1 /*! \page faq Frequently Asked Questions
3 \htmlinclude .FAQ.doc.toc
5 \section faq_simgrid I'm new to SimGrid. I have some questions. Where should I start?
7 You are at the right place... Having a look to these
8 <a href="http://www.loria.fr/~quinson/articles/simgrid-tutorial.pdf">the tutorial slides</a>
9 (or to these <a href="http://graal.ens-lyon.fr/~alegrand/articles/slides_g5k_simul.pdf">old slides</a>,
11 <a href="http://graal.ens-lyon.fr/~alegrand/articles/Simgrid-Introduction.pdf">"obsolete" slides</a>)
12 may give you some insights on what SimGrid can help you to do and what
13 are its limitations. Then you definitely should read the \ref
14 MSG_examples. The \ref GRAS_tut can also help you.
16 If you are stuck at any point and if this FAQ cannot help you, please drop us a
17 mail to the user mailing list: <simgrid-user@lists.gforge.inria.fr>.
19 \subsection faq_interfaces What is the difference between MSG, SimDag, and GRAS? Do they serve the same purpose?
21 It depend on how you define "purpose", I guess ;)
23 They all allow you to build a prototype of application which you can run
24 within the simulator afterward. They all share the same simulation kernel,
25 which is the core of the SimGrid project. They differ by the way you express
28 With SimDag, you express your code as a collection of interdependent
29 parallel tasks. So, in this model, applications can be seen as a DAG of
30 tasks. This is the interface of choice for people wanting to port old
31 code designed for SimGrid v1 or v2 to the framework current version.
33 With both GRAS and MSG, your application is seen as a set of communicating
34 processes, exchanging data by the way of messages and performing computation
37 The difference between both is that MSG is somehow easier to use, but GRAS
38 is not limited to the simulator. Once you're done writing your GRAS code,
39 you can run your code both in the simulator or on a real platform. For this,
40 there is two implementations of the GRAS interface, one for simulation, one
41 for real execution. So, you just have to relink your code to chose one of
44 \subsection faq_generic First steps with SimGrid
46 If you decide to go for the MSG interface, please read carefully the
47 \ref MSG_examples. You'll find in \ref MSG_ex_master_slave a very
48 simple consisting of a master (that owns a bunch of tasks and
49 distributes them) , some slaves (that process tasks whenever they
50 receive one) and some forwarder agents (that simply pass the tasks
51 they receive to some slaves).
53 If you decide to go for the GRAS interface, you should definitively
54 read the \ref GRAS_tut. The first section constitutes an introduction
55 to the tool and presents the model we use. The second section
56 constitutes a complete step-by-step tutorial building a distributed
57 application from the beginning and exemplifying most of the GRAS
58 features in the process. The last section groups some HOWTOS
59 highlighting a given feature of the framework in a more concise way.
61 If you decide to go for another interface, I'm afraid your only sources
62 of information will be the source code and the mailing lists...
64 \subsection faq_visualization Visualizing and analyzing the results
66 It is sometime convenient to "see" how the agents are behaving. If you
67 like colors, you can use <tt>tools/MSG_visualization/colorize.pl </tt>
68 as a filter to your MSG outputs. It works directly with INFO. Beware,
69 INFO() prints on stderr. Do not forget to redirect if you want to
70 filter (e.g. with bash):
72 ./msg_test small_platform.xml small_deployment.xml 2>&1 | ../../tools/MSG_visualization/colorize.pl
75 We also have a more graphical output. Have a look at MSG_paje_output(). It
76 generates an input to <a href="http://www-id.imag.fr/Logiciels/paje/">Paje</a>.
79 <a href="Paje_MSG_screenshot.jpg"><img src="Paje_MSG_screenshot_thn.jpg"></a>
83 Visualization with Paje can be seen as a kind of postmortem
84 analysis. However, as soon as you start playing with big simulations,
85 you'll realize that processing such output is kind of tricky. There is
86 so much generic information that it is hard to find the information
89 As a matter of fact, logging really depends on simulations (e.g. what
90 kind of events is important...). That is why we do not propose a big
91 dump of your whole simulation (it would slow everything down) but give
92 you neat tools to structure you logs. Have a look at \ref XBT_log. In
93 fact, rather than a post-mortem analysis, you may want to do it on the
94 fly. The process you are running can do whatever you want. Have you
95 thought about adding a global structure where you directly compute the
96 information that are really important rather than writing everything
97 down and then processing huge files?
99 \subsection faq_C Argh! Do I really have to code in C?
101 Up until now, there is no binding for other languages. If you use C++,
102 you should be able to use the SimGrid library as a standard C library
103 and everything should work fine (simply <i>link</i> against this
104 library; recompiling SimGrid with a C++ compiler won't work and it
105 wouldn't help if you could).
107 In fact, we are currently working on Java bindings of MSG to allow
108 all the undergrad students of the world to use this tool. This is a
109 little more tricky than I would have expected, but the work is moving
110 fast forward [2006/05/13]. More languages are evaluated, but for now,
111 we do not feel a real demand for any other language. Please speak up!
113 \section faq_installation Installing the SimGrid library
115 Many people have been asking me questions on how to use SimGrid. Quite
116 often, the questions were not really about SimGrid but on the
117 installation process. This section is intended to help people that are
118 not familiar with compiling C files under UNIX. If you follow these
119 instructions and still have some troubles, drop an e-mail to
120 <simgrid-user@lists.gforge.inria.fr>.
122 \subsection faq_compiling Compiling SimGrid from a stable archive
124 First of all, you need to download the latest version of SimGrid from
125 <a href="http://gforge.inria.fr/frs/?group_id=12">here</a>.
126 Suppose you have uncompressed SimGrid in some temporary location of
127 your home directory (say <tt>/home/joe/tmp/simgrid-3.0.1 </tt>). The
128 simplest way to use SimGrid is to install it in your home
129 directory. Change your directory to
130 <tt>/home/joe/tmp/simgrid-3.0.1</tt> and type
133 ./configure --prefix=$HOME
138 If at some point, something fails, check the section \ref faq_trouble_compil .
139 If it does not help, you can report this problem to the
140 list but, please, avoid sending a laconic mail like "There is a problem. Is it
141 okay?". Send the config.log file which is automatically generated by
142 configure. Try to capture both the standard output and the error output of the
143 <tt>make</tt> command with <tt>script</tt>. There is no way for us to help you
144 without the relevant bits of information.
146 Now, the following directory should have been created :
148 \li <tt>/home/joe/doc/simgrid/html/</tt>
149 \li <tt>/home/joe/lib/</tt>
150 \li <tt>/home/joe/include/</tt>
152 SimGrid is not a binary, it is a library. Both a static and a dynamic
153 version are available. Here is what you can find if you try a <tt>ls
156 \verbatim libsimgrid.a libsimgrid.la libsimgrid.so libsimgrid.so.0 libsimgrid.so.0.0.1
159 Thus, there is two ways to link your program with SimGrid:
160 \li Either you use the static version, e.g
161 \verbatim gcc libsimgrid.a -o MainProgram MainProgram.c
163 In this case, all the SimGrid functions are directly
164 included in <tt>MainProgram</tt> (hence a bigger binary).
165 \li Either you use the dynamic version (the preferred method)
166 \verbatim gcc -lsimgrid -o MainProgram MainProgram.c
168 In this case, the SimGrid functions are not included in
169 <tt>MainProgram</tt> and you need to set your environment
170 variable in such a way that <tt>libsimgrid.so</tt> will be
171 found at runtime. This can be done by adding the following
172 line in your .bashrc (if you use bash and if you have
173 installed the SimGrid libraries in your home directory):
174 \verbatim export LD_LIBRARY_PATH=$HOME/lib/:$LD_LIBRARY_PATH
177 \subsection faq_compiling_java Java bindings don't get compiled
179 The configure script detects automatically whether you have the
180 softwares needed to use the Java bindings or not. At the end of the
181 configure, you can see the configuration picked by the script, which
182 should look similar to
183 \verbatim Configuration of package simgrid' (version 3.3.4-svn) on
186 Compiler: gcc (version: )
188 CFlags: -O3 -finline-functions -funroll-loops -fno-strict-aliasing -Wall -Wunused -Wmissing-prototypes -Wmissing-declarations -Wpointer-arith -Wchar-subscripts -Wcomment -Wformat -Wwrite-strings -Wno-unused-function -Wno-unused-parameter -Wno-strict-aliasing -Wno-format-nonliteral -Werror -g3
192 Context backend: ucontext
199 In this example, Java backends won't be compiled.
201 On Debian-like systems (which includes ubuntu), you need the following
202 packages: sun-java6-jdk libgcj10-dev. If you cannot find the
203 libgcj10-dev, try another version, like libgcj9-dev (on Ubuntu before
204 9.10) or libgcj11-dev (not released yet, but certainly one day).
205 Please note that you need to activate the contrib and non-free
206 repositories in Debian, and the universe ones in Ubuntu. Java comes at
209 \subsection faq_compiling_snapshoot SimGrid development snapshots
211 We have very high standards on software quality, and we are reluctant releasing
212 a stable release as long as there is still some known bug in the code base. In
213 addition, we added quite an extensive test base, making sure that we correctly
214 test the most important parts of the tool.
216 As an unfortunate conclusion, there may be some time between the stable
217 releases. If you want to benefit from the most recent features we introduced,
218 but don't want to take the risk of an untested version from the SVN, then
219 development snapshots are done for you.
221 These are pre-releases of SimGrid that still fail some tests about features
222 that almost nobody use, or on platforms not being in our core target (which is
223 Linux, Mac, other Unixes and Windows, from the most important to the less
224 one). That means that using this development releases should be safe for most
227 These archives can be found on
228 <a href="http://www.loria.fr/~quinson/simgrid.html">this web page</a>. Once you
229 got the lastest archive, you can compile it just like any archive (see above).
231 \subsection faq_compiling_svn Compiling SimGrid from the SVN
233 The project development takes place in the SVN, where all changes are
234 committed when they happen. Then every once in a while, we make sure that the
235 code quality meets our standard and release an archive from the code in the
236 SVN. We afterward go back to the development in the SVN. So, if you need a
237 recently added feature and can afford some little problem with the stability
238 of the lastest features, you may want to use the SVN version instead of a
241 For that, you first need to get the "simgrid" module from
242 <a href="http://gforge.inria.fr/scm/?group_id=12">here</a>.
244 You won't find any <tt>configure</tt> and a few other things
245 (<tt>Makefile.in</tt>'s, documentation, ...) will be missing as well. The
246 reason for that is that all these files have to be regenerated using the
247 latest versions of <tt>autoconf</tt>, <tt>libtool</tt>, <tt>automake</tt>
248 (>1.9) and <tt>doxygen</tt> (>1.4). To generate the <tt>configure</tt> and
249 the <tt>Makefile.in</tt>'s, you just have to launch the <tt>bootstrap</tt>
250 command that resides in the top of the source tree. Then just follow the
251 instructions of Section \ref faq_compiling.
253 We insist on the fact that you really need the latest versions of
254 autoconf, automake and libtool. Doing this step on exotic architectures/systems
255 (i.e. anything different from a recent linux distribution) may be
256 ... uncertain. If you need to compile the SVN version on a machine where all these
257 dependencies are not met, the easiest is to do <tt>make dist</tt> in the SVN
258 directory of another machine where all dependencies are met. It will create an
259 archive you may deploy on other sites just as a regular stable release.
261 In summary, the following commands will checkout the SVN, regenerate the
262 configure script and friends, configure SimGrid and build it.
264 \verbatim svn checkout svn://scm.gforge.inria.fr/svn/simgrid/simgrid/trunk simgrid
267 ./configure --enable-maintainer-mode --prefix=<where to install SimGrid>
270 Then, if you want to install SimGrid on the current box, just do:
271 \verbatim make install \endverbatim
273 If you want to build an snapshot of the SVN to deploy it on another box (for
274 example because the other machine don't have the autotools), do:
275 \verbatim make dist \endverbatim
277 Moreover, you should never call the autotools manually since you must run
278 them in a specific order with specific arguments. Most of the times, the
279 makefiles will automatically call the tools for you. When it's not possible
280 (such as the first time you checkout the SVN), use the ./bootstrap command
281 to call them explicitly.
284 \subsection faq_setting_MSG Setting up your own MSG code
286 Do not build your simulator by modifying the SimGrid examples. Go
287 outside the SimGrid source tree and create your own working directory
288 (say <tt>/home/joe/SimGrid/MyFirstScheduler/</tt>).
290 Suppose your simulation has the following structure (remember it is
291 just an example to illustrate a possible way to compile everything;
292 feel free to organize it as you want).
294 \li <tt>sched.h</tt>: a description of the core of the
295 scheduler (i.e. which functions are can be used by the
296 agents). For example we could find the following functions
297 (master, forwarder, slave).
299 \li <tt>sched.c</tt>: a C file including <tt>sched.h</tt> and
300 implementing the core of the scheduler. Most of these
301 functions use the MSG functions defined in section \ref
304 \li <tt>masterslave.c</tt>: a C file with the main function, i.e.
305 the MSG initialization (MSG_global_init()), the platform
306 creation (e.g. with MSG_create_environment()), the
307 deployment phase (e.g. with MSG_function_register() and
308 MSG_launch_application()) and the call to
311 To compile such a program, we suggest to use the following
312 Makefile. It is a generic Makefile that we have used many times with
313 our students when we teach the C language.
317 masterslave: masterslave.o sched.o
319 INSTALL_PATH = $$HOME
321 PEDANTIC_PARANOID_FREAK = -O0 -Wshadow -Wcast-align \
322 -Waggregate-return -Wmissing-prototypes -Wmissing-declarations \
323 -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations \
324 -Wmissing-noreturn -Wredundant-decls -Wnested-externs \
325 -Wpointer-arith -Wwrite-strings -finline-functions
326 REASONABLY_CAREFUL_DUDE = -Wall
327 NO_PRAYER_FOR_THE_WICKED = -w -O2
328 WARNINGS = $(REASONABLY_CAREFUL_DUDE)
329 CFLAGS = -g $(WARNINGS)
331 INCLUDES = -I$(INSTALL_PATH)/include
332 DEFS = -L$(INSTALL_PATH)/lib/
333 LDADD = -lm -lsimgrid
337 $(CC) $(INCLUDES) $(DEFS) $(CFLAGS) $^ $(LIBS) $(LDADD) -o $@
340 $(CC) $(INCLUDES) $(DEFS) $(CFLAGS) -c -o $@ $<
343 rm -f $(BIN_FILES) *.o *~
349 The first two lines indicates what should be build when typing make
350 (<tt>masterslave</tt>) and of which files it is to be made of
351 (<tt>masterslave.o</tt> and <tt>sched.o</tt>). This makefile assumes
352 that you have set up correctly your <tt>LD_LIBRARY_PATH</tt> variable
353 (look, there is a <tt>LDADD = -lm -lsimgrid</tt>). If you prefer using
354 the static version, remove the <tt>-lsimgrid</tt> and add a
355 <tt>$(INSTALL_PATH)/lib/libsimgrid.a</tt> on the next line, right
356 after the <tt>LIBS = </tt>.
358 More generally, if you have never written a Makefile by yourself, type
359 in a terminal : <tt>info make</tt> and read the introduction. The
360 previous example should be enough for a first try but you may want to
361 perform some more complex compilations...
363 \subsection faq_setting_GRAS Setting up your own GRAS code
365 If you use the GRAS interface instead of the MSG one, then previous section
366 is not the better source of information. Instead, you should check the GRAS
367 tutorial in general, and the \ref GRAS_tut_tour_setup in particular.
369 \section faq_cmake CMAKE
371 \subsection faq_intro Some generalitty
375 CMake is a family of tools designed to build, test and package software. CMake is used to control the software compilation process using simple platform and compiler independent configuration files. CMake generates native makefiles and workspaces that can be used in the compiler environment of your choice. For more information see official web site <a href="http://www.cmake.org/">here</a>.
379 CMake permits to developers to compil projects on different plateform. Then many tools are embedded like ctest for making test, a link to cdash for vizualise results but also test coverage and bug reports.
383 CMake needs some prerequists like :
385 \li a c, c++ and java compiler regards to developers
386 \li ccmake for graphical used of CMake
387 \li cmake <a href="http://www.cmake.org/cmake/resources/software.html">download</a>
389 4 Cmake vs Autotools...
393 \subsection faq_cmakeoption Cmake options
398 "cmake -D[name]=[value] ... ./"
400 [name] disable_gtnets [value] ON/OFF or TRUE/FALSE or 1/0
401 disable_java ON/OFF or TRUE/FALSE or 1/0
402 disable_lua ON/OFF or TRUE/FALSE or 1/0
403 disable_ruby ON/OFF or TRUE/FALSE or 1/0
405 enable_compile_optimizations ON/OFF or TRUE/FALSE or 1/0
406 enable_compile_warnings ON/OFF or TRUE/FALSE or 1/0
407 enable_maintainer_mode ON/OFF or TRUE/FALSE or 1/0
409 supernovae ON/OFF or TRUE/FALSE or 1/0
411 gtnets_path <path_to_gtnets_directory>
412 prefix <path_to_install_directory>
413 with_context auto/ucontext/pthread/window
416 2 Options explaination
418 \li disable_gtnets : set to true implie that user doesn't want to use gtnets.
420 \li disable_java : set to true implie that user doesn't want to add java langage into simgrid compilation.
422 \li disable_lua : set to true implie that user doesn't want to add lua langage into simgrid compilation.
424 \li disable_ruby : set to true implie that user doesn't want to add ruby langage into simgrid compilation.
426 \li enable_compile_optimizations : add flags "-O3 -finline-functions -funroll-loops -fno-strict-aliasing"
428 \li enable_compile_warnings : add flags "-Wall -Wunused -Wmissing-prototypes -Wmissing-declarations -Wpointer-arith -Wchar-subscripts -Wcomment -Wformat -Wwrite-strings -Wno-unused-function -Wno-unused-parameter -Wno-strict-aliasing -Wno-format-nonliteral -Werror"
430 \li enable_maintainer_mode : set to true make doc and remake files with flex flexml.
433 /src/gras/DataDesc/ddt_parse.yy.c
434 /src/surf/simgrid_dtd.c
436 /src/simdag/dax_dtd.c
437 /include/surf/simgrid_dtd.h
438 /include/xbt/graphxml.h
439 /src/simdag/dax_dtd.h
441 \li supernovae : set to true make one file for each lib and compile with those generated files.
444 /src/supernovae_gras.c
445 /src/supernovae_smpi.c
447 \li gtnets_path : Path to gtnets install directory (ex /usr)
449 \li prefix : Path where are installed lib/ doc/ and include/ directories (ex /usr/local)
451 \li with context : specify which context the user wants to use.
455 Those options are initialized the first time you launch \"cmake ./\" whithout specified option.
459 enable_maintainer_mode off
463 enable_compile_warnings off
464 enable_compile_optimizations off
469 4 Option's cache and how to reset?
471 When options have been set they are keep into a cache file named \"CMakeCache.txt\". So if you want
472 reset values you just delete this file located to the project directory.
474 \subsection faq_cmakecompilation Cmake compilation
479 cmake -D[name]=[value] ... ./
488 Then follow instructions.
490 3 Resume of command line
494 cmake ./ configure the project
495 make build all tagets
496 make test test all targets and summarize
497 make package make the distrib
498 make install-simgrid install the project (doc/ lib/ include/)
499 make clean" clean all targets
502 When the project have been succesfully compiling and build you can make tests.
506 ctest launch only tests
508 ctest -D Continuous(Start|Update|Configure|Build)
509 ctest -D Continuous(Test|Coverage|MemCheck|Submit)
510 ctest -D Experimental
511 ctest -D Experimental(Start|Update|Configure|Build)
512 ctest -D Experimental(Test|Coverage|MemCheck|Submit)
514 ctest -D Nightly(Start|Update|Configure|Build)
515 ctest -D Nightly(Test|Coverage|MemCheck|Submit)
516 ctest -D NightlyMemoryCheck
519 4 Examples for different mode.
523 cmake -Denable_maintainer_mode=on ./
524 \verbatim GTnetS doesn't works : set -Ddisable_gtnets=on
525 with_context auto change to ucontext
528 Make : src/simgrid.jar with : /usr/bin/javac
529 Make examples/java with : /usr/bin/javac
531 Configuration of package `simgrid' (revision 7228M) on arch (=4):
533 SITE : Linux_Ubuntu 9.10_x86_64
535 Compiler: c++ : /usr/bin/c++
536 version: c++ (Ubuntu 4.4.1-4ubuntu9) 4.4.1
538 version: gcc (Ubuntu 4.4.1-4ubuntu9) 4.4.1
544 Context backend: ucontext
554 Simgrid dependencies: dl -llua5.1
555 Gras dependencies: pthread
559 USER_PREFIX: /usr/local
560 INSTALL_PREFIX: /usr/local
564 -- Build files have been written to: /home/navarrop/Bureau/simgrid-trunk
569 cmake -Dsupernovae=on ./
570 \verbatim GTnetS doesn't works : set -Ddisable_gtnets=on
571 with_context auto change to ucontext
574 Make : src/simgrid.jar with : /usr/bin/javac
575 Make examples/java with : /usr/bin/javac
577 Configuration of package `simgrid' (revision 7228M) on arch (=4):
578 BUILDNAME : SUPERNOVAE
579 SITE : Linux_Ubuntu 9.10_x86_64
581 Compiler: c++ : /usr/bin/c++
582 version: c++ (Ubuntu 4.4.1-4ubuntu9) 4.4.1
584 version: gcc (Ubuntu 4.4.1-4ubuntu9) 4.4.1
586 CFlags: -O3 -finline-functions -funroll-loops -fno-strict-aliasing -Wall -Wunused -Wmissing-prototypes -Wmissing-declarations -Wpointer-arith -Wchar-subscripts -Wcomment -Wformat -Wwrite-strings -Wno-unused-function -Wno-unused-parameter -Wno-strict-aliasing -Wno-format-nonliteral -Werror -g3
590 Context backend: ucontext
600 Simgrid dependencies: dl -llua5.1
601 Gras dependencies: pthread
605 USER_PREFIX: /usr/local
606 INSTALL_PREFIX: /usr/local
610 -- Build files have been written to: /home/navarrop/Bureau/simgrid-trunk
615 cmake -Dgtnets_path=/home/navarrop/Bureau/usr/ ./
616 \verbatim with_context auto change to ucontext
619 Make : src/simgrid.jar with : /usr/bin/javac
620 Make examples/java with : /usr/bin/javac
622 Configuration of package `simgrid' (revision 7228M) on arch (=4):
624 SITE : Linux_Ubuntu 9.10_x86_64
626 Compiler: c++ : /usr/bin/c++
627 version: c++ (Ubuntu 4.4.1-4ubuntu9) 4.4.1
629 version: gcc (Ubuntu 4.4.1-4ubuntu9) 4.4.1
631 CFlags: -L/home/navarrop/Bureau/usr/lib -I/home/navarrop/Bureau/usr/include/gtnets -g3
632 CPPFlags: -L/home/navarrop/Bureau/usr/lib -I/home/navarrop/Bureau/usr/include/gtnets
635 Context backend: ucontext
637 path: /home/navarrop/Bureau/usr
645 Simgrid dependencies: dl -llua5.1 -lgtnets
646 Gras dependencies: pthread
650 USER_PREFIX: /usr/local
651 INSTALL_PREFIX: /usr/local
653 INFO -->> Take care to have export LD_LIBRARY_PATH before run make command for make examples with gtnets
654 copy and paste : export LD_LIBRARY_PATH=/home/navarrop/Bureau/usr/lib/:$LD_LIBRARY_PATH
659 -- Build files have been written to: /home/navarrop/Bureau/simgrid-trunk
662 \subsection faq_cmakeinstall How to install with cmake?
667 cmake -Denable_maintainer_mode=on -Dprefix=/home/navarrop/Bureau/install_simgrid ./
675 cmake -Dprefix=/home/navarrop/Bureau/install_simgrid ./
680 \subsection faq_screenshot Screenshot
685 navarrop@caraja:~$ cd Bureau/simgrid-trunk/
686 navarrop@caraja:~/Bureau/simgrid-trunk$ cmake ./
688 GTnetS doesn't works : set -Ddisable_gtnets=on <-|some warnings are printed
689 with_context auto change to ucontext <-|
690 (skaddr) <--info (needed)
691 (sksize) <--info (needed)
692 Make : src/simgrid.jar with : /usr/bin/javac <--info (if java)
693 Make examples/java with : /usr/bin/javac <--info (if java)
695 Configuration of package `simgrid' (revision 7209M) on arch (=4):
696 BUILDNAME : UCONTEXT <-- name of the compilation regarding to cdash
697 SITE : Linux_Ubuntu 9.10_x86_64 <-- distribution of the local machine regarding to cdash
699 Compiler: c++ : /usr/bin/c++
700 version: c++ (Ubuntu 4.4.1-4ubuntu9) 4.4.1
702 version: gcc (Ubuntu 4.4.1-4ubuntu9) 4.4.1
708 Context backend: ucontext
718 Simgrid dependencies: -ldl -llua5.1
719 Gras dependencies: pthread
723 USER_PREFIX: /usr/local
724 INSTALL_PREFIX: /usr/local
728 -- Build files have been written to: /home/navarrop/Bureau/simgrid-trunk
730 Here all options are checked and printed. If it doesn't match with your configuration
731 it is probably due to a wrong configuration.
733 \subsection faq_cmakehowto How to modified sources files for developers
735 1 Add an executable or examples.
737 If you want make an executable you have to create a CMakeList.txt to the src directory.
738 You must specified where to create the executable, source list, dependencies and the name of the binary.
741 cmake_minimum_required(VERSION 2.6)
743 set(EXECUTABLE_OUTPUT_PATH "./")
744 set(LIBRARY_OUTPUT_PATH "${PROJECT_DIRECTORY}/lib")
746 add_executable(get_sender get_sender.c) #add_executable(<name_of_target> <src list>)
748 ### Add definitions for compile
749 target_link_libraries(get_sender simgrid m pthread -fprofile-arcs) #target_link_libraries(<name_of_targe> <dependencies>)
752 Then you have to modified <project/directory>/buildtools/Cmake/src/CMakeMakeExeLib.txt and add
755 add_subdirectory(${PROJECT_DIRECTORY}/<path_where_is_CMakeList.txt>)
758 2 Delete/add sources to lib.
760 If you want modified, add or delete source files from a library you have to edit <project/directory>/buildtools/Cmake/src/CMakeDefinePackages.txt
764 ${PROJECT_DIRECTORY}/src/java/simgrid/msg/MsgException.java
765 ${PROJECT_DIRECTORY}/src/java/simgrid/msg/JniException.java
766 ${PROJECT_DIRECTORY}/src/java/simgrid/msg/NativeException.java
767 ${PROJECT_DIRECTORY}/src/java/simgrid/msg/HostNotFoundException.java
768 ${PROJECT_DIRECTORY}/src/java/simgrid/msg/ProcessNotFoundException.java
769 ${PROJECT_DIRECTORY}/src/java/simgrid/msg/Msg.java
770 ${PROJECT_DIRECTORY}/src/java/simgrid/msg/Process.java
771 ${PROJECT_DIRECTORY}/src/java/simgrid/msg/Host.java
772 ${PROJECT_DIRECTORY}/src/java/simgrid/msg/Task.java
773 ${PROJECT_DIRECTORY}/src/java/simgrid/msg/MsgNative.java
774 ${PROJECT_DIRECTORY}/src/java/simgrid/msg/ApplicationHandler.java
775 ${PROJECT_DIRECTORY}/src/java/simgrid/msg/Sem.java
781 If you want modified, add or delete tests you have to edit <project/directory>/buildtools/Cmake/src/CMakeTest.txt
782 with this function : ADD_TEST(<name> <bin> <ARGS>)
785 add_test(test-simdag-1 ${PROJECT_DIRECTORY}/testsuite/simdag/sd_test --cfg=path:${PROJECT_DIRECTORY}/testsuite/simdag small_platform_variable.xml)
788 \subsection faq_cmakeExplain Explaination of sources files for cmake
791 \li CMakeCompleteInFiles.txt
793 \li CMakeMakeExeLib.txt
794 \li CMakePrintArgs.txt
795 \li CMakeDefinePackages.txt
797 \li CMakeSupernovae.txt
802 \li CTestConfig.cmake
803 \li CTestTestfile.cmake
805 Here is a list of files involved into cmake build (relative to project directory path) :
809 ./buildtools/Cmake/src/CMakeCompleteInFiles.txt
810 ./buildtools/Cmake/src/CMakeDocs.txt
811 ./buildtools/Cmake/src/CMakeMakeExeLib.txt
812 ./buildtools/Cmake/src/CMakePrintArgs.txt
813 ./buildtools/Cmake/src/CMakeDefinePackages.txt
814 ./buildtools/Cmake/src/CMakeFlags.txt
815 ./buildtools/Cmake/src/CMakeSupernovae.txt
816 ./buildtools/Cmake/src/CMakeDistrib.txt
817 ./buildtools/Cmake/src/CMakeFLEXml.txt
818 ./buildtools/Cmake/src/CMakeOption.txt
819 ./buildtools/Cmake/src/CMakeTest.txt
820 ./buildtools/Cmake/src/CTestConfig.cmake
821 ./buildtools/Cmake/src/CTestTestfile.cmake
823 Test files for define properties :
824 ./buildtools/Cmake/prog_test/prog_GRAS_ARCH.c
825 ./buildtools/Cmake/prog_test/prog_max_size.c
826 ./buildtools/Cmake/prog_test/prog_sem_init.c
827 ./buildtools/Cmake/prog_test/prog_stackgrowth.c
828 ./buildtools/Cmake/prog_test/prog_vsnprintf.c
829 ./buildtools/Cmake/prog_test/prog_AC_CHECK_MCSC.c
830 ./buildtools/Cmake/prog_test/prog_GRAS_CHECK_STRUCT_COMPACTION.c
831 ./buildtools/Cmake/prog_test/prog_mutex_timedlock.c
832 ./buildtools/Cmake/prog_test/prog_sem_timedwait.c
833 ./buildtools/Cmake/prog_test/prog_stacksetup.c
834 ./buildtools/Cmake/prog_test/prog_getline.c
835 ./buildtools/Cmake/prog_test/prog_gtnets.cpp
836 ./buildtools/Cmake/prog_test/prog_printf_null.c
837 ./buildtools/Cmake/prog_test/prog_snprintf.c
838 ./buildtools/Cmake/prog_test/prog_va_copy.c
840 CMakeLists for each binaries or examples:
842 ./teshsuite/gras/empty_main/CMakeLists.txt
843 ./teshsuite/gras/small_sleep/CMakeLists.txt
844 ./teshsuite/gras/datadesc/CMakeLists.txt
845 ./teshsuite/gras/msg_handle/CMakeLists.txt
846 ./teshsuite/simdag/CMakeLists.txt
847 ./teshsuite/simdag/partask/CMakeLists.txt
848 ./teshsuite/simdag/platforms/CMakeLists.txt
849 ./teshsuite/simdag/network/CMakeLists.txt
850 ./teshsuite/simdag/network/mxn/CMakeLists.txt
851 ./teshsuite/simdag/network/p2p/CMakeLists.txt
852 ./teshsuite/xbt/CMakeLists.txt
853 ./teshsuite/msg/CMakeLists.txt
854 ./tools/gras/CMakeLists.txt
855 ./tools/tesh/CMakeLists.txt
856 ./testsuite/simdag/CMakeLists.txt
857 ./testsuite/xbt/CMakeLists.txt
858 ./testsuite/surf/CMakeLists.txt
859 ./examples/gras/properties/CMakeLists.txt
860 ./examples/gras/ping/CMakeLists.txt
861 ./examples/gras/pmm/CMakeLists.txt
862 ./examples/gras/mmrpc/CMakeLists.txt
863 ./examples/gras/synchro/CMakeLists.txt
864 ./examples/gras/timer/CMakeLists.txt
865 ./examples/gras/mutual_exclusion/simple_token/CMakeLists.txt
866 ./examples/gras/spawn/CMakeLists.txt
867 ./examples/gras/chrono/CMakeLists.txt
868 ./examples/gras/rpc/CMakeLists.txt
869 ./examples/gras/all2all/CMakeLists.txt
870 ./examples/simdag/properties/CMakeLists.txt
871 ./examples/simdag/CMakeLists.txt
872 ./examples/simdag/metaxml/CMakeLists.txt
873 ./examples/simdag/dax/CMakeLists.txt
874 ./examples/smpi/CMakeLists.txt
875 ./examples/amok/bandwidth/CMakeLists.txt
876 ./examples/amok/saturate/CMakeLists.txt
877 ./examples/msg/priority/CMakeLists.txt
878 ./examples/msg/properties/CMakeLists.txt
879 ./examples/msg/migration/CMakeLists.txt
880 ./examples/msg/gtnets/CMakeLists.txt
881 ./examples/msg/parallel_task/CMakeLists.txt
882 ./examples/msg/trace/CMakeLists.txt
883 ./examples/msg/suspend/CMakeLists.txt
884 ./examples/msg/masterslave/CMakeLists.txt
885 ./examples/msg/actions/CMakeLists.txt
886 ./examples/msg/sendrecv/CMakeLists.txt
889 \section faq_howto Feature related questions
891 \subsection faq_MIA "Could you please add (your favorite feature here) to SimGrid?"
893 Here is the deal. The whole SimGrid project (MSG, SURF, GRAS, ...) is
894 meant to be kept as simple and generic as possible. We cannot add
895 functions for everybody's needs when these functions can easily be
896 built from the ones already in the API. Most of the time, it is
897 possible and when it was not possible we always have upgraded the API
898 accordingly. When somebody asks us a question like "How to do that?
899 Is there a function in the API to simply do this?", we're always glad
900 to answer and help. However if we don't need this code for our own
901 need, there is no chance we're going to write it... it's your job! :)
902 The counterpart to our answers is that once you come up with a neat
903 implementation of this feature (task duplication, RPC, thread
904 synchronization, ...), you should send it to us and we will be glad to
905 add it to the distribution. Thus, other people will take advantage of
906 it (and we don't have to answer this question again and again ;).
908 You'll find in this section a few "Missing In Action" features. Many
909 people have asked about it and we have given hints on how to simply do
910 it with MSG. Feel free to contribute...
912 \subsection faq_MIA_MSG MSG features
914 \subsubsection faq_MIA_examples I want some more complex MSG examples!
916 Many people have come to ask me a more complex example and each time,
917 they have realized afterward that the basics were in the previous three
920 Of course they have often been needing more complex functions like
921 MSG_process_suspend(), MSG_process_resume() and
922 MSG_process_isSuspended() (to perform synchronization), or
923 MSG_task_Iprobe() and MSG_process_sleep() (to avoid blocking
924 receptions), or even MSG_process_create() (to design asynchronous
925 communications or computations). But the examples are sufficient to
928 We know. We should add some more examples, but not really some more
929 complex ones... We should add some examples that illustrate some other
930 functionalists (like how to simply encode asynchronous
931 communications, RPC, process migrations, thread synchronization, ...)
932 and we will do it when we will have a little bit more time. We have
933 tried to document the examples so that they are understandable. Tell
934 us if something is not clear and once again feel free to participate!
937 \subsubsection faq_MIA_taskdup Missing in action: MSG Task duplication/replication
939 There is no task duplication in MSG. When you create a task, you can
940 process it or send it somewhere else. As soon as a process has sent
941 this task, he doesn't have this task anymore. It's gone. The receiver
942 process has got the task. However, you could decide upon receiving to
943 create a "copy" of a task but you have to handle by yourself the
944 semantic associated to this "duplication".
946 As we already told, we prefer keeping the API as simple as
947 possible. This kind of feature is rather easy to implement by users
948 and the semantic you associate really depends on people. Having a
949 *generic* task duplication mechanism is not that trivial (in
950 particular because of the data field). That is why I would recommand
951 that you write it by yourself even if I can give you advice on how to
954 You have the following functions to get informations about a task:
955 MSG_task_get_name(), MSG_task_get_compute_duration(),
956 MSG_task_get_remaining_computation(), MSG_task_get_data_size(),
957 and MSG_task_get_data().
959 You could use a dictionary (#xbt_dict_t) of dynars (#xbt_dynar_t). If
960 you still don't see how to do it, please come back to us...
962 \subsubsection faq_MIA_asynchronous I want to do asynchronous communications in MSG
964 Up until now, there is no asynchronous communications in MSG. However,
965 you can create as many process as you want so you should be able to do
966 whatever you want... I've written a queue module to help implementing
967 some asynchronous communications at low cost (creating thousands of
968 process only to handle communications may be problematic in term of
969 performance at some point). I'll add it in the distribution asap.
971 \subsubsection faq_MIA_thread_synchronization I need to synchronize my MSG processes
973 You obviously cannot use pthread_mutexes of pthread_conds. The best
974 thing would be to propose similar structures. Unfortunately, we
975 haven't found time to do it yet. However you can try to play with
976 MSG_process_suspend() and MSG_process_resume(). You can even do some
977 synchronization with fake communications (using MSG_task_get(),
978 MSG_task_put() and MSG_task_Iprobe()).
980 \subsubsection faq_MIA_host_load Where is the get_host_load function hidden in MSG?
982 There is no such thing because its semantic wouldn't be really
983 clear. Of course, it is something about the amount of host throughput,
984 but there is as many definition of "host load" as people asking for
985 this function. First, you have to remember that resource availability
986 may vary over time, which make any load notion harder to define.
988 It may be instantaneous value or an average one. Moreover it may be only the
989 power of the computer, or may take the background load into account, or may
990 even take the currently running tasks into account. In some SURF models,
991 communications have an influence on computational power. Should it be taken
994 First of all, it's near to impossible to predict the load beforehands in the
995 simulator since it depends on too much parameters (background load
996 variation, bandwidth sharing algorithmic complexity) some of them even being
997 not known beforehands (other task starting at the same time). So, getting
998 this information is really hard (just like in real life). It's not just that
999 we want MSG to be as painful as real life. But as it is in some way
1000 realistic, we face some of the same problems as we would face in real life.
1002 How would you do it for real? The most common option is to use something
1003 like NWS that performs active probes. The best solution is probably to do
1004 the same within MSG, as in next code snippet. It is very close from what you
1005 would have to do out of the simulator, and thus gives you information that
1006 you could also get in real settings to not hinder the realism of your
1010 double get_host_load() {
1011 m_task_t task = MSG_task_create("test", 0.001, 0, NULL);
1012 double date = MSG_get_clock();
1014 MSG_task_execute(task);
1015 date = MSG_get_clock() - date;
1016 MSG_task_destroy(task);
1017 return (0.001/date);
1021 Of course, it may not match your personal definition of "host load". In this
1022 case, please detail what you mean on the mailing list, and we will extend
1023 this FAQ section to fit your taste if possible.
1025 \subsubsection faq_MIA_communication_time How can I get the *real* communication time?
1027 Communications are synchronous and thus if you simply get the time
1028 before and after a communication, you'll only get the transmission
1029 time and the time spent to really communicate (it will also take into
1030 account the time spent waiting for the other party to be
1031 ready). However, getting the *real* communication time is not really
1032 hard either. The following solution is a good starting point.
1037 m_task_t task = MSG_task_create("Task", task_comp_size, task_comm_size,
1038 calloc(1,sizeof(double)));
1039 *((double*) task->data) = MSG_get_clock();
1040 MSG_task_put(task, slaves[i % slaves_count], PORT_22);
1041 INFO0("Send completed");
1046 m_task_t task = NULL;
1049 time1 = MSG_get_clock();
1050 a = MSG_task_get(&(task), PORT_22);
1051 time2 = MSG_get_clock();
1052 if(time1<*((double *)task->data))
1053 time1 = *((double *) task->data);
1054 INFO1("Communication time : \"%f\" ", time2-time1);
1056 MSG_task_destroy(task);
1061 \subsection faq_MIA_SimDag SimDag related questions
1063 \subsubsection faq_SG_comm Implementing communication delays between tasks.
1065 A classic question of SimDag newcomers is about how to express a
1066 communication delay between tasks. The thing is that in SimDag, both
1067 computation and communication are seen as tasks. So, if you want to
1068 model a data dependency between two DAG tasks t1 and t2, you have to
1069 create 3 SD_tasks: t1, t2 and c and add dependencies in the following
1073 SD_task_dependency_add(NULL, NULL, t1, c);
1074 SD_task_dependency_add(NULL, NULL, c, t2);
1077 This way task t2 cannot start before the termination of communication c
1078 which in turn cannot start before t1 ends.
1080 When creating task c, you have to associate an amount of data (in bytes)
1081 corresponding to what has to be sent by t1 to t2.
1083 Finally to schedule the communication task c, you have to build a list
1084 comprising the workstations on which t1 and t2 are scheduled (w1 and w2
1085 for example) and build a communication matrix that should look like
1088 \subsubsection faq_SG_DAG How to implement a distributed dynamic scheduler of DAGs.
1090 Distributed is somehow "contagious". If you start making distributed
1091 decisions, there is no way to handle DAGs directly anymore (unless I
1092 am missing something). You have to encode your DAGs in term of
1093 communicating process to make the whole scheduling process
1094 distributed. Here is an example of how you could do that. Assume T1
1095 has to be done before T2.
1098 int your_agent(int argc, char *argv[] {
1100 T1 = MSG_task_create(...);
1101 T2 = MSG_task_create(...);
1105 if(cond) MSG_task_execute(T1);
1107 if((MSG_task_get_remaining_computation(T1)=0.0) && (you_re_in_a_good_mood))
1108 MSG_task_execute(T2)
1110 /* do something else */
1116 If you decide that the distributed part is not that much important and that
1117 DAG is really the level of abstraction you want to work with, then you should
1118 give a try to \ref SD_API.
1120 \subsection faq_MIA_generic Generic features
1122 \subsubsection faq_more_processes Increasing the amount of simulated processes
1124 Here are a few tricks you can apply if you want to increase the amount
1125 of processes in your simulations.
1127 - <b>A few thousands of simulated processes</b> (soft tricks)\n
1128 SimGrid can use either pthreads library or the UNIX98 contextes. On
1129 most systems, the number of pthreads is limited and then your
1130 simulation may be limited for a stupid reason. This is especially
1131 true with the current linux pthreads, and I cannot get more than
1132 2000 simulated processes with pthreads on my box. The UNIX98
1133 contexts allow me to raise the limit to 25,000 simulated processes
1135 The <tt>--with-context</tt> option of the <tt>./configure</tt>
1136 script allows you to choose between UNIX98 contextes
1137 (<tt>--with-context=ucontext</tt>) and the pthread version
1138 (<tt>--with-context=pthread</tt>). The default value is ucontext
1139 when the script detect a working UNIX98 context implementation. On
1140 Windows boxes, the provided value is discarded and an adapted
1141 version is picked up.\n\n
1142 We experienced some issues with contextes on some rare systems
1143 (solaris 8 and lower or old alpha linuxes comes to mind). The main
1144 problem is that the configure script detect the contextes as being
1145 functional when it's not true. If you happen to use such a system,
1146 switch manually to the pthread version, and provide us with a good
1147 patch for the configure script so that it is done automatically ;)
1149 - <b>Hundred thousands of simulated processes</b> (hard-core tricks)\n
1150 As explained above, SimGrid can use UNIX98 contextes to represent
1151 and handle the simulated processes. Thanks to this, the main
1152 limitation to the number of simulated processes becomes the
1153 available memory.\n\n
1154 Here are some tricks I had to use in order to run a token ring
1155 between 25,000 processes on my laptop (1Gb memory, 1.5Gb swap).\n
1156 - First of all, make sure your code runs for a few hundreds
1157 processes before trying to push the limit. Make sure it's
1158 valgrind-clean, ie that valgrind does not report neither memory
1159 error nor memory leaks. Indeed, numerous simulated processes
1160 result in *fat* simulation hindering debugging.
1161 - It was really boring to write 25,000 entries in the deployment
1162 file, so I wrote a little script
1163 <tt>examples/gras/mutual_exclusion/simple_token/make_deployment.pl</tt>, which you may
1164 want to adapt to your case. You could also think about hijacking
1165 the SURFXML parser (have look at \ref faq_flexml_bypassing).
1166 - The deployment file became quite big, so I had to do what is in
1167 the FAQ entry \ref faq_flexml_limit
1168 - Each UNIX98 context has its own stack entry. As debugging this is
1169 quite hairly, the default value is a bit overestimated so that
1170 user don't get into trouble about this. You want to tune this
1171 size to increse the number of processes. This is the
1172 <tt>STACK_SIZE</tt> define in
1173 <tt>src/xbt/xbt_context_sysv.c</tt>, which is 128kb by default.
1174 Reduce this as much as you can, but be warned that if this value
1175 is too low, you'll get a segfault. The token ring example, which
1176 is quite simple, runs with 40kb stacks.
1177 - You may tweak the logs to reduce the stack size further. When
1178 logging something, we try to build the string to display in a
1179 char array on the stack. The size of this array is constant (and
1180 equal to XBT_LOG_BUFF_SIZE, defined in include/xbt/log/h). If the
1181 string is too large to fit this buffer, we move to a dynamically
1182 sized buffer. In which case, we have to traverse one time the log
1183 event arguments to compute the size we need for the buffer,
1184 malloc it, and traverse the argument list again to do the actual
1186 The idea here is to move XBT_LOG_BUFF_SIZE to 1, forcing the logs
1187 to use a dynamic array each time. This allows us to lower further
1188 the stack size at the price of some performance loss...\n
1189 This allowed me to run the reduce the stack size to ... 4k. Ie,
1190 on my 1Gb laptop, I can run more than 250,000 processes!
1192 \subsubsection faq_MIA_batch_scheduler Is there a native support for batch schedulers in SimGrid?
1194 No, there is no native support for batch schedulers and none is
1195 planned because this is a very specific need (and doing it in a
1196 generic way is thus very hard). However some people have implemented
1197 their own batch schedulers. Vincent Garonne wrote one during his PhD
1198 and put his code in the contrib directory of our SVN so that other can
1199 keep working on it. You may find inspiring ideas in it.
1201 \subsubsection faq_MIA_checkpointing I need a checkpointing thing
1203 Actually, it depends on whether you want to checkpoint the simulation, or to
1204 simulate checkpoints.
1206 The first one could help if your simulation is a long standing process you
1207 want to keep running even on hardware issues. It could also help to
1208 <i>rewind</i> the simulation by jumping sometimes on an old checkpoint to
1209 cancel recent calculations.\n
1210 Unfortunately, such thing will probably never exist in SG. One would have to
1211 duplicate all data structures because doing a rewind at the simulator level
1212 is very very hard (not talking about the malloc free operations that might
1213 have been done in between). Instead, you may be interested in the Libckpt
1214 library (http://www.cs.utk.edu/~plank/plank/www/libckpt.html). This is the
1215 checkpointing solution used in the condor project, for example. It makes it
1216 easy to create checkpoints (at the OS level, creating something like core
1217 files), and rerunning them on need.
1219 If you want to simulate checkpoints instead, it means that you want the
1220 state of an executing task (in particular, the progress made towards
1221 completion) to be saved somewhere. So if a host (and the task executing on
1222 it) fails (cf. #MSG_HOST_FAILURE), then the task can be restarted
1223 from the last checkpoint.\n
1225 Actually, such a thing does not exists in SimGrid either, but it's just
1226 because we don't think it is fundamental and it may be done in the user code
1227 at relatively low cost. You could for example use a watcher that
1228 periodically get the remaining amount of things to do (using
1229 MSG_task_get_remaining_computation()), or fragment the task in smaller
1232 \subsection faq_platform Platform building and Dynamic resources
1234 \subsubsection faq_platform_example Where can I find SimGrid platform files?
1236 There is several little examples in the archive, in the examples/msg
1237 directory. From time to time, we are asked for other files, but we
1238 don't have much at hand right now.
1240 You should refer to the Platform Description Archive
1241 (http://pda.gforge.inria.fr) project to see the other platform file we
1242 have available, as well as the Simulacrum simulator, meant to generate
1243 SimGrid platforms using all classical generation algorithms.
1245 \subsubsection faq_platform_alnem How can I automatically map an existing platform?
1247 We are working on a project called ALNeM (Application-Level Network
1248 Mapper) which goal is to automatically discover the topology of an
1249 existing network. Its output will be a platform description file
1250 following the SimGrid syntax, so everybody will get the ability to map
1251 their own lab network (and contribute them to the catalog project).
1252 This tool is not ready yet, but it move quite fast forward. Just stay
1255 \subsubsection faq_platform_synthetic Generating synthetic but realistic platforms
1257 The third possibility to get a platform file (after manual or
1258 automatic mapping of real platforms) is to generate synthetic
1259 platforms. Getting a realistic result is not a trivial task, and
1260 moreover, nobody is really able to define what "realistic" means when
1261 speaking of topology files. You can find some more thoughts on this
1263 <a href="http://graal.ens-lyon.fr/~alegrand/articles/Simgrid-Introduction.pdf">slides</a>.
1265 If you are looking for an actual tool, there we have a little tool to
1266 annotate Tiers-generated topologies. This perl-script is in
1267 <tt>tools/platform_generation/</tt> directory of the SVN. Dinda et Al.
1268 released a very comparable tool, and called it GridG.
1270 \subsubsection faq_SURF_dynamic Expressing dynamic resource availability in platform files
1272 A nice feature of SimGrid is that it enables you to seamlessly have
1273 resources whose availability change over time. When you build a
1274 platform, you generally declare hosts like that:
1277 <host id="host A" power="100.00"/>
1280 If you want the availability of "host A" to change over time, the only
1281 thing you have to do is change this definition like that:
1284 <host id="host A" power="100.00" availability_file="trace_A.txt" state_file="trace_A_failure.txt"/>
1287 For hosts, availability files are expressed in fraction of available
1288 power. Let's have a look at what "trace_A.txt" may look like:
1297 At time 0, our host will deliver 100 flop/s. At time 11.0, it will
1298 deliver only 50 flop/s until time 20.0 where it will will start
1299 delivering 90 flop/s. Last at time 21.0 (20.0 plus the periodicity
1300 1.0), we'll be back to the beginning and it will deliver 100 flop/s.
1302 Now let's look at the state file:
1309 A negative value means "off" while a positive one means "on". At time
1310 1.0, the host is on. At time 1.0, it is turned off and at time 2.0, it
1311 is turned on again until time 12 (2.0 plus the periodicity 10.0). It
1312 will be turned on again at time 13.0 until time 23.0, and so on.
1314 Now, let's look how the same kind of thing can be done for network
1315 links. A usual declaration looks like:
1318 <link id="LinkA" bandwidth="10.0" latency="0.2"/>
1321 You have at your disposal the following options: bandwidth_file,
1322 latency_file and state_file. The only difference with hosts is that
1323 bandwidth_file and latency_file do not express fraction of available
1324 power but are expressed directly in bytes per seconds and seconds.
1326 \subsubsection faq_platform_multipath How to express multipath routing in platform files?
1328 It is unfortunately impossible to express the fact that there is more
1329 than one routing path between two given hosts. Let's consider the
1330 following platform file:
1333 <route src="A" dst="B">
1336 <route src="B" dst="C">
1339 <route src="A" dst="C">
1344 Although it is perfectly valid, it does not mean that data traveling
1345 from A to C can either go directly (using link 3) or through B (using
1346 links 1 and 2). It simply means that the routing on the graph is not
1347 trivial, and that data do not following the shortest path in number of
1348 hops on this graph. Another way to say it is that there is no implicit
1349 in these routing descriptions. The system will only use the routes you
1350 declare (such as <route src="A" dst="C"><link:ctn
1351 id="3"/></route>), without trying to build new routes by aggregating
1354 You are also free to declare platform where the routing is not
1355 symmetric. For example, add the following to the previous file:
1358 <route src="C" dst="A">
1364 This makes sure that data from C to A go through B where data from A
1365 to C go directly. Don't worry about realism of such settings since
1366 we've seen ways more weird situation in real settings (in fact, that's
1367 the realism of very regular platforms which is questionable, but
1368 that's another story).
1370 \subsubsection faq_flexml_bypassing Bypassing the XML parser with your own C functions
1372 So you want to bypass the XML files parser, uh? Maybe doing some parameter
1373 sweep experiments on your simulations or so? This is possible, and
1374 it's not even really difficult (well. Such a brutal idea could be
1375 harder to implement). Here is how it goes.
1377 For this, you have to first remember that the XML parsing in SimGrid is done
1378 using a tool called FleXML. Given a DTD, this gives a flex-based parser. If
1379 you want to bypass the parser, you need to provide some code mimicking what
1380 it does and replacing it in its interactions with the SURF code. So, let's
1381 have a look at these interactions.
1383 FleXML parser are close to classical SAX parsers. It means that a
1384 well-formed SimGrid platform XML file might result in the following
1387 - start "platform_description" with attribute version="2"
1388 - start "host" with attributes id="host1" power="1.0"
1390 - start "host" with attributes id="host2" power="2.0"
1392 - start "link" with ...
1394 - start "route" with ...
1395 - start "link:ctn" with ...
1398 - end "platform_description"
1400 The communication from the parser to the SURF code uses two means:
1401 Attributes get copied into some global variables, and a surf-provided
1402 function gets called by the parser for each event. For example, the event
1403 - start "host" with attributes id="host1" power="1.0"
1405 let the parser do something roughly equivalent to:
1407 strcpy(A_host_id,"host1");
1412 In SURF, we attach callbacks to the different events by initializing the
1413 pointer functions to some the right surf functions. Since there can be
1414 more than one callback attached to the same event (if more than one
1415 model is in use, for example), they are stored in a dynar. Example in
1416 workstation_ptask_L07.c:
1418 /* Adding callback functions */
1419 surf_parse_reset_parser();
1420 surfxml_add_callback(STag_surfxml_host_cb_list, &parse_cpu_init);
1421 surfxml_add_callback(STag_surfxml_prop_cb_list, &parse_properties);
1422 surfxml_add_callback(STag_surfxml_link_cb_list, &parse_link_init);
1423 surfxml_add_callback(STag_surfxml_route_cb_list, &parse_route_set_endpoints);
1424 surfxml_add_callback(ETag_surfxml_link_c_ctn_cb_list, &parse_route_elem);
1425 surfxml_add_callback(ETag_surfxml_route_cb_list, &parse_route_set_route);
1427 /* Parse the file */
1428 surf_parse_open(file);
1429 xbt_assert1((!surf_parse()), "Parse error in %s", file);
1433 So, to bypass the FleXML parser, you need to write your own version of the
1434 surf_parse function, which should do the following:
1435 - Fill the A_<tag>_<attribute> variables with the wanted values
1436 - Call the corresponding STag_<tag>_fun function to simulate tag start
1437 - Call the corresponding ETag_<tag>_fun function to simulate tag end
1438 - (do the same for the next set of values, and loop)
1440 Then, tell SimGrid that you want to use your own "parser" instead of the stock one:
1442 surf_parse = surf_parse_bypass_environment;
1443 MSG_create_environment(NULL);
1444 surf_parse = surf_parse_bypass_application;
1445 MSG_launch_application(NULL);
1448 A set of macros are provided at the end of
1449 include/surf/surfxml_parse.h to ease the writing of the bypass
1450 functions. An example of this trick is distributed in the file
1451 examples/msg/masterslave/masterslave_bypass.c
1453 \subsection faq_simgrid_configuration Changing SimGrid's behavior
1455 A number of options can be given at runtime to change the default
1456 SimGrid behavior. In particular, you can change the default cpu and
1459 \subsubsection faq_simgrid_configuration_gtnets Using GTNetS
1461 It is possible to use a packet-level network simulator
1462 instead of the default flow-based simulation. You may want to use such
1463 an approach if you have doubts about the validity of the default model
1464 or if you want to perform some validation experiments. At the moment,
1465 we support the GTNetS simulator (it is still rather experimental
1466 though, so leave us a message if you play with it).
1470 To enable GTNetS model inside SimGrid it is needed to patch the GTNetS simulator source code
1471 and build/install it from scratch
1474 - <b>Download and enter the recent downloaded GTNetS directory</b>
1477 svn checkout svn://scm.gforge.inria.fr/svn/simgrid/contrib/trunk/GTNetS/
1482 - <b>Use the following commands to unzip and patch GTNetS package to work within SimGrid.</b>
1485 unzip gtnets-current.zip
1486 tar zxvf gtnets-current-patch.tgz
1488 cat ../00*.patch | patch -p1
1491 - <b>OPTIONALLY</b> you can use a patch for itanium 64bit processor family.
1494 cat ../AMD64-FATAL-Removed-DUL_SIZE_DIFF-Added-fPIC-compillin.patch | patch -p1
1497 - <b>Compile GTNetS</b>
1499 Due to portability issues it is possible that GTNetS does not compile in your architecture. The patches furnished in SimGrid SVN repository are intended for use in Linux architecture only. Unfortunately, we do not have the time, the money, neither the manpower to guarantee GTNetS portability. We advice you to use one of GTNetS communication channel to get more help in compiling GTNetS.
1503 ln -sf Makefile.linux Makefile
1509 - <b>NOTE</b> A lot of warnings are expected but the application should compile
1510 just fine. If the makefile insists in compiling some QT libraries
1511 please try a make clean before asking for help.
1514 - <b>To compile optimized version</b>
1521 - <b>Installing GTNetS</b>
1523 It is important to put the full path of your libgtsim-xxxx.so file when creating the symbolic link. Replace < userhome > by some path you have write access to.
1526 ln -sf /<absolute_path>/gtnets_current/libgtsim-debug.so /<userhome>/usr/lib/libgtnets.so
1527 export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/<userhome>/usr/lib/libgtnets.so
1528 mkdir /<userhome>/usr/include/gtnets
1529 cp -fr SRC/*.h /<userhome>/usr/include/gtnets
1533 - <b>Enable GTNetS support in SimGrid</b>
1536 ./configure --with-gtnets=/<userhome>/usr
1539 - <b>Once you have followed all the instructions for compiling and
1540 installing successfully you can activate this feature at
1541 runntime with the following options:</b>
1544 cd simgrid/example/msg/
1550 - <b>Or try the GTNetS model dogbone example with</b>
1553 gtnets/gtnets gtnets/onelink-p.xml gtnets/onelink-d.xml --cfg=network_model:GTNets
1557 A long version of this <a href="http://gforge.inria.fr/docman/view.php/12/6283/GTNetS HowTo.html">HowTo</a> it is available
1560 More about GTNetS simulator at <a href="http://www.ece.gatech.edu/research/labs/MANIACS/GTNetS/index.html">GTNetS Website</a>
1564 The patches provided by us worked successfully with GTNetS found
1565 <a href="http://www.ece.gatech.edu/research/labs/MANIACS/GTNetS/software/gtnets-current.zip">here</a>,
1566 dated from 12th June 2008. Due to the discontinuing development of
1567 GTNetS it is impossible to precise a version number. We STRONGLY recommend you
1568 to download and install the GTNetS version found in SimGrid repository as explained above.
1573 \subsubsection faq_simgrid_configuration_alternate_network Using alternative flow models
1575 The default simgrid network model uses a max-min based approach as
1576 explained in the research report
1577 <a href="ftp://ftp.ens-lyon.fr/pub/LIP/Rapports/RR/RR2002/RR2002-40.ps.gz">A Network Model for Simulation of Grid Application</a>.
1578 Other models have been proposed and implemented since then (see for example
1579 <a href="http://mescal.imag.fr/membres/arnaud.legrand/articles/simutools09.pdf">Accuracy Study and Improvement of Network Simulation in the SimGrid Framework</a>)
1580 and can be activated at runtime. For example:
1582 ./mycode platform.xml deployment.xml --cfg=workstation_model:compound --cfg=network_model:LV08 -cfg=cpu_model:Cas01
1585 Possible models for the network are currently "Constant", "CM02",
1586 "LegrandVelho", "GTNets", Reno", "Reno2", "Vegas". Others will
1587 probably be added in the future and many of the previous ones are
1588 experimental and are likely to disappear without notice...
1590 \section faq_troubleshooting Troubleshooting
1592 \subsection faq_trouble_lib_compil SimGrid compilation and installation problems
1594 \subsubsection faq_trouble_lib_config ./configure fails!
1596 We know only one reason for the configure to fail:
1598 - <b>You are using a broken build environment</b>\n
1599 If symptom is that configure complains about gcc not being able to build
1600 executables, you are probably missing the libc6-dev package. Damn Ubuntu.
1602 If you experience other kind of issue, please get in touch with us. We are
1603 always interested in improving our portability to new systems.
1605 \subsubsection faq_trouble_distcheck Dude! "make check" fails on my machine!
1607 Don't assume we never run this target, because we do. Check
1608 http://bob.loria.fr:8010 if you don't believe us.
1610 There is several reasons which may cause the make check to fail on your
1613 - <b>You are using a broken libc (probably concerning the contextes)</b>.\n
1614 The symptom is that the "make check" fails within the examples/msg directory.\n
1615 By default, SimGrid uses something called ucontexts. This is part of the
1616 libc, but it's quite undertested. For example, some (old) versions of the
1617 glibc on alpha do not implement these functions, but provide the stubs
1618 (which return ENOSYS: not implemented). It may fool our detection mechanism
1619 and leads to segfaults. There is not much we can do to fix the bug.
1620 A workaround is to compile with --with-context=pthread to avoid
1621 ucontext completely. You'll be a bit more limited in the number
1622 of simulated processes you can start concurrently, but 5000
1623 processes is still enough for most purposes, isn't it?\n
1624 This limitation is the reason why we insist on using this piece of ...
1625 software even if it's so troublesome.\n
1626 <b>=> use --with-pthread on AMD64 architecture that do not have an
1627 ultra-recent libc.</b>
1629 - <b>There is a bug in SimGrid we aren't aware of</b>.\n
1630 If none of the above apply, please drop us a mail on the mailing list so
1631 that we can check it out. Make sure to read \ref faq_bugrepport
1634 \subsection faq_trouble_compil User code compilation problems
1636 \subsubsection faq_trouble_err_logcat "gcc: _simgrid_this_log_category_does_not_exist__??? undeclared (first use in this function)"
1638 This is because you are using the log mecanism, but you didn't created
1639 any default category in this file. You should refer to \ref XBT_log
1640 for all the details, but you simply forgot to call one of
1641 XBT_LOG_NEW_DEFAULT_CATEGORY() or XBT_LOG_NEW_DEFAULT_SUBCATEGORY().
1643 \subsubsection faq_trouble_pthreadstatic "gcc: undefined reference to pthread_key_create"
1645 This indicates that one of the library SimGrid depends on (libpthread
1646 here) was missing on the linking command line. Dependencies of
1647 libsimgrid are expressed directly in the dynamic library, so it's
1648 quite impossible that you see this message when doing dynamic linking.
1650 If you compile your code statically (and if you use a pthread version
1651 of SimGrid -- see \ref faq_more_processes), you must absolutely
1652 specify <tt>-lpthread</tt> on the linker command line. As usual, this should
1653 come after <tt>-lsimgrid</tt> on this command line.
1655 \subsection faq_trouble_errors Runtime error messages
1657 \subsubsection faq_flexml_limit "surf_parse_lex: Assertion `next limit' failed."
1659 This is because your platform file is too big for the parser.
1661 Actually, the message comes directly from FleXML, the technology on top of
1662 which the parser is built. FleXML has the bad idea of fetching the whole
1663 document in memory before parsing it. And moreover, the memory buffer size
1664 must be determined at compilation time.
1666 We use a value which seems big enough for our need without bloating the
1667 simulators footprints. But of course your mileage may vary. In this case,
1668 just edit src/surf/surfxml.l modify the definition of
1669 FLEXML_BUFFERSTACKSIZE. E.g.
1672 #define FLEXML_BUFFERSTACKSIZE 1000000000
1675 Then recompile and everything should be fine, provided that your version of
1676 Flex is recent enough (>= 2.5.31). If not the compilation process should
1679 A while ago, we worked on FleXML to reduce a bit its memory consumption, but
1680 these issues remain. There is two things we should do:
1682 - use a dynamic buffer instead of a static one so that the only limit
1683 becomes your memory, not a stupid constant fixed at compilation time
1684 (maybe not so difficult).
1685 - change the parser so that it does not need to get the whole file in
1686 memory before parsing
1687 (seems quite difficult, but I'm a complete newbe wrt flex stuff).
1689 These are changes to FleXML itself, not SimGrid. But since we kinda hijacked
1690 the development of FleXML, I can grant you that any patches would be really
1691 welcome and quickly integrated.
1693 <b>Update:</b> A new version of FleXML (1.7) was released. Most of the work
1694 was done by William Dowling, who use it in his own work. The good point is
1695 that it now use a dynamic buffer, and that the memory usage was greatly
1696 improved. The downside is that William also changed some things internally,
1697 and it breaks the hack we devised to bypass the parser, as explained in
1698 \ref faq_flexml_bypassing. Indeed, this is not a classical usage of the
1699 parser, and Will didn't imagine that we may have used (and even documented)
1700 such a crude usage of FleXML. So, we now have to repair the bypassing
1701 functionality to use the lastest FleXML version and fix the memory usage in
1704 \subsubsection faq_trouble_gras_transport GRAS spits networking error messages
1706 Gras, on real platforms, naturally use regular sockets to communicate. They
1707 are deeply hidden in the gras abstraction, but when things go wrong, you may
1708 get some weird error messages. Here are some example, with the probable
1711 - <b>Transport endpoint is not connected</b>: several processes try to open
1712 a server socket on the same port number of the same machine. This is
1713 naturally bad and each process should pick its own port number for this.\n
1714 Maybe, you just have some processes remaining from a previous experiment
1716 Killing them may help, but again if you kill -KILL them, you'll have to
1717 wait for a while: they didn't close there sockets properly and the system
1718 needs a while to notice that this port is free again.
1720 - <b>Socket closed by remote side</b>: if the remote process is not
1721 supposed to close the socket at this point, it may be dead.
1723 - <b>Connection reset by peer</b>: I found this on Internet about this
1724 error. I think it's what's happening here, too:\n
1725 <i>This basically means that a network error occurred while the client was
1726 receiving data from the server. But what is really happening is that the
1727 server actually accepts the connection, processes the request, and sends
1728 a reply to the client. However, when the server closes the socket, the
1729 client believes that the connection has been terminated abnormally
1730 because the socket implementation sends a TCP reset segment telling the
1731 client to throw away the data and report an error.\n
1732 Sometimes, this problem is caused by not properly closing the
1733 input/output streams and the socket connection. Make sure you close the
1734 input/output streams and socket connection properly. If everything is
1735 closed properly, however, and the problem persists, you can work around
1736 it by adding a one-second sleep before closing the streams and the
1737 socket. This technique, however, is not reliable and may not work on all
1739 Since GRAS sockets are closed properly (repeat after me: there is no bug
1740 in GRAS), it is either that you are closing your sockets on server side
1741 before the client get a chance to read them (use gras_os_sleep() to delay
1742 the server), or the server died awfully before the client got the data.
1744 \subsubsection faq_trouble_errors_big_fat_warning I'm told that my XML files are too old.
1746 The format of the XML platform description files is sometimes
1747 improved. For example, we decided to change the units used in SimGrid
1748 from MBytes, MFlops and seconds to Bytes, Flops and seconds to ease
1749 people exchanging small messages. We also reworked the route
1750 descriptions to allow more compact descriptions.
1752 That is why the XML files are versionned using the 'version' attribute
1753 of the root tag. Currently, it should read:
1755 <platform version="2">
1758 If your files are too old, you can use the simgrid_update_xml.pl
1759 script which can be found in the tools directory of the archive.
1761 \subsection faq_trouble_valgrind Valgrind-related and other debugger issues
1763 If you don't, you really should use valgrind to debug your code, it's
1766 \subsubsection faq_trouble_vg_longjmp longjmp madness in valgrind
1768 This is when valgrind starts complaining about longjmp things, just like:
1770 \verbatim ==21434== Conditional jump or move depends on uninitialised value(s)
1771 ==21434== at 0x420DBE5: longjmp (longjmp.c:33)
1773 ==21434== Use of uninitialised value of size 4
1774 ==21434== at 0x420DC3A: __longjmp (__longjmp.S:48)
1777 This is the sign that you didn't used the exception mecanism well. Most
1778 probably, you have a <tt>return;</tt> somewhere within a <tt>TRY{}</tt>
1779 block. This is <b>evil</b>, and you must not do this. Did you read the section
1782 \subsubsection faq_trouble_vg_libc Valgrind spits tons of errors about backtraces!
1784 It may happen that valgrind, the memory debugger beloved by any decent C
1785 programmer, spits tons of warnings like the following :
1786 \verbatim ==8414== Conditional jump or move depends on uninitialised value(s)
1787 ==8414== at 0x400882D: (within /lib/ld-2.3.6.so)
1788 ==8414== by 0x414EDE9: (within /lib/tls/i686/cmov/libc-2.3.6.so)
1789 ==8414== by 0x400B105: (within /lib/ld-2.3.6.so)
1790 ==8414== by 0x414F937: _dl_open (in /lib/tls/i686/cmov/libc-2.3.6.so)
1791 ==8414== by 0x4150F4C: (within /lib/tls/i686/cmov/libc-2.3.6.so)
1792 ==8414== by 0x400B105: (within /lib/ld-2.3.6.so)
1793 ==8414== by 0x415102D: __libc_dlopen_mode (in /lib/tls/i686/cmov/libc-2.3.6.so)
1794 ==8414== by 0x412D6B9: backtrace (in /lib/tls/i686/cmov/libc-2.3.6.so)
1795 ==8414== by 0x8076446: xbt_dictelm_get_ext (dict_elm.c:714)
1796 ==8414== by 0x80764C1: xbt_dictelm_get (dict_elm.c:732)
1797 ==8414== by 0x8079010: xbt_cfg_register (config.c:208)
1798 ==8414== by 0x806821B: MSG_config (msg_config.c:42)
1801 This problem is somewhere in the libc when using the backtraces and there is
1802 very few things we can do ourselves to fix it. Instead, here is how to tell
1803 valgrind to ignore the error. Add the following to your ~/.valgrind.supp (or
1804 create this file on need). Make sure to change the obj line according to
1805 your personnal mileage (change 2.3.6 to the actual version you are using,
1806 which you can retrieve with a simple "ls /lib/ld*.so").
1809 name: Backtrace madness
1811 obj:/lib/ld-2.3.6.so
1816 fun:__libc_dlopen_mode
1819 Then, you have to specify valgrind to use this suppression file by passing
1820 the <tt>--suppressions=$HOME/.valgrind.supp</tt> option on the command line.
1821 You can also add the following to your ~/.bashrc so that it gets passed
1822 automatically. Actually, it passes a bit more options to valgrind, and this
1823 happen to be my personnal settings. Check the valgrind documentation for
1826 \verbatim export VALGRIND_OPTS="--leak-check=yes --leak-resolution=high --num-callers=40 --tool=memcheck --suppressions=$HOME/.valgrind.supp" \endverbatim
1828 \subsubsection faq_trouble_backtraces Truncated backtraces
1830 When debugging SimGrid, it's easier to pass the
1831 --disable-compiler-optimization flag to the configure if valgrind or
1832 gdb get fooled by the optimization done by the compiler. But you
1833 should remove these flag when everything works before going in
1834 production (before launching your 1252135 experiments), or everything
1835 will run only one half of the true SimGrid potential.
1837 \subsection faq_deadlock There is a deadlock in my code!!!
1839 Unfortunately, we cannot debug every code written in SimGrid. We
1840 furthermore believe that the framework provides ways enough
1841 information to debug such informations yourself. If the textual output
1842 is not enough, Make sure to check the \ref faq_visualization FAQ entry to see
1843 how to get a graphical one.
1845 Now, if you come up with a really simple example that deadlocks and
1846 you're absolutely convinced that it should not, you can ask on the
1847 list. Just be aware that you'll be severely punished if the mistake is
1848 on your side... We have plenty of FAQ entries to redact and new
1849 features to implement for the impenitents! ;)
1851 \subsection faq_surf_network_latency I get weird timings when I play with the latencies.
1853 OK, first of all, remember that units should be Bytes, Flops and
1854 Seconds. If you don't use such units, some SimGrid constants (e.g. the
1855 SG_TCP_CTE_GAMMA constant used in most network models) won't have the
1856 right unit and you'll end up with weird results.
1858 Here is what happens with a single transfer of size L on a link
1859 (bw,lat) when nothing else happens.
1862 0-----lat--------------------------------------------------t
1863 |-----|**** real_bw =min(bw,SG_TCP_CTE_GAMMA/(2*lat)) *****|
1866 In more complex situations, this min is the solution of a complex
1867 max-min linear system. Have a look
1868 <a href="http://lists.gforge.inria.fr/pipermail/simgrid-devel/2006-April/thread.html">here</a>
1869 and read the two threads "Bug in SURF?" and "Surf bug not
1870 fixed?". You'll have a few other examples of such computations. You
1871 can also read "A Network Model for Simulation of Grid Application" by
1872 Henri Casanova and Loris Marchal to have all the details. The fact
1873 that the real_bw is smaller than bw is easy to understand. The fact
1874 that real_bw is smaller than SG_TCP_CTE_GAMMA/(2*lat) is due to the
1875 window-based congestion mechanism of TCP. With TCP, you can't exploit
1876 your huge network capacity if you don't have a good round-trip-time
1877 because of the acks...
1879 Anyway, what you get is t=lat + L/min(bw,SG_TCP_CTE_GAMMA/(2*lat)).
1881 * if I you set (bw,lat)=(100 000 000, 0.00001), you get t = 1.00001 (you fully
1883 * if I you set (bw,lat)=(100 000 000, 0.0001), you get t = 1.0001 (you're on the
1885 * if I you set (bw,lat)=(100 000 000, 0.001), you get t = 10.001 (ouch!)
1887 This bound on the effective bandwidth of a flow is not the only thing
1888 that may make your result be unexpected. For example, two flows
1889 competing on a saturated link receive an amount of bandwidth inversely
1890 proportional to their round trip time.
1892 \subsection faq_bugrepport So I've found a bug in SimGrid. How to report it?
1894 We do our best to make sure to hammer away any bugs of SimGrid, but this is
1895 still an academic project so please be patient if/when you find bugs in it.
1896 If you do, the best solution is to drop an email either on the simgrid-user
1897 or the simgrid-devel mailing list and explain us about the issue. You can
1898 also decide to open a formal bug report using the
1899 <a href="https://gforge.inria.fr/tracker/?atid=165&group_id=12&func=browse">relevant
1900 interface</a>. You need to login on the server to get the ability to submit
1903 We will do our best to solve any problem repported, but you need to help us
1904 finding the issue. Just telling "it segfault" isn't enough. Telling "It
1905 segfaults when running the attached simulator" doesn't really help either.
1906 You may find the following article interesting to see how to repport
1907 informative bug repports:
1908 http://www.chiark.greenend.org.uk/~sgtatham/bugs.html (it is not SimGrid
1909 specific at all, but it's full of good advices).
1911 \author Arnaud Legrand (arnaud.legrand::imag.fr)
1912 \author Martin Quinson (martin.quinson::loria.fr)
1917 ******************************************************************
1918 * OLD CRUFT NOT USED ANYMORE *
1919 ******************************************************************
1922 \subsection faq_crosscompile Cross-compiling a Windows DLL of SimGrid from linux
1924 At the moment, we do not distribute Windows pre-compiled version of SimGrid
1925 because the support for this platform is still experimental. We know that
1926 some parts of the GRAS environment do not work, and we think that the others
1927 environments (MSG and SD) have good chances to work, but we didn't test
1928 ourselves. This section explains how we generate the SimGrid DLL so that you
1929 can build it for yourself. First of all, you need to have a version more
1930 recent than 3.1 (ie, a SVN version as time of writting).
1932 In order to cross-compile the package to windows from linux, you need to
1933 install mingw32 (minimalist gnu win32). On Debian, you can do so by
1934 installing the packages mingw32 (compiler), mingw32-binutils (linker and
1935 so), mingw32-runtime.
1937 You can use the VPATH support of configure to compile at the same time for
1938 linux and windows without dupplicating the source nor cleaning the tree
1939 between each. Just run bootstrap (if you use the SVN) to run the autotools.
1940 Then, create a linux and a win directories. Then, type:
1941 \verbatim cd linux; ../configure --srcdir=.. <usual configure flags>; make; cd ..
1942 cd win; ../configure --srcdir=.. --host=i586-mingw32msvc <flags>; make; cd ..
1944 The trick to VPATH builds is to call configure from another directory,
1945 passing it an extra --srcdir argument to tell it where all the sources are.
1946 It will understand you want to use VPATH. Then, the trick to cross-compile
1947 is simply to add a --host argument specifying the target you want to build
1948 for. The i586-mingw32msvc string is what you have to pass to use the mingw32
1949 environment as distributed in Debian.
1951 After that, you can run all make targets from both directories, and test
1952 easily that what you change for one arch does not break the other one.
1954 It is possible that this VPATH build thing breaks from time to time in the
1955 SVN since it's quite fragile, but it's granted to work in any released
1956 version. If you experience problems, drop us a mail.
1958 Another possible source of issue is that at the moment, building the
1959 examples request to use the gras_stub_generator tool, which is a compiled
1960 program, not a script. In cross-compilation, you need to cross-execute with
1961 wine for example, which is not really pleasant. We are working on this, but
1962 in the meanwhile, simply don't build the examples in cross-compilation
1963 (<tt>cd src</tt> before running make).
1965 Program (cross-)compiled with mingw32 do request an extra DLL at run-time to be
1966 usable. For example, if you want to test your build with wine, you should do
1967 the following to put this library where wine looks for DLLs.
1969 cp /usr/share/doc/mingw32-runtime/mingwm10.dll.gz ~/.wine/c/windows/system/
1970 gunzip ~/.wine/c/windows/system/mingwm10.dll.gz
1973 The DLL is built in src/.libs, and installed in the <i>prefix</i>/bin directory
1974 when you run make install.
1976 If you want to use it in a native project on windows, you need to use
1977 simgrid.dll and mingwm10.dll. For each DLL, you need to build .def file
1978 under linux (listing the defined symbols), and convert it into a .lib file
1979 under windows (specifying this in a way that windows compilers like). To
1980 generate the def files, run (under linux):
1981 \verbatim echo "LIBRARY libsimgrid-0.dll" > simgrid.def
1982 echo EXPORTS >> simgrid.def
1983 nm libsimgrid-0.dll | grep ' T _' | sed 's/.* T _//' >> simgrid.def
1984 nm libsimgrid-0.dll | grep ' D _' | sed 's/.* D _//' | sed 's/$/ DATA/' >> simgrid.def
1986 echo "LIBRARY mingwm10.dll" > mingwm10.def
1987 echo EXPORTS >> mingwm10.def
1988 nm mingwm10.dll | grep ' T _' | sed 's/.* T _//' >> mingwm10.def
1989 nm mingwm10.dll | grep ' D _' | sed 's/.* D _//' | sed 's/$/ DATA/' >> mingwm10.def
1992 To create the import .lib files, use the <tt>lib</tt> windows tool (from
1993 MSVC) the following way to produce simgrid.lib and mingwm10.lib
1994 \verbatim lib /def:simgrid.def
1995 lib /def:mingwm10.def
1998 If you happen to use Borland C Builder, the right command line is the
1999 following (note that you don't need any file.def to get this working).
2000 \verbatim implib simgrid.lib libsimgrid-0.dll
2001 implib mingwm10.lib mingwm10.dll
2004 Then, set the following parameters in Visual C++ 2005:
2005 Linker -> Input -> Additional dependencies = simgrid.lib mingwm10.lib
2007 Just in case you wonder how to generate a DLL from libtool in another
2008 project, we added -no-undefined to any lib*_la_LDFLAGS variables so that
2009 libtool accepts to generate a dynamic library under windows. Then, to make
2010 it true, we pass any dependencies (such as -lws2 under windows or -lpthread
2011 on need) on the linking line. Passing such deps is a good idea anyway so
2012 that they get noted in the library itself, avoiding the users to know about
2013 our dependencies and put them manually on their compilation line. Then we
2014 added the AC_LIBTOOL_WIN32_DLL macro just before AC_PROG_LIBTOOL in the
2015 configure.ac. It means that we exported any symbols which need to be.
2016 Nowadays, functions get automatically exported, so we don't need to load our
2017 header files with tons of __declspec(dllexport) cruft. We only need to do so
2018 for data, but there is no public data in SimGrid so we are good.
