.. raw:: html
- <object id="TOC" data="graphical-toc.svg" width="100%" type="image/svg+xml"></object>
+ <object id="TOC" data="graphical-toc.svg" type="image/svg+xml"></object>
<script>
window.onload=function() { // Wait for the SVG to be loaded before changing it
var elem=document.querySelector("#TOC").contentDocument.getElementById("ConfigBox")
Several ``--cfg`` command line arguments can naturally be used. If you
need to include spaces in the argument, don't forget to quote the
-argument. You can even escape the included quotes (write @' for ' if
-you have your argument between ').
+argument. You can even escape the included quotes (write ``@'`` for ``'`` if
+you have your argument between simple quotes).
Another solution is to use the ``<config>`` tag in the platform file. The
only restriction is that this tag must occur before the first
int main(int argc, char *argv[]) {
simgrid::s4u::Engine e(&argc, argv);
- e->set_config("Item:Value");
+ e.set_config("Item:Value");
// Rest of your code
}
- **network/bandwidth-factor:** :ref:`cfg=network/bandwidth-factor`
- **network/crosstraffic:** :ref:`cfg=network/crosstraffic`
- **network/latency-factor:** :ref:`cfg=network/latency-factor`
+- **network/loopback-lat:** :ref:`cfg=network/loopback`
+- **network/loopback-bw:** :ref:`cfg=network/loopback`
- **network/maxmin-selective-update:** :ref:`Network Optimization Level <options_model_optim>`
- **network/model:** :ref:`options_model_select`
- **network/optim:** :ref:`Network Optimization Level <options_model_optim>`
- ``cpu/model``: specify the used CPU model. We have only one model
for now:
- - **Cas01:** Simplistic CPU model (time=size/power)
+ - **Cas01:** Simplistic CPU model (time=size/speed)
- ``host/model``: The host concept is the aggregation of a CPU with a
network card. Three models exists, but actually, only 2 of them are
Note that with the default host model this option is activated by default.
+.. _cfg=network/loopback:
+
+Configuring loopback link
+^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Several network model provide an implicit loopback link to account for local
+communication on a host. By default it has a 10GBps bandwidth and a null latency.
+This can be changed with ``network/loopback-lat`` and ``network/loopback-bw``
+items.
+
.. _cfg=smpi/async-small-thresh:
Simulating Asynchronous Send
the host energy plugin by adding ``--cfg=plugin:host_energy`` to your
command line.
-Here is the full list of plugins that can be activated this way:
+Here is a partial list of plugins that can be activated this way. You can get
+the full list by passing ``--cfg=plugin:help`` to your simulator.
- - **host_energy:** keeps track of the energy dissipated by
- computations. More details in @ref plugin_energy.
- - **link_energy:** keeps track of the energy dissipated by
- communications. More details in @ref SURF_plugin_energy.
- - **host_load:** keeps track of the computational load.
- More details in @ref plugin_load.
+ - :ref:`Host Energy <plugin_host_energy>`: models the energy dissipation of the compute units.
+ - :ref:`Link Energy <plugin_link_energy>`: models the energy dissipation of the network.
+ - :ref:`Host Load <plugin_host_load>`: monitors the load of the compute units.
.. _options_modelchecking:
Configuring the Model-Checking
------------------------------
-To enable the SimGrid model-checking support the program should
+To enable SimGrid's model-checking support, the program should
be executed using the simgrid-mc wrapper:
.. code-block:: shell
If you want to specify liveness properties, you have to pass them on
the command line, specifying the name of the file containing the
-property, as formatted by the ltl2ba program. Note that ltl2ba is not
-part of SimGrid and must be installed separatly.
+property, as formatted by the `ltl2ba <https://github.com/utwente-fmt/ltl2ba>`_ program.
+Note that ltl2ba is not part of SimGrid and must be installed separatly.
.. code-block:: shell
...............................
By default, the system is backtracked to its initial state to explore
-another path instead of backtracking to the exact step before the fork
+another path, instead of backtracking to the exact step before the fork
that we want to explore (this is called stateless verification). This
is done this way because saving intermediate states can rapidly
exhaust the available memory. If you want, you can change the value of
Note that this feature may break the current implementation of the
DPOR reduction technique.
-The ``model-check/visited`` item is the maximum number of states which
+The ``model-check/visited`` item is the maximum number of states, which
are stored in memory. If the maximum number of snapshotted state is
reached, some states will be removed from the memory and some cycles
might be missed. Small values can lead to incorrect verifications, but
The ``model-check/termination`` configuration item can be used to
report if a non-termination execution path has been found. This is a
-path with a cycle which means that the program might never terminate.
+path with a cycle, which means that the program might never terminate.
This only works in safety mode, not in liveness mode.
..........
If set, the ``model-check/dot-output`` configuration item is the name
-of a file in which to write a dot file of the path leading the found
-property (safety or liveness violation) as well as the cycle for
-liveness properties. This dot file can then fed to the graphviz dot
-tool to generate an corresponding graphical representation.
+of a file in which to write a dot file of the path leading to the
+property violation discovered (safety or liveness violation), as well
+as the cycle for liveness properties. This dot file can then be fed to the
+graphviz dot tool to generate an corresponding graphical representation.
.. _cfg=model-check/max-depth:
exploration graph of the model checker. If this limit is reached, a
logging message is sent and the results might not be exact.
-By default, there is not depth limit.
+By default, there is no depth limit.
.. _cfg=model-check/timeout:
The ``model-check/communications-determinism`` and
``model-check/send-determinism`` items can be used to select the
-communication determinism mode of the model checker which checks
+communication determinism mode of the model checker, which checks
determinism properties of the communications of an application.
+.. _options_mc_perf:
+
Verification Performance Considerations
.......................................
The size of the stacks can have a huge impact on the memory
consumption when using model-checking. By default, each snapshot will
-save a copy of the whole stacks and not only of the part which is
+save a copy of the whole stacks and not only of the part that is
really meaningful: you should expect the contribution of the memory
-consumption of the snapshots to be @f$ @mbox{number of processes}
-@times @mbox{stack size} @times @mbox{number of states} @f$.
+consumption of the snapshots to be:
+:math:`\text{number of processes} \times \text{stack size} \times \text{number of states}`.
When compiled against the model checker, the stacks are not
protected with guards: if the stack size is too small for your
-application, the stack will silently overflow on other parts of the
+application, the stack will silently overflow into other parts of the
memory (see :ref:`contexts/guard-size <cfg=contexts/guard-size>`).
.. _cfg=model-check/replay:
Replaying buggy execution paths from the model checker
......................................................
-Debugging the problems reported by the model checker is challenging: First, the
-application under verification cannot be debugged with gdb because the
-model checker already traces it. Then, the model checker may explore several
-execution paths before encountering the issue, making it very difficult to
-understand the outputs. Fortunately, SimGrid provides the execution path leading
-to any reported issue so that you can replay this path out of the model checker,
-enabling the usage of classical debugging tools.
+Debugging the problems reported by the model checker is challenging:
+First, the application under verification cannot be debugged with gdb
+because the model checker already traces it. Then, the model checker may
+explore several execution paths before encountering the issue, making it
+very difficult to understand the output. Fortunately, SimGrid provides
+the execution path leading to any reported issue so that you can replay
+this path reported by the model checker, enabling the usage of classical
+debugging tools.
When the model checker finds an interesting path in the application
execution graph (where a safety or liveness property is violated), it
-generates an identifier for this path. Here is an example of output:
+generates an identifier for this path. Here is an example of the output:
.. code-block:: shell
execution path. All options (but the model checker related ones) must
remain the same. In particular, if you ran your application with
``smpirun -wrapper simgrid-mc``, then do it again. Remove all
-MC-related options, keep the other ones and add
-``--cfg=model-check/replay``.
+MC-related options, keep non-MC-related ones and add
+``--cfg=model-check/replay:???``.
Currently, if the path is of the form ``X;Y;Z``, each number denotes
the actor's pid that is selected at each indecision point. If it's of
of our own, directly implemented in assembly (only available for x86
and amd64 platforms for now) and without any unneeded system call.
-The main reason to change this setting is when the debugging tools get
+The main reason to change this setting is when the debugging tools become
fooled by the optimized context factories. Threads are the most
debugging-friendly contexts, as they allow one to set breakpoints
anywhere with gdb and visualize backtraces for all processes, in order
If you want to push the scalability limits of your code, you might
want to reduce the ``contexts/stack-size`` item. Its default value is
8192 (in KiB), while our Chord simulation works with stacks as small
-as 16 KiB, for example. This *setting is ignored* when using the
-thread factory. Instead, you should compile SimGrid and your
-application with ``-fsplit-stack``. Note that this compilation flag is
-not compatible with the model checker right now.
+as 16 KiB, for example. You can ensure that some actors have a specific
+size by simply changing the value of this configuration item before
+creating these actors. The :cpp:func:`simgrid::s4u::Engine::set_config`
+functions are handy for that.
+
+This *setting is ignored* when using the thread factory (because there
+is no way to modify the stack size with C++ system threads). Instead,
+you should compile SimGrid and your application with
+``-fsplit-stack``. Note that this compilation flag is not compatible
+with the model checker right now.
The operating system should only allocate memory for the pages of the
stack which are actually used and you might not need to use this in
.. code-block:: shell
- --cfg=tracing:yes --cfg=tracing/uncategorized:yes --cfg=triva/uncategorized:uncat.plist
+ --cfg=tracing:yes --cfg=tracing/uncategorized:yes
- The first parameter activates the tracing subsystem, the second
+ The first parameter activates the tracing subsystem, and the second
tells it to trace host and link utilization (without any
- categorization) and the third creates a graph configuration file to
- configure Triva when analysing the resulting trace file.
+ categorization).
- MSG or SimDag-based simulator and categorized traces (you need to
declare categories and classify your tasks according to them)
.. code-block:: shell
- --cfg=tracing:yes --cfg=tracing/categorized:yes --cfg=triva/categorized:cat.plist
+ --cfg=tracing:yes --cfg=tracing/categorized:yes
- The first parameter activates the tracing subsystem, the second
- tells it to trace host and link categorized utilization and the
- third creates a graph configuration file to configure Triva when
- analysing the resulting trace file.
+ The first parameter activates the tracing subsystem, and the second
+ tells it to trace host and link categorized utilization.
- SMPI simulator and traces for a space/time view:
----------------
The SMPI interface provides several specific configuration items.
-These are uneasy to see since the code is usually launched through the
+These are not easy to see, since the code is usually launched through the
``smiprun`` script directly.
.. _cfg=smpi/host-speed:
simulated machine. The variable ``smpi/host-speed`` allows one to specify
the computational speed of the host machine (in flop/s) to use when
scaling the execution times. It defaults to 20000, but you really want
-to update it to get accurate simulation results.
+to adjust it to get accurate simulation results.
-When the code is constituted of numerous consecutive MPI calls, the
+When the code consists of numerous consecutive MPI calls, the
previous mechanism feeds the simulation kernel with numerous tiny
computations. The ``smpi/cpu-threshold`` item becomes handy when this
-impacts badly the simulation performance. It specifies a threshold (in
+impacts badly on the simulation performance. It specifies a threshold (in
seconds) below which the execution chunks are not reported to the
simulation kernel (default value: 1e-6).
.. note:: The option ``smpi/cpu-threshold`` ignores any computation
time spent below this threshold. SMPI does not consider the
- `amount` of these computations; there is no offset for this. Hence,
- a value that is too small, may lead to unreliable simulation
- results.
+ `amount of time` of these computations; there is no offset for
+ this. Hence, a value that is too small, may lead to unreliable
+ simulation results.
In some cases, however, one may wish to disable simulation of
-application computation. This is the case when SMPI is used not to
-simulate an MPI applications, but instead an MPI code that performs
-"live replay" of another MPI app (e.g., ScalaTrace's replay tool,
+the computation of an application. This is the case when SMPI is used not to
+simulate an MPI application, but instead an MPI code that performs
+"live replay" of another MPI app (e.g., ScalaTrace's replay tool, or
various on-line simulators that run an app at scale). In this case the
computation of the replay/simulation logic should not be simulated by
SMPI. Instead, the replay tool or on-line simulator will issue
being replayed/simulated. At the moment, these computation events can
be simulated using SMPI by calling internal smpi_execute*() functions.
-To disable the benchmarking/simulation of computation in the simulated
+To disable the benchmarking/simulation of a computation in the simulated
application, the variable ``smpi/simulate-computation`` should be set
-to no. This option just ignores the timings in your simulation; it
+to **no**. This option just ignores the timings in your simulation; it
still executes the computations itself. If you want to stop SMPI from
doing that, you should check the SMPI_SAMPLE macros, documented in
-Section :ref:`SMPI_adapting_speed`.
+Section :ref:`SMPI_use_faster`.
+------------------------------------+-------------------------+-----------------------------+
| Solution | Computations executed? | Computations simulated? |
line means that the code between two consecutive MPI calls on line 30
in exchange_1.f and line 130 in exchange_1.f should receive a speedup
of 1.18244559422142. The value for the second column is therefore a
-speedup, if it is larger than 1 and a slow-down if it is smaller
+speedup, if it is larger than 1 and a slowdown if it is smaller
than 1. Nothing will be changed if it is equal to 1.
Of course, you can set any arbitrary filenames you want (so the start
the simulation to be more realistic. For instance, the current default
value means that messages with size 65472 and more will get a total of
MAX_BANDWIDTH*0.940694, messages of size 15424 to 65471 will get
-MAX_BANDWIDTH*0.697866 and so on (where MAX_BANDWIDTH denotes the
+MAX_BANDWIDTH*0.697866, and so on (where MAX_BANDWIDTH denotes the
bandwidth of the link).
An experimental script to compute these factors is available online. See
**Option** ``smpi/keep-temps`` **default:** 0 (false)
SMPI usually generates a lot of temporary files that are cleaned after
-use. This option request to preserve them, for example to debug or
+use. This option requests to preserve them, for example to debug or
profile your code. Indeed, the binary files are removed very early
-under the dlopen privatization schema, which tend to fool the
+under the dlopen privatization schema, which tends to fool the
debuggers.
.. _cfg=smpi/lat-factor:
**Option** ``smpi/papi-events`` **default:** unset
-When the PAPI support was compiled in SimGrid, this option takes the
+When the PAPI support is compiled into SimGrid, this option takes the
names of PAPI counters and adds their respective values to the trace
files (See Section :ref:`tracing_tracing_options`).
**Option** ``smpi/privatization`` **default:** "dlopen" (when using smpirun)
-MPI executables are usually meant to be executed in separated
+MPI executables are usually meant to be executed in separate
processes, but SMPI is executed in only one process. Global variables
-from executables will be placed in the same memory zone and shared
+from executables will be placed in the same memory region and shared
between processes, causing intricate bugs. Several options are
possible to avoid this, as described in the main `SMPI publication
<https://hal.inria.fr/hal-01415484>`_ and in the :ref:`SMPI
This configuration option can only use either full paths to libraries,
or full names. Check with ldd the name of the library you want to
-use. Example:
+use. For example:
.. code-block:: shell
This threshold specifies the size in bytes under which the send will
return immediately. This is different from the threshold detailed in
:ref:`cfg=smpi/async-small-thresh` because the message is not
-effectively sent when the send is posted. SMPI still waits for the
-correspondant receive to be posted to perform the communication
+really sent when the send is posted. SMPI still waits for the
+corresponding receive to be posted, in order to perform the communication
operation.
.. _cfg=smpi/coll-selector:
SMPI implements more than 100 different algorithms for MPI collective
communication, to accurately simulate the behavior of most of the
existing MPI libraries. The ``smpi/coll-selector`` item can be used to
-use the decision logic of either OpenMPI or MPICH libraries (by
-default SMPI uses naive version of collective operations).
+select the decision logic either of the OpenMPI or the MPICH libraries. (By
+default SMPI uses naive version of collective operations.)
Each collective operation can be manually selected with a
``smpi/collective_name:algo_name``. Available algorithms are listed in
**Option** ``smpi/iprobe-cpu-usage`` **default:** 1 (no change)
MPI_Iprobe calls can be heavily used in applications. To account
-correctly for the energy cores spend probing, it is necessary to
+correctly for the energy that cores spend probing, it is necessary to
reduce the load that these calls cause inside SimGrid.
-For instance, we measured a max power consumption of 220 W for a
+For instance, we measured a maximum power consumption of 220 W for a
particular application but only 180 W while this application was
probing. Hence, the correct factor that should be passed to this
option would be 180/220 = 0.81.
and hence accounts also for larger messages. In the first
section of the example above, this value is "2".
-Now, SMPI always checks which section it should take for a given
+Now, SMPI always checks which section it should use for a given
message; that is, if a message of size 11 is sent with the
configuration of the example above, only the second section will be
used, not the first, as the first value of the second section is
By setting this option, you can control the amount of time a process
sleeps when MPI_Test() is called; this is important, because SimGrid
normally only advances the time while communication is happening and
-thus, MPI_Test will not add to the time, resulting in a deadlock if
+thus, MPI_Test will not add to the time, resulting in deadlock if it is
used as a break-condition as in the following example:
.. code-block:: cpp
...
}
-To speed up execution, we use a counter to keep track on how often we
-already checked if the handle is now valid or not. Hence, we actually
+To speed up execution, we use a counter to keep track of how often we
+checked if the handle is now valid or not. Hence, we actually
use counter*SLEEP_TIME, that is, the time MPI_Test() causes the
process to sleep increases linearly with the number of previously
failed tests. This behavior can be disabled by setting
If your simulation consumes too much memory, you may want to modify
your code so that the working areas are shared by all MPI ranks. For
-example, in a bloc-cyclic matrix multiplication, you will only
-allocate one set of blocs, and every processes will share them.
+example, in a block-cyclic matrix multiplication, you will only
+allocate one set of blocks, and all processes will share them.
Naturally, this will lead to very wrong results, but this will save a
-lot of memory so this is still desirable for some studies. For more on
+lot of memory. So this is still desirable for some studies. For more on
the motivation for that feature, please refer to the `relevant section
<https://simgrid.github.io/SMPI_CourseWare/topic_understanding_performance/matrixmultiplication>`_
of the SMPI CourseWare (see Activity #2.2 of the pointed
-assignment). In practice, change the call to malloc() and free() into
+assignment). In practice, change the calls for malloc() and free() into
SMPI_SHARED_MALLOC() and SMPI_SHARED_FREE().
SMPI provides two algorithms for this feature. The first one, called
-``local``, allocates one bloc per call to SMPI_SHARED_MALLOC() in your
-code (each call location gets its own bloc) and this bloc is shared
-amongst all MPI ranks. This is implemented with the shm_* functions
+``local``, allocates one block per call to SMPI_SHARED_MALLOC()
+(each call site gets its own block) ,and this block is shared
+among all MPI ranks. This is implemented with the shm_* functions
to create a new POSIX shared memory object (kept in RAM, in /dev/shm)
-for each shared bloc.
+for each shared block.
With the ``global`` algorithm, each call to SMPI_SHARED_MALLOC()
-returns a new address, but it only points to a shadow bloc: its memory
-area is mapped on a 1MiB file on disk. If the returned bloc is of size
+returns a new address, but it only points to a shadow block: its memory
+area is mapped on a 1 MiB file on disk. If the returned block is of size
N MiB, then the same file is mapped N times to cover the whole bloc.
-At the end, no matter how many SMPI_SHARED_MALLOC you do, this will
+At the end, no matter how many times you call SMPI_SHARED_MALLOC, this will
only consume 1 MiB in memory.
You can disable this behavior and come back to regular mallocs (for
-example for debugging purposes) using @c "no" as a value.
+example for debugging purposes) using ``no`` as a value.
If you want to keep private some parts of the buffer, for instance if these
parts are used by the application logic and should not be corrupted, you
-can use SMPI_PARTIAL_SHARED_MALLOC(size, offsets, offsets_count). Example:
+can use SMPI_PARTIAL_SHARED_MALLOC(size, offsets, offsets_count). For example:
.. code-block:: cpp
When smpi/shared-malloc:global is used, the memory consumption problem
is solved, but it may induce too much load on the kernel's pages table.
-In this case, you should use huge pages so that we create only one
-entry per Mb of malloced data instead of one entry per 4k.
+In this case, you should use huge pages so that the kernel creates only one
+entry per MB of malloced data instead of one entry per 4 kB.
To activate this, you must mount a hugetlbfs on your system and allocate
at least one huge page:
This option controls the amount of (simulated) time spent in calls to
MPI_Wtime(), gettimeofday() and clock_gettime(). If you set this value
to 0, the simulated clock is not advanced in these calls, which leads
-to issue if your application contains such a loop:
+to issues if your application contains such a loop:
.. code-block:: cpp
}
When the option smpi/wtime is set to 0, the time advances only on
-communications and computations, so the previous code results in an
+communications and computations. So the previous code results in an
infinite loop: the current [simulated] time will never reach
``some_time_bound``. This infinite loop is avoided when that option
-is set to a small amount, as it is by default since SimGrid v3.21.
+is set to a small value, as it is by default since SimGrid v3.21.
Note that if your application does not contain any loop depending on
the current time only, then setting this option to a non-zero value
will slow down your simulations by a tiny bit: the simulation loop has
-to be broken and reset each time your code ask for the current time.
+to be broken out of and reset each time your code asks for the current time.
If the simulation speed really matters to you, you can avoid this
extra delay by setting smpi/wtime to 0.
**Option** ``debug/clean-atexit`` **default:** on
If your code is segfaulting during its finalization, it may help to
-disable this option to request SimGrid to not attempt any cleanups at
+disable this option to request that SimGrid not attempt any cleanups at
the end of the simulation. Since the Unix process is ending anyway,
the operating system will wipe it all.
**Option** ``path`` **default:** . (current dir)
-It is possible to specify a list of directories to search into for the
+It is possible to specify a list of directories to search in for the
trace files (see :ref:`pf_trace`) by using this configuration
item. To add several directory to the path, set the configuration
item several times, as in ``--cfg=path:toto --cfg=path:tutu``
By default, when Ctrl-C is pressed, the status of all existing actors
is displayed before exiting the simulation. This is very useful to
-debug your code, but it can reveal troublesome if you have many
+debug your code, but it can become troublesome if you have many
actors. Set this configuration item to **off** to disable this
feature.
This configuration option is used to remove the path from the
backtrace shown when an exception is thrown. This is mainly useful for
-the tests: the full file path makes the tests not reproducible because
-the path of source files depend of the build settings. That would
-break most of our tests as we keep comparing output.
+the tests: the full file path would makes the tests non-reproducible because
+the paths of source files depend of the build settings. That would
+break most of the tests since their output is continually compared.
Logging Configuration
---------------------
-It can be done by using XBT. Go to :ref:`XBT_log` for more details.
+This can be done by using XBT. Go to :ref:`XBT_log` for more details.
.. |br| raw:: html
