8 <object id="TOC" data="graphical-toc.svg" width="100%" type="image/svg+xml"></object>
10 window.onload=function() { // Wait for the SVG to be loaded before changing it
11 var elem=document.querySelector("#TOC").contentDocument.getElementById("ConfigBox")
12 elem.style="opacity:0.93999999;fill:#ff0000;fill-opacity:0.1;stroke:#000000;stroke-width:0.35277778;stroke-linecap:round;stroke-linejoin:round;stroke-miterlimit:4;stroke-dasharray:none;stroke-dashoffset:0;stroke-opacity:1";
18 A number of options can be given at runtime to change the default
19 SimGrid behavior. For a complete list of all configuration options
20 accepted by the SimGrid version used in your simulator, simply pass
21 the --help configuration flag to your program. If some of the options
22 are not documented on this page, this is a bug that you should please
23 report so that we can fix it. Note that some of the options presented
24 here may not be available in your simulators, depending on the
25 :ref:`compile-time options <install_src_config>` that you used.
27 Setting Configuration Items
28 ---------------------------
30 There is several way to pass configuration options to the simulators.
31 The most common way is to use the ``--cfg`` command line argument. For
32 example, to set the item ``Item`` to the value ``Value``, simply
33 type the following on the command-line:
37 my_simulator --cfg=Item:Value (other arguments)
39 Several ``--cfg`` command line arguments can naturally be used. If you
40 need to include spaces in the argument, don't forget to quote the
41 argument. You can even escape the included quotes (write @' for ' if
42 you have your argument between ').
44 Another solution is to use the ``<config>`` tag in the platform file. The
45 only restriction is that this tag must occure before the first
46 platform element (be it ``<zone>``, ``<cluster>``, ``<peer>`` or whatever).
47 The ``<config>`` tag takes an ``id`` attribute, but it is currently
48 ignored so you don't really need to pass it. The important part is that
49 within that tag, you can pass one or several ``<prop>`` tags to specify
50 the configuration to use. For example, setting ``Item`` to ``Value``
51 can be done by adding the following to the beginning of your platform
57 <prop id="Item" value="Value"/>
60 A last solution is to pass your configuration directly in your program
61 with :cpp:func:`simgrid::s4u::Engine::set_config` or :cpp:func:`MSG_config`.
65 #include <simgrid/s4u.hpp>
67 int main(int argc, char *argv[]) {
68 simgrid::s4u::Engine e(&argc, argv);
70 e->set_config("Item:Value");
77 Existing Configuration Items
78 ----------------------------
81 The full list can be retrieved by passing ``--help`` and
82 ``--help-cfg`` to an executable that uses SimGrid. Try passing
83 ``help`` as a value to get the list of values accepted by a given
84 option. For example, ``--cfg=plugin:help`` will give you the list
85 of plugins available in your installation of SimGrid.
87 - **contexts/factory:** :ref:`cfg=contexts/factory`
88 - **contexts/guard-size:** :ref:`cfg=contexts/guard-size`
89 - **contexts/nthreads:** :ref:`cfg=contexts/nthreads`
90 - **contexts/parallel-threshold:** :ref:`cfg=contexts/parallel-threshold`
91 - **contexts/stack-size:** :ref:`cfg=contexts/stack-size`
92 - **contexts/synchro:** :ref:`cfg=contexts/synchro`
94 - **cpu/maxmin-selective-update:** :ref:`Cpu Optimization Level <options_model_optim>`
95 - **cpu/model:** :ref:`options_model_select`
96 - **cpu/optim:** :ref:`Cpu Optimization Level <options_model_optim>`
98 - **debug/breakpoint:** :ref:`cfg=debug/breakpoint`
99 - **debug/clean-atexit:** :ref:`cfg=debug/clean-atexit`
100 - **debug/verbose-exit:** :ref:`cfg=debug/verbose-exit`
102 - **exception/cutpath:** :ref:`cfg=exception/cutpath`
104 - **host/model:** :ref:`options_model_select`
106 - **maxmin/precision:** :ref:`cfg=maxmin/precision`
107 - **maxmin/concurrency-limit:** :ref:`cfg=maxmin/concurrency-limit`
109 - **msg/debug-multiple-use:** :ref:`cfg=msg/debug-multiple-use`
111 - **model-check:** :ref:`options_modelchecking`
112 - **model-check/checkpoint:** :ref:`cfg=model-check/checkpoint`
113 - **model-check/communications-determinism:** :ref:`cfg=model-check/communications-determinism`
114 - **model-check/dot-output:** :ref:`cfg=model-check/dot-output`
115 - **model-check/max-depth:** :ref:`cfg=model-check/max-depth`
116 - **model-check/property:** :ref:`cfg=model-check/property`
117 - **model-check/reduction:** :ref:`cfg=model-check/reduction`
118 - **model-check/replay:** :ref:`cfg=model-check/replay`
119 - **model-check/send-determinism:** :ref:`cfg=model-check/send-determinism`
120 - **model-check/termination:** :ref:`cfg=model-check/termination`
121 - **model-check/timeout:** :ref:`cfg=model-check/timeout`
122 - **model-check/visited:** :ref:`cfg=model-check/visited`
124 - **network/bandwidth-factor:** :ref:`cfg=network/bandwidth-factor`
125 - **network/crosstraffic:** :ref:`cfg=network/crosstraffic`
126 - **network/latency-factor:** :ref:`cfg=network/latency-factor`
127 - **network/maxmin-selective-update:** :ref:`Network Optimization Level <options_model_optim>`
128 - **network/model:** :ref:`options_model_select`
129 - **network/optim:** :ref:`Network Optimization Level <options_model_optim>`
130 - **network/TCP-gamma:** :ref:`cfg=network/TCP-gamma`
131 - **network/weight-S:** :ref:`cfg=network/weight-S`
133 - **ns3/TcpModel:** :ref:`options_pls`
134 - **path:** :ref:`cfg=path`
135 - **plugin:** :ref:`cfg=plugin`
137 - **storage/max_file_descriptors:** :ref:`cfg=storage/max_file_descriptors`
139 - **surf/precision:** :ref:`cfg=surf/precision`
141 - **For collective operations of SMPI,** please refer to Section :ref:`cfg=smpi/coll-selector`
142 - **smpi/async-small-thresh:** :ref:`cfg=smpi/async-small-thresh`
143 - **smpi/bw-factor:** :ref:`cfg=smpi/bw-factor`
144 - **smpi/coll-selector:** :ref:`cfg=smpi/coll-selector`
145 - **smpi/comp-adjustment-file:** :ref:`cfg=smpi/comp-adjustment-file`
146 - **smpi/cpu-threshold:** :ref:`cfg=smpi/cpu-threshold`
147 - **smpi/display-timing:** :ref:`cfg=smpi/display-timing`
148 - **smpi/grow-injected-times:** :ref:`cfg=smpi/grow-injected-times`
149 - **smpi/host-speed:** :ref:`cfg=smpi/host-speed`
150 - **smpi/IB-penalty-factors:** :ref:`cfg=smpi/IB-penalty-factors`
151 - **smpi/iprobe:** :ref:`cfg=smpi/iprobe`
152 - **smpi/iprobe-cpu-usage:** :ref:`cfg=smpi/iprobe-cpu-usage`
153 - **smpi/init:** :ref:`cfg=smpi/init`
154 - **smpi/keep-temps:** :ref:`cfg=smpi/keep-temps`
155 - **smpi/lat-factor:** :ref:`cfg=smpi/lat-factor`
156 - **smpi/ois:** :ref:`cfg=smpi/ois`
157 - **smpi/or:** :ref:`cfg=smpi/or`
158 - **smpi/os:** :ref:`cfg=smpi/os`
159 - **smpi/papi-events:** :ref:`cfg=smpi/papi-events`
160 - **smpi/privatization:** :ref:`cfg=smpi/privatization`
161 - **smpi/privatize-libs:** :ref:`cfg=smpi/privatize-libs`
162 - **smpi/send-is-detached-thresh:** :ref:`cfg=smpi/send-is-detached-thresh`
163 - **smpi/shared-malloc:** :ref:`cfg=smpi/shared-malloc`
164 - **smpi/shared-malloc-hugepage:** :ref:`cfg=smpi/shared-malloc-hugepage`
165 - **smpi/simulate-computation:** :ref:`cfg=smpi/simulate-computation`
166 - **smpi/test:** :ref:`cfg=smpi/test`
167 - **smpi/wtime:** :ref:`cfg=smpi/wtime`
169 - **Tracing configuration options** can be found in Section :ref:`tracing_tracing_options`
171 - **storage/model:** :ref:`options_model_select`
173 - **vm/model:** :ref:`options_model_select`
177 Configuring the Platform Models
178 -------------------------------
180 .. _options_model_select:
182 Choosing the Platform Models
183 ............................
185 SimGrid comes with several network, CPU and storage models built in,
186 and you can change the used model at runtime by changing the passed
187 configuration. The three main configuration items are given below.
188 For each of these items, passing the special ``help`` value gives you
189 a short description of all possible values (for example,
190 ``--cfg=network/model:help`` will present all provided network
191 models). Also, ``--help-models`` should provide information about all
192 models for all existing resources.
194 - ``network/model``: specify the used network model. Possible values:
196 - **LV08 (default one):** Realistic network analytic model
197 (slow-start modeled by multiplying latency by 13.01, bandwidth by
198 .97; bottleneck sharing uses a payload of S=20537 for evaluating
199 RTT). Described in `Accuracy Study and Improvement of Network
200 Simulation in the SimGrid Framework
201 <http://mescal.imag.fr/membres/arnaud.legrand/articles/simutools09.pdf>`_.
202 - **Constant:** Simplistic network model where all communication
203 take a constant time (one second). This model provides the lowest
204 realism, but is (marginally) faster.
205 - **SMPI:** Realistic network model specifically tailored for HPC
206 settings (accurate modeling of slow start with correction factors on
207 three intervals: < 1KiB, < 64 KiB, >= 64 KiB). This model can be
208 :ref:`further configured <options_model_network>`.
209 - **IB:** Realistic network model specifically tailored for HPC
210 settings with InfiniBand networks (accurate modeling contention
211 behavior, based on the model explained in `this PhD work
212 <http://mescal.imag.fr/membres/jean-marc.vincent/index.html/PhD/Vienne.pdf>`_.
213 This model can be :ref:`further configured <options_model_network>`.
214 - **CM02:** Legacy network analytic model. Very similar to LV08, but
215 without corrective factors. The timings of small messages are thus
216 poorly modeled. This model is described in `A Network Model for
217 Simulation of Grid Application
218 <https://hal.inria.fr/inria-00071989/document>`_.
219 - **ns-3** (only available if you compiled SimGrid accordingly):
220 Use the packet-level network
221 simulators as network models (see :ref:`model_ns3`).
222 This model can be :ref:`further configured <options_pls>`.
224 - ``cpu/model``: specify the used CPU model. We have only one model
227 - **Cas01:** Simplistic CPU model (time=size/power)
229 - ``host/model``: The host concept is the aggregation of a CPU with a
230 network card. Three models exists, but actually, only 2 of them are
231 interesting. The "compound" one is simply due to the way our
232 internal code is organized, and can easily be ignored. So at the
233 end, you have two host models: The default one allows to aggregate
234 an existing CPU model with an existing network model, but does not
235 allow parallel tasks because these beasts need some collaboration
236 between the network and CPU model. That is why, ptask_07 is used by
237 default when using SimDag.
239 - **default:** Default host model. Currently, CPU:Cas01 and
240 network:LV08 (with cross traffic enabled)
241 - **compound:** Host model that is automatically chosen if
242 you change the network and CPU models
243 - **ptask_L07:** Host model somehow similar to Cas01+CM02 but
244 allowing "parallel tasks", that are intended to model the moldable
245 tasks of the grid scheduling literature.
247 - ``storage/model``: specify the used storage model. Only one model is
249 - ``vm/model``: specify the model for virtual machines. Only one model
252 .. todo: make 'compound' the default host model.
254 .. _options_model_optim:
259 The network and CPU models that are based on lmm_solve (that
260 is, all our analytical models) accept specific optimization
263 - items ``network/optim`` and ``cpu/optim`` (both default to 'Lazy'):
265 - **Lazy:** Lazy action management (partial invalidation in lmm +
266 heap in action remaining).
267 - **TI:** Trace integration. Highly optimized mode when using
268 availability traces (only available for the Cas01 CPU model for
270 - **Full:** Full update of remaining and variables. Slow but may be
271 useful when debugging.
273 - items ``network/maxmin-selective-update`` and
274 ``cpu/maxmin-selective-update``: configure whether the underlying
275 should be lazily updated or not. It should have no impact on the
276 computed timings, but should speed up the computation. |br| It is
277 still possible to disable this feature because it can reveal
278 counter-productive in very specific scenarios where the
279 interaction level is high. In particular, if all your
280 communication share a given backbone link, you should disable it:
281 without it, a simple regular loop is used to update each
282 communication. With it, each of them is still updated (because of
283 the dependency induced by the backbone), but through a complicated
284 and slow pattern that follows the actual dependencies.
286 .. _cfg=maxmin/precision:
287 .. _cfg=surf/precision:
292 **Option** ``maxmin/precision`` **Default:** 0.00001 (in flops or bytes) |br|
293 **Option** ``surf/precision`` **Default:** 0.00001 (in seconds)
295 The analytical models handle a lot of floating point values. It is
296 possible to change the epsilon used to update and compare them through
297 this configuration item. Changing it may speedup the simulation by
298 discarding very small actions, at the price of a reduced numerical
299 precision. You can modify separately the precision used to manipulate
300 timings (in seconds) and the one used to manipulate amounts of work
303 .. _cfg=maxmin/concurrency-limit:
308 **Option** ``maxmin/concurrency-limit`` **Default:** -1 (no limit)
310 The maximum number of variables per resource can be tuned through this
311 option. You can have as many simultaneous actions per resources as you
312 want. If your simulation presents a very high level of concurrency, it
313 may help to use e.g. 100 as a value here. It means that at most 100
314 actions can consume a resource at a given time. The extraneous actions
315 are queued and wait until the amount of concurrency of the considered
316 resource lowers under the given boundary.
318 Such limitations help both to the simulation speed and simulation accuracy
319 on highly constrained scenarios, but the simulation speed suffers of this
320 setting on regular (less constrained) scenarios so it is off by default.
322 .. _options_model_network:
324 Configuring the Network Model
325 .............................
327 .. _cfg=network/TCP-gamma:
329 Maximal TCP Window Size
330 ^^^^^^^^^^^^^^^^^^^^^^^
332 **Option** ``network/TCP-gamma`` **Default:** 4194304
334 The analytical models need to know the maximal TCP window size to take
335 the TCP congestion mechanism into account. On Linux, this value can
336 be retrieved using the following commands. Both give a set of values,
337 and you should use the last one, which is the maximal size.
339 .. code-block:: shell
341 cat /proc/sys/net/ipv4/tcp_rmem # gives the sender window
342 cat /proc/sys/net/ipv4/tcp_wmem # gives the receiver window
344 .. _cfg=smpi/IB-penalty-factors:
345 .. _cfg=network/bandwidth-factor:
346 .. _cfg=network/latency-factor:
347 .. _cfg=network/weight-S:
349 Correcting Important Network Parameters
350 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
352 SimGrid can take network irregularities such as a slow startup or
353 changing behavior depending on the message size into account. You
354 should not change these values unless you really know what you're
355 doing. The corresponding values were computed through data fitting
356 one the timings of packet-level simulators, as described in `Accuracy
357 Study and Improvement of Network Simulation in the SimGrid Framework
358 <http://mescal.imag.fr/membres/arnaud.legrand/articles/simutools09.pdf>`_.
361 If you are using the SMPI model, these correction coefficients are
362 themselves corrected by constant values depending on the size of the
363 exchange. By default SMPI uses factors computed on the Stampede
364 Supercomputer at TACC, with optimal deployment of processes on
365 nodes. Again, only hardcore experts should bother about this fact.
367 InfiniBand network behavior can be modeled through 3 parameters
368 ``smpi/IB-penalty-factors:"βe;βs;γs"``, as explained in `this PhD
370 <http://mescal.imag.fr/membres/jean-marc.vincent/index.html/PhD/Vienne.pdf>`_.
372 .. todo:: This section should be rewritten, and actually explain the
373 options network/bandwidth-factor, network/latency-factor,
376 .. _cfg=network/crosstraffic:
378 Simulating Cross-Traffic
379 ^^^^^^^^^^^^^^^^^^^^^^^^
381 Since SimGrid v3.7, cross-traffic effects can be taken into account in
382 analytical simulations. It means that ongoing and incoming
383 communication flows are treated independently. In addition, the LV08
384 model adds 0.05 of usage on the opposite direction for each new
385 created flow. This can be useful to simulate some important TCP
386 phenomena such as ack compression.
388 For that to work, your platform must have two links for each
389 pair of interconnected hosts. An example of usable platform is
390 available in ``examples/platforms/crosstraffic.xml``.
392 This is activated through the ``network/crosstraffic`` item, that
393 can be set to 0 (disable this feature) or 1 (enable it).
395 Note that with the default host model this option is activated by default.
397 .. _cfg=smpi/async-small-thresh:
399 Simulating Asyncronous Send
400 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
402 (this configuration item is experimental and may change or disapear)
404 It is possible to specify that messages below a certain size will be
405 sent as soon as the call to MPI_Send is issued, without waiting for
406 the correspondant receive. This threshold can be configured through
407 the ``smpi/async-small-thresh`` item. The default value is 0. This
408 behavior can also be manually set for mailboxes, by setting the
409 receiving mode of the mailbox with a call to
410 :cpp:func:`MSG_mailbox_set_async`. After this, all messages sent to
411 this mailbox will have this behavior regardless of the message size.
413 This value needs to be smaller than or equals to the threshold set at
414 @ref options_model_smpi_detached , because asynchronous messages are
415 meant to be detached as well.
422 **Option** ``ns3/TcpModel`` **Default:** "default" (ns-3 default)
424 When using ns-3, there is an extra item ``ns3/TcpModel``, corresponding
425 to the ``ns3::TcpL4Protocol::SocketType`` configuration item in
426 ns-3. The only valid values (enforced on the SimGrid side) are
427 'default' (no change to the ns-3 configuration), 'NewReno' or 'Reno' or
430 Configuring the Storage model
431 .............................
433 .. _cfg=storage/max_file_descriptors:
435 File Descriptor Cound per Host
436 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
438 **Option** ``storage/max_file_descriptors`` **Default:** 1024
440 Each host maintains a fixed-size array of its file descriptors. You
441 can change its size through this item to either enlarge it if your
442 application requires it or to reduce it to save memory space.
449 SimGrid plugins allow to extend the framework without changing its
450 source code directly. Read the source code of the existing plugins to
451 learn how to do so (in ``src/plugins``), and ask your questions to the
452 usual channels (Stack Overflow, Mailing list, IRC). The basic idea is
453 that plugins usually register callbacks to some signals of interest.
454 If they need to store some information about a given object (Link, CPU
455 or Actor), they do so through the use of a dedicated object extension.
457 Some of the existing plugins can be activated from the command line,
458 meaning that you can activate them from the command line without any
459 modification to your simulation code. For example, you can activate
460 the host energy plugin by adding ``--cfg=plugin:host_energy`` to your
463 Here is the full list of plugins that can be activated this way:
465 - **host_energy:** keeps track of the energy dissipated by
466 computations. More details in @ref plugin_energy.
467 - **link_energy:** keeps track of the energy dissipated by
468 communications. More details in @ref SURF_plugin_energy.
469 - **host_load:** keeps track of the computational load.
470 More details in @ref plugin_load.
472 .. _options_modelchecking:
474 Configuring the Model-Checking
475 ------------------------------
477 To enable the SimGrid model-checking support the program should
478 be executed using the simgrid-mc wrapper:
480 .. code-block:: shell
482 simgrid-mc ./my_program
484 Safety properties are expressed as assertions using the function
485 :cpp:func:`void MC_assert(int prop)`.
487 .. _cfg=model-check/property:
489 Specifying a liveness property
490 ..............................
492 **Option** ``model-check/property`` **Default:** unset
494 If you want to specify liveness properties, you have to pass them on
495 the command line, specifying the name of the file containing the
496 property, as formatted by the ltl2ba program.
499 .. code-block:: shell
501 simgrid-mc ./my_program --cfg=model-check/property:<filename>
503 .. _cfg=model-check/checkpoint:
505 Going for Stateful Verification
506 ...............................
508 By default, the system is backtracked to its initial state to explore
509 another path instead of backtracking to the exact step before the fork
510 that we want to explore (this is called stateless verification). This
511 is done this way because saving intermediate states can rapidly
512 exhaust the available memory. If you want, you can change the value of
513 the ``model-check/checkpoint`` item. For example,
514 ``--cfg=model-check/checkpoint:1`` asks to take a checkpoint every
515 step. Beware, this will certainly explode your memory. Larger values
516 are probably better, make sure to experiment a bit to find the right
517 setting for your specific system.
519 .. _cfg=model-check/reduction:
521 Specifying the kind of reduction
522 ................................
524 The main issue when using the model-checking is the state space
525 explosion. To counter that problem, you can chose a exploration
526 reduction techniques with
527 ``--cfg=model-check/reduction:<technique>``. For now, this
528 configuration variable can take 2 values:
530 - **none:** Do not apply any kind of reduction (mandatory for now for
532 - **dpor:** Apply Dynamic Partial Ordering Reduction. Only valid if
533 you verify local safety properties (default value for safety
536 There is unfortunately no silver bullet here, and the most efficient
537 reduction techniques cannot be applied to any properties. In
538 particular, the DPOR method cannot be applied on liveness properties
539 since our implementation of DPOR may break some cycles, while cycles
540 are very important to the soundness of the exploration for liveness
543 .. _cfg=model-check/visited:
545 Size of Cycle Detection Set
546 ...........................
548 In order to detect cycles, the model-checker needs to check if a new
549 explored state is in fact the same state than a previous one. For
550 that, the model-checker can take a snapshot of each visited state:
551 this snapshot is then used to compare it with subsequent states in the
554 The ``model-check/visited`` item is the maximum number of states which
555 are stored in memory. If the maximum number of snapshotted state is
556 reached, some states will be removed from the memory and some cycles
557 might be missed. Small values can lead to incorrect verifications, but
558 large value can exhaust your memory, so choose carefully.
560 By default, no state is snapshotted and cycles cannot be detected.
562 .. _cfg=model-check/termination:
564 Non-Termination Detection
565 .........................
567 The ``model-check/termination`` configuration item can be used to
568 report if a non-termination execution path has been found. This is a
569 path with a cycle which means that the program might never terminate.
571 This only works in safety mode, not in liveness mode.
573 This options is disabled by default.
575 .. _cfg=model-check/dot-output:
580 If set, the ``model-check/dot-output`` configuration item is the name
581 of a file in which to write a dot file of the path leading the found
582 property (safety or liveness violation) as well as the cycle for
583 liveness properties. This dot file can then fed to the graphviz dot
584 tool to generate an corresponding graphical representation.
586 .. _cfg=model-check/max-depth:
588 Exploration Depth Limit
589 .......................
591 The ``model-checker/max-depth`` can set the maximum depth of the
592 exploration graph of the model-checker. If this limit is reached, a
593 logging message is sent and the results might not be exact.
595 By default, there is not depth limit.
597 .. _cfg=model-check/timeout:
602 By default, the model-checker does not handle timeout conditions: the `wait`
603 operations never time out. With the ``model-check/timeout`` configuration item
604 set to **yes**, the model-checker will explore timeouts of `wait` operations.
606 .. _cfg=model-check/communications-determinism:
607 .. _cfg=model-check/send-determinism:
609 Communication Determinism
610 .........................
612 The ``model-check/communications-determinism`` and
613 ``model-check/send-determinism`` items can be used to select the
614 communication determinism mode of the model-checker which checks
615 determinism properties of the communications of an application.
617 Verification Performance Considerations
618 .......................................
620 The size of the stacks can have a huge impact on the memory
621 consumption when using model-checking. By default, each snapshot will
622 save a copy of the whole stacks and not only of the part which is
623 really meaningful: you should expect the contribution of the memory
624 consumption of the snapshots to be @f$ @mbox{number of processes}
625 @times @mbox{stack size} @times @mbox{number of states} @f$.
627 When compiled against the model checker, the stacks are not
628 protected with guards: if the stack size is too small for your
629 application, the stack will silently overflow on other parts of the
630 memory (see :ref:`contexts/guard-size <cfg=contexts/guard-size>`).
632 .. _cfg=model-check/replay:
634 Replaying buggy execution paths out of the model-checker
635 ........................................................
637 Debugging the problems reported by the model-checker is challenging: First, the
638 application under verification cannot be debugged with gdb because the
639 model-checker already traces it. Then, the model-checker may explore several
640 execution paths before encountering the issue, making it very difficult to
641 understand the outputs. Fortunately, SimGrid provides the execution path leading
642 to any reported issue so that you can replay this path out of the model checker,
643 enabling the usage of classical debugging tools.
645 When the model-checker finds an interesting path in the application
646 execution graph (where a safety or liveness property is violated), it
647 generates an identifier for this path. Here is an example of output:
649 .. code-block:: shell
651 [ 0.000000] (0:@) Check a safety property
652 [ 0.000000] (0:@) **************************
653 [ 0.000000] (0:@) *** PROPERTY NOT VALID ***
654 [ 0.000000] (0:@) **************************
655 [ 0.000000] (0:@) Counter-example execution trace:
656 [ 0.000000] (0:@) [(1)Tremblay (app)] MC_RANDOM(3)
657 [ 0.000000] (0:@) [(1)Tremblay (app)] MC_RANDOM(4)
658 [ 0.000000] (0:@) Path = 1/3;1/4
659 [ 0.000000] (0:@) Expanded states = 27
660 [ 0.000000] (0:@) Visited states = 68
661 [ 0.000000] (0:@) Executed transitions = 46
663 The interesting line is ``Path = 1/3;1/4``, which means that you should use
664 `--cfg=model-check/replay:1/3;1/4`` to replay your application on the buggy
665 execution path. The other options should be the same (but the model-checker
666 should be disabled). Note that format and meaning of the path may change between
669 Configuring the User Code Virtualization
670 ----------------------------------------
672 .. _cfg=contexts/factory:
674 Selecting the Virtualization Factory
675 ....................................
677 **Option** contexts/factory **Default:** "raw"
679 In SimGrid, the user code is virtualized in a specific mechanism that
680 allows the simulation kernel to control its execution: when a user
681 process requires a blocking action (such as sending a message), it is
682 interrupted, and only gets released when the simulated clock reaches
683 the point where the blocking operation is done. This is explained
684 graphically in the `relevant tutorial, available online
685 <https://simgrid.org/tutorials/simgrid-simix-101.pdf>`_.
687 In SimGrid, the containers in which user processes are virtualized are
688 called contexts. Several context factory are provided, and you can
689 select the one you want to use with the ``contexts/factory``
690 configuration item. Some of the following may not exist on your
691 machine because of portability issues. In any case, the default one
692 should be the most effcient one (please report bugs if the
693 auto-detection fails for you). They are approximately sorted here from
694 the slowest to the most efficient:
696 - **thread:** very slow factory using full featured threads (either
697 pthreads or windows native threads). They are slow but very
698 standard. Some debuggers or profilers only work with this factory.
699 - **java:** Java applications are virtualized onto java threads (that
700 are regular pthreads registered to the JVM)
701 - **ucontext:** fast factory using System V contexts (Linux and FreeBSD only)
702 - **boost:** This uses the `context
703 implementation <http://www.boost.org/doc/libs/1_59_0/libs/context/doc/html/index.html>`_
704 of the boost library for a performance that is comparable to our
706 |br| Install the relevant library (e.g. with the
707 libboost-contexts-dev package on Debian/Ubuntu) and recompile
709 - **raw:** amazingly fast factory using a context switching mechanism
710 of our own, directly implemented in assembly (only available for x86
711 and amd64 platforms for now) and without any unneeded system call.
713 The main reason to change this setting is when the debugging tools get
714 fooled by the optimized context factories. Threads are the most
715 debugging-friendly contextes, as they allow to set breakpoints
716 anywhere with gdb and visualize backtraces for all processes, in order
717 to debug concurrency issues. Valgrind is also more comfortable with
718 threads, but it should be usable with all factories (Exception: the
719 callgrind tool really dislikes raw and ucontext factories).
721 .. _cfg=contexts/stack-size:
723 Adapting the Stack Size
724 .......................
726 **Option** ``contexts/stack-size`` **Default:** 8192 KiB
728 Each virtualized used process is executed using a specific system
729 stack. The size of this stack has a huge impact on the simulation
730 scalability, but its default value is rather large. This is because
731 the error messages that you get when the stack size is too small are
732 rather disturbing: this leads to stack overflow (overwriting other
733 stacks), leading to segfaults with corrupted stack traces.
735 If you want to push the scalability limits of your code, you might
736 want to reduce the ``contexts/stack-size`` item. Its default value is
737 8192 (in KiB), while our Chord simulation works with stacks as small
738 as 16 KiB, for example. This *setting is ignored* when using the
739 thread factory. Instead, you should compile SimGrid and your
740 application with ``-fsplit-stack``. Note that this compilation flag is
741 not compatible with the model-checker right now.
743 The operating system should only allocate memory for the pages of the
744 stack which are actually used and you might not need to use this in
745 most cases. However, this setting is very important when using the
746 model checker (see :ref:`options_mc_perf`).
748 .. _cfg=contexts/guard-size:
750 Disabling Stack Guard Pages
751 ...........................
753 **Option** ``contexts/guard-size`` **Default** 1 page in most case (0 pages on Windows or with MC)
755 Unless you use the threads context factory (see
756 :ref:`cfg=contexts/factory`), a stack guard page is usually used
757 which prevents the stack of a given actor from overflowing on another
758 stack. But the performance impact may become prohibitive when the
759 amount of actors increases. The option ``contexts/guard-size`` is the
760 number of stack guard pages used. By setting it to 0, no guard pages
761 will be used: in this case, you should avoid using small stacks (with
762 :ref:`contexts/stack-size <cfg=contexts/stack-size>`) as the stack
763 will silently overflow on other parts of the memory.
765 When no stack guard page is created, stacks may then silently overflow
766 on other parts of the memory if their size is too small for the
769 .. _cfg=contexts/nthreads:
770 .. _cfg=contexts/parallel-threshold:
771 .. _cfg=contexts/synchro:
773 Running User Code in Parallel
774 .............................
776 Parallel execution of the user code is only considered stable in
777 SimGrid v3.7 and higher, and mostly for MSG simulations. SMPI
778 simulations may well fail in parallel mode. It is described in
779 `INRIA RR-7653 <http://hal.inria.fr/inria-00602216/>`_.
781 If you are using the **ucontext** or **raw** context factories, you can
782 request to execute the user code in parallel. Several threads are
783 launched, each of them handling as much user contexts at each run. To
784 actiave this, set the ``contexts/nthreads`` item to the amount of
785 cores that you have in your computer (or lower than 1 to have
786 the amount of cores auto-detected).
788 Even if you asked several worker threads using the previous option,
789 you can request to start the parallel execution (and pay the
790 associated synchronization costs) only if the potential parallelism is
791 large enough. For that, set the ``contexts/parallel-threshold``
792 item to the minimal amount of user contexts needed to start the
793 parallel execution. In any given simulation round, if that amount is
794 not reached, the contexts will be run sequentially directly by the
795 main thread (thus saving the synchronization costs). Note that this
796 option is mainly useful when the grain of the user code is very fine,
797 because our synchronization is now very efficient.
799 When parallel execution is activated, you can choose the
800 synchronization schema used with the ``contexts/synchro`` item,
801 which value is either:
803 - **futex:** ultra optimized synchronisation schema, based on futexes
804 (fast user-mode mutexes), and thus only available on Linux systems.
805 This is the default mode when available.
806 - **posix:** slow but portable synchronisation using only POSIX
808 - **busy_wait:** not really a synchronisation: the worker threads
809 constantly request new contexts to execute. It should be the most
810 efficient synchronisation schema, but it loads all the cores of
811 your machine for no good reason. You probably prefer the other less
814 Configuring the Tracing
815 -----------------------
817 The :ref:`tracing subsystem <outcomes_vizu>` can be configured in
818 several different ways depending on the nature of the simulator (MSG,
819 SimDag, SMPI) and the kind of traces that need to be obtained. See the
820 :ref:`Tracing Configuration Options subsection
821 <tracing_tracing_options>` to get a detailed description of each
822 configuration option.
824 We detail here a simple way to get the traces working for you, even if
825 you never used the tracing API.
828 - Any SimGrid-based simulator (MSG, SimDag, SMPI, ...) and raw traces:
830 .. code-block:: shell
832 --cfg=tracing:yes --cfg=tracing/uncategorized:yes --cfg=triva/uncategorized:uncat.plist
834 The first parameter activates the tracing subsystem, the second
835 tells it to trace host and link utilization (without any
836 categorization) and the third creates a graph configuration file to
837 configure Triva when analysing the resulting trace file.
839 - MSG or SimDag-based simulator and categorized traces (you need to
840 declare categories and classify your tasks according to them)
842 .. code-block:: shell
844 --cfg=tracing:yes --cfg=tracing/categorized:yes --cfg=triva/categorized:cat.plist
846 The first parameter activates the tracing subsystem, the second
847 tells it to trace host and link categorized utilization and the
848 third creates a graph configuration file to configure Triva when
849 analysing the resulting trace file.
851 - SMPI simulator and traces for a space/time view:
853 .. code-block:: shell
857 The `-trace` parameter for the smpirun script runs the simulation
858 with ``--cfg=tracing:yes --cfg=tracing/smpi:yes``. Check the
859 smpirun's `-help` parameter for additional tracing options.
861 Sometimes you might want to put additional information on the trace to
862 correctly identify them later, or to provide data that can be used to
863 reproduce an experiment. You have two ways to do that:
865 - Add a string on top of the trace file as comment:
867 .. code-block:: shell
869 --cfg=tracing/comment:my_simulation_identifier
871 - Add the contents of a textual file on top of the trace file as comment:
873 .. code-block:: shell
875 --cfg=tracing/comment-file:my_file_with_additional_information.txt
877 Please, use these two parameters (for comments) to make reproducible
878 simulations. For additional details about this and all tracing
879 options, check See the :ref:`tracing_tracing_options`.
884 .. _cfg=msg/debug-multiple-use:
889 **Option** ``msg/debug-multiple-use`` **Default:** off
891 Sometimes your application may try to send a task that is still being
892 executed somewhere else, making it impossible to send this task. However,
893 for debugging purposes, one may want to know what the other host is/was
894 doing. This option shows a backtrace of the other process.
899 The SMPI interface provides several specific configuration items.
900 These are uneasy to see since the code is usually launched through the
901 ``smiprun`` script directly.
903 .. _cfg=smpi/host-speed:
904 .. _cfg=smpi/cpu-threshold:
905 .. _cfg=smpi/simulate-computation:
907 Automatic Benchmarking of SMPI Code
908 ...................................
910 In SMPI, the sequential code is automatically benchmarked, and these
911 computations are automatically reported to the simulator. That is to
912 say that if you have a large computation between a ``MPI_Recv()`` and
913 a ``MPI_Send()``, SMPI will automatically benchmark the duration of
914 this code, and create an execution task within the simulator to take
915 this into account. For that, the actual duration is measured on the
916 host machine and then scaled to the power of the corresponding
917 simulated machine. The variable ``smpi/host-speed`` allows to specify
918 the computational speed of the host machine (in flop/s) to use when
919 scaling the execution times. It defaults to 20000, but you really want
920 to update it to get accurate simulation results.
922 When the code is constituted of numerous consecutive MPI calls, the
923 previous mechanism feeds the simulation kernel with numerous tiny
924 computations. The ``smpi/cpu-threshold`` item becomes handy when this
925 impacts badly the simulation performance. It specifies a threshold (in
926 seconds) below which the execution chunks are not reported to the
927 simulation kernel (default value: 1e-6).
929 .. note:: The option ``smpi/cpu-threshold`` ignores any computation
930 time spent below this threshold. SMPI does not consider the
931 `amount` of these computations; there is no offset for this. Hence,
932 a value that is too small, may lead to unreliable simulation
935 In some cases, however, one may wish to disable simulation of
936 application computation. This is the case when SMPI is used not to
937 simulate an MPI applications, but instead an MPI code that performs
938 "live replay" of another MPI app (e.g., ScalaTrace's replay tool,
939 various on-line simulators that run an app at scale). In this case the
940 computation of the replay/simulation logic should not be simulated by
941 SMPI. Instead, the replay tool or on-line simulator will issue
942 "computation events", which correspond to the actual MPI simulation
943 being replayed/simulated. At the moment, these computation events can
944 be simulated using SMPI by calling internal smpi_execute*() functions.
946 To disable the benchmarking/simulation of computation in the simulated
947 application, the variable ``smpi/simulate-computation`` should be set
948 to no. This option just ignores the timings in your simulation; it
949 still executes the computations itself. If you want to stop SMPI from
950 doing that, you should check the SMPI_SAMPLE macros, documented in
951 Section :ref:`SMPI_adapting_speed`.
953 +------------------------------------+-------------------------+-----------------------------+
954 | Solution | Computations executed? | Computations simulated? |
955 +====================================+=========================+=============================+
956 | --cfg=smpi/simulate-computation:no | Yes | Never |
957 +------------------------------------+-------------------------+-----------------------------+
958 | --cfg=smpi/cpu-threshold:42 | Yes, in all cases | If it lasts over 42 seconds |
959 +------------------------------------+-------------------------+-----------------------------+
960 | SMPI_SAMPLE() macro | Only once per loop nest | Always |
961 +------------------------------------+-------------------------+-----------------------------+
963 .. _cfg=smpi/comp-adjustment-file:
965 Slow-down or speed-up parts of your code
966 ........................................
968 **Option** ``smpi/comp-adjustment-file:`` **Default:** unset
970 This option allows you to pass a file that contains two columns: The
971 first column defines the section that will be subject to a speedup;
972 the second column is the speedup. For instance:
974 .. code-block:: shell
977 "exchange_1.f:30:exchange_1.f:130",1.18244559422142
979 The first line is the header - you must include it. The following
980 line means that the code between two consecutive MPI calls on line 30
981 in exchange_1.f and line 130 in exchange_1.f should receive a speedup
982 of 1.18244559422142. The value for the second column is therefore a
983 speedup, if it is larger than 1 and a slow-down if it is smaller
984 than 1. Nothing will be changed if it is equal to 1.
986 Of course, you can set any arbitrary filenames you want (so the start
987 and end don't have to be in the same file), but be aware that this
988 mechanism only supports `consecutive calls!`
990 Please note that you must pass the ``-trace-call-location`` flag to
991 smpicc or smpiff, respectively. This flag activates some internal
992 macro definitions that help with obtaining the call location.
994 .. _cfg=smpi/bw-factor:
999 **Option** ``smpi/bw-factor``
1000 |br| **Default:** 65472:0.940694;15424:0.697866;9376:0.58729;5776:1.08739;3484:0.77493;1426:0.608902;732:0.341987;257:0.338112;0:0.812084
1002 The possible throughput of network links is often dependent on the
1003 message sizes, as protocols may adapt to different message sizes. With
1004 this option, a series of message sizes and factors are given, helping
1005 the simulation to be more realistic. For instance, the current default
1006 value means that messages with size 65472 and more will get a total of
1007 MAX_BANDWIDTH*0.940694, messages of size 15424 to 65471 will get
1008 MAX_BANDWIDTH*0.697866 and so on (where MAX_BANDWIDTH denotes the
1009 bandwidth of the link).
1011 An experimental script to compute these factors is available online. See
1012 https://framagit.org/simgrid/platform-calibration/
1013 https://simgrid.org/contrib/smpi-saturation-doc.html
1015 .. _cfg=smpi/display-timing:
1017 Reporting Simulation Time
1018 .........................
1020 **Option** ``smpi/display-timing`` **Default:** 0 (false)
1022 Most of the time, you run MPI code with SMPI to compute the time it
1023 would take to run it on a platform. But since the code is run through
1024 the ``smpirun`` script, you don't have any control on the launcher
1025 code, making it difficult to report the simulated time when the
1026 simulation ends. If you enable the ``smpi/display-timing`` item,
1027 ``smpirun`` will display this information when the simulation
1030 .. _cfg=smpi/keep-temps:
1032 Keeping temporary files after simulation
1033 ........................................
1035 **Option** ``smpi/keep-temps`` **default:** 0 (false)
1037 SMPI usually generates a lot of temporary files that are cleaned after
1038 use. This option request to preserve them, for example to debug or
1039 profile your code. Indeed, the binary files are removed very early
1040 under the dlopen privatization schema, which tend to fool the
1043 .. _cfg=smpi/lat-factor:
1048 **Option** ``smpi/lat-factor`` |br|
1049 **default:** 65472:11.6436;15424:3.48845;9376:2.59299;5776:2.18796;3484:1.88101;1426:1.61075;732:1.9503;257:1.95341;0:2.01467
1051 The motivation and syntax for this option is identical to the motivation/syntax
1052 of :ref:`cfg=smpi/bw-factor`.
1054 There is an important difference, though: While smpi/bw-factor `reduces` the
1055 actual bandwidth (i.e., values between 0 and 1 are valid), latency factors
1056 increase the latency, i.e., values larger than or equal to 1 are valid here.
1058 .. _cfg=smpi/papi-events:
1060 Trace hardware counters with PAPI
1061 .................................
1063 **Option** ``smpi/papi-events`` **default:** unset
1065 When the PAPI support was compiled in SimGrid, this option takes the
1066 names of PAPI counters and adds their respective values to the trace
1067 files (See Section :ref:`tracing_tracing_options`).
1071 This feature currently requires superuser privileges, as registers
1072 are queried. Only use this feature with code you trust! Call
1073 smpirun for instance via ``smpirun -wrapper "sudo "
1074 <your-parameters>`` or run ``sudo sh -c "echo 0 >
1075 /proc/sys/kernel/perf_event_paranoid"`` In the later case, sudo
1076 will not be required.
1078 It is planned to make this feature available on a per-process (or per-thread?) basis.
1079 The first draft, however, just implements a "global" (i.e., for all processes) set
1080 of counters, the "default" set.
1082 .. code-block:: shell
1084 --cfg=smpi/papi-events:"default:PAPI_L3_LDM:PAPI_L2_LDM"
1086 .. _cfg=smpi/privatization:
1088 Automatic Privatization of Global Variables
1089 ...........................................
1091 **Option** ``smpi/privatization`` **default:** "dlopen" (when using smpirun)
1093 MPI executables are usually meant to be executed in separated
1094 processes, but SMPI is executed in only one process. Global variables
1095 from executables will be placed in the same memory zone and shared
1096 between processes, causing intricate bugs. Several options are
1097 possible to avoid this, as described in the main `SMPI publication
1098 <https://hal.inria.fr/hal-01415484>`_ and in the :ref:`SMPI
1099 documentation <SMPI_what_globals>`. SimGrid provides two ways of
1100 automatically privatizing the globals, and this option allows to
1101 choose between them.
1103 - **no** (default when not using smpirun): Do not automatically
1104 privatize variables. Pass ``-no-privatize`` to smpirun to disable
1106 - **dlopen** or **yes** (default when using smpirun): Link multiple
1107 times against the binary.
1108 - **mmap** (slower, but maybe somewhat more stable):
1109 Runtime automatic switching of the data segments.
1112 This configuration option cannot be set in your platform file. You can only
1113 pass it as an argument to smpirun.
1115 .. _cfg=smpi/privatize-libs:
1117 Automatic privatization of global variables inside external libraries
1118 .....................................................................
1120 **Option** ``smpi/privatize-libs`` **default:** unset
1122 **Linux/BSD only:** When using dlopen (default) privatization,
1123 privatize specific shared libraries with internal global variables, if
1124 they can't be linked statically. For example libgfortran is usually
1125 used for Fortran I/O and indexes in files can be mixed up.
1127 Multiple libraries can be given, semicolon separated.
1129 This configuration option can only use either full paths to libraries,
1130 or full names. Check with ldd the name of the library you want to
1133 .. code-block:: shell
1137 libgfortran.so.3 => /usr/lib/x86_64-linux-gnu/libgfortran.so.3 (0x00007fbb4d91b000)
1140 Then you can use ``--cfg=smpi/privatize-libs:libgfortran.so.3``
1141 or ``--cfg=smpi/privatize-libs:/usr/lib/x86_64-linux-gnu/libgfortran.so.3``,
1142 but not ``libgfortran`` nor ``libgfortran.so``.
1144 .. _cfg=smpi/send-is-detached-thresh:
1146 Simulating MPI detached send
1147 ............................
1149 **Option** ``smpi/send-is-detached-thresh`` **default:** 65536
1151 This threshold specifies the size in bytes under which the send will
1152 return immediately. This is different from the threshold detailed in
1153 :ref:`options_model_network_asyncsend` because the message is not
1154 effectively sent when the send is posted. SMPI still waits for the
1155 correspondant receive to be posted to perform the communication
1158 .. _cfg=smpi/coll-selector:
1160 Simulating MPI collective algorithms
1161 ....................................
1163 **Option** ``smpi/coll-selector`` **Possible values:** naive (default), ompi, mpich
1165 SMPI implements more than 100 different algorithms for MPI collective
1166 communication, to accurately simulate the behavior of most of the
1167 existing MPI libraries. The ``smpi/coll-selector`` item can be used to
1168 use the decision logic of either OpenMPI or MPICH libraries (by
1169 default SMPI uses naive version of collective operations).
1171 Each collective operation can be manually selected with a
1172 ``smpi/collective_name:algo_name``. Available algorithms are listed in
1173 :ref:`SMPI_use_colls`.
1175 .. TODO:: All available collective algorithms will be made available
1176 via the ``smpirun --help-coll`` command.
1178 .. _cfg=smpi/iprobe:
1180 Inject constant times for MPI_Iprobe
1181 ....................................
1183 **Option** ``smpi/iprobe`` **default:** 0.0001
1185 The behavior and motivation for this configuration option is identical
1186 with :ref:`smpi/test <cfg=smpi/test>`, but for the function
1189 .. _cfg=smpi/iprobe-cpu-usage:
1191 Reduce speed for iprobe calls
1192 .............................
1194 **Option** ``smpi/iprobe-cpu-usage`` **default:** 1 (no change)
1196 MPI_Iprobe calls can be heavily used in applications. To account
1197 correctly for the energy cores spend probing, it is necessary to
1198 reduce the load that these calls cause inside SimGrid.
1200 For instance, we measured a max power consumption of 220 W for a
1201 particular application but only 180 W while this application was
1202 probing. Hence, the correct factor that should be passed to this
1203 option would be 180/220 = 0.81.
1207 Inject constant times for MPI_Init
1208 ..................................
1210 **Option** ``smpi/init`` **default:** 0
1212 The behavior and motivation for this configuration option is identical
1213 with :ref:`smpi/test <cfg=smpi/test>`, but for the function ``MPI_Init()``.
1217 Inject constant times for MPI_Isend()
1218 .....................................
1220 **Option** ``smpi/ois``
1222 The behavior and motivation for this configuration option is identical
1223 with :ref:`smpi/os <cfg=smpi/os>`, but for the function ``MPI_Isend()``.
1227 Inject constant times for MPI_send()
1228 ....................................
1230 **Option** ``smpi/os``
1232 In several network models such as LogP, send (MPI_Send, MPI_Isend) and
1233 receive (MPI_Recv) operations incur costs (i.e., they consume CPU
1234 time). SMPI can factor these costs in as well, but the user has to
1235 configure SMPI accordingly as these values may vary by machine. This
1236 can be done by using ``smpi/os`` for MPI_Send operations; for MPI_Isend
1237 and MPI_Recv, use ``smpi/ois`` and ``smpi/or``, respectively. These work
1238 exactly as ``smpi/ois``.
1240 This item can consist of multiple sections; each section takes three
1241 values, for example ``1:3:2;10:5:1``. The sections are divided by ";"
1242 so this example contains two sections. Furthermore, each section
1243 consists of three values.
1245 1. The first value denotes the minimum size for this section to take effect;
1246 read it as "if message size is greater than this value (and other section has a larger
1247 first value that is also smaller than the message size), use this".
1248 In the first section above, this value is "1".
1250 2. The second value is the startup time; this is a constant value that will always
1251 be charged, no matter what the size of the message. In the first section above,
1254 3. The third value is the `per-byte` cost. That is, it is charged for every
1255 byte of the message (incurring cost messageSize*cost_per_byte)
1256 and hence accounts also for larger messages. In the first
1257 section of the example above, this value is "2".
1259 Now, SMPI always checks which section it should take for a given
1260 message; that is, if a message of size 11 is sent with the
1261 configuration of the example above, only the second section will be
1262 used, not the first, as the first value of the second section is
1263 closer to the message size. Hence, when ``smpi/os=1:3:2;10:5:1``, a
1264 message of size 11 incurs the following cost inside MPI_Send:
1265 ``5+11*1`` because 5 is the startup cost and 1 is the cost per byte.
1267 Note that the order of sections can be arbitrary; they will be ordered internally.
1271 Inject constant times for MPI_Recv()
1272 ....................................
1274 **Option** ``smpi/or``
1276 The behavior and motivation for this configuration option is identical
1277 with :ref:`smpi/os <cfg=smpi/os>`, but for the function ``MPI_Recv()``.
1280 .. _cfg=smpi/grow-injected-times:
1282 Inject constant times for MPI_Test
1283 ..................................
1285 **Option** ``smpi/test`` **default:** 0.0001
1287 By setting this option, you can control the amount of time a process
1288 sleeps when MPI_Test() is called; this is important, because SimGrid
1289 normally only advances the time while communication is happening and
1290 thus, MPI_Test will not add to the time, resulting in a deadlock if
1291 used as a break-condition as in the following example:
1296 MPI_Test(request, flag, status);
1300 To speed up execution, we use a counter to keep track on how often we
1301 already checked if the handle is now valid or not. Hence, we actually
1302 use counter*SLEEP_TIME, that is, the time MPI_Test() causes the
1303 process to sleep increases linearly with the number of previously
1304 failed tests. This behavior can be disabled by setting
1305 ``smpi/grow-injected-times`` to **no**. This will also disable this
1306 behavior for MPI_Iprobe.
1308 .. _cfg=smpi/shared-malloc:
1309 .. _cfg=smpi/shared-malloc-hugepage:
1314 **Option** ``smpi/shared-malloc`` **Possible values:** global (default), local
1316 If your simulation consumes too much memory, you may want to modify
1317 your code so that the working areas are shared by all MPI ranks. For
1318 example, in a bloc-cyclic matrix multiplication, you will only
1319 allocate one set of blocs, and every processes will share them.
1320 Naturally, this will lead to very wrong results, but this will save a
1321 lot of memory so this is still desirable for some studies. For more on
1322 the motivation for that feature, please refer to the `relevant section
1323 <https://simgrid.github.io/SMPI_CourseWare/topic_understanding_performance/matrixmultiplication>`_
1324 of the SMPI CourseWare (see Activity #2.2 of the pointed
1325 assignment). In practice, change the call to malloc() and free() into
1326 SMPI_SHARED_MALLOC() and SMPI_SHARED_FREE().
1328 SMPI provides two algorithms for this feature. The first one, called
1329 ``local``, allocates one bloc per call to SMPI_SHARED_MALLOC() in your
1330 code (each call location gets its own bloc) and this bloc is shared
1331 amongst all MPI ranks. This is implemented with the shm_* functions
1332 to create a new POSIX shared memory object (kept in RAM, in /dev/shm)
1333 for each shared bloc.
1335 With the ``global`` algorithm, each call to SMPI_SHARED_MALLOC()
1336 returns a new adress, but it only points to a shadow bloc: its memory
1337 area is mapped on a 1MiB file on disk. If the returned bloc is of size
1338 N MiB, then the same file is mapped N times to cover the whole bloc.
1339 At the end, no matter how many SMPI_SHARED_MALLOC you do, this will
1340 only consume 1 MiB in memory.
1342 You can disable this behavior and come back to regular mallocs (for
1343 example for debugging purposes) using @c "no" as a value.
1345 If you want to keep private some parts of the buffer, for instance if these
1346 parts are used by the application logic and should not be corrupted, you
1347 can use SMPI_PARTIAL_SHARED_MALLOC(size, offsets, offsets_count). Example:
1351 mem = SMPI_PARTIAL_SHARED_MALLOC(500, {27,42 , 100,200}, 2);
1353 This will allocate 500 bytes to mem, such that mem[27..41] and
1354 mem[100..199] are shared while other area remain private.
1356 Then, it can be deallocated by calling SMPI_SHARED_FREE(mem).
1358 When smpi/shared-malloc:global is used, the memory consumption problem
1359 is solved, but it may induce too much load on the kernel's pages table.
1360 In this case, you should use huge pages so that we create only one
1361 entry per Mb of malloced data instead of one entry per 4k.
1362 To activate this, you must mount a hugetlbfs on your system and allocate
1363 at least one huge page:
1365 .. code-block:: shell
1368 sudo mount none /home/huge -t hugetlbfs -o rw,mode=0777
1369 sudo sh -c 'echo 1 > /proc/sys/vm/nr_hugepages' # echo more if you need more
1371 Then, you can pass the option
1372 ``--cfg=smpi/shared-malloc-hugepage:/home/huge`` to smpirun to
1373 actually activate the huge page support in shared mallocs.
1377 Inject constant times for MPI_Wtime, gettimeofday and clock_gettime
1378 ...................................................................
1380 **Option** ``smpi/wtime`` **default:** 10 ns
1382 This option controls the amount of (simulated) time spent in calls to
1383 MPI_Wtime(), gettimeofday() and clock_gettime(). If you set this value
1384 to 0, the simulated clock is not advanced in these calls, which leads
1385 to issue if your application contains such a loop:
1389 while(MPI_Wtime() < some_time_bound) {
1390 /* some tests, with no communication nor computation */
1393 When the option smpi/wtime is set to 0, the time advances only on
1394 communications and computations, so the previous code results in an
1395 infinite loop: the current [simulated] time will never reach
1396 ``some_time_bound``. This infinite loop is avoided when that option
1397 is set to a small amount, as it is by default since SimGrid v3.21.
1399 Note that if your application does not contain any loop depending on
1400 the current time only, then setting this option to a non-zero value
1401 will slow down your simulations by a tiny bit: the simulation loop has
1402 to be broken and reset each time your code ask for the current time.
1403 If the simulation speed really matters to you, you can avoid this
1404 extra delay by setting smpi/wtime to 0.
1406 Other Configurations
1407 --------------------
1409 .. _cfg=debug/clean-atexit:
1411 Cleanup at Termination
1412 ......................
1414 **Option** ``debug/clean-atexit`` **default:** on
1416 If your code is segfaulting during its finalization, it may help to
1417 disable this option to request SimGrid to not attempt any cleanups at
1418 the end of the simulation. Since the Unix process is ending anyway,
1419 the operating system will wipe it all.
1426 **Option** ``path`` **default:** . (current dir)
1428 It is possible to specify a list of directories to search into for the
1429 trace files (see :ref:`pf_trace`) by using this configuration
1430 item. To add several directory to the path, set the configuration
1431 item several times, as in ``--cfg=path:toto --cfg=path:tutu``
1433 .. _cfg=debug/breakpoint:
1438 **Option** ``debug/breakpoint`` **default:** unset
1440 This configuration option sets a breakpoint: when the simulated clock
1441 reaches the given time, a SIGTRAP is raised. This can be used to stop
1442 the execution and get a backtrace with a debugger.
1444 It is also possible to set the breakpoint from inside the debugger, by
1445 writing in global variable simgrid::simix::breakpoint. For example,
1448 .. code-block:: shell
1450 set variable simgrid::simix::breakpoint = 3.1416
1452 .. _cfg=debug/verbose-exit:
1457 **Option** ``debug/verbose-exit`` **default:** on
1459 By default, when Ctrl-C is pressed, the status of all existing actors
1460 is displayed before exiting the simulation. This is very useful to
1461 debug your code, but it can reveal troublesome if you have many
1462 actors. Set this configuration item to **off** to disable this
1465 .. _cfg=exception/cutpath:
1467 Truncate local path from exception backtrace
1468 ............................................
1470 **Option** ``exception/cutpath`` **default:** off
1472 This configuration option is used to remove the path from the
1473 backtrace shown when an exception is thrown. This is mainly useful for
1474 the tests: the full file path makes the tests not reproducible because
1475 the path of source files depend of the build settings. That would
1476 break most of our tests as we keep comparing output.
1478 Logging Configuration
1479 ---------------------
1481 It can be done by using XBT. Go to :ref:`XBT_log` for more details.