Commit 8dfdfb31 authored by Nicholas Breen's avatar Nicholas Breen

Update MPI man pages

parent 08373aaa
......@@ -2,6 +2,7 @@ gromacs (2018-1) UNRELEASED; urgency=medium
* New upstream release.
- Remove test_GMX_ALIGNED.patch, incorporated.
- Update debian/man/*mdrun* man pages.
* Bump Standards-Version to 4.1.3.
-- Nicholas Breen <nbreen@debian.org> Wed, 10 Jan 2018 19:04:30 -0800
......
.\" Man page generated from reStructuredText.
.
.TH "MDRUN_MPI" "1" "Feb 07, 2017" "2016.2" "GROMACS"
.TH "MDRUN_MPI" "1" "Jan 11, 2018" "2018" "GROMACS"
.SH NAME
mdrun_mpi \- Perform a simulation, do a normal mode analysis or an energy minimization
.
......@@ -39,24 +39,26 @@ level margin: \\n[rst2man-indent\\n[rst2man-indent-level]]
mdrun_mpi [\fB\-s\fP \fI[<.tpr>]\fP] [\fB\-cpi\fP \fI[<.cpt>]\fP] [\fB\-table\fP \fI[<.xvg>]\fP]
[\fB\-tablep\fP \fI[<.xvg>]\fP] [\fB\-tableb\fP \fI[<.xvg> [...]]\fP]
[\fB\-rerun\fP \fI[<.xtc/.trr/...>]\fP] [\fB\-ei\fP \fI[<.edi>]\fP]
[\fB\-multidir\fP \fI[<dir> [...]]\fP] [\fB\-membed\fP \fI[<.dat>]\fP] [\fB\-mp\fP \fI[<.top>]\fP]
[\fB\-mn\fP \fI[<.ndx>]\fP] [\fB\-o\fP \fI[<.trr/.cpt/...>]\fP] [\fB\-x\fP \fI[<.xtc/.tng>]\fP]
[\fB\-cpo\fP \fI[<.cpt>]\fP] [\fB\-c\fP \fI[<.gro/.g96/...>]\fP] [\fB\-e\fP \fI[<.edr>]\fP]
[\fB\-g\fP \fI[<.log>]\fP] [\fB\-dhdl\fP \fI[<.xvg>]\fP] [\fB\-field\fP \fI[<.xvg>]\fP]
[\fB\-tpi\fP \fI[<.xvg>]\fP] [\fB\-tpid\fP \fI[<.xvg>]\fP] [\fB\-eo\fP \fI[<.xvg>]\fP]
[\fB\-devout\fP \fI[<.xvg>]\fP] [\fB\-runav\fP \fI[<.xvg>]\fP] [\fB\-px\fP \fI[<.xvg>]\fP]
[\fB\-pf\fP \fI[<.xvg>]\fP] [\fB\-ro\fP \fI[<.xvg>]\fP] [\fB\-ra\fP \fI[<.log>]\fP] [\fB\-rs\fP \fI[<.log>]\fP]
[\fB\-rt\fP \fI[<.log>]\fP] [\fB\-mtx\fP \fI[<.mtx>]\fP] [\fB\-if\fP \fI[<.xvg>]\fP]
[\fB\-swap\fP \fI[<.xvg>]\fP] [\fB\-deffnm\fP \fI<string>\fP] [\fB\-xvg\fP \fI<enum>\fP]
[\fB\-dd\fP \fI<vector>\fP] [\fB\-ddorder\fP \fI<enum>\fP] [\fB\-npme\fP \fI<int>\fP] [\fB\-nt\fP \fI<int>\fP]
[\fB\-ntmpi\fP \fI<int>\fP] [\fB\-ntomp\fP \fI<int>\fP] [\fB\-ntomp_pme\fP \fI<int>\fP]
[\fB\-pin\fP \fI<enum>\fP] [\fB\-pinoffset\fP \fI<int>\fP] [\fB\-pinstride\fP \fI<int>\fP]
[\fB\-gpu_id\fP \fI<string>\fP] [\fB\-[no]ddcheck\fP] [\fB\-rdd\fP \fI<real>\fP]
[\fB\-multidir\fP \fI[<dir> [...]]\fP] [\fB\-awh\fP \fI[<.xvg>]\fP]
[\fB\-membed\fP \fI[<.dat>]\fP] [\fB\-mp\fP \fI[<.top>]\fP] [\fB\-mn\fP \fI[<.ndx>]\fP]
[\fB\-o\fP \fI[<.trr/.cpt/...>]\fP] [\fB\-x\fP \fI[<.xtc/.tng>]\fP] [\fB\-cpo\fP \fI[<.cpt>]\fP]
[\fB\-c\fP \fI[<.gro/.g96/...>]\fP] [\fB\-e\fP \fI[<.edr>]\fP] [\fB\-g\fP \fI[<.log>]\fP]
[\fB\-dhdl\fP \fI[<.xvg>]\fP] [\fB\-field\fP \fI[<.xvg>]\fP] [\fB\-tpi\fP \fI[<.xvg>]\fP]
[\fB\-tpid\fP \fI[<.xvg>]\fP] [\fB\-eo\fP \fI[<.xvg>]\fP] [\fB\-devout\fP \fI[<.xvg>]\fP]
[\fB\-runav\fP \fI[<.xvg>]\fP] [\fB\-px\fP \fI[<.xvg>]\fP] [\fB\-pf\fP \fI[<.xvg>]\fP]
[\fB\-ro\fP \fI[<.xvg>]\fP] [\fB\-ra\fP \fI[<.log>]\fP] [\fB\-rs\fP \fI[<.log>]\fP] [\fB\-rt\fP \fI[<.log>]\fP]
[\fB\-mtx\fP \fI[<.mtx>]\fP] [\fB\-if\fP \fI[<.xvg>]\fP] [\fB\-swap\fP \fI[<.xvg>]\fP]
[\fB\-deffnm\fP \fI<string>\fP] [\fB\-xvg\fP \fI<enum>\fP] [\fB\-dd\fP \fI<vector>\fP]
[\fB\-ddorder\fP \fI<enum>\fP] [\fB\-npme\fP \fI<int>\fP] [\fB\-nt\fP \fI<int>\fP] [\fB\-ntmpi\fP \fI<int>\fP]
[\fB\-ntomp\fP \fI<int>\fP] [\fB\-ntomp_pme\fP \fI<int>\fP] [\fB\-pin\fP \fI<enum>\fP]
[\fB\-pinoffset\fP \fI<int>\fP] [\fB\-pinstride\fP \fI<int>\fP] [\fB\-gpu_id\fP \fI<string>\fP]
[\fB\-gputasks\fP \fI<string>\fP] [\fB\-[no]ddcheck\fP] [\fB\-rdd\fP \fI<real>\fP]
[\fB\-rcon\fP \fI<real>\fP] [\fB\-dlb\fP \fI<enum>\fP] [\fB\-dds\fP \fI<real>\fP] [\fB\-gcom\fP \fI<int>\fP]
[\fB\-nb\fP \fI<enum>\fP] [\fB\-nstlist\fP \fI<int>\fP] [\fB\-[no]tunepme\fP] [\fB\-[no]v\fP]
[\fB\-pforce\fP \fI<real>\fP] [\fB\-[no]reprod\fP] [\fB\-cpt\fP \fI<real>\fP] [\fB\-[no]cpnum\fP]
[\fB\-[no]append\fP] [\fB\-nsteps\fP \fI<int>\fP] [\fB\-maxh\fP \fI<real>\fP] [\fB\-multi\fP \fI<int>\fP]
[\fB\-replex\fP \fI<int>\fP] [\fB\-nex\fP \fI<int>\fP] [\fB\-reseed\fP \fI<int>\fP]
[\fB\-nb\fP \fI<enum>\fP] [\fB\-nstlist\fP \fI<int>\fP] [\fB\-[no]tunepme\fP] [\fB\-pme\fP \fI<enum>\fP]
[\fB\-pmefft\fP \fI<enum>\fP] [\fB\-[no]v\fP] [\fB\-pforce\fP \fI<real>\fP] [\fB\-[no]reprod\fP]
[\fB\-cpt\fP \fI<real>\fP] [\fB\-[no]cpnum\fP] [\fB\-[no]append\fP] [\fB\-nsteps\fP \fI<int>\fP]
[\fB\-maxh\fP \fI<real>\fP] [\fB\-multi\fP \fI<int>\fP] [\fB\-replex\fP \fI<int>\fP] [\fB\-nex\fP \fI<int>\fP]
[\fB\-reseed\fP \fI<int>\fP]
.ft P
.fi
.UNINDENT
......@@ -121,7 +123,7 @@ interaction functions are read using the \fB\-tableb\fP option.
For each different tabulated interaction type used, a table file name must
be given. For the topology to work, a file name given here must match a
character sequence before the file extension. That sequence is: an underscore,
then a \(aqb\(aq for bonds, an \(aqa\(aq for angles or a \(aqd\(aq for dihedrals,
then a ‘b’ for bonds, an ‘a’ for angles or a ‘d’ for dihedrals,
and finally the matching table number index used in the topology.
.sp
The options \fB\-px\fP and \fB\-pf\fP are used for writing pull COM
......@@ -142,7 +144,8 @@ the index and topology files used for the embedding.
The option \fB\-pforce\fP is useful when you suspect a simulation
crashes due to too large forces. With this option coordinates and
forces of atoms with a force larger than a certain value will
be printed to stderr.
be printed to stderr. It will also terminate the run when non\-finite
forces are present.
.sp
Checkpoints containing the complete state of the system are written
at regular intervals (option \fB\-cpt\fP) to the file \fB\-cpo\fP,
......@@ -194,12 +197,16 @@ combination with setting \fBnsteps\fP to \-1 either in the mdp or using the
similarly named command line option. This results in an infinite run,
terminated only when the time limit set by \fB\-maxh\fP is reached (if any)or upon receiving a signal.
.sp
When \fBmdrun\fP receives a TERM signal, it will stop as soon as
checkpoint file can be written, i.e. after the next global communication step.
When \fBmdrun\fP receives an INT signal (e.g. when ctrl+C is
When \fBmdrun\fP receives a TERM or INT signal (e.g. when ctrl+C is
pressed), it will stop at the next neighbor search step or at the
second global communication step, whichever happens later.
In both cases all the usual output will be written to file.
When \fBmdrun\fP receives a second TERM or INT signal and
reproducibility is not requested, it will stop at the first global
communication step.
In both cases all the usual output will be written to file and
a checkpoint file is written at the last step.
When \fBmdrun\fP receives an ABRT signal or the third TERM or INT signal,
it will abort directly without writing a new checkpoint file.
When running with MPI, a signal to one of the \fBmdrun\fP ranks
is sufficient, this signal should not be sent to mpirun or
the \fBmdrun\fP process that is the parent of the others.
......@@ -231,18 +238,21 @@ xvgr/xmgr file
.B \fB\-tablep\fP [<.xvg>] (tablep.xvg) (Optional)
xvgr/xmgr file
.TP
.B \fB\-tableb\fP [<.xvg> [...]] (table.xvg) (Optional)
.B \fB\-tableb\fP [<.xvg> []] (table.xvg) (Optional)
xvgr/xmgr file
.TP
.B \fB\-rerun\fP [<.xtc/.trr/...>] (rerun.xtc) (Optional)
.B \fB\-rerun\fP [<.xtc/.trr/>] (rerun.xtc) (Optional)
Trajectory: xtc trr cpt gro g96 pdb tng
.TP
.B \fB\-ei\fP [<.edi>] (sam.edi) (Optional)
ED sampling input
.TP
.B \fB\-multidir\fP [<dir> [...]] (rundir) (Optional)
.B \fB\-multidir\fP [<dir> []] (rundir) (Optional)
Run directory
.TP
.B \fB\-awh\fP [<.xvg>] (awhinit.xvg) (Optional)
xvgr/xmgr file
.TP
.B \fB\-membed\fP [<.dat>] (membed.dat) (Optional)
Generic data file
.TP
......@@ -256,7 +266,7 @@ Index file
Options to specify output files:
.INDENT 0.0
.TP
.B \fB\-o\fP [<.trr/.cpt/...>] (traj.trr)
.B \fB\-o\fP [<.trr/.cpt/>] (traj.trr)
Full precision trajectory: trr cpt tng
.TP
.B \fB\-x\fP [<.xtc/.tng>] (traj_comp.xtc) (Optional)
......@@ -265,7 +275,7 @@ Compressed trajectory (tng format or portable xdr format)
.B \fB\-cpo\fP [<.cpt>] (state.cpt) (Optional)
Checkpoint file
.TP
.B \fB\-c\fP [<.gro/.g96/...>] (confout.gro)
.B \fB\-c\fP [<.gro/.g96/>] (confout.gro)
Structure file: gro g96 pdb brk ent esp
.TP
.B \fB\-e\fP [<.edr>] (ener.edr)
......@@ -345,7 +355,7 @@ Number of separate ranks to be used for PME, \-1 is guess
Total number of threads to start (0 is guess)
.TP
.B \fB\-ntmpi\fP <int> (0)
Number of thread\-MPI threads to start (0 is guess)
Number of thread\-MPI ranks to start (0 is guess)
.TP
.B \fB\-ntomp\fP <int> (0)
Number of OpenMP threads per MPI rank to start (0 is guess)
......@@ -363,7 +373,10 @@ The lowest logical core number to which mdrun should pin the first thread
Pinning distance in logical cores for threads, use 0 to minimize the number of threads per physical core
.TP
.B \fB\-gpu_id\fP <string>
List of GPU device id\-s to use, specifies the per\-node PP rank to GPU mapping
List of unique GPU device IDs available to use
.TP
.B \fB\-gputasks\fP <string>
List of GPU device IDs, mapping each PP task on each node to a device
.TP
.B \fB\-[no]ddcheck\fP (yes)
Check for all bonded interactions with DD
......@@ -384,13 +397,19 @@ Fraction in (0,1) by whose reciprocal the initial DD cell size will be increased
Global communication frequency
.TP
.B \fB\-nb\fP <enum> (auto)
Calculate non\-bonded interactions on: auto, cpu, gpu, gpu_cpu
Calculate non\-bonded interactions on: auto, cpu, gpu
.TP
.B \fB\-nstlist\fP <int> (0)
Set nstlist when using a Verlet buffer tolerance (0 is guess)
.TP
.B \fB\-[no]tunepme\fP (yes)
Optimize PME load between PP/PME ranks or GPU/CPU
Optimize PME load between PP/PME ranks or GPU/CPU (only with the Verlet cut\-off scheme)
.TP
.B \fB\-pme\fP <enum> (auto)
Perform PME calculations on: auto, cpu, gpu
.TP
.B \fB\-pmefft\fP <enum> (auto)
Perform PME FFT calculations on: auto, cpu, gpu
.TP
.B \fB\-[no]v\fP (no)
Be loud and noisy
......@@ -434,6 +453,6 @@ Seed for replica exchange, \-1 is generate a seed
.sp
More information about GROMACS is available at <\fI\%http://www.gromacs.org/\fP>.
.SH COPYRIGHT
2017, GROMACS development team
2018, GROMACS development team
.\" Generated by docutils manpage writer.
.
.\" Man page generated from reStructuredText.
.
.TH "MDRUN_MPI" "1" "Feb 07, 2017" "2016.2" "GROMACS"
.TH "MDRUN_MPI" "1" "Jan 11, 2018" "2018" "GROMACS"
.SH NAME
mdrun_mpi \- Perform a simulation, do a normal mode analysis or an energy minimization
.
......@@ -39,24 +39,26 @@ level margin: \\n[rst2man-indent\\n[rst2man-indent-level]]
mdrun_mpi [\fB\-s\fP \fI[<.tpr>]\fP] [\fB\-cpi\fP \fI[<.cpt>]\fP] [\fB\-table\fP \fI[<.xvg>]\fP]
[\fB\-tablep\fP \fI[<.xvg>]\fP] [\fB\-tableb\fP \fI[<.xvg> [...]]\fP]
[\fB\-rerun\fP \fI[<.xtc/.trr/...>]\fP] [\fB\-ei\fP \fI[<.edi>]\fP]
[\fB\-multidir\fP \fI[<dir> [...]]\fP] [\fB\-membed\fP \fI[<.dat>]\fP] [\fB\-mp\fP \fI[<.top>]\fP]
[\fB\-mn\fP \fI[<.ndx>]\fP] [\fB\-o\fP \fI[<.trr/.cpt/...>]\fP] [\fB\-x\fP \fI[<.xtc/.tng>]\fP]
[\fB\-cpo\fP \fI[<.cpt>]\fP] [\fB\-c\fP \fI[<.gro/.g96/...>]\fP] [\fB\-e\fP \fI[<.edr>]\fP]
[\fB\-g\fP \fI[<.log>]\fP] [\fB\-dhdl\fP \fI[<.xvg>]\fP] [\fB\-field\fP \fI[<.xvg>]\fP]
[\fB\-tpi\fP \fI[<.xvg>]\fP] [\fB\-tpid\fP \fI[<.xvg>]\fP] [\fB\-eo\fP \fI[<.xvg>]\fP]
[\fB\-devout\fP \fI[<.xvg>]\fP] [\fB\-runav\fP \fI[<.xvg>]\fP] [\fB\-px\fP \fI[<.xvg>]\fP]
[\fB\-pf\fP \fI[<.xvg>]\fP] [\fB\-ro\fP \fI[<.xvg>]\fP] [\fB\-ra\fP \fI[<.log>]\fP] [\fB\-rs\fP \fI[<.log>]\fP]
[\fB\-rt\fP \fI[<.log>]\fP] [\fB\-mtx\fP \fI[<.mtx>]\fP] [\fB\-if\fP \fI[<.xvg>]\fP]
[\fB\-swap\fP \fI[<.xvg>]\fP] [\fB\-deffnm\fP \fI<string>\fP] [\fB\-xvg\fP \fI<enum>\fP]
[\fB\-dd\fP \fI<vector>\fP] [\fB\-ddorder\fP \fI<enum>\fP] [\fB\-npme\fP \fI<int>\fP] [\fB\-nt\fP \fI<int>\fP]
[\fB\-ntmpi\fP \fI<int>\fP] [\fB\-ntomp\fP \fI<int>\fP] [\fB\-ntomp_pme\fP \fI<int>\fP]
[\fB\-pin\fP \fI<enum>\fP] [\fB\-pinoffset\fP \fI<int>\fP] [\fB\-pinstride\fP \fI<int>\fP]
[\fB\-gpu_id\fP \fI<string>\fP] [\fB\-[no]ddcheck\fP] [\fB\-rdd\fP \fI<real>\fP]
[\fB\-multidir\fP \fI[<dir> [...]]\fP] [\fB\-awh\fP \fI[<.xvg>]\fP]
[\fB\-membed\fP \fI[<.dat>]\fP] [\fB\-mp\fP \fI[<.top>]\fP] [\fB\-mn\fP \fI[<.ndx>]\fP]
[\fB\-o\fP \fI[<.trr/.cpt/...>]\fP] [\fB\-x\fP \fI[<.xtc/.tng>]\fP] [\fB\-cpo\fP \fI[<.cpt>]\fP]
[\fB\-c\fP \fI[<.gro/.g96/...>]\fP] [\fB\-e\fP \fI[<.edr>]\fP] [\fB\-g\fP \fI[<.log>]\fP]
[\fB\-dhdl\fP \fI[<.xvg>]\fP] [\fB\-field\fP \fI[<.xvg>]\fP] [\fB\-tpi\fP \fI[<.xvg>]\fP]
[\fB\-tpid\fP \fI[<.xvg>]\fP] [\fB\-eo\fP \fI[<.xvg>]\fP] [\fB\-devout\fP \fI[<.xvg>]\fP]
[\fB\-runav\fP \fI[<.xvg>]\fP] [\fB\-px\fP \fI[<.xvg>]\fP] [\fB\-pf\fP \fI[<.xvg>]\fP]
[\fB\-ro\fP \fI[<.xvg>]\fP] [\fB\-ra\fP \fI[<.log>]\fP] [\fB\-rs\fP \fI[<.log>]\fP] [\fB\-rt\fP \fI[<.log>]\fP]
[\fB\-mtx\fP \fI[<.mtx>]\fP] [\fB\-if\fP \fI[<.xvg>]\fP] [\fB\-swap\fP \fI[<.xvg>]\fP]
[\fB\-deffnm\fP \fI<string>\fP] [\fB\-xvg\fP \fI<enum>\fP] [\fB\-dd\fP \fI<vector>\fP]
[\fB\-ddorder\fP \fI<enum>\fP] [\fB\-npme\fP \fI<int>\fP] [\fB\-nt\fP \fI<int>\fP] [\fB\-ntmpi\fP \fI<int>\fP]
[\fB\-ntomp\fP \fI<int>\fP] [\fB\-ntomp_pme\fP \fI<int>\fP] [\fB\-pin\fP \fI<enum>\fP]
[\fB\-pinoffset\fP \fI<int>\fP] [\fB\-pinstride\fP \fI<int>\fP] [\fB\-gpu_id\fP \fI<string>\fP]
[\fB\-gputasks\fP \fI<string>\fP] [\fB\-[no]ddcheck\fP] [\fB\-rdd\fP \fI<real>\fP]
[\fB\-rcon\fP \fI<real>\fP] [\fB\-dlb\fP \fI<enum>\fP] [\fB\-dds\fP \fI<real>\fP] [\fB\-gcom\fP \fI<int>\fP]
[\fB\-nb\fP \fI<enum>\fP] [\fB\-nstlist\fP \fI<int>\fP] [\fB\-[no]tunepme\fP] [\fB\-[no]v\fP]
[\fB\-pforce\fP \fI<real>\fP] [\fB\-[no]reprod\fP] [\fB\-cpt\fP \fI<real>\fP] [\fB\-[no]cpnum\fP]
[\fB\-[no]append\fP] [\fB\-nsteps\fP \fI<int>\fP] [\fB\-maxh\fP \fI<real>\fP] [\fB\-multi\fP \fI<int>\fP]
[\fB\-replex\fP \fI<int>\fP] [\fB\-nex\fP \fI<int>\fP] [\fB\-reseed\fP \fI<int>\fP]
[\fB\-nb\fP \fI<enum>\fP] [\fB\-nstlist\fP \fI<int>\fP] [\fB\-[no]tunepme\fP] [\fB\-pme\fP \fI<enum>\fP]
[\fB\-pmefft\fP \fI<enum>\fP] [\fB\-[no]v\fP] [\fB\-pforce\fP \fI<real>\fP] [\fB\-[no]reprod\fP]
[\fB\-cpt\fP \fI<real>\fP] [\fB\-[no]cpnum\fP] [\fB\-[no]append\fP] [\fB\-nsteps\fP \fI<int>\fP]
[\fB\-maxh\fP \fI<real>\fP] [\fB\-multi\fP \fI<int>\fP] [\fB\-replex\fP \fI<int>\fP] [\fB\-nex\fP \fI<int>\fP]
[\fB\-reseed\fP \fI<int>\fP]
.ft P
.fi
.UNINDENT
......@@ -121,7 +123,7 @@ interaction functions are read using the \fB\-tableb\fP option.
For each different tabulated interaction type used, a table file name must
be given. For the topology to work, a file name given here must match a
character sequence before the file extension. That sequence is: an underscore,
then a \(aqb\(aq for bonds, an \(aqa\(aq for angles or a \(aqd\(aq for dihedrals,
then a ‘b’ for bonds, an ‘a’ for angles or a ‘d’ for dihedrals,
and finally the matching table number index used in the topology.
.sp
The options \fB\-px\fP and \fB\-pf\fP are used for writing pull COM
......@@ -142,7 +144,8 @@ the index and topology files used for the embedding.
The option \fB\-pforce\fP is useful when you suspect a simulation
crashes due to too large forces. With this option coordinates and
forces of atoms with a force larger than a certain value will
be printed to stderr.
be printed to stderr. It will also terminate the run when non\-finite
forces are present.
.sp
Checkpoints containing the complete state of the system are written
at regular intervals (option \fB\-cpt\fP) to the file \fB\-cpo\fP,
......@@ -194,12 +197,16 @@ combination with setting \fBnsteps\fP to \-1 either in the mdp or using the
similarly named command line option. This results in an infinite run,
terminated only when the time limit set by \fB\-maxh\fP is reached (if any)or upon receiving a signal.
.sp
When \fBmdrun\fP receives a TERM signal, it will stop as soon as
checkpoint file can be written, i.e. after the next global communication step.
When \fBmdrun\fP receives an INT signal (e.g. when ctrl+C is
When \fBmdrun\fP receives a TERM or INT signal (e.g. when ctrl+C is
pressed), it will stop at the next neighbor search step or at the
second global communication step, whichever happens later.
In both cases all the usual output will be written to file.
When \fBmdrun\fP receives a second TERM or INT signal and
reproducibility is not requested, it will stop at the first global
communication step.
In both cases all the usual output will be written to file and
a checkpoint file is written at the last step.
When \fBmdrun\fP receives an ABRT signal or the third TERM or INT signal,
it will abort directly without writing a new checkpoint file.
When running with MPI, a signal to one of the \fBmdrun\fP ranks
is sufficient, this signal should not be sent to mpirun or
the \fBmdrun\fP process that is the parent of the others.
......@@ -231,18 +238,21 @@ xvgr/xmgr file
.B \fB\-tablep\fP [<.xvg>] (tablep.xvg) (Optional)
xvgr/xmgr file
.TP
.B \fB\-tableb\fP [<.xvg> [...]] (table.xvg) (Optional)
.B \fB\-tableb\fP [<.xvg> []] (table.xvg) (Optional)
xvgr/xmgr file
.TP
.B \fB\-rerun\fP [<.xtc/.trr/...>] (rerun.xtc) (Optional)
.B \fB\-rerun\fP [<.xtc/.trr/>] (rerun.xtc) (Optional)
Trajectory: xtc trr cpt gro g96 pdb tng
.TP
.B \fB\-ei\fP [<.edi>] (sam.edi) (Optional)
ED sampling input
.TP
.B \fB\-multidir\fP [<dir> [...]] (rundir) (Optional)
.B \fB\-multidir\fP [<dir> []] (rundir) (Optional)
Run directory
.TP
.B \fB\-awh\fP [<.xvg>] (awhinit.xvg) (Optional)
xvgr/xmgr file
.TP
.B \fB\-membed\fP [<.dat>] (membed.dat) (Optional)
Generic data file
.TP
......@@ -256,7 +266,7 @@ Index file
Options to specify output files:
.INDENT 0.0
.TP
.B \fB\-o\fP [<.trr/.cpt/...>] (traj.trr)
.B \fB\-o\fP [<.trr/.cpt/>] (traj.trr)
Full precision trajectory: trr cpt tng
.TP
.B \fB\-x\fP [<.xtc/.tng>] (traj_comp.xtc) (Optional)
......@@ -265,7 +275,7 @@ Compressed trajectory (tng format or portable xdr format)
.B \fB\-cpo\fP [<.cpt>] (state.cpt) (Optional)
Checkpoint file
.TP
.B \fB\-c\fP [<.gro/.g96/...>] (confout.gro)
.B \fB\-c\fP [<.gro/.g96/>] (confout.gro)
Structure file: gro g96 pdb brk ent esp
.TP
.B \fB\-e\fP [<.edr>] (ener.edr)
......@@ -345,7 +355,7 @@ Number of separate ranks to be used for PME, \-1 is guess
Total number of threads to start (0 is guess)
.TP
.B \fB\-ntmpi\fP <int> (0)
Number of thread\-MPI threads to start (0 is guess)
Number of thread\-MPI ranks to start (0 is guess)
.TP
.B \fB\-ntomp\fP <int> (0)
Number of OpenMP threads per MPI rank to start (0 is guess)
......@@ -363,7 +373,10 @@ The lowest logical core number to which mdrun should pin the first thread
Pinning distance in logical cores for threads, use 0 to minimize the number of threads per physical core
.TP
.B \fB\-gpu_id\fP <string>
List of GPU device id\-s to use, specifies the per\-node PP rank to GPU mapping
List of unique GPU device IDs available to use
.TP
.B \fB\-gputasks\fP <string>
List of GPU device IDs, mapping each PP task on each node to a device
.TP
.B \fB\-[no]ddcheck\fP (yes)
Check for all bonded interactions with DD
......@@ -384,13 +397,19 @@ Fraction in (0,1) by whose reciprocal the initial DD cell size will be increased
Global communication frequency
.TP
.B \fB\-nb\fP <enum> (auto)
Calculate non\-bonded interactions on: auto, cpu, gpu, gpu_cpu
Calculate non\-bonded interactions on: auto, cpu, gpu
.TP
.B \fB\-nstlist\fP <int> (0)
Set nstlist when using a Verlet buffer tolerance (0 is guess)
.TP
.B \fB\-[no]tunepme\fP (yes)
Optimize PME load between PP/PME ranks or GPU/CPU
Optimize PME load between PP/PME ranks or GPU/CPU (only with the Verlet cut\-off scheme)
.TP
.B \fB\-pme\fP <enum> (auto)
Perform PME calculations on: auto, cpu, gpu
.TP
.B \fB\-pmefft\fP <enum> (auto)
Perform PME FFT calculations on: auto, cpu, gpu
.TP
.B \fB\-[no]v\fP (no)
Be loud and noisy
......@@ -434,6 +453,6 @@ Seed for replica exchange, \-1 is generate a seed
.sp
More information about GROMACS is available at <\fI\%http://www.gromacs.org/\fP>.
.SH COPYRIGHT
2017, GROMACS development team
2018, GROMACS development team
.\" Generated by docutils manpage writer.
.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment