentation issue described in redmine issue 681.
Added warning line to terminal output about interpretation of results.
Change-Id: I92fdc8a1a641b7db534b0da1f483ecdcb5592b6b
Rossen Apostolov [Tue, 20 Sep 2011 07:01:42 +0000 (09:01 +0200)]
New release: 4.5.5
Change-Id: I6130c171c98c187ec233ee9910cf072cef00ce47
Justin A. Lemkul [Mon, 19 Sep 2011 21:15:27 +0000 (17:15 -0400)]
Fixed typos in documentation.
IssueID #814
Change-Id: If05a4a47509ad087c8de41305666324287e7a70a
Sander Pronk [Mon, 19 Sep 2011 15:51:31 +0000 (17:51 +0200)]
C89 fixes.
Change-Id: I04bfe69d1e97ae7b3160209a68421c9f88468177
David van der Spoel [Mon, 19 Sep 2011 14:10:57 +0000 (16:10 +0200)]
Fixed compilation issues
Change-Id: Ic2b724627a994f968fad9bcc1caf6a852d5e5f24
Rossen Apostolov [Mon, 19 Sep 2011 13:38:49 +0000 (15:38 +0200)]
Merge "Resolved gerrit issue https://gerrit.gromacs.org/#change,126" into release-4-5-patches
David van der Spoel [Mon, 19 Sep 2011 12:09:04 +0000 (14:09 +0200)]
Resolved gerrit issue https://gerrit.gromacs.org/#change,126
Change-Id: I21190310734c8ccab645606515a9a7265c805ccc
Rossen Apostolov [Mon, 19 Sep 2011 11:06:35 +0000 (13:06 +0200)]
Added support for g_dos, g_hydorder and g_densorder in CMake.
Change-Id: I501747d6d0c41f88427c7f0bfa181d77adece9f0
Rossen Apostolov [Mon, 19 Sep 2011 10:16:21 +0000 (12:16 +0200)]
Fixed a bug in egbOBC and double precission.
Change-Id: I016342ebe0f7758ffe0373f906e7095d2c9a1a06
Berk Hess [Mon, 19 Sep 2011 09:30:51 +0000 (11:30 +0200)]
added grompp check for posres+pcoupl+refcoord_scaling=no
Change-Id: Iec18eb3192a0fbbd44883e0963df471e441cb8bb
Rossen Apostolov [Fri, 16 Sep 2011 20:24:23 +0000 (22:24 +0200)]
Merge "g_wham: fixed pullx.xvg reading issue with pull_geometry=cylinder" into release-4-5-patches
Erik Lindahl [Fri, 16 Sep 2011 15:13:37 +0000 (17:13 +0200)]
Merge "made allocation for LINCS with particle decomposition dynamic, fixes #777" into release-4-5-patches
Jochen Hub [Fri, 16 Sep 2011 13:33:09 +0000 (15:33 +0200)]
g_wham: fixed pullx.xvg reading issue with pull_geometry=cylinder
Change-Id: Ica6c4a4685efa8e3900ea893f11f379bfcece718
Berk Hess [Thu, 15 Sep 2011 13:47:43 +0000 (15:47 +0200)]
mdrun -multi -cpi now checks if simulations parts are identical
Change-Id: I7163d5042c7841eece3957df87859e8b1e7ceae0
Berk Hess [Thu, 15 Sep 2011 13:47:38 +0000 (15:47 +0200)]
Merge "Introduced different options for pdb2gmx termini introduction (-chainsep) and merging into single moleculetype (-merge)" into release-4-5-patches
Mark Abraham [Thu, 15 Sep 2011 07:46:31 +0000 (17:46 +1000)]
Fix trjconv -sub
close_trx doesn't set the defunct t_trxstatus * to NULL, so the
subsequent test for clust_status[i] to be non-NULL was erroneous.
Change-Id: I38f3b0404c82122cdd74f0af11738155a5e34013
Mark Abraham [Thu, 15 Sep 2011 07:45:20 +0000 (17:45 +1000)]
Warning about trjconv -sub needing many files open
Change-Id: I6645b7ccdca47b9eed61d2e937cad99d43e816ac
Mark Abraham [Thu, 15 Sep 2011 04:26:14 +0000 (14:26 +1000)]
g_sham -mdata is an input file, not output file
Change-Id: I2795de08bc8049a95f921300b325b5942a74d516
Mark Abraham [Wed, 14 Sep 2011 00:28:46 +0000 (10:28 +1000)]
Format warning better
Change-Id: I1a18ee94a238fc0c4690da4f33e7a95b071d15c6
Rossen Apostolov [Tue, 13 Sep 2011 20:44:17 +0000 (22:44 +0200)]
Merge "Added warning line." into release-4-5-patches
Erik Lindahl [Tue, 13 Sep 2011 17:24:06 +0000 (19:24 +0200)]
Introduced different options for pdb2gmx termini introduction (-chainsep) and merging into single moleculetype (-merge)
Change-Id: I041f152f157f6f6b70145582ee5624bcbcf9bda2
Berk Hess [Tue, 13 Sep 2011 15:56:07 +0000 (17:56 +0200)]
Merge "Fix pdb2gmx -debug and .hdb files" into release-4-5-patches
Berk Hess [Tue, 13 Sep 2011 15:55:51 +0000 (17:55 +0200)]
Merge "Fixed charge groups of CHARMM27 DNA/RNA terminal H" into release-4-5-patches
Berk Hess [Tue, 13 Sep 2011 15:55:35 +0000 (17:55 +0200)]
Merge "Added ACE and CT3 to CHARMM27 aminoacids.rtp" into release-4-5-patches
David van der Spoel [Tue, 13 Sep 2011 14:34:10 +0000 (16:34 +0200)]
Added warning line.
Change-Id: Ia301e2515cba3ccebb7f35dbe0c7a3d0ed5f5093
Erik Lindahl [Tue, 13 Sep 2011 13:37:37 +0000 (15:37 +0200)]
Fixed incorrection AmberGS force field (redmine #728)
Change-Id: Iec340d97851bc08bc8555667df859cb6effc4378
Roland Schulz [Tue, 13 Sep 2011 13:05:26 +0000 (15:05 +0200)]
Merge "fixed grompp issue with decoupling and virtual sites" into release-4-5-patches
Rossen Apostolov [Tue, 13 Sep 2011 12:14:06 +0000 (14:14 +0200)]
Merge "Fixes for CMake finding FORTRAN library names on POWER6" into release-4-5-patches
Berk Hess [Tue, 13 Sep 2011 07:50:51 +0000 (09:50 +0200)]
Merge "Remove segv in xpm2ps, fixes #611" into release-4-5-patches
Berk Hess [Tue, 13 Sep 2011 07:50:11 +0000 (09:50 +0200)]
fixed grompp issue with decoupling and virtual sites
grompp would give a fatal error while generating virtual site
parameters when state A was decoupled using the decouple option.
Change-Id: I8ea099d93d859537869aa32a2f8d6d825f277bde
Mark Abraham [Tue, 13 Sep 2011 05:11:28 +0000 (15:11 +1000)]
Fix pdb2gmx -debug and .hdb files
For some archaic reason, pdb2gmx produces sanitized .hdb files when
you use -debug. However, these can run afoul of the wildcard search
for .hdb files on subsequent calls to pdb2gmx. So I have commented
out the sanitizing feature.
Change-Id: Ie0efada2960dd62260f6be5c796eed18a684b3e7
Mark Abraham [Fri, 9 Sep 2011 06:46:34 +0000 (16:46 +1000)]
Fix for CMake finding LAPACK
Suggested by Marcin Zielinski on gmx-users:
http://lists.gromacs.org/pipermail/gmx-users/2011-September/064392.html
Change-Id: I4af8637ecdf0258ff3229a49bf7b4a3b51edc3ec
Mark Abraham [Fri, 9 Sep 2011 01:48:14 +0000 (11:48 +1000)]
Added ACE and CT3 to CHARMM27 aminoacids.rtp
Also supporting changes to .hdb and residetypes.dat
Change-Id: If35afd88d9bb4f71ef5ef8556bf6db245a16bbea
Mark Abraham [Thu, 8 Sep 2011 01:05:15 +0000 (11:05 +1000)]
Fixed charge groups of CHARMM27 DNA/RNA terminal H
The nucleic acid termini databases added hydrogen atoms by default in
the same charge group as the atom to which they were bound, however
in CHARMM force fields, each atom should be its own charge group.
IssueID #744
Change-Id: Ic608065008cc20e6c452bb5cf1d116db6bfde679
Mark Abraham [Fri, 9 Sep 2011 14:46:21 +0000 (00:46 +1000)]
Fixes for CMake finding FORTRAN library names on POWER6
Change-Id: I99a5d406a8c2059574e17e8664686ae9b09e0c6f
Rossen Apostolov [Fri, 9 Sep 2011 14:28:50 +0000 (16:28 +0200)]
Added support for custom external BLAS/LAPACK libraries.
To use them give give to cmake
-DGMX_EXTERNAL_LAPACK=TRUE and
-DGMX_LAPACK_USER=/path/to/external/lapack
Same for BLAS.
Change-Id: I4ca1c4ca365788fcafc4219934fdfe8758db84f1
Mark Abraham [Tue, 6 Sep 2011 01:44:43 +0000 (11:44 +1000)]
Fix bug with large ints without MPI_IN_PLACE
Communication of gmx_large_int_t arrays when MPI_IN_PLACE was disabled
never worked when it was introduced just before 4.5.4, because it used
an int-sized buffer. This affected replica exchange sanity tests on the
number of steps.
IssueID #801 #803
Change-Id: I48f04b1c4dd80646711db99c5dab89cb910e79b4
Berk Hess [Mon, 5 Sep 2011 15:13:28 +0000 (17:13 +0200)]
made allocation for LINCS with particle decomposition dynamic, fixes #777
Change-Id: I070b5f8917d12ab64896f62ecc80ed38b011f066
Michael Shirts [Mon, 5 Sep 2011 14:32:12 +0000 (16:32 +0200)]
Merge "fixed incorrect pV term in the energy file" into release-4-5-patches
Mark Abraham [Thu, 1 Sep 2011 14:25:31 +0000 (00:25 +1000)]
Made grompp note about non-zero charge more useful
Non-programmers need not know what -6.
300000e-1 actually means,
so we should avoid the %e printf format. URL provided to discussion
of why the user needs to pay some attention here.
Change-Id: Ie593919540db8728b898cd7d0904659c7f3d9345
Mark Abraham [Thu, 25 Aug 2011 04:58:38 +0000 (14:58 +1000)]
Re-implemented generating default protein index groups
Hopefully using a struct to associate data that is used in association
will make future maintenance easier. Some general improvements
to the code quality, also.
Change-Id: I37017e1dd3612d143df935ada79374dd9a8fbcd3
Mark Abraham [Thu, 25 Aug 2011 02:01:17 +0000 (12:01 +1000)]
Adjusted construction of default index groups
The CHARMM convention of naming the H atom bonded to backbone N as
"HN" led to erroneous construction the MainChain+H and (complementary)
SideChain groups. Protein, Protein-H and SideChain-H were fortuitously
correct.
IssueID #788
Change-Id: Ib8525dba61e4f072ae5074a9f6b523ac13892833
Mark Abraham [Thu, 25 Aug 2011 02:00:45 +0000 (12:00 +1000)]
Fixed comment
Change-Id: I42e6292e7c658c69810626cf4e6c861394546a40
Carsten Kutzner [Fri, 26 Aug 2011 14:05:41 +0000 (16:05 +0200)]
Remove segv in xpm2ps, fixes #611
Issue a fatal error if matrix elements are to be
added/subtracted/multiplied/divided (with xpm2ps -compine) but no
proper matrix data could be extracted from one of the xpm files.
Berk Hess [Mon, 29 Aug 2011 13:24:10 +0000 (15:24 +0200)]
fixed incorrect pV term in the energy file
Change-Id: I157b0468907802e29c7dec763ed2e5db99c02922
Carsten Kutzner [Wed, 24 Aug 2011 15:20:38 +0000 (17:20 +0200)]
Instead of a segv, mdrun now gives an error msg if npme>np, fixes #795
Instead of a segv, mdrun now gives an error msg if npme>np, fixes #795
Change-Id: I2e93bccf45ace215d6237354559b8750628a13d6
Roland Schulz [Tue, 23 Aug 2011 07:22:17 +0000 (09:22 +0200)]
Merge "Fixed g_disre dereferencing cr" into release-4-5-patches
Roland Schulz [Tue, 23 Aug 2011 07:21:34 +0000 (09:21 +0200)]
Merge "Typo fix" into release-4-5-patches
Roland Schulz [Tue, 23 Aug 2011 07:21:00 +0000 (09:21 +0200)]
Merge "fixed operation on closed file structure in gmx_rmsdist, fixes #780" into release-4-5-patches
Roland Schulz [Tue, 23 Aug 2011 07:19:31 +0000 (09:19 +0200)]
Merge "fixed possible infinite loop when reading atom types, fixes #791" into release-4-5-patches
Mark Abraham [Tue, 23 Aug 2011 05:58:13 +0000 (15:58 +1000)]
Fixed g_disre dereferencing cr
mdrun uses some code in common with g_disre, and some code that deals
with PD vs DD has to use cr. g_disre sets cr to NULL, of course, and
the common code just needs to be sensitive to that and test for NULL
before using it.
IssueID #766
Change-Id: Ia92bd1d18ba8085d285f37b14897f384a71c64ce
Mark Abraham [Tue, 23 Aug 2011 05:58:04 +0000 (15:58 +1000)]
Typo fix
Change-Id: I0fd1e3f6e21c3fedcb43151935bbfc7fe69d84bd
Berk Hess [Mon, 22 Aug 2011 15:20:34 +0000 (17:20 +0200)]
fixed operation on closed file structure in gmx_rmsdist, fixes #780
Change-Id: Iddac26debf616d962df2fd40b79d01dc01942c1c
Berk Hess [Mon, 22 Aug 2011 14:33:36 +0000 (16:33 +0200)]
fixed possible infinite loop when reading atom types, fixes #791
Change-Id: If5ddcc0b5238cdfa625b1d85d05e1dabbe4eb32a
Mark Abraham [Fri, 19 Aug 2011 15:59:09 +0000 (01:59 +1000)]
Explained terminal residues and AMBER in pdb2gmx -h
The -ter mechanism doesn't work because naming terminal residues with
suitable prefixes is required.
Change-Id: Idde88d84988ef4d1d0752e6f854a0d18c194ea88
Mark Abraham [Tue, 16 Aug 2011 11:37:00 +0000 (21:37 +1000)]
Made free energy perturbation work with CMAP
That is, FEP of other parameters works when CMAP is active, not
FEP of CMAP parameters themselves (if that even makes sense!)
Berk Hess [Mon, 15 Aug 2011 10:00:45 +0000 (12:00 +0200)]
fixed incorrect pointer assignment in g_protonate
Change-Id: I719c24921157077aca60afae1e4579b7dcf670e3
Berk Hess [Fri, 12 Aug 2011 08:16:37 +0000 (10:16 +0200)]
implemented full exclusion support for the charge-group innerloops
Change-Id: Ib662addaf9e33288c072ee05446d883bac09394c
Roland Schulz [Thu, 11 Aug 2011 17:25:55 +0000 (19:25 +0200)]
Merge "removed npmenodes mismatch warning in read_checkpoint without DD" into release-4-5-patches
Berk Hess [Thu, 11 Aug 2011 13:44:54 +0000 (15:44 +0200)]
Merge "Make sure that ftell works correctly for files opened in append mode fixing Issue #762" into release-4-5-patches
Roland Schulz [Tue, 26 Jul 2011 07:24:33 +0000 (03:24 -0400)]
Make sure that ftell works correctly for files opened in append mode
fixing Issue #762
Change-Id: I019fa6a2b51a2156c900e1a31a0b521e62b79961
Berk Hess [Wed, 10 Aug 2011 07:52:15 +0000 (09:52 +0200)]
removed npmenodes mismatch warning in read_checkpoint without DD
Berk Hess [Wed, 10 Aug 2011 07:50:39 +0000 (09:50 +0200)]
removed fatal_error for prime number of nodes when -npme is set
Mark Abraham [Tue, 2 Aug 2011 23:08:01 +0000 (09:08 +1000)]
Fixed output bug in demux.pl
Simulation time written to the output index files was limited to
8 characters, which was too small. Now 20 characters.
Carsten Kutzner [Tue, 2 Aug 2011 13:10:41 +0000 (15:10 +0200)]
Typo.
Berk Hess [Tue, 2 Aug 2011 10:24:44 +0000 (12:24 +0200)]
made trjconv -novel work again, fixes #722
Carsten Kutzner [Fri, 29 Jul 2011 08:41:25 +0000 (10:41 +0200)]
Fixed calculation of max and min number of PME-only nodes. Fixes #787.
Michael Shirts [Fri, 22 Jul 2011 04:41:14 +0000 (00:41 -0400)]
number of degrees of freedom can be fractional -- in NHC_trotter, 'nd' has to be double, not int.
Justin A. Lemkul [Wed, 20 Jul 2011 16:50:05 +0000 (12:50 -0400)]
Removed jargon and abbreviations from pdb2gmx.
Not everyone will know what 'iso' means. These little tweaks
will hopefully lead to better understanding of the pdb2gmx options.
David van der Spoel [Wed, 20 Jul 2011 09:24:05 +0000 (11:24 +0200)]
Merge branch 'release-4-5-patches' of git@git.gromacs.org:gromacs into release-4-5-patches
David van der Spoel [Wed, 20 Jul 2011 09:21:30 +0000 (11:21 +0200)]
Added initialization of arrays after srenew call.
Mark Abraham [Sun, 17 Jul 2011 22:06:13 +0000 (08:06 +1000)]
Streamlined small part of pdb2gmx
In the absence of manual assignment of histidine protonation, the
automated assignment mechanism analysed the existence of hydrogen
bonds before determining whether any histidine residues existed in
the current chain. This led to confusing output on (for example)
DNA chains.
This improvement checks for the existence of at least one histidine
before analysing and reporting on the hydrogen-bonding network.
Thus pdb2gmx will process chains lacking histidine slightly faster,
and be negligibly slower for those including a histidine.
It's also a little more verbose about why it is searching for
hydrogen-bonding networks.
Mark Abraham [Fri, 15 Jul 2011 12:14:43 +0000 (22:14 +1000)]
Reading a .g96 trajectory now causes fatal error
When someone fixes the .g96 reading code to be thread-safe and
correct, this functionality can be re-instated.
IssueID #781
Berk Hess [Wed, 13 Jul 2011 15:48:06 +0000 (17:48 +0200)]
fixed div by 0 in compute_io, fixes #778
Carsten Kutzner [Mon, 11 Jul 2011 11:02:39 +0000 (13:02 +0200)]
Added two missing headers
Mark Abraham [Mon, 11 Jul 2011 07:31:31 +0000 (17:31 +1000)]
Fix some identifier naming issues in last commit
That'll teach me not to avoid doing a test compilation! :-(
Mark Abraham [Mon, 11 Jul 2011 07:00:47 +0000 (17:00 +1000)]
Avoided some integer overflows
With very large numbers of PME nodes and/or Fourier grid points,
integer overflows could occur in comm_cost_est(). Now fixed.
Mark Abraham [Sat, 2 Jul 2011 01:15:47 +0000 (11:15 +1000)]
Clarified use of OPLS/AA zwitterion termini
These are only for use with single-residue amino acids
Mark Abraham [Tue, 5 Jul 2011 06:24:34 +0000 (16:24 +1000)]
Fixed improper file close with GB and mdrun -debug
Mark Abraham [Sat, 2 Jul 2011 01:15:21 +0000 (11:15 +1000)]
Made pdb2gmx error message less cryptic
Carsten Kutzner [Wed, 6 Jul 2011 07:43:33 +0000 (09:43 +0200)]
Added globsig.h to include/types/Makefile.am
David van der Spoel [Fri, 1 Jul 2011 21:14:37 +0000 (23:14 +0200)]
Merge branch 'release-4-5-patches' of git@git.gromacs.org:gromacs into release-4-5-patches
Christoph Junghans [Fri, 1 Jul 2011 11:51:29 +0000 (13:51 +0200)]
dl lib was missing in libgmx.pc if GMX_DLOPEN is enabled
Justin A. Lemkul [Fri, 1 Jul 2011 00:45:10 +0000 (20:45 -0400)]
Fixed PME citation.
IssueID #769
Justin A. Lemkul [Wed, 29 Jun 2011 19:32:38 +0000 (15:32 -0400)]
Fixes for g_bar output. Units are now correct.
Mark Abraham [Wed, 29 Jun 2011 03:08:57 +0000 (13:08 +1000)]
Revert "draft"
This reverts commit
89abdeb7fa078c7b8dd4391aad3df2c03a60a2f0.
Oops, stuff from the gmx_packed_t branch leaked in. Sorry.
Mark Abraham [Wed, 29 Jun 2011 01:14:05 +0000 (11:14 +1000)]
Improved error message and documentation
-pbc mol now notes the need for a -s .tpr file. Also, when the index
groups (perhaps implicitly defined by a .tpr file) do not match the
input trajectory, the error message now hints that a mismatch exists.
For example, this wil help users trying to visualize the results of
genion by way of "trjconv -pbc mol -ur compact", by erroneously
using the .tpr file that was input to genion in combination with its
output file.
Mark Abraham [Sun, 19 Jun 2011 13:21:03 +0000 (23:21 +1000)]
draft
Berk Hess [Thu, 23 Jun 2011 10:58:12 +0000 (12:58 +0200)]
tpbconv can now read cpt files, fixes #741
Berk Hess [Wed, 22 Jun 2011 14:49:16 +0000 (16:49 +0200)]
fixed convergence of multiple pull constraints
With multiple constraint pull groups, only the convergence
of the last group was checked. This fixes #657
Justin A. Lemkul [Tue, 21 Jun 2011 16:40:23 +0000 (12:40 -0400)]
Fixed typos.
Justin A. Lemkul [Tue, 21 Jun 2011 14:36:33 +0000 (10:36 -0400)]
Made grompp fatal error a little more explicit.
Mark Abraham [Mon, 23 May 2011 01:14:37 +0000 (11:14 +1000)]
Expanded LINCS warning text
It now indicates which simulation in a multi-sim is the culprit
Sander Pronk [Mon, 20 Jun 2011 09:26:09 +0000 (11:26 +0200)]
Fixed symtab buffer error message.
Symtab error message now a little less cryptic. Buffer size increased to 1024.
Christoph Junghans [Thu, 16 Jun 2011 07:29:24 +0000 (09:29 +0200)]
cmake: rename LIB to GMXLIB to avoid conflict with FindMPI.cmake
IssueID #763
Berk Hess [Fri, 10 Jun 2011 16:52:50 +0000 (18:52 +0200)]
fixed grompp enx Nose-Hoover Xi reading, fixes #759
David van der Spoel [Fri, 10 Jun 2011 06:26:11 +0000 (08:26 +0200)]
Merge branch 'release-4-5-patches' of git@git.gromacs.org:gromacs into release-4-5-patches