|
@@ -66,8 +66,8 @@ which will be concurrently running on the devices. The default value is 1.
|
|
|
<dd>
|
|
|
\anchor STARPU_CUDA_THREAD_PER_WORKER
|
|
|
\addindex __env__STARPU_CUDA_THREAD_PER_WORKER
|
|
|
-Specify if the cuda driver should provide a thread per stream or a single thread
|
|
|
-dealing with all the streams. 0 if one thread per stream, 1 otherwise. The default
|
|
|
+Specify if the cuda driver should provide a thread per stream or a single thread
|
|
|
+dealing with all the streams. 0 if one thread per stream, 1 otherwise. The default
|
|
|
value is 0. Setting it to 1 is contradictory with setting STARPU_CUDA_THREAD_PER_DEV to 1.
|
|
|
</dd>
|
|
|
|
|
@@ -75,8 +75,8 @@ value is 0. Setting it to 1 is contradictory with setting STARPU_CUDA_THREAD_PER
|
|
|
<dd>
|
|
|
\anchor STARPU_CUDA_THREAD_PER_DEV
|
|
|
\addindex __env__STARPU_CUDA_THREAD_PER_DEV
|
|
|
-Specify if the cuda driver should provide a thread per device or a single thread
|
|
|
-dealing with all the devices. 0 if one thread per device, 1 otherwise. The default
|
|
|
+Specify if the cuda driver should provide a thread per device or a single thread
|
|
|
+dealing with all the devices. 0 if one thread per device, 1 otherwise. The default
|
|
|
value is 1, unless STARPU_CUDA_THREAD_PER_WORKER is set to 1. Setting it to 1 is
|
|
|
contradictory with setting STARPU_CUDA_THREAD_PER_WORKER to 1.
|
|
|
</dd>
|
|
@@ -369,7 +369,7 @@ Disable asynchronous copies between CPU and MIC devices.
|
|
|
|
|
|
<dt>STARPU_DISABLE_ASYNCHRONOUS_MPI_MS_COPY</dt>
|
|
|
<dd>
|
|
|
-\anchor STARPU_DISABLE_ASYNCHRONOUS_MPI_MS_COPY
|
|
|
+\anchor STARPU_DISABLE_ASYNCHRONOUS_MPI_MS_COPY
|
|
|
\addindex __env__STARPU_DISABLE_ASYNCHRONOUS_MPI_MS_COPY
|
|
|
Disable asynchronous copies between CPU and MPI Slave devices.
|
|
|
</dd>
|
|
@@ -946,7 +946,7 @@ full.
|
|
|
This specifies then backend to be used by StarPU to push data when the main
|
|
|
memory is getting full. The default is unistd (i.e. using read/write functions),
|
|
|
other values are stdio (i.e. using fread/fwrite), unistd_o_direct (i.e. using
|
|
|
-read/write with O_DIRECT), leveldb (i.e. using a leveldb database), and hdf5
|
|
|
+read/write with O_DIRECT), leveldb (i.e. using a leveldb database), and hdf5
|
|
|
(i.e. using HDF5 library).
|
|
|
</dd>
|
|
|
|
|
@@ -961,7 +961,7 @@ memory is getting full. The default is unlimited.
|
|
|
<dt>STARPU_LIMIT_MAX_SUBMITTED_TASKS</dt>
|
|
|
<dd>
|
|
|
\anchor STARPU_LIMIT_MAX_SUBMITTED_TASKS
|
|
|
-\addindex __env__STARPU_LIMIT_MAX_SUBMITTED_TASKS
|
|
|
+\addindex __env__STARPU_LIMIT_MAX_SUBMITTED_TASKS
|
|
|
This variable allows the user to control the task submission flow by specifying
|
|
|
to StarPU a maximum number of submitted tasks allowed at a given time, i.e. when
|
|
|
this limit is reached task submission becomes blocking until enough tasks have
|
|
@@ -972,7 +972,7 @@ Setting it enables allocation cache buffer reuse in main memory.
|
|
|
<dt>STARPU_LIMIT_MIN_SUBMITTED_TASKS</dt>
|
|
|
<dd>
|
|
|
\anchor STARPU_LIMIT_MIN_SUBMITTED_TASKS
|
|
|
-\addindex __env__STARPU_LIMIT_MIN_SUBMITTED_TASKS
|
|
|
+\addindex __env__STARPU_LIMIT_MIN_SUBMITTED_TASKS
|
|
|
This variable allows the user to control the task submission flow by specifying
|
|
|
to StarPU a submitted task threshold to wait before unblocking task submission. This
|
|
|
variable has to be used in conjunction with \ref STARPU_LIMIT_MAX_SUBMITTED_TASKS
|
|
@@ -997,6 +997,15 @@ When set to <c>1</c>, this variable indicates that StarPU should automatically
|
|
|
generate a Paje trace when starpu_shutdown() is called.
|
|
|
</dd>
|
|
|
|
|
|
+<dt>STARPU_GENERATE_TRACE_OPTIONS</dt>
|
|
|
+<dd>
|
|
|
+\anchor STARPU_GENERATE_TRACE_OPTIONS
|
|
|
+\addindex __env__STARPU_GENERATE_TRACE_OPTIONS
|
|
|
+When the variable \ref STARPU_GENERATE_TRACE is set to <c>1</c> to
|
|
|
+generate a Paje trace, this variable can be set to specify options (see
|
|
|
+<c>starpu_fxt_tool --help</c>).
|
|
|
+</dd>
|
|
|
+
|
|
|
<dt>STARPU_ENABLE_STATS</dt>
|
|
|
<dd>
|
|
|
\anchor STARPU_ENABLE_STATS
|
|
@@ -1162,7 +1171,7 @@ accesses (see \ref ConcurrentDataAccess).
|
|
|
|
|
|
<dt>STARPU_USE_NUMA</dt>
|
|
|
<dd>
|
|
|
-\anchor STARPU_USE_NUMA
|
|
|
+\anchor STARPU_USE_NUMA
|
|
|
\addindex __env__STARPU_USE_NUMA
|
|
|
When defined, NUMA nodes are taking into account by StarPU. Otherwise, memory
|
|
|
is considered as only one node. This is experimental for now.
|
|
@@ -1191,7 +1200,7 @@ The file will contain the sum of the idle times of all the workers.
|
|
|
<dd>
|
|
|
\anchor SC_HYPERVISOR_POLICY
|
|
|
\addindex __env__SC_HYPERVISOR_POLICY
|
|
|
-Choose between the different resizing policies proposed by StarPU for the hypervisor:
|
|
|
+Choose between the different resizing policies proposed by StarPU for the hypervisor:
|
|
|
idle, app_driven, feft_lp, teft_lp; ispeed_lp, throughput_lp etc.
|
|
|
|
|
|
Use <c>SC_HYPERVISOR_POLICY=help</c> to get the list of available policies for the hypervisor
|
|
@@ -1221,7 +1230,7 @@ of the total time of execution of the application. The default value is the resi
|
|
|
\addindex __env__SC_HYPERVISOR_MAX_SPEED_GAP
|
|
|
Indicate the ratio of speed difference between contexts that should trigger the hypervisor.
|
|
|
This situation may occur only when a theoretical speed could not be computed and the hypervisor
|
|
|
-has no value to compare the speed to. Otherwise the resizing of a context is not influenced by the
|
|
|
+has no value to compare the speed to. Otherwise the resizing of a context is not influenced by the
|
|
|
the speed of the other contexts, but only by the the value that a context should have.
|
|
|
</dd>
|
|
|
|
|
@@ -1239,7 +1248,7 @@ is not done.
|
|
|
\anchor SC_HYPERVISOR_LAZY_RESIZE
|
|
|
\addindex __env__SC_HYPERVISOR_LAZY_RESIZE
|
|
|
By default the hypervisor resizes the contexts in a lazy way, that is workers are firstly added to a new context
|
|
|
-before removing them from the previous one. Once this workers are clearly taken into account
|
|
|
+before removing them from the previous one. Once this workers are clearly taken into account
|
|
|
into the new context (a task was poped there) we remove them from the previous one. However if the application
|
|
|
would like that the change in the distribution of workers should change right away this variable should be set to 0
|
|
|
</dd>
|