A last solution is to pass your configuration directly using the C
interface. If you happen to use the MSG interface, this is very easy
-with the MSG_config() function. If you do not use MSG, that's a bit
+with the simgrid::s4u::Engine::setConfig() or MSG_config() functions. If you do not use MSG, that's a bit
more complex, as you have to mess with the internal configuration set
directly as follows. Check the \ref XBT_config "relevant page" for
details on all the functions you can use in this context, \c
@code
#include <xbt/config.h>
-extern xbt_cfg_t _sg_cfg_set;
-
int main(int argc, char *argv[]) {
SD_init(&argc, argv);
/* Prefer MSG_config() if you use MSG!! */
- xbt_cfg_set_parse(_sg_cfg_set,"Item:Value");
+ xbt_cfg_set_parse("Item:Value");
// Rest of your code
}
@endcode
+\section options_index Index of all existing configuration options
+
+\note
+ The full list can be retrieved by passing "--help" and
+ "--help-cfg" to an executable that uses SimGrid.
+
+- \c clean-atexit: \ref options_generic_clean_atexit
+
+- \c contexts/factory: \ref options_virt_factory
+- \c contexts/guard-size: \ref options_virt_guard_size
+- \c contexts/nthreads: \ref options_virt_parallel
+- \c contexts/parallel-threshold: \ref options_virt_parallel
+- \c contexts/stack-size: \ref options_virt_stacksize
+- \c contexts/synchro: \ref options_virt_parallel
+
+- \c cpu/maxmin-selective-update: \ref options_model_optim
+- \c cpu/model: \ref options_model_select
+- \c cpu/optim: \ref options_model_optim
+
+- \c exception/cutpath: \ref options_exception_cutpath
+
+- \c host/model: \ref options_model_select
+
+- \c maxmin/precision: \ref options_model_precision
+- \c maxmin/concurrency-limit: \ref options_concurrency_limit
+
+- \c msg/debug-multiple-use: \ref options_msg_debug_multiple_use
+
+- \c model-check: \ref options_modelchecking
+- \c model-check/checkpoint: \ref options_modelchecking_steps
+- \c model-check/communications-determinism: \ref options_modelchecking_comm_determinism
+- \c model-check/dot-output: \ref options_modelchecking_dot_output
+- \c model-check/hash: \ref options_modelchecking_hash
+- \c model-check/property: \ref options_modelchecking_liveness
+- \c model-check/max-depth: \ref options_modelchecking_max_depth
+- \c model-check/record: \ref options_modelchecking_recordreplay
+- \c model-check/reduction: \ref options_modelchecking_reduction
+- \c model-check/replay: \ref options_modelchecking_recordreplay
+- \c model-check/send-determinism: \ref options_modelchecking_comm_determinism
+- \c model-check/sparse-checkpoint: \ref options_modelchecking_sparse_checkpoint
+- \c model-check/termination: \ref options_modelchecking_termination
+- \c model-check/timeout: \ref options_modelchecking_timeout
+- \c model-check/visited: \ref options_modelchecking_visited
+
+- \c network/bandwidth-factor: \ref options_model_network_coefs
+- \c network/crosstraffic: \ref options_model_network_crosstraffic
+- \c network/latency-factor: \ref options_model_network_coefs
+- \c network/maxmin-selective-update: \ref options_model_optim
+- \c network/model: \ref options_model_select
+- \c network/optim: \ref options_model_optim
+- \c network/TCP-gamma: \ref options_model_network_gamma
+- \c network/weight-S: \ref options_model_network_coefs
+
+- \c ns3/TcpModel: \ref options_pls
+- \c path: \ref options_generic_path
+- \c plugin: \ref options_generic_plugin
+
+- \c storage/max_file_descriptors: \ref option_model_storage_maxfd
+
+- \c surf/precision: \ref options_model_precision
+
+- \c <b>For collective operations of SMPI, please refer to Section \ref options_index_smpi_coll</b>
+- \c smpi/async-small-thresh: \ref options_model_network_asyncsend
+- \c smpi/bw-factor: \ref options_model_smpi_bw_factor
+- \c smpi/coll-selector: \ref options_model_smpi_collectives
+- \c smpi/comp-adjustment-file: \ref options_model_smpi_adj_file
+- \c smpi/cpu-threshold: \ref options_smpi_bench
+- \c smpi/display-timing: \ref options_smpi_timing
+- \c smpi/grow-injected-times: \ref options_model_smpi_test
+- \c smpi/host-speed: \ref options_smpi_bench
+- \c smpi/IB-penalty-factors: \ref options_model_network_coefs
+- \c smpi/iprobe: \ref options_model_smpi_iprobe
+- \c smpi/iprobe-cpu-usage: \ref options_model_smpi_iprobe_cpu_usage
+- \c smpi/init: \ref options_model_smpi_init
+- \c smpi/keep-temps: \ref options_smpi_temps
+- \c smpi/lat-factor: \ref options_model_smpi_lat_factor
+- \c smpi/ois: \ref options_model_smpi_ois
+- \c smpi/or: \ref options_model_smpi_or
+- \c smpi/os: \ref options_model_smpi_os
+- \c smpi/papi-events: \ref options_smpi_papi_events
+- \c smpi/privatization: \ref options_smpi_privatization
+- \c smpi/send-is-detached-thresh: \ref options_model_smpi_detached
+- \c smpi/shared-malloc: \ref options_model_smpi_shared_malloc
+- \c smpi/shared-malloc-hugepage: \ref options_model_smpi_shared_malloc
+- \c smpi/simulate-computation: \ref options_smpi_bench
+- \c smpi/test: \ref options_model_smpi_test
+- \c smpi/wtime: \ref options_model_smpi_wtime
+
+- \c <b>Tracing configuration options can be found in Section \ref tracing_tracing_options</b>.
+
+- \c storage/model: \ref options_storage_model
+- \c verbose-exit: \ref options_generic_exit
+
+- \c vm/model: \ref options_vm_model
+
+\subsection options_index_smpi_coll Index of SMPI collective algorithms options
+
+TODO: All available collective algorithms will be made available via the ``smpirun --help-coll`` command.
+
\section options_model Configuring the platform models
\anchor options_storage_model
- \b storage/model: specify the used storage model (there is currently only one such model - this option is hence only useful for future releases)
- \b vm/model: specify the model for virtual machines (there is currently only one such model - this option is hence only useful for future releases)
-%As of writing, the following network models are accepted. Over
+As of writing, the following network models are accepted. Over
the time new models can be added, and some experimental models can be
removed; check the values on your simulators for an uptodate
information. Note that the CM02 model is described in the research report
lmm_solve (experts only; check the code for more info).
If you compiled SimGrid accordingly, you can use packet-level network
-simulators as network models (see \ref pls). In that case, you have
-two extra models, described below, and some \ref options_pls "specific
-additional configuration flags".
- - \b NS3: Network pseudo-model using the NS3 tcp model instead of an
- analytic model
+simulators as network models (see \ref pls_ns3). In that case, you have
+two extra models, described below, and some
+\ref options_pls "specific additional configuration flags".
+ - \b NS3: Network pseudo-model using the NS3 tcp model
Concerning the CPU, we have only one model for now:
- \b Cas01: Simplistic CPU model (time=size/power)
- \b compound: Host model that is automatically chosen if
you change the network and CPU models
- \b ptask_L07: Host model somehow similar to Cas01+CM02 but
- allowing parallel tasks
+ allowing "parallel tasks", that are intended to model the moldable
+ tasks of the grid scheduling literature.
\subsection options_generic_plugin Plugins
The network and CPU models that are based on lmm_solve (that
is, all our analytical models) accept specific optimization
configurations.
- - items \b network/optim and \b CPU/optim (both default to 'Lazy'):
+ - items \b network/optim and \b cpu/optim (both default to 'Lazy'):
- \b Lazy: Lazy action management (partial invalidation in lmm +
heap in action remaining).
- \b TI: Trace integration. Highly optimized mode when using
now).
- \b Full: Full update of remaining and variables. Slow but may be
useful when debugging.
- - items \b network/maxmin_selective_update and
- \b cpu/maxmin_selective_update: configure whether the underlying
+ - items \b network/maxmin-selective-update and
+ \b cpu/maxmin-selective-update: configure whether the underlying
should be lazily updated or not. It should have no impact on the
computed timings, but should speed up the computation.
-It is still possible to disable the \c maxmin_selective_update feature
+It is still possible to disable the \c maxmin-selective-update feature
because it can reveal counter-productive in very specific scenarios
where the interaction level is high. In particular, if all your
communication share a given backbone link, you should disable it:
-without \c maxmin_selective_update, every communications are updated
+without \c maxmin-selective-update, every communications are updated
at each step through a simple loop over them. With that feature
enabled, every communications will still get updated in this case
(because of the dependency induced by the backbone), but through a
may speedup the simulation by discarding very small actions, at the
price of a reduced numerical precision.
-\subsection options_model_nthreads Parallel threads for model updates
+\subsection options_concurrency_limit Concurrency limit
-By default, Surf computes the analytical models sequentially to share their
-resources and update their actions. It is possible to run them in parallel,
-using the \b surf/nthreads item (default value: 1). If you use a
-negative or null value, the amount of available cores is automatically
-detected and used instead.
+The maximum number of variables per resource can be tuned through
+the \b maxmin/concurrency-limit item. The default value is -1, meaning that
+there is no such limitation. You can have as many simultaneous actions per
+resources as you want. If your simulation presents a very high level of
+concurrency, it may help to use e.g. 100 as a value here. It means that at
+most 100 actions can consume a resource at a given time. The extraneous actions
+are queued and wait until the amount of concurrency of the considered resource
+lowers under the given boundary.
-Depending on the workload of the models and their complexity, you may get a
-speedup or a slowdown because of the synchronization costs of threads.
+Such limitations help both to the simulation speed and simulation accuracy
+on highly constrained scenarios, but the simulation speed suffers of this
+setting on regular (less constrained) scenarios so it is off by default.
\subsection options_model_network Configuring the Network model
\subsubsection options_model_network_gamma Maximal TCP window size
The analytical models need to know the maximal TCP window size to take
-the TCP congestion mechanism into account. This is set to 20000 by
-default, but can be changed using the \b network/TCP_gamma item.
+the TCP congestion mechanism into account. This is set to 4194304 by
+default, but can be changed using the \b network/TCP-gamma item.
On linux, this value can be retrieved using the following
commands. Both give a set of values, and you should use the last one,
These factors can be changed through the following option:
\verbatim
-smpi/IB_penalty_factors:"βe;βs;γs"
+smpi/IB-penalty-factors:"βe;βs;γs"
\endverbatim
By default SMPI uses factors computed on the Stampede Supercomputer at TACC, with optimal
\subsubsection options_model_network_crosstraffic Simulating cross-traffic
-%As of SimGrid v3.7, cross-traffic effects can be taken into account in
+As of SimGrid v3.7, cross-traffic effects can be taken into account in
analytical simulations. It means that ongoing and incoming
communication flows are treated independently. In addition, the LV08
model adds 0.05 of usage on the opposite direction for each new
Note that with the default host model this option is activated by default.
-\subsubsection options_model_network_coord Coordinated-based network models
-
-When you want to use network coordinates, as it happens when you use
-an \<AS\> in your platform file with \c Vivaldi as a routing (see also
-Section \ref pf_routing_model_vivaldi "Vivaldi Routing Model"), you must
-set the \b network/coordinates to \c yes so that all mandatory
-initialization are done in the simulator.
-
-\subsubsection options_model_network_sendergap Simulating sender gap
-
-(this configuration item is experimental and may change or disapear)
-
-It is possible to specify a timing gap between consecutive emission on
-the same network card through the \b network/sender_gap item. This
-is still under investigation as of writting, and the default value is
-to wait 10 microseconds (1e-5 seconds) between emissions.
-
\subsubsection options_model_network_asyncsend Simulating asyncronous send
(this configuration item is experimental and may change or disapear)
It is possible to specify that messages below a certain size will be sent
as soon as the call to MPI_Send is issued, without waiting for the
correspondant receive. This threshold can be configured through the
-\b smpi/async_small_thresh item. The default value is 0. This behavior can also be
+\b smpi/async-small-thresh item. The default value is 0. This behavior can also be
manually set for MSG mailboxes, by setting the receiving mode of the mailbox
with a call to \ref MSG_mailbox_set_async . For MSG, all messages sent to this
mailbox will have this behavior, so consider using two mailboxes if needed.
in NS3. The only valid values (enforced on the SimGrid side) are
'NewReno' or 'Reno' or 'Tahoe'.
+\subsection options_model_storage Configuring the Storage model
+
+\subsubsection option_model_storage_maxfd Maximum amount of file descriptors per host
+
+Each host maintains a fixed-size array of its file descriptors. You
+can change its size (1024 by default) through the \b
+storage/max_file_descriptors item to either enlarge it if your
+application requires it or to reduce it to save memory space.
+
\section options_modelchecking Configuring the Model-Checking
To enable the SimGrid model-checking support the program should
* none: Do not apply any kind of reduction (mandatory for now for
liveness properties)
* dpor: Apply Dynamic Partial Ordering Reduction. Only valid if you
- verify local safety properties.
+ verify local safety properties (default value for safety checks).
\subsection options_modelchecking_visited model-check/visited, Cycle detection
In order to detect cycles, the model-checker needs to check if a new explored
-state is in fact the same state than a previous one. In order to do this,
+state is in fact the same state than a previous one. For that,
the model-checker can take a snapshot of each visited state: this snapshot is
then used to compare it with subsequent states in the exploration graph.
-The \b model-check/visited 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.
+The \b model-check/visited option 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 large value can
+exhaust your memory, so choose carefully.
By default, no state is snapshotted and cycles cannot be detected.
This options is disabled by default.
-\subsection options_modelchecking_dot_output model-check/dot_output, Dot output
+\subsection options_modelchecking_dot_output model-check/dot-output, Dot output
-If set, the \b model-check/dot_output configuration item is the name of a file
+If set, the \b 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.
-\subsection options_modelchecking_max_depth model-check/max_depth, Depth limit
+\subsection options_modelchecking_max_depth model-check/max-depth, Depth limit
-The \b model-checker/max_depth can set the maximum depth of the exploration
+The \b model-checker/max-depth can set the maximum depth of the exploration
graph of the model-checker. If this limit is reached, a logging message is
sent and the results might not be exact.
\subsection options_modelchecking_comm_determinism Communication determinism
-The \b model-check/communications_determinism and
-\b model-check/send_determinism items can be used to select the communication
+The \b model-check/communications-determinism and
+\b model-check/send-determinism items can be used to select the communication
determinism mode of the model-checker which checks determinism properties of
the communications of an application.
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
-memory.
+memory (see \ref options_virt_guard_size).
\subsection options_modelchecking_hash Hashing of the state (experimental)
that allows the simulation kernel to control its execution: when a user
process requires a blocking action (such as sending a message), it is
interrupted, and only gets released when the simulated clock reaches
-the point where the blocking operation is done.
+the point where the blocking operation is done. This is explained
+graphically in the [relevant tutorial, available online](http://simgrid.gforge.inria.fr/tutorials/simgrid-simix-101.pdf).
In SimGrid, the containers in which user processes are virtualized are
called contexts. Several context factory are provided, and you can
configuration item. Some of the following may not exist on your
machine because of portability issues. In any case, the default one
should be the most effcient one (please report bugs if the
-auto-detection fails for you). They are sorted here from the slowest
-to the most effient:
+auto-detection fails for you). They are approximately sorted here from
+the slowest to the most efficient:
+
- \b thread: very slow factory using full featured threads (either
- pthreads or windows native threads)
- - \b ucontext: fast factory using System V contexts (or a portability
- layer of our own on top of Windows fibers)
- - \b raw: amazingly fast factory using a context switching mecanism
- of our own, directly implemented in assembly (only available for x86
- and amd64 platforms for now)
+ pthreads or windows native threads). They are slow but very
+ standard. Some debuggers or profilers only work with this factory.
+ - \b java: Java applications are virtualized onto java threads (that
+ are regular pthreads registered to the JVM)
+ - \b ucontext: fast factory using System V contexts (Linux and FreeBSD only)
- \b boost: This uses the [context implementation](http://www.boost.org/doc/libs/1_59_0/libs/context/doc/html/index.html)
- of the boost library; you must have this library installed before
- you compile SimGrid. (On Debian GNU/Linux based systems, this is
- provided by the libboost-contexts-dev package.)
+ of the boost library for a performance that is comparable to our
+ raw implementation.\nInstall the relevant library (e.g. with the
+ libboost-contexts-dev package on Debian/Ubuntu) and recompile
+ SimGrid. Note that our implementation is not compatible with recent
+ implementations of the library, and it will be hard to fix this since
+ the library's author decided to hide an API that we were using.
+ - \b raw: amazingly fast factory using a context switching mechanism
+ of our own, directly implemented in assembly (only available for x86
+ and amd64 platforms for now) and without any unneeded system call.
-The only reason to change this setting is when the debugging tools get
+The main reason to change this setting is when the debugging tools get
fooled by the optimized context factories. Threads are the most
-debugging-friendly contextes, as they allow to set breakpoints anywhere with gdb
- and visualize backtraces for all processes, in order to debug concurrency issues.
-Valgrind is also more comfortable with threads, but it should be usable with all factories.
+debugging-friendly contextes, as they allow to set breakpoints
+anywhere with gdb and visualize backtraces for all processes, in order
+to debug concurrency issues. Valgrind is also more comfortable with
+threads, but it should be usable with all factories (but the callgrind
+tool that really don't like raw and ucontext factories).
\subsection options_virt_stacksize Adapting the used stack size
stacks), leading to segfaults with corrupted stack traces.
If you want to push the scalability limits of your code, you might
-want to reduce the \b contexts/stack_size item. Its default value
+want to reduce the \b 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. For the thread factory, the default value
-is the one of the system, if it is too large/small, it has to be set
-with this parameter.
+is the one of the system but you can still change it with this parameter.
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
most cases. However, this setting is very important when using the
model checker (see \ref options_mc_perf).
-In some cases, no stack guard page is used and the stack will silently
-overflow on other parts of the memory if the stack size is too small
-for your application. This happens :
+\subsection options_virt_guard_size Disabling stack guard pages
+
+A stack guard page is usually used which prevents the stack of a given
+actor from overflowing on another stack. But the performance impact
+may become prohibitive when the amount of actors increases. The
+option \b contexts:guard-size is the number of stack guard pages used.
+By setting it to 0, no guard pages will be used: in this case, you
+should avoid using small stacks (\b stack-size) as the stack will
+silently overflow on other parts of the memory.
+
+When no stack guard page is created, stacks may then silently overflow
+on other parts of the memory if their size is too small for the
+application. This happens:
- on Windows systems;
- when the model checker is enabled;
-- when stack guard pages are explicitely disabled (see \ref options_perf_guard_size).
+- and of course when guard pages are explicitely disabled (with \b contexts:guard-size=0).
\subsection options_virt_parallel Running user code in parallel
Parallel execution of the user code is only considered stable in
-SimGrid v3.7 and higher. It is described in
+SimGrid v3.7 and higher, and mostly for MSG simulations. SMPI
+simulations may well fail in parallel mode. It is described in
<a href="http://hal.inria.fr/inria-00602216/">INRIA RR-7653</a>.
If you are using the \c ucontext or \c raw context factories, you can
Even if you asked several worker threads using the previous option,
you can request to start the parallel execution (and pay the
associated synchronization costs) only if the potential parallelism is
-large enough. For that, set the \b contexts/parallel_threshold
+large enough. For that, set the \b contexts/parallel-threshold
item to the minimal amount of user contexts needed to start the
parallel execution. In any given simulation round, if that amount is
not reached, the contexts will be run sequentially directly by the
\section options_tracing Configuring the tracing subsystem
-The \ref tracing "tracing subsystem" can be configured in several
+The \ref outcomes_vizu "tracing subsystem" can be configured in several
different ways depending on the nature of the simulator (MSG, SimDag,
SMPI) and the kind of traces that need to be obtained. See the \ref
tracing_tracing_options "Tracing Configuration Options subsection" to
- Add the contents of a textual file on top of the trace file as comment:
\verbatim
---cfg=tracing/comment_file:my_file_with_additional_information.txt
+--cfg=tracing/comment-file:my_file_with_additional_information.txt
\endverbatim
Please, use these two parameters (for comments) to make reproducible
Enable this option by adding
\verbatim
---cfg=msg/debug_multiple_use:on
+--cfg=msg/debug-multiple-use:on
\endverbatim
\section options_smpi Configuring SMPI
code, and create an execution task within the simulator to take this
into account. For that, the actual duration is measured on the host
machine and then scaled to the power of the corresponding simulated
-machine. The variable \b smpi/running_power allows to specify the
-computational power of the host machine (in flop/s) to use when
+machine. The variable \b smpi/host-speed allows 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.
When the code is constituted of numerous consecutive MPI calls, the
previous mechanism feeds the simulation kernel with numerous tiny
-computations. The \b smpi/cpu_threshold item becomes handy when this
+computations. The \b smpi/cpu-threshold item becomes handy when this
impacts badly 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
+ The option smpi/cpu-threshold ignores any computation time spent
below this threshold. SMPI does not consider the \a amount of these
computations; there is no offset for this. Hence, by using a
value that is too low, you may end up with unreliable simulation
results.
- In some cases, however, one may wish to disable simulation of
+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,
be simulated using SMPI by calling internal smpi_execute*() functions.
To disable the benchmarking/simulation of computation in the simulated
-application, the variable \b smpi/simulate_computation should be set to no.
-Equivalently, setting \b smpi/cpu_threshold to -1 also ignores all
-computation.
+application, the variable \b smpi/simulate-computation should be set to no.
\note
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 the chapter
+ you should check the SMPI_SAMPLE macros, documented in the section
\ref SMPI_adapting_speed.
-\subsection options_model_smpi_bw_factor smpi/bw_factor: Bandwidth factors
+Solution | Computations actually executed? | Computations simulated ?
+---------------------------------- | ------------------------------- | ------------------------
+--cfg=smpi/simulate-computation:no | Yes | No, never
+--cfg=smpi/cpu-threshold:42 | Yes, in all cases | Only if it lasts more than 42 seconds
+SMPI_SAMPLE() macro | Only once per loop nest (see @ref SMPI_adapting_speed "documentation") | Always
+
+\subsection options_model_smpi_adj_file smpi/comp-adjustment-file: Slow-down or speed-up parts of your code.
+
+This option allows you to pass a file that contains two columns: The first column
+defines the section that will be subject to a speedup; the second column is the speedup.
+
+For instance:
+
+\verbatim
+"start:stop","ratio"
+"exchange_1.f:30:exchange_1.f:130",1.18244559422142
+\endverbatim
+
+The first line is the header - you must include it.
+The following 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 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 and end don't have to be
+in the same file), but be aware that this mechanism only supports @em consecutive calls!
+
+\note
+ Please note that you must pass the \b -trace-call-location flag to smpicc
+ or smpiff, respectively! This flag activates some macro definitions in our
+ mpi.h / mpi.f files that help with obtaining the call location.
+
+\subsection options_model_smpi_bw_factor smpi/bw-factor: Bandwidth factors
The possible throughput of network links is often dependent on the
message sizes, as protocols may adapt to different message sizes. With
1. http://simgrid.gforge.inria.fr/contrib/smpi-calibration-doc.html
2. http://simgrid.gforge.inria.fr/contrib/smpi-saturation-doc.html
-\subsection options_smpi_timing smpi/display_timing: Reporting simulation time
+\subsection options_smpi_timing smpi/display-timing: Reporting simulation time
\b Default: 0 (false)
would take to run it on a platform. But since the
code is run through the \c smpirun script, you don't have any control
on the launcher code, making it difficult to report the simulated time
-when the simulation ends. If you set the \b smpi/display_timing item
+when the simulation ends. If you set the \b smpi/display-timing item
to 1, \c smpirun will display this information when the simulation ends. \verbatim
Simulation time: 1e3 seconds.
\endverbatim
-\subsection options_model_smpi_lat_factor smpi/lat_factor: Latency factors
+\subsection options_smpi_temps smpi/keep-temps: not cleaning up after simulation
+
+\b Default: 0 (false)
+
+Under some conditions, SMPI generates a lot of temporary files. They
+usually get cleaned, but you may use this option to not erase these
+files. This is for example useful when debugging or profiling
+executions using the dlopen privatization schema, as missing binary
+files tend to fool the debuggers.
+
+\subsection options_model_smpi_lat_factor smpi/lat-factor: Latency factors
The motivation and syntax for this option is identical to the motivation/syntax
-of smpi/bw_factor, see \ref options_model_smpi_bw_factor for details.
+of smpi/bw-factor, see \ref options_model_smpi_bw_factor for details.
-There is an important difference, though: While smpi/bw_factor \a reduces the
+There is an important difference, though: While smpi/bw-factor \a reduces the
actual bandwidth (i.e., values between 0 and 1 are valid), latency factors
increase the latency, i.e., values larger than or equal to 1 are valid here.
1. http://simgrid.gforge.inria.fr/contrib/smpi-calibration-doc.html
2. http://simgrid.gforge.inria.fr/contrib/smpi-saturation-doc.html
-\subsection options_smpi_global smpi/privatize_global_variables: Automatic privatization of global variables
+\subsection options_smpi_papi_events smpi/papi-events: Trace hardware counters with PAPI
-MPI executables are meant to be executed in separated processes, but SMPI is
-executed in only one process. Global variables from executables will be placed
-in the same memory zone and shared between processes, causing hard to find bugs.
-To avoid this, several options are possible :
- - Manual edition of the code, for example to add __thread keyword before data
- declaration, which allows the resulting code to work with SMPI, but only
- if the thread factory (see \ref options_virt_factory) is used, as global
- variables are then placed in the TLS (thread local storage) segment.
- - Source-to-source transformation, to add a level of indirection
- to the global variables. SMPI does this for F77 codes compiled with smpiff,
- and used to provide coccinelle scripts for C codes, which are not functional anymore.
- - Compilation pass, to have the compiler automatically put the data in
- an adapted zone.
- - Runtime automatic switching of the data segments. SMPI stores a copy of
- each global data segment for each process, and at each context switch replaces
- the actual data with its copy from the right process. This mechanism uses mmap,
- and is for now limited to systems supporting this functionnality (all Linux
- and some BSD should be compatible).
- Another limitation is that SMPI only accounts for global variables defined in
- the executable. If the processes use external global variables from dynamic
- libraries, they won't be switched correctly. To avoid this, using static
- linking is advised (but not with the simgrid library, to avoid replicating
- its own global variables).
-
- To use this runtime automatic switching, the variable \b smpi/privatize_global_variables
- should be set to yes
+\warning
+ This option is experimental and will be subject to change.
+ This feature currently requires superuser privileges, as registers are queried.
+ Only use this feature with code you trust! Call smpirun for instance via
+ smpirun -wrapper "sudo " <your-parameters>
+ or run sudo sh -c "echo 0 > /proc/sys/kernel/perf_event_paranoid"
+ In the later case, sudo will not be required.
+
+\note
+ This option is only available when SimGrid was compiled with PAPI support.
+
+This option takes the names of PAPI counters and adds their respective values
+to the trace files. (See Section \ref tracing_tracing_options.)
+
+It is planned to make this feature available on a per-process (or per-thread?) basis.
+The first draft, however, just implements a "global" (i.e., for all processes) set
+of counters, the "default" set.
+
+\verbatim
+--cfg=smpi/papi-events:"default:PAPI_L3_LDM:PAPI_L2_LDM"
+\endverbatim
+\subsection options_smpi_privatization smpi/privatization: Automatic privatization of global variables
+MPI executables are usually meant to be executed in separated processes, but SMPI is
+executed in only one process. Global variables from executables will be placed
+in the same memory zone and shared between processes, causing intricate bugs.
+Several options are possible to avoid this, as described in the main
+<a href="https://hal.inria.fr/hal-01415484">SMPI publication</a>.
+SimGrid provides two ways of automatically privatizing the globals,
+and this option allows to choose between them.
+
+ - <b>no</b> (default): Do not automatically privatize variables.
+ - <b>mmap</b> or <b>yes</b>: Runtime automatic switching of the data segments.\n
+ SMPI stores a copy of each global data segment for each process,
+ and at each context switch replaces the actual data with its copy
+ from the right process. No copy actually occures as this mechanism
+ uses mmap for efficiency. As such, it is for now limited to
+ systems supporting this functionnality (all Linux and most BSD).\n
+ Another limitation is that SMPI only accounts for global variables
+ defined in the executable. If the processes use external global
+ variables from dynamic libraries, they won't be switched
+ correctly. The easiest way to solve this is to statically link
+ against the library with these globals (but you should never
+ statically link against the simgrid library itself).
+ - <b>dlopen</b>: Link multiple times against the binary.\n
+ SMPI loads several copy of the same binary in memory, resulting in
+ the natural duplication global variables. Since the dynamic linker
+ refuses to link the same file several times, the binary is copied
+ in a temporary file before being dl-loaded (it is erased right
+ after loading).\n
+ Note that this feature is somewhat experimental at time of writing
+ (v3.16) but seems to work.\n
+ This approach greatly speeds up the context switching, down to
+ about 40 CPU cycles with our raw contextes, instead of requesting
+ several syscalls with the \c mmap approach. Another advantage is
+ that it permits to run the SMPI contexts in parallel, which is
+ obviously not possible with the \c mmap approach.\n
+ Further work may be possible to alleviate the memory and disk
+ overconsumption. It seems that we could
+ <a href="https://lwn.net/Articles/415889/">punch holes</a>
+ in the files before dl-loading them to remove the code and
+ constants, and mmap these area onto a unique copy. This require
+ to understand the ELF layout of the file, but would
+ reduce the disk- and memory- usage to the bare minimum. In
+ addition, this would reduce the pressure on the CPU caches (in
+ particular on instruction one).
+
+\warning
+ This configuration option cannot be set in your platform file. You can only
+ pass it as an argument to smpirun.
\subsection options_model_smpi_detached Simulating MPI detached send
immediately. This is different from the threshold detailed in \ref options_model_network_asyncsend
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 operation. This threshold can be set
-by changing the \b smpi/send_is_detached item. The default value is 65536.
+by changing the \b smpi/send-is-detached-thresh item. The default value is 65536.
\subsection options_model_smpi_collectives Simulating MPI collective algorithms
SMPI implements more than 100 different algorithms for MPI collective communication, to accurately
-simulate the behavior of most of the existing MPI libraries. The \b smpi/coll_selector item can be used
+simulate the behavior of most of the existing MPI libraries. The \b smpi/coll-selector item can be used
to use the decision logic of either OpenMPI or MPICH libraries (values: ompi or mpich, by default SMPI
uses naive version of collective operations). Each collective operation can be manually selected with a
-\b smpi/collective_name:algo_name. Available algorithms are listed in \ref SMPI_collective_algorithms .
+\b smpi/collective_name:algo_name. Available algorithms are listed in \ref SMPI_use_colls .
\subsection options_model_smpi_iprobe smpi/iprobe: Inject constant times for calls to MPI_Iprobe
The behavior and motivation for this configuration option is identical with \a smpi/test, see
Section \ref options_model_smpi_test for details.
+\subsection options_model_smpi_iprobe_cpu_usage smpi/iprobe-cpu-usage: Reduce speed for iprobe calls
+
+\b Default value: 1 (no change from default behavior)
+
+MPI_Iprobe calls can be heavily used in applications. To account correctly for the energy
+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 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.
+
+\subsection options_model_smpi_init smpi/init: Inject constant times for calls to MPI_Init
+
+\b Default value: 0
+
+The behavior for this configuration option is identical with \a smpi/test, see
+Section \ref options_model_smpi_test for details.
+
\subsection options_model_smpi_ois smpi/ois: Inject constant times for asynchronous send operations
This configuration option works exactly as \a smpi/os, see Section \ref options_model_smpi_os.
5+11*1
\endverbatim
-%As 5 is the startup cost and 1 is the cost per byte.
+As 5 is the startup cost and 1 is the cost per byte.
\note
The order of sections can be arbitrary; they will be ordered internally.
Internally, in order 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 use counter*SLEEP_TIME, that is, the time MPI_Test() causes the process
- to sleep increases linearly with the number of previously failed testk.
+ to sleep increases linearly with the number of previously failed tests. This
+ behavior can be disabled by setting smpi/grow-injected-times to no. This will
+ also disable this behavior for MPI_Iprobe.
+
+
+\subsection options_model_smpi_shared_malloc smpi/shared-malloc: Factorize malloc()s
+
+\b Default: global
+
+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.
+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
+the motivation for that feature, please refer to the
+<a href="https://simgrid.github.io/SMPI_CourseWare/topic_understanding_performance/matrixmultiplication/">relevant
+section</a> of the SMPI CourseWare (see Activity #2.2 of the pointed
+assignment). In practice, change the call to malloc() and free() into
+SMPI_SHARED_MALLOC() and SMPI_SHARED_FREE().
+
+SMPI provides 2 algorithms for this feature. The first one, called \c
+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
+to create a new POSIX shared memory object (kept in RAM, in /dev/shm)
+for each shared bloc.
+
+With the \c global algorithm, each call to SMPI_SHARED_MALLOC()
+returns a new adress, 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
+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
+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.
+
+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).
+
+As an example,
+
+\code{.C}
+ mem = SMPI_PARTIAL_SHARED_MALLOC(500, {27,42 , 100,200}, 2);
+\endcode
+will allocate 500 bytes to mem, such that mem[27..41] and mem[100..199]
+are shared and other area remain private.
-\subsection options_model_smpi_use_shared_malloc smpi/use_shared_malloc: Use shared memory
+Then, it can be deallocated by calling SMPI_SHARED_FREE(mem).
-\b Default: 1
+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.
+To activate this, you must mount a hugetlbfs on your system and allocate
+at least one huge page:
-SMPI can use shared memory by calling shm_* functions; this might speed up the simulation.
-This opens or creates a new POSIX shared memory object, kept in RAM, in /dev/shm.
+\code{.sh}
+ mkdir /home/huge
+ sudo mount none /home/huge -t hugetlbfs -o rw,mode=0777
+ sudo sh -c 'echo 1 > /proc/sys/vm/nr_hugepages' # echo more if you need more
+\endcode
-If you want to disable this behavior, set the value to 0.
+Then, you can pass the option --cfg=smpi/shared-malloc-hugepage:/home/huge
+to smpirun to actually activate the huge page support in shared mallocs.
\subsection options_model_smpi_wtime smpi/wtime: Inject constant times for calls to MPI_Wtime
The C / C++ standard contains a function called \b [atexit](http://www.cplusplus.com/reference/cstdlib/atexit/).
atexit registers callbacks, which are called just before the program terminates.
-By setting the configuration option clean_atexit to 1 (true), a callback
+By setting the configuration option clean-atexit to 1 (true), a callback
is registered and will clean up some variables and terminate/cleanup the tracing.
TODO: Add when this should be used.
-\subsection options_generic_path XML file inclusion path
+\subsection options_generic_path Profile files' search path
It is possible to specify a list of directories to search into for the
-\<include\> tag in XML files by using the \b path configuration
+trace files (see @ref pf_trace) by using the \b path configuration
item. To add several directory to the path, set the configuration
item several times, as in \verbatim
--cfg=path:toto --cfg=path:tutu
\subsection options_exception_cutpath Truncate local path from exception backtrace
-<b>This configuration option is an internal option and should normally not be used
-by the user.</b> It is used to remove the path from the backtrace
-shown when an exception is thrown; if we didn't remove this part, the tests
-testing the exception parts of simgrid would fail on most machines, as we are
-currently comparing output. Clearly, the path used on different machines are almost
-guaranteed to be different and hence, the output would
-mismatch, causing the test to fail.
+\verbatim
+--cfg=exception/cutpath:1
+\endverbatim
+
+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, and
+thus failing as we are currently comparing output. Clearly, the path
+used on different machines are almost guaranteed to be different and
+hence, the output would mismatch, causing the test to fail.
\section options_log Logging Configuration
It can be done by using XBT. Go to \ref XBT_log for more details.
-\section options_perf Performance optimizations
-
-\subsection options_perf_context Context factory
-
-In order to achieve higher performance, you might want to use the raw
-context factory which avoids any system call when switching between
-tasks. If it is not possible you might use ucontext instead.
-
-\subsection options_perf_guard_size Disabling stack guard pages
-
-A stack guard page is usually used which prevents the stack from
-overflowing on other parts of the memory. However this might have a
-performance impact if a huge number of processes is created. The
-option \b contexts:guard_size is the number of stack guard pages
-used. By setting it to 0, no guard pages will be used: in this case,
-you should avoid using small stacks (\b stack_size) as the stack will
-silently overflow on other parts of the memory.
-
-\section options_index Index of all existing configuration options
-
-\note
- Almost all options are defined in <i>src/simgrid/sg_config.c</i>. You may
- want to check this file, too, but this index should be somewhat complete
- for the moment (May 2015).
-
-\note
- \b Please \b note: You can also pass the command-line option "--help" and
- "--help-cfg" to an executable that uses simgrid.
-
-- \c clean_atexit: \ref options_generic_clean_atexit
-
-- \c contexts/factory: \ref options_virt_factory
-- \c contexts/guard_size: \ref options_virt_parallel
-- \c contexts/nthreads: \ref options_virt_parallel
-- \c contexts/parallel_threshold: \ref options_virt_parallel
-- \c contexts/stack_size: \ref options_virt_stacksize
-- \c contexts/synchro: \ref options_virt_parallel
-
-- \c cpu/maxmin_selective_update: \ref options_model_optim
-- \c cpu/model: \ref options_model_select
-- \c cpu/optim: \ref options_model_optim
-
-- \c exception/cutpath: \ref options_exception_cutpath
-
-- \c host/model: \ref options_model_select
-
-- \c maxmin/precision: \ref options_model_precision
-
-- \c msg/debug_multiple_use: \ref options_msg_debug_multiple_use
-
-- \c model-check: \ref options_modelchecking
-- \c model-check/checkpoint: \ref options_modelchecking_steps
-- \c model-check/communications_determinism: \ref options_modelchecking_comm_determinism
-- \c model-check/dot_output: \ref options_modelchecking_dot_output
-- \c model-check/hash: \ref options_modelchecking_hash
-- \c model-check/property: \ref options_modelchecking_liveness
-- \c model-check/max_depth: \ref options_modelchecking_max_depth
-- \c model-check/record: \ref options_modelchecking_recordreplay
-- \c model-check/reduction: \ref options_modelchecking_reduction
-- \c model-check/replay: \ref options_modelchecking_recordreplay
-- \c model-check/send_determinism: \ref options_modelchecking_comm_determinism
-- \c model-check/sparse-checkpoint: \ref options_modelchecking_sparse_checkpoint
-- \c model-check/termination: \ref options_modelchecking_termination
-- \c model-check/timeout: \ref options_modelchecking_timeout
-- \c model-check/visited: \ref options_modelchecking_visited
-
-- \c network/bandwidth_factor: \ref options_model_network_coefs
-- \c network/coordinates: \ref options_model_network_coord
-- \c network/crosstraffic: \ref options_model_network_crosstraffic
-- \c network/latency_factor: \ref options_model_network_coefs
-- \c network/maxmin_selective_update: \ref options_model_optim
-- \c network/model: \ref options_model_select
-- \c network/optim: \ref options_model_optim
-- \c network/sender_gap: \ref options_model_network_sendergap
-- \c network/TCP_gamma: \ref options_model_network_gamma
-- \c network/weight_S: \ref options_model_network_coefs
-
-- \c ns3/TcpModel: \ref options_pls
-- \c path: \ref options_generic_path
-- \c plugin: \ref options_generic_plugin
-
-- \c surf/nthreads: \ref options_model_nthreads
-- \c surf/precision: \ref options_model_precision
-
-- \c <b>For collective operations of SMPI, please refer to Section \ref options_index_smpi_coll</b>
-- \c smpi/async_small_thresh: \ref options_model_network_asyncsend
-- \c smpi/bw_factor: \ref options_model_smpi_bw_factor
-- \c smpi/coll_selector: \ref options_model_smpi_collectives
-- \c smpi/cpu_threshold: \ref options_smpi_bench
-- \c smpi/display_timing: \ref options_smpi_timing
-- \c smpi/lat_factor: \ref options_model_smpi_lat_factor
-- \c smpi/IB_penalty_factors: \ref options_model_network_coefs
-- \c smpi/iprobe: \ref options_model_smpi_iprobe
-- \c smpi/ois: \ref options_model_smpi_ois
-- \c smpi/or: \ref options_model_smpi_or
-- \c smpi/os: \ref options_model_smpi_os
-- \c smpi/privatize_global_variables: \ref options_smpi_global
-- \c smpi/running_power: \ref options_smpi_bench
-- \c smpi/send_is_detached_thresh: \ref options_model_smpi_detached
-- \c smpi/simulate_computation: \ref options_smpi_bench
-- \c smpi/test: \ref options_model_smpi_test
-- \c smpi/use_shared_malloc: \ref options_model_smpi_use_shared_malloc
-- \c smpi/wtime: \ref options_model_smpi_wtime
-
-- \c <b>Tracing configuration options can be found in Section \ref tracing_tracing_options</b>.
-
-- \c storage/model: \ref options_storage_model
-- \c verbose-exit: \ref options_generic_exit
-
-- \c vm/model: \ref options_vm_model
-
-\subsection options_index_smpi_coll Index of SMPI collective algorithms options
-
-TODO: All available collective algorithms will be made available via the ``smpirun --help-coll`` command.
-
*/
