''

git-svn-id: svn://svn.icms.temple.edu/lammps-ro/trunk@14090 f3b2605a-c512-4ea7-a41b-209d697bcdaa

doc/Section_start.html

@@ -1651,16 +1651,23 @@ writes screen information to a file.N.  For both one-partition and
 multi-partition mode, if the specified file is “none”, then no screen
 output is performed. Option -pscreen will override the name of the
 partition screen files file.N.</p>
-<div class="highlight-python"><div class="highlight"><pre>-suffix style
+<div class="highlight-python"><div class="highlight"><pre>-suffix style args
 </pre></div>
 </div>
 <p>Use variants of various styles if they exist.  The specified style can
-be <em>cuda</em>, <em>gpu</em>, <em>intel</em>, <em>kk</em>, <em>omp</em>, or <em>opt</em>.  These refer to
+be <em>cuda</em>, <em>gpu</em>, <em>intel</em>, <em>kk</em>, <em>omp</em>, <em>opt</em>, or <em>hybrid</em>.  These refer
-optional packages that LAMMPS can be built with, as described above in
+to optional packages that LAMMPS can be built with, as described above in
 <a class="reference internal" href="#start-3"><span>Section 2.3</span></a>.  The &#8220;cuda&#8221; style corresponds to the USER-CUDA
 package, the &#8220;gpu&#8221; style to the GPU package, the &#8220;intel&#8221; style to the
 USER-INTEL package, the &#8220;kk&#8221; style to the KOKKOS package, the &#8220;opt&#8221;
-style to the OPT package, and the &#8220;omp&#8221; style to the USER-OMP package.</p>
+style to the OPT package, and the &#8220;omp&#8221; style to the USER-OMP package. The
+hybrid style is the only style that accepts arguments. It allows for two
+packages to be specified. The first package specified is the default and
+will be used if it is available. If no style is available for the first
+package, the style for the second package will be used if available. For
+example, &#8220;-suffix hybrid intel omp&#8221; will use styles from the USER-INTEL
+package if they are installed and available, but styles for the USER-OMP
+package otherwise.</p>
 <p>Along with the &#8220;-package&#8221; command-line switch, this is a convenient
 mechanism for invoking accelerator packages and their options without
 having to edit an input script.</p>
@@ -1681,14 +1688,14 @@ the &#8220;-package gpu&#8221; command-line switch or the <a class="reference in
 invokes the default USER-INTEL settings, as if the command &#8220;package
 intel 1&#8221; were used at the top of your input script.  These settings
 can be changed by using the &#8220;-package intel&#8221; command-line switch or
-the <a class="reference internal" href="package.html"><em>package intel</em></a> command in your script.  If the
+the <a class="reference internal" href="package.html"><em>package intel</em></a> command in your script. If the
-USER-OMP package is also installed, the intel suffix will make the omp
+USER-OMP package is also installed, the hybrid style with &#8220;intel omp&#8221;
-suffix a second choice, if a requested style is not available in the
+arguments can be used to make the omp suffix a second choice, if a
-USER-INTEL package.  It will also invoke the default USER-OMP
+requested style is not available in the USER-INTEL package.  It will
-settings, as if the command &#8220;package omp 0&#8221; were used at the top of
+also invoke the default USER-OMP settings, as if the command &#8220;package
-your input script.  These settings can be changed by using the
+omp 0&#8221; were used at the top of your input script.  These settings can
-&#8220;-package omp&#8221; command-line switch or the <a class="reference internal" href="package.html"><em>package omp</em></a>
+be changed by using the &#8220;-package omp&#8221; command-line switch or the
-command in your script.</p>
+<a class="reference internal" href="package.html"><em>package omp</em></a> command in your script.</p>
 <p>For the KOKKOS package, using this command-line switch also invokes
 the default KOKKOS settings, as if the command &#8220;package kokkos&#8221; were
 used at the top of your input script.  These settings can be changed

doc/Section_start.txt

@@ -1645,15 +1645,22 @@ multi-partition mode, if the specified file is "none", then no screen
 output is performed. Option -pscreen will override the name of the
 partition screen files file.N.
 
--suffix style :pre
+-suffix style args :pre
 
 Use variants of various styles if they exist.  The specified style can
-be {cuda}, {gpu}, {intel}, {kk}, {omp}, or {opt}.  These refer to
+be {cuda}, {gpu}, {intel}, {kk}, {omp}, {opt}, or {hybrid}.  These refer 
-optional packages that LAMMPS can be built with, as described above in
+to optional packages that LAMMPS can be built with, as described above in
 "Section 2.3"_#start_3.  The "cuda" style corresponds to the USER-CUDA
 package, the "gpu" style to the GPU package, the "intel" style to the
 USER-INTEL package, the "kk" style to the KOKKOS package, the "opt"
-style to the OPT package, and the "omp" style to the USER-OMP package.
+style to the OPT package, and the "omp" style to the USER-OMP package. The
+hybrid style is the only style that accepts arguments. It allows for two 
+packages to be specified. The first package specified is the default and 
+will be used if it is available. If no style is available for the first 
+package, the style for the second package will be used if available. For 
+example, "-suffix hybrid intel omp" will use styles from the USER-INTEL 
+package if they are installed and available, but styles for the USER-OMP 
+package otherwise.
 
 Along with the "-package" command-line switch, this is a convenient
 mechanism for invoking accelerator packages and their options without
@@ -1680,14 +1687,14 @@ For the USER-INTEL package, using this command-line switch also
 invokes the default USER-INTEL settings, as if the command "package
 intel 1" were used at the top of your input script.  These settings
 can be changed by using the "-package intel" command-line switch or
-the "package intel"_package.html command in your script.  If the
+the "package intel"_package.html command in your script. If the
-USER-OMP package is also installed, the intel suffix will make the omp
+USER-OMP package is also installed, the hybrid style with "intel omp" 
-suffix a second choice, if a requested style is not available in the
+arguments can be used to make the omp suffix a second choice, if a 
-USER-INTEL package.  It will also invoke the default USER-OMP
+requested style is not available in the USER-INTEL package.  It will 
-settings, as if the command "package omp 0" were used at the top of
+also invoke the default USER-OMP settings, as if the command "package 
-your input script.  These settings can be changed by using the
+omp 0" were used at the top of your input script.  These settings can 
-"-package omp" command-line switch or the "package omp"_package.html
+be changed by using the "-package omp" command-line switch or the 
-command in your script.
+"package omp"_package.html command in your script.
 
 For the KOKKOS package, using this command-line switch also invokes
 the default KOKKOS settings, as if the command "package kokkos" were

doc/accelerate_intel.html

@@ -145,15 +145,15 @@ cores.  Since less MPI tasks than CPU cores will typically be invoked
 when running with coprocessors, this enables the extra CPU cores to be
 used for useful computation.</p>
 <p>If LAMMPS is built with both the USER-INTEL and USER-OMP packages
-intsalled, this mode of operation is made easier to use, because the
+installed, this mode of operation is made easier to use, with the
-&#8220;-suffix intel&#8221; <a class="reference internal" href="Section_start.html#start-7"><span>command-line switch</span></a> or
+&#8220;-suffix hybrid intel omp&#8221; <a class="reference internal" href="Section_start.html#start-7"><span>command-line switch</span></a>
-the <a class="reference internal" href="suffix.html"><em>suffix intel</em></a> command will both set a second-choice
+or the <a class="reference internal" href="suffix.html"><em>suffix hybrid intel omp</em></a> command will both set a
-suffix to &#8220;omp&#8221; so that styles from the USER-OMP package will be used
+second-choice suffix to &#8220;omp&#8221; so that styles from the USER-OMP package will be
-if available, after first testing if a style from the USER-INTEL
+used if available, after first testing if a style from the USER-INTEL
 package is available.</p>
-<p>When using the USER-INTEL package, you must choose at build time
+<p>When using the USER-INTEL package, you must choose at build time whether the
-whether you are building for CPU-only acceleration or for using the
+binary will support offload to Xeon Phi coprocessors. Binaries supporting
-Xeon Phi in offload mode.</p>
+offload can still be run in CPU-only (host-only) mode.</p>
 <p>Here is a quick overview of how to use the USER-INTEL package
 for CPU-only acceleration:</p>
 <ul class="simple">
@@ -259,15 +259,14 @@ the physical number of cores (on a node), otherwise performance will
 suffer.</p>
 <p>If LAMMPS was built with coprocessor support for the USER-INTEL
 package, you also need to specify the number of coprocessor/node and
-the number of coprocessor threads per MPI task to use.  Note that
+optionally the number of coprocessor threads per MPI task to use.  Note that
 coprocessor threads (which run on the coprocessor) are totally
 independent from OpenMP threads (which run on the CPU).  The default
 values for the settings that affect coprocessor threads are typically
 fine, as discussed below.</p>
 <p>Use the &#8220;-sf intel&#8221; <a class="reference internal" href="Section_start.html#start-7"><span>command-line switch</span></a>,
-which will automatically append &#8220;intel&#8221; to styles that support it.  If
+which will automatically append &#8220;intel&#8221; to styles that support it.
-a style does not support it, an &#8220;omp&#8221; suffix is tried next.  OpenMP
+OpenMP threads per MPI task can be set via the &#8220;-pk intel Nphi omp Nt&#8221; or
-threads per MPI task can be set via the &#8220;-pk intel Nphi omp Nt&#8221; or
 &#8220;-pk omp Nt&#8221; <a class="reference internal" href="Section_start.html#start-7"><span>command-line switches</span></a>, which
 set Nt = # of OpenMP threads per MPI task to use.  The &#8220;-pk omp&#8221; form
 is only allowed if LAMMPS was also built with the USER-OMP package.</p>
@@ -278,38 +277,36 @@ MPI tasks, unless the <em>tptask</em> option of the &#8220;-pk intel&#8221; <a c
 threads per MPI task.  See the <a class="reference internal" href="package.html"><em>package intel</em></a> command
 for details.</p>
 <div class="highlight-python"><div class="highlight"><pre>CPU-only without USER-OMP (but using Intel vectorization on CPU):
-lmp_machine -sf intel -in in.script                 # 1 MPI task
+mpirun -np 32 lmp_machine -sf intel -in in.script         # 32 MPI tasks on as many nodes as needed (e.g. 2 16-core nodes)
-mpirun -np 32 lmp_machine -sf intel -in in.script   # 32 MPI tasks on as many nodes as needed (e.g. 2 16-core nodes)
+lmp_machine -sf intel -pk intel 0 omp 16 -in in.script    # 1 MPI task and 16 threads
 </pre></div>
 </div>
 <div class="highlight-python"><div class="highlight"><pre>CPU-only with USER-OMP (and Intel vectorization on CPU):
-lmp_machine -sf intel -pk intel 16 0 -in in.script             # 1 MPI task on a 16-core node
+lmp_machine -sf hybrid intel omp -pk intel 0 omp 16 -in in.script         # 1 MPI task on a 16-core node with 16 threads
-mpirun -np 4 lmp_machine -sf intel -pk omp 4 -in in.script     # 4 MPI tasks each with 4 threads on a single 16-core node
+mpirun -np 4 lmp_machine -sf hybrid intel omp -pk omp 4 -in in.script     # 4 MPI tasks each with 4 threads on a single 16-core node
-mpirun -np 32 lmp_machine -sf intel -pk omp 4 -in in.script    # ditto on 8 16-core nodes
 </pre></div>
 </div>
 <div class="highlight-python"><div class="highlight"><pre>CPUs + Xeon Phi(TM) coprocessors with or without USER-OMP:
-lmp_machine -sf intel -pk intel 1 omp 16 -in in.script                       # 1 MPI task, 16 OpenMP threads on CPU, 1 coprocessor, all 240 coprocessor threads
+mpirun -np 32 -ppn 16 lmp_machine -sf intel -pk intel 1 -in in.script                 # 2 nodes with 16 MPI tasks on each, 240 total threads on coprocessor
-lmp_machine -sf intel -pk intel 1 omp 16 tptask 32 -in in.script             # 1 MPI task, 16 OpenMP threads on CPU, 1 coprocessor, only 32 coprocessor threads
+mpirun -np 16 -ppn 8 lmp_machine -sf intel -pk intel 1 omp 2 -in in.script            # 2 nodes, 8 MPI tasks on each node, 2 threads for each task, 240 total threads on coprocessor
-mpirun -np 4 lmp_machine -sf intel -pk intel 1 omp 4 -in in.script           # 4 MPI tasks, 4 OpenMP threads/task, 1 coprocessor, 60 coprocessor threads/task
+mpirun -np 16 -ppn 8 lmp_machine -sf hybrid intel omp -pk intel 1 omp 2 -in in.script # 2 nodes, 8 MPI tasks on each node, 2 threads for each task, 240 total threads on coprocessor, USER-OMP package for some styles
-mpirun -np 32 -ppn 4 lmp_machine -sf intel -pk intel 1 omp 4 -in in.script   # ditto on 8 16-core nodes
-mpirun -np 8 lmp_machine -sf intel -pk intel 4 omp 2 -in in.script           # 8 MPI tasks, 2 OpenMP threads/task, 4 coprocessors, 120 coprocessor threads/task
 </pre></div>
 </div>
-<p>Note that if the &#8220;-sf intel&#8221; switch is used, it also invokes two
+<p>Note that if the &#8220;-sf intel&#8221; switch is used, it also invokes a
-default commands: <a class="reference internal" href="package.html"><em>package intel 1</em></a>, followed by <a class="reference internal" href="package.html"><em>package omp 0</em></a>.  These both set the number of OpenMP threads per
+default command: <a class="reference internal" href="package.html"><em>package intel 1</em></a>. If the &#8220;-sf hybrid intel omp&#8221;
-MPI task via the OMP_NUM_THREADS environment variable.  The first
+switch is used, the default USER-OMP command <a class="reference internal" href="package.html"><em>package omp 0</em></a> is
-command sets the number of Xeon Phi(TM) coprocessors/node to 1 (and
+also invoked. Both set the number of OpenMP threads per MPI task via the
-the precision mode to &#8220;mixed&#8221;, as one of its option defaults).  The
+OMP_NUM_THREADS environment variable.  The first command sets the number of
-latter command is not invoked if LAMMPS was not built with the
+Xeon Phi(TM) coprocessors/node to 1 (and the precision mode to &#8220;mixed&#8221;, as one
-USER-OMP package.  The Nphi = 1 value for the first command is ignored
+of its option defaults).  The latter command is not invoked if LAMMPS was not
-if LAMMPS was not built with coprocessor support.</p>
+built with the USER-OMP package.  The Nphi = 1 value for the first command is
+ignored if LAMMPS was not built with coprocessor support.</p>
 <p>Using the &#8220;-pk intel&#8221; or &#8220;-pk omp&#8221; switches explicitly allows for
 direct setting of the number of OpenMP threads per MPI task, and
 additional options for either of the USER-INTEL or USER-OMP packages.
 In particular, the &#8220;-pk intel&#8221; switch sets the number of
 coprocessors/node and can limit the number of coprocessor threads per
-MPI task.  The syntax for these two switches is the same as the
+MPI task. The syntax for these two switches is the same as the
 <a class="reference internal" href="package.html"><em>package omp</em></a> and <a class="reference internal" href="package.html"><em>package intel</em></a> commands.
 See the <a class="reference internal" href="package.html"><em>package</em></a> command doc page for details, including
 the default values used for all its options if these switches are not
@@ -331,7 +328,7 @@ coprocessor options.  Its doc page explains how to set the number of
 OpenMP threads via an environment variable if desired.</p>
 <p>If LAMMPS was also built with the USER-OMP package, you must also use
 the <a class="reference internal" href="package.html"><em>package omp</em></a> command to enable that package, unless
-the &#8220;-sf intel&#8221; or &#8220;-pk omp&#8221; <a class="reference internal" href="Section_start.html#start-7"><span>command-line switches</span></a> were used.  It specifies how many
+the &#8220;-sf hybrid intel omp&#8221; or &#8220;-pk omp&#8221; <a class="reference internal" href="Section_start.html#start-7"><span>command-line switches</span></a> were used.  It specifies how many
 OpenMP threads per MPI task to use, as well as other options.  Its doc
 page explains how to set the number of OpenMP threads via an
 environment variable if desired.</p>

doc/accelerate_intel.txt

@@ -31,16 +31,16 @@ when running with coprocessors, this enables the extra CPU cores to be
 used for useful computation.
 
 If LAMMPS is built with both the USER-INTEL and USER-OMP packages
-intsalled, this mode of operation is made easier to use, because the
+installed, this mode of operation is made easier to use, with the
-"-suffix intel" "command-line switch"_Section_start.html#start_7 or
+"-suffix hybrid intel omp" "command-line switch"_Section_start.html#start_7 
-the "suffix intel"_suffix.html command will both set a second-choice
+or the "suffix hybrid intel omp"_suffix.html command will both set a 
-suffix to "omp" so that styles from the USER-OMP package will be used
+second-choice suffix to "omp" so that styles from the USER-OMP package will be 
-if available, after first testing if a style from the USER-INTEL
+used if available, after first testing if a style from the USER-INTEL
 package is available.
 
-When using the USER-INTEL package, you must choose at build time
+When using the USER-INTEL package, you must choose at build time whether the 
-whether you are building for CPU-only acceleration or for using the
+binary will support offload to Xeon Phi coprocessors. Binaries supporting
-Xeon Phi in offload mode.
+offload can still be run in CPU-only (host-only) mode.
 
 Here is a quick overview of how to use the USER-INTEL package
 for CPU-only acceleration:
@@ -167,16 +167,15 @@ suffer.
 
 If LAMMPS was built with coprocessor support for the USER-INTEL
 package, you also need to specify the number of coprocessor/node and
-the number of coprocessor threads per MPI task to use.  Note that
+optionally the number of coprocessor threads per MPI task to use.  Note that
 coprocessor threads (which run on the coprocessor) are totally
 independent from OpenMP threads (which run on the CPU).  The default
 values for the settings that affect coprocessor threads are typically
 fine, as discussed below.
 
 Use the "-sf intel" "command-line switch"_Section_start.html#start_7,
-which will automatically append "intel" to styles that support it.  If
+which will automatically append "intel" to styles that support it. 
-a style does not support it, an "omp" suffix is tried next.  OpenMP
+OpenMP threads per MPI task can be set via the "-pk intel Nphi omp Nt" or
-threads per MPI task can be set via the "-pk intel Nphi omp Nt" or
 "-pk omp Nt" "command-line switches"_Section_start.html#start_7, which
 set Nt = # of OpenMP threads per MPI task to use.  The "-pk omp" form
 is only allowed if LAMMPS was also built with the USER-OMP package.
@@ -191,37 +190,34 @@ threads per MPI task.  See the "package intel"_package.html command
 for details.
 
 CPU-only without USER-OMP (but using Intel vectorization on CPU):
-lmp_machine -sf intel -in in.script                 # 1 MPI task
+mpirun -np 32 lmp_machine -sf intel -in in.script         # 32 MPI tasks on as many nodes as needed (e.g. 2 16-core nodes)
-mpirun -np 32 lmp_machine -sf intel -in in.script   # 32 MPI tasks on as many nodes as needed (e.g. 2 16-core nodes) :pre
+lmp_machine -sf intel -pk intel 0 omp 16 -in in.script    # 1 MPI task and 16 threads :pre
 
 CPU-only with USER-OMP (and Intel vectorization on CPU):
-lmp_machine -sf intel -pk intel 16 0 -in in.script             # 1 MPI task on a 16-core node
+lmp_machine -sf hybrid intel omp -pk intel 0 omp 16 -in in.script         # 1 MPI task on a 16-core node with 16 threads
-mpirun -np 4 lmp_machine -sf intel -pk omp 4 -in in.script     # 4 MPI tasks each with 4 threads on a single 16-core node
+mpirun -np 4 lmp_machine -sf hybrid intel omp -pk omp 4 -in in.script     # 4 MPI tasks each with 4 threads on a single 16-core node :pre
-mpirun -np 32 lmp_machine -sf intel -pk omp 4 -in in.script    # ditto on 8 16-core nodes :pre
 
 CPUs + Xeon Phi(TM) coprocessors with or without USER-OMP:
-lmp_machine -sf intel -pk intel 1 omp 16 -in in.script                       # 1 MPI task, 16 OpenMP threads on CPU, 1 coprocessor, all 240 coprocessor threads
+mpirun -np 32 -ppn 16 lmp_machine -sf intel -pk intel 1 -in in.script                 # 2 nodes with 16 MPI tasks on each, 240 total threads on coprocessor
-lmp_machine -sf intel -pk intel 1 omp 16 tptask 32 -in in.script             # 1 MPI task, 16 OpenMP threads on CPU, 1 coprocessor, only 32 coprocessor threads
+mpirun -np 16 -ppn 8 lmp_machine -sf intel -pk intel 1 omp 2 -in in.script            # 2 nodes, 8 MPI tasks on each node, 2 threads for each task, 240 total threads on coprocessor
-mpirun -np 4 lmp_machine -sf intel -pk intel 1 omp 4 -in in.script           # 4 MPI tasks, 4 OpenMP threads/task, 1 coprocessor, 60 coprocessor threads/task
+mpirun -np 16 -ppn 8 lmp_machine -sf hybrid intel omp -pk intel 1 omp 2 -in in.script # 2 nodes, 8 MPI tasks on each node, 2 threads for each task, 240 total threads on coprocessor, USER-OMP package for some styles :pre
-mpirun -np 32 -ppn 4 lmp_machine -sf intel -pk intel 1 omp 4 -in in.script   # ditto on 8 16-core nodes
-mpirun -np 8 lmp_machine -sf intel -pk intel 4 omp 2 -in in.script           # 8 MPI tasks, 2 OpenMP threads/task, 4 coprocessors, 120 coprocessor threads/task :pre 
 
-Note that if the "-sf intel" switch is used, it also invokes two
+Note that if the "-sf intel" switch is used, it also invokes a
-default commands: "package intel 1"_package.html, followed by "package
+default command: "package intel 1"_package.html. If the "-sf hybrid intel omp"
-omp 0"_package.html.  These both set the number of OpenMP threads per
+switch is used, the default USER-OMP command "package omp 0"_package.html is
-MPI task via the OMP_NUM_THREADS environment variable.  The first
+also invoked. Both set the number of OpenMP threads per MPI task via the 
-command sets the number of Xeon Phi(TM) coprocessors/node to 1 (and
+OMP_NUM_THREADS environment variable.  The first command sets the number of 
-the precision mode to "mixed", as one of its option defaults).  The
+Xeon Phi(TM) coprocessors/node to 1 (and the precision mode to "mixed", as one 
-latter command is not invoked if LAMMPS was not built with the
+of its option defaults).  The latter command is not invoked if LAMMPS was not 
-USER-OMP package.  The Nphi = 1 value for the first command is ignored
+built with the USER-OMP package.  The Nphi = 1 value for the first command is 
-if LAMMPS was not built with coprocessor support.
+ignored if LAMMPS was not built with coprocessor support.
 
 Using the "-pk intel" or "-pk omp" switches explicitly allows for
 direct setting of the number of OpenMP threads per MPI task, and
 additional options for either of the USER-INTEL or USER-OMP packages.
 In particular, the "-pk intel" switch sets the number of
 coprocessors/node and can limit the number of coprocessor threads per
-MPI task.  The syntax for these two switches is the same as the
+MPI task. The syntax for these two switches is the same as the
 "package omp"_package.html and "package intel"_package.html commands.
 See the "package"_package.html command doc page for details, including
 the default values used for all its options if these switches are not
@@ -248,7 +244,7 @@ OpenMP threads via an environment variable if desired.
 
 If LAMMPS was also built with the USER-OMP package, you must also use
 the "package omp"_package.html command to enable that package, unless
-the "-sf intel" or "-pk omp" "command-line
+the "-sf hybrid intel omp" or "-pk omp" "command-line
 switches"_Section_start.html#start_7 were used.  It specifies how many
 OpenMP threads per MPI task to use, as well as other options.  Its doc
 page explains how to set the number of OpenMP threads via an

doc/package.html

@@ -467,7 +467,7 @@ USER-OMP packages, be aware that both packages allow setting of the
 single global <em>Nthreads</em> value used by OpenMP.  Thus if both package
 commands are invoked, you should insure the two values are consistent.
 If they are not, the last one invoked will take precedence, for both
-packages.  Also note that if the &#8220;-sf intel&#8221; <a class="reference internal" href="Section_start.html#start-7"><span>command-line switch</span></a> is used, it invokes a &#8220;package
+packages.  Also note that if the &#8220;-sf hybrid intel omp&#8221; <a class="reference internal" href="Section_start.html#start-7"><span>command-line switch</span></a> is used, it invokes a &#8220;package
 intel&#8221; command, followed by a &#8220;package omp&#8221; command, both with a
 setting of <em>Nthreads</em> = 0.</p>
 </div>

doc/package.txt

@@ -376,7 +376,7 @@ USER-OMP packages, be aware that both packages allow setting of the
 single global {Nthreads} value used by OpenMP.  Thus if both package
 commands are invoked, you should insure the two values are consistent.
 If they are not, the last one invoked will take precedence, for both
-packages.  Also note that if the "-sf intel" "command-line
+packages.  Also note that if the "-sf hybrid intel omp" "command-line
 switch"_"_Section_start.html#start_7 is used, it invokes a "package
 intel" command, followed by a "package omp" command, both with a
 setting of {Nthreads} = 0.

doc/suffix.html

@@ -128,11 +128,12 @@
 <span id="index-0"></span><h1>suffix command<a class="headerlink" href="#suffix-command" title="Permalink to this headline">¶</a></h1>
 <div class="section" id="syntax">
 <h2>Syntax<a class="headerlink" href="#syntax" title="Permalink to this headline">¶</a></h2>
-<div class="highlight-python"><div class="highlight"><pre>suffix style
+<div class="highlight-python"><div class="highlight"><pre>suffix style args
 </pre></div>
 </div>
 <ul class="simple">
-<li>style = <em>off</em> or <em>on</em> or <em>cuda</em> or <em>gpu</em> or <em>intel</em> or <em>kk</em> or <em>omp</em> or <em>opt</em></li>
+<li>style = <em>off</em> or <em>on</em> or <em>cuda</em> or <em>gpu</em> or <em>intel</em> or <em>kk</em> or <em>omp</em> or <em>opt</em> or <em>hybrid</em></li>
+<li>args = for hybrid style, default suffix to be used and alternative suffix</li>
 </ul>
 </div>
 <div class="section" id="examples">
@@ -141,6 +142,7 @@
 suffix on
 suffix gpu
 suffix intel
+suffix hybrid intel omp
 suffix kk
 </pre></div>
 </div>
@@ -150,13 +152,13 @@ suffix kk
 <p>This command allows you to use variants of various styles if they
 exist.  In that respect it operates the same as the <a class="reference internal" href="Section_start.html#start-7"><span>-suffix command-line switch</span></a>.  It also has options
 to turn off or back on any suffix setting made via the command line.</p>
-<p>The specified style can be <em>cuda</em>, <em>gpu</em>, <em>intel</em>, <em>kk</em>, <em>omp</em>, or
+<p>The specified style can be <em>cuda</em>, <em>gpu</em>, <em>intel</em>, <em>kk</em>, <em>omp</em>, <em>opt</em>
-<em>opt</em>.  These refer to optional packages that LAMMPS can be built
+or <em>hybrid</em>. These refer to optional packages that LAMMPS can be built
 with, as described in <a class="reference internal" href="Section_start.html#start-3"><span>this section of the manual</span></a>.  The &#8220;cuda&#8221; style corresponds to
 the USER-CUDA package, the &#8220;gpu&#8221; style to the GPU package, the &#8220;intel&#8221;
 style to the USER-INTEL package, the &#8220;kk&#8221; style to the KOKKOS package,
 the &#8220;omp&#8221; style to the USER-OMP package, and the &#8220;opt&#8221; style to the
-OPT package,</p>
+OPT package.</p>
 <p>These are the variants these packages provide:</p>
 <ul class="simple">
 <li>USER-CUDA = a collection of atom, pair, fix, compute, and intergrate
@@ -174,6 +176,7 @@ kspace, compute, and fix styles with support for OpenMP
 multi-threading</li>
 <li>OPT = a handful of pair styles, cache-optimized for faster CPU
 performance</li>
+<li>HYBRID = a combination of two packages can be specified (see below)</li>
 </ul>
 <p>As an example, all of the packages provide a <a class="reference internal" href="pair_lj.html"><em>pair_style lj/cut</em></a> variant, with style names lj/cut/opt, lj/cut/omp,
 lj/cut/gpu, lj/cut/intel, lj/cut/cuda, or lj/cut/kk.  A variant styles
@@ -188,10 +191,12 @@ input script command creates a new <a class="reference internal" href="atom_styl
 <a class="reference internal" href="fix.html"><em>fix</em></a>, <a class="reference internal" href="compute.html"><em>compute</em></a>, or <a class="reference internal" href="run_style.html"><em>run</em></a> style.
 If the variant version does not exist, the standard version is
 created.</p>
-<p>When using the intel suffix, LAMMPS will first attempt to use a style
+<p>For &#8220;hybrid&#8221;, two packages are specified. The first is used whenever
-with the intel suffix.  If the USER-OMP package is installed, the the
+available. If a style with the first suffix is not available, the style
-omp suffix will be tried as a second choice, if a requested style is
+with the suffix for the second package will be used if available. For
-not available in the USER-INTEL package.</p>
+example, &#8220;hybrid intel omp&#8221; will use styles from the USER-INTEL package
+as a first choice and styles from the USER-OMP package as a second choice
+if no USER-INTEL variant is available.</p>
 <p>If the specified style is <em>off</em>, then any previously specified suffix
 is temporarily disabled, whether it was specified by a command-line
 switch or a previous suffix command.  If the specified style is <em>on</em>,

doc/suffix.txt

@@ -10,9 +10,10 @@ suffix command :h3
 
 [Syntax:]
 
-suffix style :pre
+suffix style args :pre
 
-style = {off} or {on} or {cuda} or {gpu} or {intel} or {kk} or {omp} or {opt} :ul
+style = {off} or {on} or {cuda} or {gpu} or {intel} or {kk} or {omp} or {opt} or {hybrid}
+args = for hybrid style, default suffix to be used and alternative suffix :ul
 
 [Examples:]
 
@@ -20,6 +21,7 @@ suffix off
 suffix on
 suffix gpu
 suffix intel
+suffix hybrid intel omp
 suffix kk :pre
 
 [Description:]
@@ -29,14 +31,14 @@ exist.  In that respect it operates the same as the "-suffix
 command-line switch"_Section_start.html#start_7.  It also has options
 to turn off or back on any suffix setting made via the command line.
 
-The specified style can be {cuda}, {gpu}, {intel}, {kk}, {omp}, or
+The specified style can be {cuda}, {gpu}, {intel}, {kk}, {omp}, {opt} 
-{opt}.  These refer to optional packages that LAMMPS can be built
+or {hybrid}. These refer to optional packages that LAMMPS can be built
 with, as described in "this section of the
 manual"_Section_start.html#start_3.  The "cuda" style corresponds to
 the USER-CUDA package, the "gpu" style to the GPU package, the "intel"
 style to the USER-INTEL package, the "kk" style to the KOKKOS package,
 the "omp" style to the USER-OMP package, and the "opt" style to the
-OPT package,
+OPT package.  
 
 These are the variants these packages provide:
 
@@ -59,7 +61,9 @@ kspace, compute, and fix styles with support for OpenMP
 multi-threading :l
 
 OPT = a handful of pair styles, cache-optimized for faster CPU
-performance :ule,l
+performance :l
+
+HYBRID = a combination of two packages can be specified (see below) :ule,l
 
 As an example, all of the packages provide a "pair_style
 lj/cut"_pair_lj.html variant, with style names lj/cut/opt, lj/cut/omp,
@@ -76,10 +80,12 @@ input script command creates a new "atom"_atom_style.html,
 If the variant version does not exist, the standard version is
 created.
 
-When using the intel suffix, LAMMPS will first attempt to use a style
+For "hybrid", two packages are specified. The first is used whenever 
-with the intel suffix.  If the USER-OMP package is installed, the the
+available. If a style with the first suffix is not available, the style 
-omp suffix will be tried as a second choice, if a requested style is
+with the suffix for the second package will be used if available. For 
-not available in the USER-INTEL package.
+example, "hybrid intel omp" will use styles from the USER-INTEL package 
+as a first choice and styles from the USER-OMP package as a second choice 
+if no USER-INTEL variant is available.
 
 If the specified style is {off}, then any previously specified suffix
 is temporarily disabled, whether it was specified by a command-line