Paul Bauer [Mon, 16 Dec 2019 11:39:33 +0000 (12:39 +0100)]
Prepare for GROMACS 2019.6
Reset doi strings
Change-Id: I38fa39c52cde5d967525770ec2a6381b7fcee33a
Paul Bauer [Mon, 16 Dec 2019 11:36:38 +0000 (12:36 +0100)]
GROMACS version 2019.5
Added doi strings
Updated regressiontest hash.
Change-Id: If6c34638caf40f562fd8a744af767be4941ff73b
Paul Bauer [Sun, 22 Dec 2019 16:04:37 +0000 (17:04 +0100)]
Revert "Fix that tests intended for multiple ranks run that way"
This reverts commit
1ebc7b0bff4a32ce957fd79a1d1bd20b535ea518.
Reason for revert: The parallel running of tests doesn't work the way we want here and the effort for fixing things in a soon to be critical fixes only branch is not worth it in my opinion
Change-Id: I59fb6f9daf7019592de7e8db202ccd050c133610
Christian Blau [Mon, 25 Nov 2019 15:42:25 +0000 (16:42 +0100)]
Remove problematic output of gmx angle
It could happen that the calculation of the standard deviation
for angles caused a divide by zero error for empty populations.
Because this standard deviation of per frame averages over all
frames was anyhow meaningless, it has been removed.
Refs #3206
Change-Id: Icf541a8b3bef502da9a22b55a8e8154b2d4fc8f8
Paul Bauer [Mon, 16 Dec 2019 12:51:02 +0000 (13:51 +0100)]
Fix formatting error
Fix originally proposed by Boris Timofeev.
Refs #3225
Change-Id: Ie742f2c751473d17fc8f69e9ed79faacdd6fc56c
Mark Abraham [Sat, 28 Sep 2019 08:25:41 +0000 (10:25 +0200)]
Fix that tests intended for multiple ranks run that way
Added the flag that triggers ctest to use the flags that will produce
multiple ranks in both thread-MPI and real MPI cases.
Change-Id: I4ef3af4fd1750ab7cc231a29191fc4042385c309
Szilárd Páll [Sat, 14 Dec 2019 22:15:44 +0000 (23:15 +0100)]
Fix the GPU bonded stream with 1 PP + 1 PME rank
WIth 1 PP + 1 PME rank the GpuBonded constructor gets passed the
non-local nonbonded stream which is nullptr and as a result the bonded
kernel launch happens in the default stream blocking concurrent
kernel execution.
This change makes sure that only when there is PP domain decomposition
is the GpuBonded constructor passed the nonlocal stream.
Fixes #3241
Change-Id: I858401b78c620adc3bea176e40e6fa179e583483
Paul Bauer [Fri, 6 Dec 2019 11:18:23 +0000 (12:18 +0100)]
Fix torsion angle calculation
Fix originally contributed by Boris Timofeev.
Fixes #3225
Change-Id: I4862679e4aeae514736df53f0eff44bf3b85f928
Paul Bauer [Wed, 27 Nov 2019 15:35:16 +0000 (16:35 +0100)]
Check for using correct hwloc headers and runtime
Also add assertion in the code to prevent errors from linking
against the wrong library while running.
Fixes #3200
Change-Id: Ib2f2861702e111f67c38b0c9d65ccbe4c81a0ccd
Mark Abraham [Wed, 31 Oct 2018 13:44:55 +0000 (14:44 +0100)]
Simplify gmxapi RPATH handling
Fixes #2722
Change-Id: Ic697f596285acdb5c7728b7ebc995ea44c82f3ac
Christian Blau [Fri, 29 Nov 2019 16:31:00 +0000 (17:31 +0100)]
Change arbitrary units to counts in RMSD distance distribution output
Users were confused by the "a.u." label in the RMSD distance
distribution output. This patch changes the label to "counts"
to better state that this is the number of events captured in the
histogram column.
Change-Id: I6c082147ae9f1647ec743275af6b72564401de8d
Berk Hess [Thu, 28 Nov 2019 10:58:48 +0000 (11:58 +0100)]
Remove assertion failure in AWH without initial stage
When not using the initial stage in AWH, we should not check
for coverage and histogram anomalies at step 0.
Fixes #3217
Change-Id: Iee19c4740c02ce56c0b43ad89efb6e37f4ad1a0d
Berk Hess [Mon, 25 Nov 2019 14:43:05 +0000 (15:43 +0100)]
Fix issues with AWH with periodic pulling
Removed fatal error with AWH with periodic pull-geometry 'direction'
when the distance was within 2% of half the box size.
Changed an assertion failure when the AWH interval was larger than
the box size to a fatal error.
Clarified the documentation for pull geometry 'direction-periodic'.
Refs #2946
Change-Id: I447a9a490a5883139afc268fd367b8b249591fbc
Christian Blau [Mon, 25 Nov 2019 15:18:40 +0000 (16:18 +0100)]
Fix duplicate CONNECT records in pdb files
When adding connect records in pdb files, it was not checked if records
already existed and were thus duplicated in output. This patch fixes
this behavior.
Fix originally provided by Boris Timofeev.
refs #3206
Change-Id: I5850f4b6279b37df07282acc4416f433099d90c8
Berk Hess [Mon, 18 Nov 2019 15:33:29 +0000 (16:33 +0100)]
Fix intermolecular interactions with DD
The links for atoms with intermolecular interactions would use
incorrect atom indices. This meant that mdrun could exit with
a missing interactions errors when using domain decompostion
with intermolecular interactions at distance larger than the cutoff.
Fixes #3204
Change-Id: I0c78810204f116f028b18c83bc6b51859ed7245a
Berk Hess [Tue, 5 Nov 2019 08:41:02 +0000 (09:41 +0100)]
Fix .gro large box writing
Contrary to the format specification in the manual, the box components
in the .gro file were not separated by whitespace when components
(except for the first one) were >= 1000 or <= -100. Now a space is
always printed. This might not match the original GROMOS format,
but most codes reading .gro file assume space separation.
Note formatting of boxes that did have spacing is not changed.
Fixes #3176
Change-Id: I8cfcde98b46550b16b9f6bc299dc0b67454b2437
Erik Lindahl [Wed, 6 Nov 2019 13:00:37 +0000 (14:00 +0100)]
Work around broken Apple compilers in Mac OS 10.15
The Xcode compiler in Mac OS Catalina checks and enforces
stack alignment by default, but embarrasingly enough the
C library provided by Apple fails to adhere to the
aligned stack for AVX despite bug reports being filed during
their beta cycle. We work around this sloppiness by removing
the check, since AVX does not require the alignment.
Fixes #3199.
Change-Id: I9c16b156cc5b3a5d9fc542335cbf63a6caf6fb1b
Berk Hess [Wed, 23 Oct 2019 09:50:02 +0000 (11:50 +0200)]
Fix illegal memory access in FE calculations
With free-energy calculations not using lambda states, an output
buffer would be accessed one element beyond it's allocated size.
Note that this code should be completely refactored, but not
in a release branch.
Fixes #3173
Change-Id: I677e602ba96c9f64fbf79a626e43c9e590c18bea
Berk Hess [Sun, 29 Sep 2019 20:21:15 +0000 (22:21 +0200)]
Increase spc minimization test tolerance
This is needed because of larger differences when running
this test on multiple ranks.
Change-Id: I3056ae5933800e30e274791e8d27e2b8e79f6e04
Berk Hess [Wed, 2 Oct 2019 12:20:25 +0000 (14:20 +0200)]
Fix harmless OpenMP write race
Change-Id: I77e014ec2005e5289a0bd13ec608c73641928b54
Berk Hess [Wed, 2 Oct 2019 09:28:06 +0000 (11:28 +0200)]
Fix GPU atom data init timer issue
The GPU atom data init timer was read conditionally on the timing
of the local pairlist transfer. But the local pairlist transfer
is not timed with an empty list, leading to an inconsistent timer
state.
Change-Id: Ifc2a63c7273ae65ae66708c6a8b0fb526041ee38
Berk Hess [Fri, 4 Oct 2019 07:43:19 +0000 (09:43 +0200)]
Fix random energy and virial with PME on GPU on a PME-only rank
Missing zero initialization of never computed LJ PME energy and virial
terms with PME on the GPU could lead to random energy, virial and
pressure numbers.
The effect of this bug was that the potential and total energy could
be off (not the Coulomb mesh energy). This didn't affect sampling.
The pressure could be off, which would affect sampling when pressure
coupling is used, but likely the simulation would explode after
a few steps.
Fixes #3120
Change-Id: I309dde958f1b73e7f71f87f4f5ad016d16f8d16b
Carsten Kutzner [Wed, 2 Oct 2019 14:09:45 +0000 (16:09 +0200)]
Update the www link to the Doxygen documentation in the dev guide
The doxygen manual seems to have been relocated from
http://www.stack.nl/~dimitri/doxygen/manual/index.html
to http://www.doxygen.nl/manual/
Change-Id: I321072065576c52a104a119abb5e05dc90eece3a
Paul Bauer [Wed, 2 Oct 2019 05:19:24 +0000 (07:19 +0200)]
Prepare for 2019.5
Reset DOI strings
Change-Id: I259bcc910b6676a9443090197f9209a399061a4b
Paul Bauer [Wed, 25 Sep 2019 09:47:18 +0000 (11:47 +0200)]
GROMACS version 2019.4
Set DOI strings.
Updated regressiontest hash
Change-Id: Ic63b9942083b122a3d5ce509da51a29a65aa89a6
Paul Bauer [Wed, 25 Sep 2019 14:36:01 +0000 (16:36 +0200)]
Fix chain ID handling in make_ndx
The tool was still using the oldest possible way to read in structure
files, leading to issues with the chain IDs being overwritten and
replaced with the default id.
Fixes #3070
Refs #3113
Change-Id: I25112103a043d0266fe64c7619fe8e4abc56ffa5
Berk Hess [Mon, 30 Sep 2019 09:19:23 +0000 (11:19 +0200)]
Always output correct kinetic at last step
When the last step did not coincide with an energy calculation step,
the kinetic energy and temperature written to the energy and log
file were incorrect when using the leap frog integrator.
Fixes #2950
Change-Id: I9a0bc2280235f233a37cd3b2cd03d9d5067bd4f9
Berk Hess [Mon, 30 Sep 2019 07:38:44 +0000 (09:38 +0200)]
Actually fix gmx wham with angle geometries
A previous fix for converting the force constant in gmx wham
for angle and dihedral geometry applied an incorrect conversion
factor.
Fixes #2609 and #3094
Change-Id: I0aa9b848c7fa62306ccf15b327a650b1ad88e03f
Mark Abraham [Thu, 26 Sep 2019 17:54:20 +0000 (19:54 +0200)]
Fix gro file output with index groups
The bug was found with editconf, but the issue was present in writing
of any gro file for which index groups were used (not just index
groups supplied by the user).
Fixes #3107
Change-Id: I3fc586c69066a354b3210d9616125ef666f1ce26
Berk Hess [Mon, 23 Sep 2019 11:33:21 +0000 (13:33 +0200)]
Limit Zen nbnxm kernel choice to Zen 1
Changed the Zen flag in gmx_hw_info_t to a Zen1 flag, so Zen 2,
and newer, is treated as a generic x86 AVX2 architecture.
Change-Id: I5a0fd73973416bbd8cd46422a6f22f1ad95d9924
Christoph Junghans [Thu, 19 Sep 2019 22:54:41 +0000 (16:54 -0600)]
workaround for rhel bug #
1749463, cmake bug #18349
MPI_C_COMPILE_FLAGS is a list instead of a string
see: https://gitlab.kitware.com/cmake/cmake/issues/18349
We mainly want to workaround this issue as RHEL 8 triggers it:
https://bugzilla.redhat.com/show_bug.cgi?id=
1749463
This issue is fixed in cmake-3.12.3
Change-Id: I14c1c0a4f33f26b128d55baa1b0455f8a463a999
Berk Hess [Sat, 21 Sep 2019 09:36:32 +0000 (11:36 +0200)]
Fix incorrect shift forces for CMAP
The shift force indices were inverted for the second and third
atom in the CMAP term, leading to incorrect virial and pressure
contributions when these atoms resided in different periodic images.
Fixes #2845 and #2867
Change-Id: I1946a1d375d6c62e4e6d23ee25b92b42a3d4a6f7
Erik Lindahl [Fri, 30 Aug 2019 19:39:21 +0000 (21:39 +0200)]
Fix CPU detection and SIMD on modern Mac/cmake combos
With recent versions of cmake on Mac, the CPU detection
program failed to compile since it was very fragile (buggy)
with respect to how compiler flags were defined - and that
in turn will mean losing all SIMD support.
This will fix the build, and it also adds a more vocal warning
if the CPU detection code cannot be compiled at all.
Change-Id: Idcdc8b59a4252b3faf8572311d2ca63a4d2a1bc0
Berk Hess [Mon, 19 Aug 2019 15:50:20 +0000 (17:50 +0200)]
Fix incorrect rvdw on GPU with rvdw<rcoulomb
When rvdw < rcoulomb was set in the mdp file, rvdw would initially
be set to rcoulomb on the GPU. With default mdrun settings,
the correct rvdw would be set after 2*nstlist steps by PME tuning.
TODO: Add an mdrun test case with rvdw<rcoulomb, refs #3062
Fixes #3056
Change-Id: I7243f27e75e46adedd668822dcd6b9045ef98a3f
Berk Hess [Mon, 26 Aug 2019 14:46:00 +0000 (16:46 +0200)]
Fix grompp constraints=all-angles with constraints
When using the mdp option constraints=all-angles, angles involving
bonds supplied as constraints in the topology would be removed,
but not replaced by angle constraints.
Fixes #3067
Change-Id: I7db390785a23c67abb30ae98e12272f88f5d5af7
Erik Lindahl [Thu, 22 Aug 2019 21:42:15 +0000 (23:42 +0200)]
Default to AVX2_256 SIMD for Zen2
From Zen2, we should no longer use the previous
hack with 128-bit AVX2 since the microarchitecture
can now execute two full-width AVX2 instructions
per cycle. Rather than specializing for Zen2, the
logic has been changed so we only apply the 128-bit
optimization for the chips where we know it helps
(Zen and Zen+, based on the model numbers), while
we default to full-width AVX2 for all other AMD
CPUs - which for now is only Zen2.
Fixes #3061.
Change-Id: I66017b200cd627bb9792f53ee39dd80d8e05965a
Berk Hess [Tue, 20 Aug 2019 06:56:43 +0000 (08:56 +0200)]
Fix missing bonded forces with CUDA GPUs and DD
Fixes issue #3063
Change-Id: Ibce7fa741a7d2171f61968f22a758b40f79a8759
Mark Abraham [Tue, 23 Jul 2019 19:46:51 +0000 (21:46 +0200)]
Fix for MSVC+CUDA
MSVC finds it hard to deal with template parameter packs in a few
situations. We don't test with MSVC+CUDA, but a user reports this works in that case.
Change-Id: Id576482facaaf40972ba4fd55b2afcba3113d0e8
Berk Hess [Tue, 14 May 2019 14:17:21 +0000 (16:17 +0200)]
Made gmx disre work with non-consecutively labeled restraints
gmx disre assumed restraint were consecutively labeled starting at 0.
Refs #2953
Change-Id: I7d9c14d43854c88cb1cdec89fc297c64cbea8783
yongwangCPH [Tue, 9 Jul 2019 13:57:59 +0000 (15:57 +0200)]
Solve issue with demux.pl
Changes printout format to resolve discontinuous trajectories
after 100ns simulation length when exchange stride is not a multiple
of 1 ps.
Change-Id: If6bd4d578ba08f9d4fd814b80e69896caf9079eb
Berk Hess [Mon, 29 Jul 2019 11:45:34 +0000 (13:45 +0200)]
Fix segmentation fault with cosine pulling
Both grompp and mdrun would fail with a segmentation fault when
using cosine-weighted COM pulling.
Fixes #3023
Change-Id: I6342a579251d21ba07a42d23ce578ae1e20c3286
David van der Spoel [Mon, 8 Jul 2019 08:24:29 +0000 (09:24 +0100)]
Fixed small bug in gmx anaeig
Poor use of variable names led to a small bug.
Fixes #2972
Change-Id: I244eac13dbaeeb0eac2be6531e579a125f4a66d3
Paul Bauer [Tue, 9 Jul 2019 08:31:45 +0000 (10:31 +0200)]
Fix link to acpype repository
Fix as reported on the user mailing list.
Change-Id: Ifdba6a60857360346653d3c9caaa929094bebae5
Paul Bauer [Mon, 8 Jul 2019 08:57:39 +0000 (10:57 +0200)]
Fix issues found by clang-tidy 7
Same issues as found for the master branch after switching the build
to run on a different machine.
Change-Id: Icf7589eae5ebc77d7974181e4faeea9e9fe8b9d7
Mark Abraham [Tue, 2 Jul 2019 13:19:54 +0000 (15:19 +0200)]
Clarify wording of release notes
A query on gmx-users indicated that the previous wording was confusing
- someone thought it might have meant normal nonbonded interactions
didn't work.
Change-Id: I0481c42ae1667bb5841e9541551dd98f5650c3ae
Paul Bauer [Mon, 1 Jul 2019 09:17:04 +0000 (11:17 +0200)]
Fix bug in gmx xpm2ps
When not providing the name for a library input file the program would
try to create a null string and fail with an assertion failure.
Fixes #3012
Change-Id: Ifd69af5420cdac2f624246aee09f36219b3f6aeb
Szilárd Páll [Thu, 25 Apr 2019 09:16:29 +0000 (11:16 +0200)]
Disable PME GPU on Apple OpenCL
Apple's OpenCL compiler seems to be unable to build a functional clFFT
which leads to a runtime error that aborts the runs rather than falling
back to PME-CPU. Hence this commit disabled PME-GPU on all Apple
platform builds.
Fixes #2941 #2719
Change-Id: I6afd19a56ced2e49d16ab598cd91527e6f73a1db
Paul Bauer [Wed, 12 Jun 2019 07:40:58 +0000 (09:40 +0200)]
Prepare for 2019.4
Reset DOI strings.
Change-Id: I1b5112bda4beb17e0f33276f60b8354b159858f0
Paul Bauer [Wed, 12 Jun 2019 07:36:58 +0000 (09:36 +0200)]
Version 2019.3
Added DOI strings.
Updated regressiontest hash.
Change-Id: If0a3389653dc3594a69886c8a9031af0dc9ddff2
Paul Bauer [Fri, 14 Jun 2019 04:54:44 +0000 (06:54 +0200)]
Merge branch release-2018 into release-2019
Change-Id: I50c1c93898246a8cbb9198227e1e77d483393bab
Paul Bauer [Wed, 12 Jun 2019 13:44:08 +0000 (15:44 +0200)]
Fix qsort not being defined on Apple Clang
Change-Id: Iffcf002d44ef8f6d174886afc67d6926aa8f2faf
Mark Abraham [Wed, 29 May 2019 16:11:02 +0000 (18:11 +0200)]
Initialize all of gmx_domdec_comm_t
Many fields were left uninitizalized when C++ new was used, rather
than snew. All fields are now initialized to zero, false, or nullptr,
in some cases via aggregate initializations of arrays. Moved the
field documentation to the line before the field and its initializer,
which reads better.
The initializers require another call to snew be replaced with one to
new[].
Fixes #2960
Change-Id: Ide917a47c667a8237519c2f4600bbce48c48f948
Paul Bauer [Tue, 14 May 2019 12:46:05 +0000 (14:46 +0200)]
Fix residue and molecule indexing in selections
The residue and molecule indexing did not account for the global
indices when building blocks from selections.
Also fixed a docs formatting issue noticed while working on the fix.
Fixes #2951
Change-Id: Ifa16f032822f5d96b6ff747333a7aed60b9a414d
Paul Bauer [Wed, 5 Jun 2019 14:04:17 +0000 (16:04 +0200)]
Extend information for gmxcpp syntax error.
Fixes #2911
Change-Id: I6d22c95d32993d5d640bf20031bfdc9c04c99604
Berk Hess [Mon, 13 May 2019 12:39:16 +0000 (14:39 +0200)]
Allow AWH geometry 'direction' to be periodic
When an AWH dimension with pull geometry 'direction' has an interval
length (nearly) matching the periodic unit cell dimension, it is now
made periodic.
Fixes #2946
Change-Id: I9a09b8eb7464e04bd3d72059324bceb590e531f5
Paul Bauer [Mon, 3 Jun 2019 15:30:37 +0000 (17:30 +0200)]
Add check for too many grid points in hbond
Fixes #2962
Change-Id: I61e80f8f254f1b5a0d968af087a87361fe4554ae
Berk Hess [Mon, 10 Jun 2019 09:12:28 +0000 (11:12 +0200)]
Fix gmx wham with angle geometries
Fixes #2609
Change-Id: I36156e1d8bb98499e4357eadc4ca56384da433a5
Berk Hess [Tue, 14 May 2019 09:53:24 +0000 (11:53 +0200)]
Fix reference SIMD memory alignment
Fixes #2952
Change-Id: I86b82c3259986c12c59c0b741df705b4774efe05
Berk Hess [Wed, 15 May 2019 11:43:07 +0000 (13:43 +0200)]
Fix missing intermolecular interactions with DD
When running with domain decomposition, all intermolecular interactions
(when present) were ignored.
Fixes #2953
Change-Id: I98783f4175b40fdfa6ad035323e0897e8caaee5c
Paul Bauer [Mon, 13 May 2019 08:40:23 +0000 (10:40 +0200)]
Version 2018.7
Updated regressiontest hash as usual.
Change-Id: Ibdd47ea72906e7696c67b5e2be6d327bf0d5deae
Berk Hess [Thu, 16 May 2019 07:54:02 +0000 (09:54 +0200)]
Work around gcc 7 avx512 bug
Due to an avx512 loop vectorization bug in gcc 7, many non-bonded
interactions could be ignored when running with more than 16 OpenMP
threads.
Fixes #2762
Change-Id: I3e03fde7114542bbd43069166a6c74937fc0f986
Paul Bauer [Tue, 21 May 2019 08:31:42 +0000 (10:31 +0200)]
Add more info for wrong line endings
Adds some more info for users that encounter preprocessing errors due
to wring line endings.
Change-Id: Id76404befdf598438f1b1fa5da98812cc31e3c5f
Paul Bauer [Tue, 21 May 2019 08:14:36 +0000 (10:14 +0200)]
Fix PQR formatting
Was still broken after the last fix a few months ago.
Fixes #2955
Change-Id: I55b1296918c723654db505d1797114e1358e5aeb
Paul Bauer [Thu, 9 May 2019 14:17:37 +0000 (16:17 +0200)]
Fix membrane embedding
Refactoring forgot to allocate some data, causing a segmentation fault.
More refactoring used wrong data, causing a crash further on.
Even more refactoring lead to invalid memory access.
Fixes #2947
Change-Id: I61ad3125102b50c0338e0935a75cd7b1de95bc3f
Berk Hess [Thu, 23 May 2019 08:27:03 +0000 (10:27 +0200)]
Fix SHAKE dH/dlambda contributions
The constraint contributions to dH/dlambda would be incorrect
with more than one SHAKE block.
Refs #2434 abd #2879
Change-Id: I0cb30a9f61893ce57d76bac34e7352fe307efe4e
Mark Abraham [Thu, 16 May 2019 07:17:04 +0000 (09:17 +0200)]
Revert "Fixed dvdlambda for SHAKE + FE"
This broke normal simulations with SHAKE and does not seem
to fix anything with FEP simulations either.
This reverts commit
33093601ff229ed8c8a40e73866a7ff351e8963b.
Fixes #2879
Change-Id: Ib6e9b4fe2b65811e9f64c37e842fd0ed9cdb055c
Paul Bauer [Tue, 21 May 2019 07:57:47 +0000 (09:57 +0200)]
Add check for allocating too many interactions
Also change argument to nbnxn reallocation function from int to
std::size_t to avoid signed integer overflow.
Fixes #2932
Change-Id: I97bdd825ba1e4d81b65b4e4c037f980164c571f2
Mark Abraham [Mon, 20 May 2019 21:28:33 +0000 (23:28 +0200)]
Fix equation typo
Change-Id: Ica8b72f0e423255c410aee3238aadfdc52f4d317
Berk Hess [Mon, 13 May 2019 13:56:19 +0000 (15:56 +0200)]
Correct user guide postion restraint virial
Corrected comment in the user guide note on virial and pressure
with position restraints and refcoord-scaling none or com.
Refs #2942
Change-Id: I3cdb5a01640c388f1c4222b1a43b342ac55fd882
Paul Bauer [Mon, 13 May 2019 08:30:54 +0000 (10:30 +0200)]
Prepare for 2018.7
Also add the release note update for Redmine issue 2845.
Change-Id: I63c6ef4bf4ea2d78741ab3bf77079ea257f1a205
Mark Abraham [Tue, 7 May 2019 07:09:56 +0000 (09:09 +0200)]
Add an example of using an MPI wrapper compiler
A long-time GROMACS user reported not being able to make an MPI build,
and this might help clue people in.
Change-Id: Ie3f442e126561bb5c6e1ae71fee650f2e18c21aa
Berk Hess [Fri, 3 May 2019 13:23:37 +0000 (15:23 +0200)]
Removed non-existent AWH mdp option
Removed non-existent mdp option awh1-dim1-period from user guide.
Fixes #2940
Change-Id: I96c4f1e636e0d171561dcfcf1e17e7f0ce419f8b
Paul Bauer [Thu, 25 Apr 2019 11:50:14 +0000 (13:50 +0200)]
Fix FPE in do_steep
Fixes error when running non-interacting system in steepest decent
minimization.
Fixes #2917
Change-Id: Iaff794ab6c2d41e8c4b3d89feeafae6ae2372ddf
Paul Bauer [Thu, 11 Apr 2019 08:49:50 +0000 (10:49 +0200)]
Prepare for version 2019.3
Reset DOI strings
Change-Id: Iddbafb48832b28eb862a9a7cbe632e9e7a6fdbf7
Paul Bauer [Thu, 11 Apr 2019 08:35:46 +0000 (10:35 +0200)]
Version 2019.2
Set DOI strings.
Updated regressiontest hash.
Change-Id: I87dd50fd154c4f10a04e48be0c4fbe60964395ff
Paul Bauer [Mon, 15 Apr 2019 15:51:57 +0000 (17:51 +0200)]
Fix off-by-one error in gmx cluster
The writing of indices for linking frames to clusters was of by one.
Also fixed naming to refer to indices and not indexes
Fixes #2926
Change-Id: I67d7f8cae56734e425c84460080de8896f6239cb
Berk Hess [Fri, 8 Mar 2019 12:53:04 +0000 (13:53 +0100)]
Make mdrun -nsteps work again.
The mdrun -nsteps option was only functional in some cases,
because it modified ir->nsteps after the checkpoint loading
modified it.
Note that this option is deprecated, but is should either work
or be removed.
Fixes #2881
Change-Id: Ib611b0e007171f8a351f61f3ecc9b1bda5248486
Berk Hess [Mon, 15 Apr 2019 11:31:31 +0000 (13:31 +0200)]
Update 2019 release notes for issue 2845
We now know that issue reported in 2845 is not actually related to
LJ force switching. So the "fix" in 2019.1 fixes a bug, but the
issue 2845. This is now reported in the release notes for 2019.1
and 2019.2.
Change-Id: I000c518850f970c0f9cc331df22afaa24d9a27bc
Szilárd Páll [Mon, 15 Apr 2019 09:50:35 +0000 (11:50 +0200)]
Add user guide note on AVX512 throttle in GPU runs
Refs #2793
Change-Id: I161f5b92a8a34989ecc1d49410033f9ea8f91164
Berk Hess [Sat, 13 Apr 2019 10:00:06 +0000 (12:00 +0200)]
Fix pdb-related issues
In editconf, if a PDB input was read, chain IDs are known, and should
be written if the output is also PDB. This now works again, after
being broken in commit
8dd3c9ae88004054b3.
This required adding missing tpr support to readConfAndAtoms(), which
also fixes that pdb2gmx, insert-molecules, and solvate could not read
tpr files. Those are the only tools that directly or indirectly called
readConfAndAtoms with a .tpr file.
Fixes #2900
Change-Id: I883777be945023a7e69260b22e76df54477828ee
Berk Hess [Mon, 15 Apr 2019 09:31:19 +0000 (11:31 +0200)]
Disallow pull geometry direction-periodic with AWH
Fixes #2923
Change-Id: Ic309a5598c69e82824d80270466a2de7b00e7d41
Mark Abraham [Tue, 9 Apr 2019 12:51:44 +0000 (14:51 +0200)]
Fix self-consistency tests of hwloc data structures
These relied on assumptions of regularity that have been shown to be
violated in practice. The new tests check that there is a bijective
mapping of logical processors to valid hardware thread descriptors.
Fixes #2921
Change-Id: I31e998b7a2881c05dfb5c7c8b46550489cbdefd6
Mark Abraham [Fri, 12 Apr 2019 16:27:29 +0000 (18:27 +0200)]
Improve warning messages about duplicate definitions
Because the warning is fatal, the wording with "overriding" is
misleading. An override only occurs if the warning is suppressed.
Email on gmx-users made clear that the word "overriding" is not clear
enough for all users to understand that more than one definition was
provided. They do not necessarily know that grompp has already read in
the force field definition. Typically a self-contained molecule .itp
file might override definitions found in the force-field (whether
duplicate or different). In such cases, the user needs to consider
what their intentions are and to express them clearly.
Change-Id: Ie73f1dbe1233d6fd0df41628c770b8f2ad1f1b1b
Mark Abraham [Tue, 9 Apr 2019 12:50:47 +0000 (14:50 +0200)]
Fix typo in AVX512 detection code
Change-Id: I2fb425a1649ac7d06480831016d6c35e7e63ce6f
David van der Spoel [Fri, 29 Mar 2019 12:31:11 +0000 (13:31 +0100)]
Fixed manual for dihedral restraints.
The equations in the manual for dihedral restraints were incorrect
and inconsistent with the code in bonded.cpp.
Change-Id: I335f72cf81990e636366ae75f890df85c5053388
Морозов Дмитрий [Wed, 3 Apr 2019 12:02:21 +0000 (15:02 +0300)]
Fixes bug in l-bfgs.
l-bfgs should now work excatly as it was in 4.6.5,
up-to some numerical precision in forces and energies.
Fixes #2641
Change-Id: I11a85a4241cc933fef94e2095dd8c3cbbb28b01b
Szilárd Páll [Thu, 4 Apr 2019 16:59:20 +0000 (18:59 +0200)]
Revert "Temporarily disable ARM HPC compiler in post-submit"
This reverts commit
2e63850dc17bc9f718344e68cd245c9c96a2256b.
Reason for revert: build slave software issues resolved.
Change-Id: I2a04afeeb51fe0177631c1bf55bed52ea2b0ea63
Paul Bauer [Thu, 28 Feb 2019 09:39:57 +0000 (10:39 +0100)]
Fix segmentation fault with simulated annealing
Fix contributed by Daniel Fragiadakis.
Fixes #2871
Change-Id: I237d49f9996eb2ae869d59e4c39a6c2d33c9ece7
Szilárd Páll [Mon, 1 Apr 2019 19:51:32 +0000 (21:51 +0200)]
Temporarily disable ARM HPC compiler in post-submit
Change-Id: I27a6699109e5d7ce3576200a3a977b7a63a305aa
Szilárd Páll [Mon, 25 Feb 2019 15:29:57 +0000 (16:29 +0100)]
Better document GMX_OPENMP_MAX_THREADS
Change-Id: I2c44f8b2331d69fde5fa242b5e53132813a2cb31
Roland Schulz [Tue, 26 Mar 2019 00:24:39 +0000 (17:24 -0700)]
Fix UB vector usage
Fixes a buffer overflow for c_simdBestPairAlignment=2.
Found with _LIBCPP_DEBUG=1.
Change-Id: Ib21ca875244673b27748a01373e7fc10252a7c44
Mark Abraham [Tue, 19 Mar 2019 09:23:18 +0000 (10:23 +0100)]
Prevented FFTW build errors with clang and AVX-512
FFTW hard-codes the inclusion of files that does not work with its
AVX-512 flags when using thee clang compiler, so in this case
GROMACS should compile without AVX-512 support.
Updated release notes, but didn't update the install guide for this
niche case.
Fixes #2892
Change-Id: Ida18fad66e654511937d51ee58f42440de2a321d
Mark Abraham [Wed, 13 Mar 2019 08:59:27 +0000 (09:59 +0100)]
Provide reminder of which nvcc is in use
FindCUDA.cmake + the user might have combined to find an unexpected
nvcc, so we help everyone help themselves by outputting it here,
particularly if they report problems on the users list.
Change-Id: Ic951fa219c9133760a3c97ad2a92d65d64f03bfb
Paul Bauer [Tue, 12 Mar 2019 10:27:58 +0000 (11:27 +0100)]
Fix formatting error in user guide reference
Change-Id: I02a25bb81f9c30ee7ddf7ed6f3757cb494ac4f06
Berk Hess [Thu, 7 Mar 2019 17:41:25 +0000 (18:41 +0100)]
Add grompp warning with use of GROMOS FF
Replaced GROMOS-based simulation database entry with similar
OPLS-based one, since our test systems should not trigger warnings
unless that is the intent of the test.
Refs #2884
Change-Id: I55afcf7ea5b423fa27792acebe09ea520b475974
Roland Schulz [Fri, 8 Mar 2019 15:52:29 +0000 (07:52 -0800)]
Fix ICC warnings
Change-Id: I5170eacf9ac7590d2d72dbb1dac3316a1b2e8ab0
Paul Bauer [Sun, 24 Feb 2019 06:44:49 +0000 (07:44 +0100)]
Add error when manual cannot be built
Added error message informing the user why a manual build is not
possible if Sphinx is not installed.
Change-Id: I74765960ea5f5f642771e75ac349cf2b55d360a9
Szilárd Páll [Tue, 9 Oct 2018 00:56:23 +0000 (02:56 +0200)]
Relax OpenCL gather kernel barrier on AMD
Not needed on arch with >32 execution width.
Refs #2519
Change-Id: I4c8f2cff405cd3044bd60b99f01bcdd918dc5d0e
Jukka Maatta [Wed, 27 Feb 2019 15:03:52 +0000 (17:03 +0200)]
Make OpenCL dummyKernel to require a input parameter
This is a tentative fix for OpenCL error with mdrun on MacOS.
MacOS seems to require that kernel functions have at least one parameter.
Otherwise clBuildProgram fails with error -11.
Here dummyKernel takes an empty pointer in the definition.
We then pass an empty pointer to dummyKernel with clSetKernelArg.
Fixes #2865
Change-Id: Ib7c08eeeb2ec6d8a43bfe703cf1b273819a45a29
Mark Abraham [Tue, 26 Feb 2019 21:41:26 +0000 (13:41 -0800)]
Keep niche CMake variables as advanced
Change-Id: I99781749d9f4f481f7943a27c415af6b5c4ef5bc