Assuming you have built LAMMPS with the appropriate package, these styles can be invoked by specifying them explicitly in your input -script. Or you can use the -accelerator command-line +script. Or you can use the -suffix command-line switch to invoke the accelerated versions -automatically. See the acclerator command for info -on how to turn off/on the suffix associated with this switch within -your input script. +automatically. See the suffix command for info on how +to turn off/on the suffix associated with this switch within your +input script.
Styles with an "opt" suffix are part of the OPT package and typically speed-up the pairwise portion of your simulation by 5-25%. diff --git a/doc/Section_accelerate.txt b/doc/Section_accelerate.txt index 17b079691b..4950476be5 100644 --- a/doc/Section_accelerate.txt +++ b/doc/Section_accelerate.txt @@ -29,11 +29,11 @@ style exist in LAMMPS: Assuming you have built LAMMPS with the appropriate package, these styles can be invoked by specifying them explicitly in your input -script. Or you can use the "-accelerator command-line +script. Or you can use the "-suffix command-line switch"_Section_start.html#2_6 to invoke the accelerated versions -automatically. See the "acclerator"_accelerator.html command for info -on how to turn off/on the suffix associated with this switch within -your input script. +automatically. See the "suffix"_suffix.html command for info on how +to turn off/on the suffix associated with this switch within your +input script. Styles with an "opt" suffix are part of the OPT package and typically speed-up the pairwise portion of your simulation by 5-25%. diff --git a/doc/Section_commands.html b/doc/Section_commands.html index a711e75bcc..e67081e6ab 100644 --- a/doc/Section_commands.html +++ b/doc/Section_commands.html @@ -311,20 +311,20 @@ default LAMMPS build. These dependencies are listed as Restrictions in the command's documentation.
For example, lmp_ibm might be launched as follows:
-mpirun -np 16 lmp_ibm -v f tmp.out -l my.log -s none < in.alloy +mpirun -np 16 lmp_ibm -v f tmp.out -l my.log -sc none < in.alloy mpirun -np 16 lmp_ibm -var f tmp.out -log my.log -screen none < in.alloyHere are the details on the options:
--accelerator style +-cuda on/off-Use accelerated variants of various styles if they exist. The -specified style can be none or opt or gpu or cuda. None -means do not attempt to use an accelerated style, unless it is -explicitly specified in the input script. Opt, gpu, and cuda -refer to optional packages that LAMMPS can be built with, as described -above in Section 2.3. The "opt" style corrsponds to the OPT -package, the "gpu" style to the GPU package, and the "cuda" style to -the USER-CUDA package. -
-As an example, all of the packages provide a pair_style -lj/cut variant, with style names lj/cut/opt or -lj/cut/gpu or lj/cut/cuda. Any accelerated styles can be specified -explicitly in your input script, e.g. pair_style lj/cut/gpu. If the --accelerator switch is used, you do not need to modify your input -script. The accelerator suffix (opt,gpu,cuda) is automatically -appended when the style is created for atom, pair, fix, compute, and -integrate styles. If an accelerated version does not exist, the -standard version is created. -
-The default value of this switch is "none", unless LAMMPS was built -with the USER-CUDA package, in which case the default value is "cuda". -See the acclerator command for info on how to turn -off/on the suffix associated with this switch within your input -script. +
Explicitly enable or disable CUDA support, as provided by the +USER-CUDA package. If LAMMPS is built with this package, as described +above in Section 2.3, then by default LAMMPS will run in CUDA +mode. If this switch is set to "off", then it will not, even if it +was built with the USER-CUDA package, which means you can run standard +LAMMPS or with the GPU package for testing or benchmarking purposes. +The only reason to set the switch to "on", is to check if LAMMPS was +built with the USER-CUDA package, since an error will be generated if +it was not.
-echo style@@ -915,6 +901,27 @@ 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. +-suffix style ++Use variants of various styles if they exist. The specified style can +be opt or gpu or cuda. These refer to optional packages that +LAMMPS can be built with, as described above in Section 2.3. +The "opt" style corrsponds to the OPT package, the "gpu" style to the +GPU package, and the "cuda" style to the USER-CUDA package. +
+As an example, all of the packages provide a pair_style +lj/cut variant, with style names lj/cut/opt or +lj/cut/gpu or lj/cut/cuda. Any variant styles can be specified +explicitly in your input script, e.g. pair_style lj/cut/gpu. If the +-suffix switch is used, you do not need to modify your input script. +The specified suffix (opt,gpu,cuda) is automatically appended whenever +an input script command creates a new atom, pair, fix, compute, or +integrate style. If the variant version does not exist, the standard +version is created. +
+The suffix command can also set the suffix and it can +also turn off/on the suffix associated with this command-line switch. +
-var name value1 value2 ...Specify a variable that will be defined for substitution purposes when diff --git a/doc/Section_start.txt b/doc/Section_start.txt index a22a770060..1524541d9f 100644 --- a/doc/Section_start.txt +++ b/doc/Section_start.txt @@ -792,47 +792,33 @@ At run time, LAMMPS recognizes several optional command-line switches which may be used in any order. Either the full word or a one-letter abbreviation can be used: --a or -accelerator +-c or -cuda -e or -echo -i or -in -l or -log -p or -partition --s or -screen +-sc or -screen +-sf or -suffix -v or -var :ul For example, lmp_ibm might be launched as follows: -mpirun -np 16 lmp_ibm -v f tmp.out -l my.log -s none < in.alloy +mpirun -np 16 lmp_ibm -v f tmp.out -l my.log -sc none < in.alloy mpirun -np 16 lmp_ibm -var f tmp.out -log my.log -screen none < in.alloy :pre Here are the details on the options: --accelerator style :pre +-cuda on/off :pre -Use accelerated variants of various styles if they exist. The -specified style can be {none} or {opt} or {gpu} or {cuda}. {None} -means do not attempt to use an accelerated style, unless it is -explicitly specified in the input script. {Opt}, {gpu}, and {cuda} -refer to optional packages that LAMMPS can be built with, as described -above in "Section 2.3"_#2_3. The "opt" style corrsponds to the OPT -package, the "gpu" style to the GPU package, and the "cuda" style to -the USER-CUDA package. - -As an example, all of the packages provide a "pair_style -lj/cut"_pair_lj.html variant, with style names lj/cut/opt or -lj/cut/gpu or lj/cut/cuda. Any accelerated styles can be specified -explicitly in your input script, e.g. pair_style lj/cut/gpu. If the --accelerator switch is used, you do not need to modify your input -script. The accelerator suffix (opt,gpu,cuda) is automatically -appended when the style is created for atom, pair, fix, compute, and -integrate styles. If an accelerated version does not exist, the -standard version is created. - -The default value of this switch is "none", unless LAMMPS was built -with the USER-CUDA package, in which case the default value is "cuda". -See the "acclerator"_accelerator.html command for info on how to turn -off/on the suffix associated with this switch within your input -script. +Explicitly enable or disable CUDA support, as provided by the +USER-CUDA package. If LAMMPS is built with this package, as described +above in "Section 2.3"_#2_3, then by default LAMMPS will run in CUDA +mode. If this switch is set to "off", then it will not, even if it +was built with the USER-CUDA package, which means you can run standard +LAMMPS or with the GPU package for testing or benchmarking purposes. +The only reason to set the switch to "on", is to check if LAMMPS was +built with the USER-CUDA package, since an error will be generated if +it was not. -echo style :pre @@ -905,6 +891,27 @@ 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. +-suffix style :pre + +Use variants of various styles if they exist. The specified style can +be {opt} or {gpu} or {cuda}. These refer to optional packages that +LAMMPS can be built with, as described above in "Section 2.3"_#2_3. +The "opt" style corrsponds to the OPT package, the "gpu" style to the +GPU package, and the "cuda" style to the USER-CUDA package. + +As an example, all of the packages provide a "pair_style +lj/cut"_pair_lj.html variant, with style names lj/cut/opt or +lj/cut/gpu or lj/cut/cuda. Any variant styles can be specified +explicitly in your input script, e.g. pair_style lj/cut/gpu. If the +-suffix switch is used, you do not need to modify your input script. +The specified suffix (opt,gpu,cuda) is automatically appended whenever +an input script command creates a new atom, pair, fix, compute, or +integrate style. If the variant version does not exist, the standard +version is created. + +The "suffix"_suffix.html command can also set the suffix and it can +also turn off/on the suffix associated with this command-line switch. + -var name value1 value2 ... :pre Specify a variable that will be defined for substitution purposes when diff --git a/doc/package.html b/doc/package.html new file mode 100644 index 0000000000..6ca6d46ac4 --- /dev/null +++ b/doc/package.html @@ -0,0 +1,54 @@ + +
LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Commands + + + + + + + +
+ +package command +
+Syntax: +
+package style args ++
cuda args = to be determined ++ +
Examples: +
+package cuda blah ++
Description: +
+This command is to set package-specific settings. Currently only +the USER-CUDA package uses it. +
+The cuda style invokes options associated with the use of the +USER-CUDA package. These will be described when the USER-CUDA package +is released with LAMMPS. +
+Restrictions: +
+The cuda style of this command can only be invoked if LAMMPS was built +with the USER-CUDA package. See the Making +LAMMPS section for more info. +
+Obviously, you must have GPU hardware and associated software to build +and use LAMMPS with either the GPU or USER-CUDA packages. +
+Related commands: +
+fix gpu +
+Default: none +
+ diff --git a/doc/package.txt b/doc/package.txt new file mode 100644 index 0000000000..e92eb1549d --- /dev/null +++ b/doc/package.txt @@ -0,0 +1,46 @@ +"LAMMPS WWW Site"_lws - "LAMMPS Documentation"_ld - "LAMMPS Commands"_lc :c + +:link(lws,http://lammps.sandia.gov) +:link(ld,Manual.html) +:link(lc,Section_commands.html#comm) + +:line + +package command :h3 + +[Syntax:] + +package style args :pre + +style = {cuda} :ulb,l +args = 0 or more args specific to the style :l + {cuda} args = to be determined :pre +:ule + +[Examples:] + +package cuda blah :pre + +[Description:] + +This command is to set package-specific settings. Currently only +the USER-CUDA package uses it. + +The {cuda} style invokes options associated with the use of the +USER-CUDA package. These will be described when the USER-CUDA package +is released with LAMMPS. + +[Restrictions:] + +The cuda style of this command can only be invoked if LAMMPS was built +with the USER-CUDA package. See the "Making +LAMMPS"_Section_start.html#2_3 section for more info. + +Obviously, you must have GPU hardware and associated software to build +and use LAMMPS with either the GPU or USER-CUDA packages. + +[Related commands:] + +"fix gpu"_fix_gpu.html + +[Default:] none diff --git a/doc/accelerator.html b/doc/suffix.html similarity index 61% rename from doc/accelerator.html rename to doc/suffix.html index dc95be8725..d19049e8fa 100644 --- a/doc/accelerator.html +++ b/doc/suffix.html @@ -9,30 +9,44 @@Syntax:
-accelerator style args +suffix style-
off args = none - on args = none - cuda args = to be determined -- +
Examples:
-accelerator off -accelerator on -accelerator cuda blah +suffix off +suffix on +suffix gpuDescription:
+This command allows you to set a suffix which will +be added +
+Use variants of various styles if they exist. The specified style can +be opt or gpu or cuda. These refer to optional packages that +LAMMPS can be built with, as described above in Section 2.3. +The "opt" style corrsponds to the OPT package, the "gpu" style to the +GPU package, and the "cuda" style to the USER-CUDA package. +
+As an example, all of the packages provide a pair_style +lj/cut variant, with style names lj/cut/opt or +lj/cut/gpu or lj/cut/cuda. Any variant styles can be specified +explicitly in your input script, e.g. pair_style lj/cut/gpu. If the +-suffix switch is used, you do not need to modify your input script. +The specified suffix (opt,gpu,cuda) is automatically appended whenever +an input script command creates a new atom, pair, fix, compute, or +integrate style. If the variant version does not exist, the standard +version is created. +
+The suffix command can also set the suffix. It can also +turn off/on the suffix associated with this command-line switch. +
Alter settings for use of accelerated versions of various styles. LAMMPS can be built with optional packages which provide accelerated versions of specific atom, pair, @@ -66,22 +80,11 @@ be useful for performance testing or debugging. command-line switch is effectively turned off until another accelerator command is used with the on style.
-The cuda style invokes options associated with the use of the -USER-CUDA package. These will be described when the USER-CUDA package -is released with LAMMPS. -
-Restrictions: -
-This cuda style can only be invoked if LAMMPS was built with the -USER-CUDA package. See the Making LAMMPS -section for more info. -
-Obviously, you must have GPU hardware and associated software to build -and use LAMMPS with either the GPU or USER_CUDA packages. +
Restrictions: none
Related commands:
-fix gpu +
Default: none
diff --git a/doc/accelerator.txt b/doc/suffix.txt similarity index 60% rename from doc/accelerator.txt rename to doc/suffix.txt index 9be285f466..86d49d2406 100644 --- a/doc/accelerator.txt +++ b/doc/suffix.txt @@ -6,27 +6,46 @@ :line -accelerator command :h3 +suffix command :h3 [Syntax:] -accelerator style args :pre +suffix style :pre -style = {off} or {on} or {cuda} :ulb,l -args = 0 or more args specific to the style :l - {off} args = none - {on} args = none - {cuda} args = to be determined :pre -:ule +style = {off} or {on} or {opt} or {gpu} or {cuda} :ul [Examples:] -accelerator off -accelerator on -accelerator cuda blah :pre +suffix off +suffix on +suffix gpu :pre [Description:] +This command allows you to set a suffix which will +be added + +Use variants of various styles if they exist. The specified style can +be {opt} or {gpu} or {cuda}. These refer to optional packages that +LAMMPS can be built with, as described above in "Section 2.3"_#2_3. +The "opt" style corrsponds to the OPT package, the "gpu" style to the +GPU package, and the "cuda" style to the USER-CUDA package. + +As an example, all of the packages provide a "pair_style +lj/cut"_pair_lj.html variant, with style names lj/cut/opt or +lj/cut/gpu or lj/cut/cuda. Any variant styles can be specified +explicitly in your input script, e.g. pair_style lj/cut/gpu. If the +-suffix switch is used, you do not need to modify your input script. +The specified suffix (opt,gpu,cuda) is automatically appended whenever +an input script command creates a new atom, pair, fix, compute, or +integrate style. If the variant version does not exist, the standard +version is created. + +The "suffix"_suffix.html command can also set the suffix. It can also +turn off/on the suffix associated with this command-line switch. + + + Alter settings for use of accelerated versions of various styles. LAMMPS can be built with optional packages which provide accelerated versions of specific "atom"_atom_style.html, "pair"_pair_style.html, @@ -60,21 +79,10 @@ The {off} style allows you to do this. The effect of the -accelerator command-line switch is effectively turned off until another accelerator command is used with the {on} style. -The {cuda} style invokes options associated with the use of the -USER-CUDA package. These will be described when the USER-CUDA package -is released with LAMMPS. - -[Restrictions:] - -This cuda style can only be invoked if LAMMPS was built with the -USER-CUDA package. See the "Making LAMMPS"_Section_start.html#2_3 -section for more info. - -Obviously, you must have GPU hardware and associated software to build -and use LAMMPS with either the GPU or USER_CUDA packages. +[Restrictions:] none [Related commands:] -"fix gpu"_fix_gpu.html +"Command-line switch -suffix"_Section_start.html_2_6 [Default:] none