Shared library: put in FIELD/lib or install a library?

Similar Questions

• Call to a shared library a VI on a target NI Linux RT (cRIO 9068)

  Hi, I wrote a very small shared library (.so) with Eclipse (toolchain 2014), which essentially serves as a wrapper for more complex, but I am having problems with the call go to in LabVIEW RT on a target Linux RT - specifically, the cRIO 9068. First: the library was copied in/usr/lib and ldconfig has been correctly implemented. More important yet, I have also written a C program (using Eclipse as well) that calls the single function currently implemented in the shared library: this works perfectly, both in Eclipse and you connecting directly on the cRIO with a Putty terminal. So I guess the shared library itself is ok and can be called from code/external programs. Now, dating back to LabVIEW (2014 here btw). Here's where things get difficult, I guess. Initially, I had the symptoms listed here--> http://forums.ni.com/t5/LabVIEW/How-to-create-a-c-shared-library-so-for-linux-real-time-for/td-p/302... who prevented me from actually run any code on the target of RT. Then I changed the call library function node: various tutorials suggest to put name_of_library.* in the path text box or the name of library, but unfortunately it doesn't seem to work, so I had to put the name and the extension too. But it still produces the error below ("the name of the function is not found" etc.) So I check the box 'specify path on diagram' and add/usr/lib as a parameter to the node library function call: now the VI can work and is actually transferred to the target of the RT... but the cluster of output error returns error 7, which is a kind of "File not found" error. However, I believe that this error message is misleading: indeed, if I try to debug remotely this library shared under Eclipse, I am actually able to pause it. and when I press the Pause button on the debugger, the goal of RT VI pauses, then continues as soon as I press the Resume button, I'm stuck... I tried searching forums and Google as well, but I have not yet found a solution. Any ideas on what's going on? Moreover, I can add more details if needed.

  Problem solved, it was a bad configuration of the node library function call which prevented the actual library (.so) to be called.

• myRIO can't find my shared library .so

  Hi all

  I have developed a shared .so library to run on a device myRIO. I moved my shared library for the path/usr/lib of the myRIO. When I run the .VI I get this error:

  Deployment RT.viRT.vi loaded with errors on the target and was closed.
  LabVIEW: Unable to load the shared library libsolve.so:solve:C. make sure that the library is present on the target of RT. Either MAX allows you to install software from OR or FTP to transfer custom RT target libraries.

  Any idea? I reinstalled the software on the myRIO but I still have this error (I comproved if the shared library is on the right track with filezilla and Eclipse).

  Thanks in advance

  Well lets try a little more detailed explanation:

  Firstly the soname parameter is a parameter of time to compile/link in the project of the shared library. I don't have a computer with a handy Eclipse installation but it's somewhere in the additional indicators in probably the settings of the linker where to a shared library named libfoo.so.1.3.6 add something like:

  -Wl,-soname,libfoo.so.1
  

  You would then copy the shared library resulting on the linux system.

  cp libfoo.* /usr/local/lib
  

  After that, you must do a few extra things:

  (1) change the access rights for the file, otherwise ldconfig may not recognize it as an executable file and simply ignore it:

• Is Linux - possible to build a shared library that calls other libraries shared without having to make a static link to them?

  Hello

  Operating system = Linux (Ubuntu 14.04 LTS 32-bit)

  I am building a shared library *. so.  This *.so should be used by LabVIEW.  This *.so is calling other functions in shared libraries.  In my case, the other library share is libxml2 and libz.  The only way I could compile my shared library and operate was a static link from libxml2.a and libz.a.  Is there a way to make a dynamic link?

  Here's the makefile file that I use.  The goal of 'TestValidateSchema' compiles the shared without the static link library and this shared library is called by an application to test 'c' and it works.  Now, to make it work in LabVIEW, I need to make a static link, as you can see in the target ' install: «otherwise the function "call library node" throws me a symbol unresolve...»»

  This is the only way to do it and if yes why?

  Thank you

  --------------------------------

  CC = gcc

  all: TestValidateSchema

  TestValidateSchema: testValidateSchema.c
  $(CC) o Release/validateschema.o - c - wall - Werror - fpic Src / validateschema.c - I. / Include-I/usr/include/libxml2
  $(CC) - shared-o./Release/libvalidateschema.so./Release/validateschema.o
  $(CC) - wall o ./release/testvalidateschema testValidateSchema.c./Release/validateschema.o-lxml2

  clean:
  $(RM) $(OBJS)./Release/testvalidateschema./Release/libvalidateschema.so./Release/validateschema.o /usr/include/validateschema.h /usr/lib/libvalidateschema.so

  install:
  $(CC) o Release/validateschema.o - c - wall - Werror - fpic Src / validateschema.c - I. / Include-I/usr/include/libxml2
  $(CC) - shared-o./Release/libvalidateschema.so./Release/validateschema.o Src/libxml2.a Src/libz.a

  --------------------------------------

  In addition to what said Daniel on the specification of an external shared library in the liaison stage of your version of the shared library, together with the - shared option, there is still one thing to consider:

  Your library of hello.so then contains references to functions in external shared libraries libxml.so and zlib.so. When LabVIEW calls dlopen ("/ path/dir/hello.so",...) it basically hands of any control in the linker Linux running, ld.so that takes care of the rest. LD.so then try to load the shared library, recognizing that there are external references and try to solve.

  First he tries to match the missing external references to any symbols that have been previously loaded into the current process already. If this is not successful for a particular system, it queries the ldcache for a library and a symbol. ldcache is updated with the ldconfig program. When you add a new shared library on the system that you want the linker ld.so runtime in order to find, run ldconfig.

  NOTE: Running ldconfig is not normally necessary to shared libraries that access you directly via the node call library, as you specify here the full access to your custom library shared already, so path ld.so will fail immediately if it is not the library specified in the given absolute position and will not attempt to query ldcache at all.

  If you are installing packages through your package manager platform update ldcache for all publicly accessible libraries shared in these packages is usually done automatically for you as part of the package installation script. If however you manually install the libraries you need to run ldconfig yourself. In your case, you could add the libxml.so and/usr/lib zlib.so and simply run ldconfig from the command-line as it default search/lib and/usr/lib for libraries shared and rebuilt the ldcache from that. If you add your new libraries in a different directory on your computer you need to run ldconfig with - n option, saying what directory the new shared libraries have been installed.

• Shared library

  Hello!
  
  
  Article "install Netbackup for Oracle on Unix agent."
  

  http://www.Symantec.com/connect/articles/installing-NetBackup-Oracle-agent-UNIX
  Stop all oracle instances on this client. The reason: sometimes Oracle takes a library shared (like ours) and place it in its shared memory space

  my question is how I can control the shared library of existence in a memory space to define the real need of Shuting down of all oracle instances?
  I'm usung 11.2.0.3
  
  Thank you and best regards,
  Pavel

  Pavel,

  Even until now, I've never worked with Netbackup tool, but that when this line says "stop all instances oracle on this customer.", then I think that we must avoid any confusion, because when Oracle puts the shared library in the shared memory, then when you install netbackup, there are chances that some internal conflicts can happen and avoid these conflicts and corruption in the installation its safe to stop all instances on the client.

  In the same connection, they said two or three MOS note ID, so please take a look at them also.

  Concerning
  Girish Sharma

• RSL Error 2046 when TERM-shared-library-path using Ant mxmlc

  Hi,.

  I build my project using the ant mxmlc with SDK 3.4 target

  If I < flex: mxmlc...-link-TIME-shared-library static = 'false' static-RSLS = "false"... >

  It works very well. My SWF is a bit lighter but not enough!

  If in my mxmlc, I add the following nodes (all of them or one of them)

  < runtime-shared-library-path

  Path-Element="${FLEX_HOME}/frameworks/libs/Framework.swc" >

  "< url rsl-url="${contextroot}/client/framework_3.4.0.6955.swz "/ >

  "< url rsl-url="${contextroot}/client/framework_3.4.0.6955.swf "/ >

  < / DURATION-shared-library-path >

  < runtime-shared-library-path

  Path-Element="${FLEX_HOME}/frameworks/libs/RPC.swc" >

  "< url rsl-url="${contextroot}/client/rpc_3.4.0.6955.swz "/ >

  "< url rsl-url="${contextroot}/client/rpc_3.4.0.6955.swf "/ >

  < / DURATION-shared-library-path >

  < runtime-shared-library-path

  Path-Element="${FLEX_HOME}/frameworks/libs/DataVisualization.swc" >

  "< url rsl-url="${contextroot}/client/datavisualization_3.4.0.9271.swz "/ >

  "< url rsl-url="${contextroot}/client/datavisualization_3.4.0.9271.swf "/ >

  < / DURATION-shared-library-path >

  I have a: Error #2046: the loaded file did not have a valid signature

  I thought: why not add summaries check = false to my target mxmlx?

  = > Unfortunately the ant task does not support this option

  I also read somewhere that this may be caused by the Decree of the library.

  It is true that in the case of the flex plugin for eclipse

  I put the first framwork.swc
  DataVisualization.swc and configure rpc.swc
  the EFA resulting works

  = > This makes me think that my swz are duly signed

  = > I can't find a way to make the swz ordering the ant target

  So is my only solution use the mxmlc commed-line within the ant?

  What can possibly be wrong?

  Only builds "Milestone" should have signed to RSL. The stable and nightly versions have not signed RSL. All signed RSL you might find in an Add ons download "Adobe" is there by mistake and should be ignored.

  -Darrell

• tables as input for the shared library function

  I played with the call of LV VI using matlab.

  So I built a simple VI which returns an integer as input and multiply it by 10, I then put in a shared library and named it MATLAB - well.

  Now, I tried to do the same thing with a table - I want to send a picture to the VI and multiplied it by 10, but when I build the .h file, it seems that the function expects get table of entry AND exit of table as inputs.

  so, how I can build a shared library VI who gets an array (of a constant size if this is important) and multiply it by 10?

  Thank you!

  A function in a DLL can only return a scalar value, not a table, no matter what languages are used. To return an array, instead calling it allocates the array pass a reference to the DLL and then after the function called the referenced table table contains the new data. This is why there are two parameters - the input array (actually a pointer to it) and a pointer (reference) to the output array. You must change your code in MATLAB, there is nothing you can do about it in LabVIEW. EDIT: Also note that it allows to re-use of the input as an output table, optionally passing a reference only to the table of entry and then by changing that. You can do it in LabVIEW by configuring the setting table as input and output.

• Unable to load the shared library LVASPT_WA.*

  Hello

  I have a hope that easy to fix the error: I want to use the function "WA Detrend' to the signal processing tool advanced in a real-time application running on a cRio 9030. When I try to run the VI I get the following error:

  ...

  Deployment NI_AdvSigProcWA.lvlib:WA decimation Filter.vi

  NI_AdvSigProcWA.lvlib:WA decimation Filter.vi loaded with errors on the target and was closed.
  LabVIEW: Unable to load the shared library LVASPT_WA.*ptDecimationFilterH:C. make sure that the library is present on the target of RT. Either MAX allows you to install software from OR or FTP to transfer custom RT target libraries.

  ...

  I reinstalled signals Advanced processing toolkit and linked things 'add new software' MAX, but the error persists.

  A picture of what the system looks like to the MAX is attached.

  Also, I Fund this post:

  http://forums.NI.com/T5/LabVIEW/problem-error-deploying-item-on-RT/m-p/3201345/highlight/true#M92822...

  could someone tell me what he means by "Please check the installation of the custom software all mode '?

  Can you telll me what I am doing wrong?

  Thank you very much in advance!

  I ran your situation in R & D, and the problem seems to be as a result of Advanced Signal Processing Toolkit is not not compatible on the x 64 under Linux RT OS in the 9030.  The product page could be clearer on this gap.  I am currently a web application to get this fixed!

  Best,

  David

• shared library problem MathScript

  I have a shared library of labview created in labview 8.2.1.  This labview dll contains a three functions, having to turn a number of mathscript embeded nodes inside of them.  The reason for their compilation in a shared library dll is because I want to call this dll of labview 7.0.  I have a test program of work created in labview 8.2.1 and labview 7.0.

  The problem I had is that these executables both run ok on my machine development, but when I move them to another machine with only the installed runtimes (7.0 and 8.2.1) labview dll function calls fail to do anything.  No errors appear, so it is very difficult to find where the problem occurs.    I have two different dll called in the example program.  The first dll just manages the file access and has not all the mathscript in bits, but the second, which is the one that I'm really interested in, has the mathscript node.  The memory usage in looking, I see it's the appeal of the file access dll but do not do the mathscript one.

  I have installed an evaluation version of labview 8.2.1 on this laptop with only the runtime, run the program and it works fine.  I think there must be something not in the dll but I don't see anything in the options to add something.  I have the mathscript check box selected in the build option.

  I know there are a number of functions that do not work with just running, but I'm not all there that use my program.

  Does anyone have advice?

  OK, I found the solution to my problem.  It was a problem with the version of the runtime that I used.  Labview CDs development is only a partial performance and include all of the files to run applications full (time to install it on his own).  Go to the website of national instruments, there is a period of standard edition which includes everything you need.  Download and install this version seem to have solved the problem.

• Warning message about the shared library file

  I got this warning message when I try to open the sample xml configuration.

  C:\Program NIUninstaller Instruments\LabVIEW reference (Impl) 8.6\vi.lib\xml\Close .vi (NI_XML.lvlib: Close Reference (Impl) .vi)
  -The shared library should happen to "": \addons\internet\dom\DOMUserDefRef.dll "has been charged with"DOMUserDefRef.dll"."
  C:\Program NIUninstaller Instruments\LabVIEW .vi reference (NdList) 8.6\vi.lib\xml\Close (NI_XML.lvlib: Close Reference (NdList) .vi)
  -The shared library should happen to "": \addons\internet\dom\DOMUserDefRef.dll "has been charged with"DOMUserDefRef.dll"."
  C:\Program NIUninstaller Instruments\LabVIEW .vi reference (NNMap) 8.6\vi.lib\xml\Close (NI_XML.lvlib: Close Reference (NNMap) .vi)
  -The shared library should happen to "": \addons\internet\dom\DOMUserDefRef.dll "has been charged with"DOMUserDefRef.dll"."
  C:\Program NIUninstaller Instruments\LabVIEW reference (node) 8.6\vi.lib\xml\Close .vi (NI_XML.lvlib: Close Reference (Node) .vi)
  -The shared library should happen to "": \addons\internet\dom\DOMUserDefRef.dll "has been charged with"DOMUserDefRef.dll"."
  C:\Program NIUninstaller Instruments\LabVIEW 8.6\vi.lib\xml\Get Child.vi first non-text (NI_XML.lvlib:Get first non-text Child.vi)
  -The shared library should happen to "": \addons\internet\dom\DOMUserDefRef.dll "has been charged with"DOMUserDefRef.dll"."
  C:\Program NIUninstaller Instruments\LabVIEW 8.6\vi.lib\xml\Get Sibling.vi non-text next (NI_XML.lvlib:Get next non-text Sibling.vi)
  -The shared library should happen to "": \addons\internet\dom\DOMUserDefRef.dll "has been charged with"DOMUserDefRef.dll"."
  C:\Program NIUninstaller Instruments\LabVIEW 8.6\vi.lib\xml\Get text Content.vi (NI_XML.lvlib:Get text Content.vi node) node
  -The shared library should happen to "": \addons\internet\dom\DOMUserDefRef.dll "has been charged with"DOMUserDefRef.dll"."
  C:\Program NIUninstaller Instruments\LabVIEW 8.6\vi.lib\xml\Load XML File.vi (NI_XML.lvlib:Load XML File.vi)
  -The shared library should happen to "": \addons\internet\dom\DOMUserDefRef.dll "has been charged with"DOMUserDefRef.dll"."
  C:\Program NIUninstaller Instruments\LabVIEW 8.6\vi.lib\xml\New.vi (NI_XML.lvlib:New.vi)
  -The shared library should happen to "": \addons\internet\dom\DOMUserDefRef.dll "has been charged with"DOMUserDefRef.dll"."

  I've recently updated to 8.6 and I try to compile an executable, but b/c of this warning my executable is not running sends me weird errors. I took the code that calls the XML that is read, and it is fixed. How should I do? Thank you!

  I get this error message when I tried to run the .exe with the xml configuration reader.

  "The application failed to start because xerces - c_2_7.dll was not found. Reinstalling the application may fix this problem. »

• Error during the generation of shared library for OSX

  Hello

  I am trying to build a shared library (.framework) using LabVIEW 2011 as Mac.

  My Version of OSX is 10.7.3.

  When I run the build process, TI allways give up showing the error message:

  There is 1 error in DLL - C compile Source File.vi Possible reason (s): you must have the Apple Developer tools installed to build a shared library. Install the Apple Developer tools and attempts to reconstruct the shared library.

  I don't know what developer tool is missing or if there is another problem with my environment.

  Thank you

  Gschwert

  Hi Gschwert,

  in the meantime, I received a response from our developers in Austin.

  Developer Tools Apple which refers to should be X-Code.

  They work with X-Code Version 3.0 and have no problem. They will try to work with your versions (4.2 and 4.3) too and take a look, if there is something special to look for.

  Have you tried using X-Code 3.0? Then you get the same problems?

  Good day

• Error message "unable to load the shared library.

  I try to get my programs for school to work, and I get these error message whenever I try to run anything. The last line of the State of deployment is as follows:

  Deployment time get NLVSim and isAcceptedStep.vi IsAcceptedStep.vi and NILVSim get busy with errors on the target and has been closed.

  LabVIEW: Unable to load the shared library NILVSim.*: SIM_GetTimeAndIsAccepted:C. make sure the library is present on the target of RT. Either MAX allows you to install software from OR or FTP to transfer custom RT target libraries.

  I tried everything I can think of to fix this. I checked NEITHER Max and I'm all for date on the software. I reinstalled LabVIEW twice, and I reset my myRIO. My classmates do not experience problems with getting theres to work. I start to lose hope and hope someone might know what the question. Thank you for your help.

  WubaDee,

  NILVSim is part of the LV Control Design and Simulation Module. This one is correctly installed on your Windows development system and you installed it on your myRIO?

  Norbert

• Failure to build LabVIEW application on Linux using "Shared library".

  In order to create a LabVIEW application on linux without X display, I have respected the LabVIEW VI using the shared library for Linux. I did it by clicking with the right button on construction Specifications and selecting New > Shared Library, and in the section Advanced , check the box labeled embedded version using the runtime engine. At the end of the build process, a message to inform that the build failed with the following message:

  "Error 127 to the Exec System: sh: gcc: command not found".

  The error code is not set. No one has provided a description for this code, or you could have wired a number which is not an admission of error code error code. »

  I have had no compling the LabVIEW VI in normal mode, so am confused about which is the problem.

  Hi Julian,.

  GCC is a compiler C and C++ under Linux.  A simple way to check if it is installed is to open a terminal and type 'gcc', if the command is recognized then the package is already installed, and the problem must lie elsewhere.  If this slot of Linux has a package manager, then it must be posted in here also.  If not then it could point to him being not installed or missing a dependency makes not accessible from the command line.  Here is a link that should hopefully scroll you through installation of gcc with Redhat: http://www.cyberciti.biz/faq/centos-linux-install-gcc-c-c-compiler/ .

• WMP Shared Library work

  I searched for a few days now and I can't find an answer to a question I have about Windows Media Player and work for music in a shared library, so I will try to explain here in the hope that someone might be able to help.

  There are two involved PC - one, machine 'A' is currently running Windows XP Pro and currently hosts all of my photos, video, music, etc. Machine 'B' runs Windows Vista Edition full and is generally used to hear or see the media of the machine.

  After a few questions of works of art in general I decied to try to pin down exactly what was going on.

  I erased the data files WMP on both machines and made sure there is no folders added to automatic records / default for both the libraries began as empty (initially find only 'local' music that comes with Windows itself).  Also, two WMP is set to not download any information from the internet media. Fine so far.

  On computer A, I created a test folder and put it in a folder with the tracks of an album and another folder with a few individual pieces of music. I also created an easily identifiable artwork for the album and for some of the individual tracks and made sure this work has been the unique work stored in the relevant music files ID3 tags.  I have also made sure there is no WMP work files in this folder to test (so not of Folder.jpg, AlbumArtSmall.jpg or AlbumArt_ {GUID} .jpg).

  I started up WMP again on computer A, and added this test folder to the library.  WMP correctly analysed and displayed a few items, he created the usual mess of it is work of course, files but it used * my * work from the ID3 tags.  Once again, great so far.  I put it to share its library.

  On computer B, I asked "look for other media share" and he immediately found the library shared by a machine.  However, if I look at the items in the shared library, I see the content of my file test, but with different illustrations - the art of actual coverage of the relevant albums.

  My first assumption was that, even if the machine B has been fixed to not download anything on the internet recovering always the album artwork for the items in the shared library.  To test this, I have blocked access to internet on my router and again allowed out WMP files on computer B.  When he again accesses the library shared on computer A, it immediately displays the work of 'bad' again - but he could not have received this Web this time, if he get all this?

  I assume that the work is cached somewhere and picked up images of a previous connection to my comprehensive library on the computer B the machine, but I don't want to.  I want to use the work in the files of current music, like WMP on machine made.

  I'm a bit puzzled, because I don't even know if machine B becomes the ('bad') work from somewhere in itself, or if it comes from the machine.  I can't imagine that it's from a machine - why WMP on this machine would use the 'proper' work itself but then give another machine a copy cached in the work of 'bad '.  So presumably, these images come from somewhere on computer B, or is there another explanation?

  Sorry that took so long to explain, I can't believe Microsoft had done this so complicated.

  Anyone can shed some light on this?

  Thank you very much

  Foam

  Update:

  After a better looking PC in question, I could have answered my own question, I thought I would add here in case this is useful to someone else...

  It seems that WMP (and Media Center) on the Vista machine at least keep several copies of the updated album art cache in multiple places and using different format of file names - that's why the failure of my first attempt to locate them.

  So after having erased the following folders:

  %LocalAppData%/Microsoft/eHome/art cache

  Cache drive/Art %LocalAppData%/Microsoft/Media/LocalMLS

  %LocalAppData%/Microsoft/Media Player/MCH/Albums

  Both Windows Media Player and Media Center on my Vista machines, reloaded the correct album artwork - I don't know if they use the ID3 tags on music files or work remote library files, but as they were all based on my work now, I don't actually mind.

  Foam

• Shared library with 10.2 NDK

  I have some classes of Qt/stunts, I'm working on that I want to incorporate a shared library.

  Originally, I was able to follow http://supportforums.blackberry.com/t5/Native-Development-Knowledge/How-to-create-a-shared-or-static... to create the library, but in the latest releases of the NDK, this does not work.

  Method B of this article does not work for Qt\Cascades classes since moc is never called, even if the used compiler is QCC. If I want to do a project of Qt\Cascades, I can only do an application project. If I tape everything that is not related to the library (assets, translations, master, etc.). I got some errors.

  Is first with the missing translation files. This can be corrected by changing the make files and the file *.pri (although I guess it will be readded by the NDK...)

  The second and more important question, is that even when I add "TEMPLATE = lib" in the file * .pro, which appears to have ignored, I get an error which is described as 'missing main function', which tells me that it is still trying to apply as opposed to a *. so.

  How can I produce a shared Qt\Cascades within the current NDK 10.2 library?

  Discover my stunts/Qt library template https://github.com/blackberry/Cascades-Community-Samples/tree/master/Cascades-Library-Template.

  It should have everything you need.