The device keyword can be used to tune parameters to optimize for a specific +accelerator when using OpenCL. For CUDA, the device keyword is ignored. +Currently, the device type is limited to NVIDIA Kepler, NVIDIA Fermi, +AMD Cypress, or a generic device. More devices will be added soon. The default +device type can be specified when building LAMMPS with the GPU library. +
The cuda style invokes options associated with the use of the diff --git a/doc/package.txt b/doc/package.txt index fb2431f8ad..a4e33f1181 100644 --- a/doc/package.txt +++ b/doc/package.txt @@ -20,11 +20,13 @@ args = arguments specific to the style :l last = ID of last GPU to be used on each node split = fraction of particles assigned to the GPU zero or more keyword/value pairs may be appended - keywords = {threads_per_atom} or {cellsize} + keywords = {threads_per_atom} or {cellsize} or {device} {threads_per_atom} value = Nthreads Nthreads = # of GPU threads used per atom {cellsize} value = dist dist = length (distance units) in each dimension for neighbor bins + {device} value = device_type + device_type = {kepler} or {fermi} or {cypress} or {generic} {cuda} args = keyword value ... one or more keyword/value pairs may be appended keywords = {gpu/node} or {gpu/node/special} or {timing} or {test} or {override/bpa} @@ -141,6 +143,12 @@ it can be more efficient to use smaller values for cellsize in parallel simulations. For example, with a cutoff of 20*sigma and a neighbor skin of sigma, a cellsize of 5.25*sigma can be efficient for parallel simulations. +The {device} keyword can be used to tune parameters to optimize for a specific +accelerator when using OpenCL. For CUDA, the {device} keyword is ignored. +Currently, the device type is limited to NVIDIA Kepler, NVIDIA Fermi, +AMD Cypress, or a generic device. More devices will be added soon. The default +device type can be specified when building LAMMPS with the GPU library. + :line The {cuda} style invokes options associated with the use of the diff --git a/doc/pair_beck.html b/doc/pair_beck.html index 1bdd0e5ed8..bd781dad54 100644 --- a/doc/pair_beck.html +++ b/doc/pair_beck.html @@ -11,6 +11,8 @@
Syntax: diff --git a/doc/pair_beck.txt b/doc/pair_beck.txt index 8c3691156c..a0b2f15c8d 100644 --- a/doc/pair_beck.txt +++ b/doc/pair_beck.txt @@ -7,6 +7,7 @@ :line pair_style beck command :h3 +pair_style beck/gpu command :h3 pair_style beck/omp command :h3 [Syntax:] diff --git a/doc/pair_lj.html b/doc/pair_lj.html index 88da42e1db..ea153066fe 100644 --- a/doc/pair_lj.html +++ b/doc/pair_lj.html @@ -53,6 +53,8 @@
Syntax:
pair_style mie/cut cutoff diff --git a/doc/pair_mie.txt b/doc/pair_mie.txt index 6b576ff4ce..438742021c 100644 --- a/doc/pair_mie.txt +++ b/doc/pair_mie.txt @@ -7,6 +7,7 @@ :line pair_style mie/cut command :h3 +pair_style mie/cut/gpu command :h3 [Syntax:] diff --git a/doc/pair_soft.html b/doc/pair_soft.html index 8d3b45cc09..cc2351b816 100644 --- a/doc/pair_soft.html +++ b/doc/pair_soft.html @@ -11,6 +11,8 @@pair_style soft command
+pair_style soft/gpu command +
pair_style soft/omp command
Syntax: diff --git a/doc/pair_soft.txt b/doc/pair_soft.txt index 3acb16d602..722302c283 100644 --- a/doc/pair_soft.txt +++ b/doc/pair_soft.txt @@ -7,6 +7,7 @@ :line pair_style soft command :h3 +pair_style soft/gpu command :h3 pair_style soft/omp command :h3 [Syntax:] diff --git a/doc/pair_sw.html b/doc/pair_sw.html index b81eb546dd..ecca24ae58 100644 --- a/doc/pair_sw.html +++ b/doc/pair_sw.html @@ -13,6 +13,8 @@
pair_style sw/cuda command
+pair_style sw/gpu command +
pair_style sw/omp command
Syntax: @@ -201,6 +203,8 @@ You can use the SW potential with any LAMMPS units, but you would need to create your own SW potential file with coefficients listed in the appropriate units if your simulation doesn't use "metal" units.
+The sw/gpu style is currently limited to a single element. +
Related commands:
pair_coeff diff --git a/doc/pair_sw.txt b/doc/pair_sw.txt index 6941c61cb5..51a0e307b2 100644 --- a/doc/pair_sw.txt +++ b/doc/pair_sw.txt @@ -8,6 +8,7 @@ pair_style sw command :h3 pair_style sw/cuda command :h3 +pair_style sw/gpu command :h3 pair_style sw/omp command :h3 [Syntax:] @@ -196,6 +197,8 @@ You can use the SW potential with any LAMMPS units, but you would need to create your own SW potential file with coefficients listed in the appropriate units if your simulation doesn't use "metal" units. +The sw/gpu style is currently limited to a single element. + [Related commands:] "pair_coeff"_pair_coeff.html