I am trying to build a debug build of boost serialization/filesystem/system using boost buildsystem. Here is what I use:
b2 serialization filesystem system variant=debug link=static threading=multi
runtimelink=static define=_HAS_ITERATOR_DEBUGGING=0 toolset=msvc-11.0
However, when I include the library in my debug visual studio projects I still get errors like this:
(basic_iarchive.obj) : error LNK2038: mismatch detected for '_ITERATOR_DEBUG_LEVEL': value '2' doesn't match value '0' in Game.obj
Which leads me to belive that my 'define' in the boost build did not have any effect. What is the proper syntax? I cannot find any examples, and the documentation dosnt say anything else on the matter.
EDIT: I would like to enforce that I want to define _HAS_ITERATOR_DEBUGGING=0 in my boost build
Adding define=_ITERATOR_DEBUG_LEVEL=0 works for me, as Igor commented.
_HAS_ITERATOR_DEBUGGING is then defined by the MS-headers based on the above setting, so you shouldn't try to set it directly.
Setting it to 0 improves debug performance dramatically in some cases (real-time 3d) so there are definitely use cases that speak for it, even though the pain of recompiling all DLLs that expose STL-containers/iterators in their interfaces.
_ITERATOR_DEBUG_LEVEL is not boost's thing but part of standard library implementation that ships with MSVC. You must take greatest care to compile everything in your build with a consistent value of this setting. In many cases a violation it's diagnosed by the linker like the message you quote, unfortunately it can go undetected. As it changes the size and layout of most std collections and related stuff, guess what follows.
By default it is not set to anything, and when you include some standard header it looks for _DEBUG and sets itself to 2. Good way if nothing in the system ever sets it and you consistently use _DEBUG.
With boost I'm not familiar, but would guess the setting you quote would force the macro with value 0. If so, no wonder you have a discrepancy. Set it to 2, or if you mean 0, go ahead and use that in all your projects.
Related
I have a C++ project that I build using GCC and CMake.
Generally I like to compile with -fmax-errors=1. My normal workflow is to fix the first error and then rebuild since subsequent errors are often caused by the first one.
But unfortunately, with C++20, an error involving a constraint failure is often treated as multiple "errors" by GCC. In order to see why the constraint failed, I need to see more than one error.
So occasionally I like to set -fmax-errors to a higher number, probably 2, when such an error occurs.
But changing the compiler flags (by manually changing CMakeLists.txt or passing a cache variable to cmake on the command line) invalidates the CMake cache and makes the build start from scratch.
This behavior generally makes sense of course; arbitrary configuration changes could require a rebuild. But we humans understand that changing the compiler's error-formatting behavior doesn't require a rebuild. Is there a way of expressing this distinction to CMake?
Or, failing that, is there a clever way of working around this? I thought of having CMake read an environment variable at the time when the compiler is invoked (not at the time when cmake is run), but I can't find any documentation suggesting that this is actually possible.
(I could probably create a script that forwards most of its arguments to g++ but also adds -fmax-errors="$MY_COOL_ENV_VARIABLE" and then tell CMake that the script in question is the C++ compiler to build with, but I imagine that this might violate some of CMake's expectations about the "compiler.")
At the advice of Marc Glisse, I gave the approach that I hypothesized in parentheses at the end of the question a try. This in fact works quite well.
I now have a Python script called invoke_compiler.py in the top-level directory of my project. I point CMake to the script in the usual way in which one specifies a C++ compiler.
set(CMAKE_CXX_COMPILER ${PROJECT_SOURCE_DIR}/invoke_compiler.py)
Now I was actually a bit terse in the question for the sake of exposition. In actuality I regularly build this project with both GCC and Clang. So I want to be able to specify the C++ compiler to CMake when I invoke cmake.
cmake my-src-dir -DCMAKE_CXX_COMPILER=g++
So invoke_compiler.py has to take a few extra command-line flags, --my-project-cxx-compiler and --my-project-num-errors-flag.
Then the script invokes the compiler (whose executable is the one specified by --my-project-cxx-compiler), forwarding all of its command-line arguments except the extra ones mentioned above and adding f"{num_errors_flag}={os.environ.get('MY_PROJECT_NUM_ERRORS', 1)}". (The name of the compiler flag specifying the number of errors to display must itself be passed as an argument to the script because GCC and Clang call the flag --fmax-errors and --ferror-limit respectively.)
The trick is simply that the logic in CMakeLists.txt that determines which arguments to pass to invoke_compiler.py needs to execute before CMAKE_CXX_COMPILER is overwritten.
The user just interacts with CMake as usual and changes the value of the MY_PROJECT_NUM_ERRORS variable at any time in order to get a different number of errors from the compiler.
I tried to link a library (let's call it LibB) to a program (TheProgram) that already linked another library (LibA).
The debug configuration builds fine, but in release mode I get this error:
error LNK2038: mismatch detected for 'RuntimeLibrary': value 'MT_StaticRelease' doesn't match value 'MTd_StaticDebug' in TheProgram.obj
I thought that's easy to fix, because I only need to make sure that each of the linked objects are built with /MT. I did, and they all are. I also checked if there are any per-file settings that have the runtime flag differently in this configuration - there aren't. Also checked the final command line in the build - it's /MT everywhere.
Even more interestingly, if I link the debug version of LibB (while building the Release configuration), this error disappears, but I get other errors.
What could cause the mismatch despite the matching flags everywhere?
This drove me crazy and took days to discover. I wanted to share this here, because there is an abundance of questions where the problem is actually mismatching /M* flags, but that was not the case here.
The mistake was that I added a new preprocessor define in Debug mode, then copied over the whole field to the Release configuration (not wanting to fiddle with the mouse selection). This field contained _DEBUG (instead of NDEBUG) and that caused the linker to detect that /MTd was being linked. When I changed _DEBUG to NDEBUG in the Preprocessor Definitions in Release mode, the problem was solved.
As a side note, I tried to find the error in git diff, because I had a pretty good idea which commit introduced it, but it was very hard to see without a word-by-word diff due to the nature of the VS project file format. One colleague spotted this define for me, but he also pointed me to this answer about highlighting word differences in git diff, I think this can be very useful to find problems like this.
I want to create shared map objects that multiple processes can access. The most promising approach I've found is this demo code from Boost.Interprocess, which allocates map objects in a managed shared memory segment. This question will mostly be about the boost problems I'm having, but I'd also be grateful if anyone has non-boost alternative approaches.
I'm completely new to boost: it looks amazing, if huge, and I was encouraged by its claim that "often, there's nothing to build". But in this case that promise is broken in what seems to be a senseless way, and I'm failing to compile the demo because of dependency problems internal to boost.
I'm on Windows, with Visual C++ Express 2010 installed. After saving the demo code as shmap.cpp I do the following:
"%VS100COMNTOOLS%\..\..\VC\vcvarsall.bat"
cl /EHsc /I boost_1_57_0 shmap.cpp
It compiles OK, but then I get this:
LINK : fatal error LNK1104: cannot open file 'libboost_date_time-vc100-mt-s-1_57.lib'
This surprises me on a number of levels. (Q1): I didn't ask for libraries---where and how is boost leading the linker to expect them? (Q2): Why would it be asking for date_time in particular? At no point in the code is anything as functionally specific as a date or time computed, referenced or included. Is this a case of overzealous blanket dependency, and if so is there a way I can weed it out?
Regardless, the obvious first thing to try was to play the game: in the boost_1_57_0 directory I ran bootstrap.bat followed by b2. The Earth turned a good few degrees, boost was built successfully, and I retried with:
cl /EHsc /I boost_1_57_0 shmap.cpp /link /LIBPATH:boost_1_57_0\stage\lib
I still get the same linker error. This is because b2 seems to have built libs with -mt- and with -mt-gd- in their names, but not with the -mt-s- that the linker is looking for. Boost's "Getting Started" webpage tells me what these stand for but doesn't tell me (Q3): how can I change either the type of library that gets built, or the type that the linker expects?
"At no point in the code is anything as functionally specific as a date or time computed, referenced or included."
(Q2): Why would it be asking for date_time in particular?
Apparently the things you used depend on it.
E.g the mutex operations have timed_lock function
(Q1): I didn't add libraries to the project---where and how is boost leading the linker to expect them?
Boost does autolinking by default. This uses MSVC++ specific pragmas to indicate the right flavour of the right link libraries. This is an awesome feature.
You just have to make sure the import libraries are on the library path for your project.
There are ways to disable auto-linking in boost (I think it involves defining BOOST_ALL_NO_LIB)
There might be ways to
disable dependency on boost date_time (dropping features); see the autl-link description in the Getting Started guide
linking to date_Time statically (or make it header-only)
I'd refer to the documentation for this.
Here's what I've learned, in large part thanks to sehe:
Q1: It's magic---specifically, MSVC-specific magic---and it happens because it's necessary.
Q2: It becomes unnecessary---i.e. the demo can be compiled without needing to look for a binary date_time lib---if I add /DBOOST_ALL_NO_LIB to the compile flags. But it's unclear whether that will still be true once I start to use additional IPC functionality like time-dependent mutexing.
Q3: Strings from the "Boost.Build option" column of this table can be passed to b2, so the way to create *-mt-s-*.lib is to say b2 runtime-link=static. This finally lets me compile without the /DBOOST_ALL_NO_LIB flag, and discover that date_time is the only library the demo seems to need.
I also discovered that the dependencies can be tracked with the bcp tool, and (eventually) also how to build bcp in the first place, as follows:
build:
cd boost_1_57_0
bjam tools\bcp
cd ..
report:
boost_1_57_0\dist\bin\bcp.exe --boost=boost_1_57_0 --report --scan shmap.cpp report.html
The result is that the maps-in-shared-memory demo needs 1421 files from boost 1.57.0.
I'm trying to use vtk in my code, but I'm having problems running an example. I have almost no clue about the reasons since it's the first time I'm using it and I'm not very experienced.
I'm using visual studio 2012 and x64 platform.
Since I don't really know which libs should I use I added all of them to the "Additional Dependencies".
The example is given in this link.
The problem is that when I run it, the window shows this message
Generic Warning: In C:\location\VTK6.0.0\Rendering\Core\vtkPolyDataMapper.cxx, line 27
Error: no override found for 'vtkPolyDataMapper'.
which corresponds to this line
// Return NULL if no override is supplied.
vtkAbstractObjectFactoryNewMacro(vtkPolyDataMapper)
And the error that visual studio shows is
First-chance exception at 0x000007F7AA106C8F in Test.exe: 0xC0000005: Access violation reading location 0x0000000000000000.
Does anyone know how to solve this problem or at least what does this error mean?
I too was getting this error. The error means that the linker can't find the definition for the vtkPolyDataMapper method. One has to note which vtk rendering backend they used, during build. It will probably be either vtkRenderingOpenGL, or vtkRenderingOpenGL2. Go to your build/lib folder and search for either one of these. I have VS 2015 Community and had the vtkRenderingOpenGL2, with vtk-7.1 built on Windows 8.1, x86_64 Platform, Release configuration.
I fixed the issue by inserting the 3 following lines at the very top of my source files, before any other preprocessor directives:
#include "vtkAutoInit.h"
VTK_MODULE_INIT(vtkRenderingOpenGL2); // VTK was built with vtkRenderingOpenGL2
VTK_MODULE_INIT(vtkInteractionStyle);
This initializes the specified VTK modules. CMake includes these by default, but other compilers such as VS do not.
The last two lines can be combined into the following:
#define vtkRenderingCore_AUTOINIT 2(vtkRenderingOpenGL2, vtkInteractionStyle)
According to the VTK migration guide, if you are not using CMake to compile your code, you need to add some #defines. For VTK 6.0, these lines need to go before any other VTK #includes:
#define vtkRenderingCore_AUTOINIT 4(vtkInteractionStyle,vtkRenderingFreeType,vtkRenderingFreeTypeOpenGL,vtkRenderingOpenGL)
#define vtkRenderingVolume_AUTOINIT 1(vtkRenderingVolumeOpenGL)
You are missing include(${VTK_USE_FILE}) in your CMakeLists.txt file.
Assuming your are using OpenGL2, you should initialise the vtkRenderingOpenGL2 module, ensuring its object factory is correctly registered:
VTK_MODULE_INIT(vtkRenderingOpenGL2)
You should call this macro in the global scope (ex. main.cpp) as documented in the source code:
Initialize the named module, ensuring its object factory
is correctly registered and unregistered. This call must be made in
global scope in the translation unit of your executable (which can
include a shared library, but will not work as expected in a static
library).
#include "vtkAutoInit.h"
VTK_MODULE_INIT(vtkRenderingOpenGL);
The above snippet if included in the global scope will ensure the
object factories for vtkRenderingOpenGL are correctly registered and
unregistered.
How do you know which module to include?
The easiest method is to search in the VTK build folder for "vtkClassThatNeedsAnOverride", i.e. "vtkPolyDataMapper" in your case (note the use of quotes ".) and looking for a *ObjectFactory in your search results:
Rendering/OpenGL2/vtkRenderingOpenGL2ObjectFactory.cxx:
this->RegisterOverride("vtkPolyDataMapper",
"vtkOpenGLPolyDataMapper",
"Override for vtkRenderingOpenGL2 module", 1,
vtkObjectFactoryCreatevtkOpenGLPolyDataMapper);
It may be even more beneficial to look for RegisterOverride("vtkPolyDataMapper".
Which object factories exist?
To obtain a list of all existing modules that you could initialise, you can search for _AutoInit_Construct. *_AutoInit_Construct is the method that is called by VTK_MODULE_INIT.
As an alternative, you can look at all classes that derive from vtkObjectFactory.
A second alternative is to look for all calls to RegisterOverride.
Further information
VTK 6 Migration: Factories now require defines
Build System Migration: You do not need to call VTK_MODULE_INIT manually using cmake by calling include(${VTK_USE_FILE}) in your CMakeLists.txt
Note that I originally wrote this answer for a duplicate question, but I think the general information about solving this problem may be of interest for other people with the same error message.
I would recommend following the guide here, with the VTK_MODULE_INIT macro being the most reliable, with the guide here providing a high level overview of the changes needed. You must link to vtkRenderingOpenGL for example to get most of the standard overrides. If you use CMake then specifying it on the COMPONENTS argument to find_package would cause it to be added to VTK_LIBRARIES, and including VTK_USE_FILE would cause the correct compiler definitions to be added.
I had the same issue at my platform;
Visual Studio 2015
Windows 7
VTK 6.3
I followed VTK/Build System Migration from Marcus D. Hanwell's post, and it works. My additonal lines are;
#include <vtkAutoInit.h>
VTK_MODULE_INIT(vtkRenderingOpenGL);
VTK_MODULE_INIT(vtkInteractionStyle);
on the top of preprocessor. The difference from RestlessC0bra's post is probably OpenGL version.
When using ParaView's Catalyst libraries you have to add the following in addition to include("${PARAVIEW_USE_FILE}"):
set_property(DIRECTORY APPEND PROPERTY COMPILE_DEFINITIONS ${VTK_DEFINITIONS})
A quick hack solution: In CMakeList.txt file, replace vtkRendering${VTK_RENDERING_BACKEND} with vtkRenderingOpenGL2. The reason why we need this is because Cmake does not know where the rendering core is. By specifying it, we can use the rendering core to override the proper method.
The proper solution should be replace the whole find_package paragraph with:
find_package(VTK REQUIRED COMPONENTS vtkCommonCore)
find_package(VTK COMPONENTS
vtkFiltersSources
vtkInteractionStyle
vtkRendering${VTK_RENDERING_BACKEND})
The first find_package lets the CMake know where to find the packages, then second find_package would know where to find vtkRendering${VTK_RENDERING_BACKEND}.
Sometimes VS autos/locals/watches break and instead of variables/values all I have is different kinds of:
CXX0029: Error: not struct pointer
CXX0033: Error: error in OMF type information
CXX0072: Error: type information missing or unknown
CXX0025: Error: operator needs class/struct/union
Rebuilding project, cleaning PDB/NCB etc doesn't solve it. What can I do?
Look at this Microsoft support note on: FIX: CXX0033 Error in OMF Type from Forward Class Declaration
Once you fix the PCH problem cited in the support note, I think all your errors will go away.
There is in fact a solution that lets you keep using precompiled headers: check out this more recent KB article and the documentation of the /Yl switch - which seems specifically tailored to this error.
Just add to the stdafx.cpp (or your own custom /Yc file) command line '/Ylxxxx', where xxxx stands for an arbitrary function name in your lib.
I recently faced symptoms identical to yours (in VS2010), and that solved it for me.
Are you trying to debug the "release" build? If so, many local variables will not exist as "debuggable" elements. You can get around this (if you must debug the release build) by debugging at the assembly level and look at the register values (vs. stack values, where auto/local would be in the debug build) and cast them appropriately in the "watch window".
Otherwise, build the Debug build and debug that build version. You'll get assertions where preconditions are not met, relevant/irrelevant stuff dumped to your output window, and more straight-forward debug single stepping.
It helped me to switch from using a program database (/ZI) to "c7 compatible" (/Z7). Switching off precompiled headers did not make a difference. Neither did rebuilding.