date:20120919

Re: [CMake] Visual Studio 2012 Express for WD and cmake 2.8.9

2012-09-19 Thread David Cole

On Wed, Sep 19, 2012 at 1:48 PM, Bogdan Cristea  wrote:

> On Wednesday 19 September 2012 08:56:02 you wrote:
> > Premium and Express are two very different beasts.
>
> Indeed, just learned that: I have used cmake 2.8.8 and it detects VS 2012
> ultimate, while cmake 2.8.9 does not detect VS 2012 Express for Win8.
>
> Are there any plans to support VS Express 2012 ?
>
> regards
> --
> Bogdan
>


Yes, of course.

Please see my email from yesterday:
http://www.cmake.org/pipermail/cmake/2012-September/052053.html
--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake

Re: [CMake] Visual Studio 2012 Express for WD and cmake 2.8.9

2012-09-19 Thread Bogdan Cristea

On Wednesday 19 September 2012 08:56:02 you wrote:
> Premium and Express are two very different beasts.

Indeed, just learned that: I have used cmake 2.8.8 and it detects VS 2012 
ultimate, while cmake 2.8.9 does not detect VS 2012 Express for Win8.

Are there any plans to support VS Express 2012 ?

regards
-- 
Bogdan
--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake

Re: [CMake] check_function_exists failing for gcc built-in functions?

2012-09-19 Thread Rolf Eike Beer

Andreas Naumann wrote:
> Hi,
> 
> it is an interesting problem. If you tell the gcc to use the math library,
> it works. The interesting thing is the way of testing.
> 
> The cmake macro simply declares a function foo with the signature
> char foo();
> and call this one without arguments. If you tell cmake to link against the
> math library, it works.
> 
> But, if you simply call sqrt(5.0) the library is ommitted (I think the gcc
> optimizes the code?)

As it also does with e.g. strlen("something"). Functions that are marked to 
return the same value if the same value is passed in and which are used with a 
constant input are calculated during compilation and the result is used in the 
binary. So if you write something like:

if (sqrt(strlen("foobar")) > 4)
  return 1;
else
  return 0;

I'm rather sure the code generated by gcc (if optimizations are turned on) 
will be just "return 0".

Eike
-- 


signature.asc
Description: This is a digitally signed message part.
--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake

Re: [CMake] check_function_exists failing for gcc built-in functions?

2012-09-19 Thread Andreas Naumann





Hi,

it is an interesting problem. If you tell the gcc to use the math
library, it works. The interesting thing is the way of testing.

The cmake macro simply declares a function foo with the signature
    char foo();
and call this one without arguments. If you tell cmake to link against
the math library, it works.

But, if you simply call sqrt(5.0) the library is ommitted (I think the
gcc optimizes the code?)

So, if you always know the math library for the specific platform, you
should use it.

Hth,

Andreas

Am 19.09.2012 17:08, schrieb Jack Stalnaker:
Hi,
  
  
  I'm having trouble with check_function_exists() when it comes to
built-in functions in gcc. If I create just a simple C++ file with a
call to sqrt() or pow() and have a CMakeLists.txt with the following
lines,
  
  
  include (CheckFunctionExists)
  check_function_exists(sqrt HAVE_SQRT)
  
  
  cmake fails to find sqrt (or pow, log, exp, etc). In the sqrt
case, CMakeError.log shows 
  
  
  /usr/share/cmake/Modules/CheckFunctionExists.c:3: warning:
conflicting types for built-in function 'sqrt'
  Linking C executable cmTryCompileExec
  /usr/bin/cmake -E cmake_link_script
CMakeFiles/cmTryCompileExec.dir/link.txt --verbose=1
  /usr/bin/gcc -DCHECK_FUNCTION_EXISTS=sqrt -fPIC
CMakeFiles/cmTryCompileExec.dir/CheckFunctionExists.c.o -o
cmTryCompileExec -rdynamic 
  CMakeFiles/cmTryCompileExec.dir/CheckFunctionExists.c.o: In
function 'main': 
  CheckFunctionExists.c:(.text+0x15): undefined reference to 'sqrt'
  
  
  What is the workaround for this, in the situation where the
function is built in, but I want to test for it on other platforms?
  
  
  Thanks
  

--

Powered by www.kitware.com

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at: http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake




--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake

Re: [CMake] check_function_exists failing for gcc built-in functions?

2012-09-19 Thread Rolf Eike Beer

Jack Stalnaker wrote:
> Hi,
> 
> I'm having trouble with check_function_exists() when it comes to built-in
> functions in gcc. If I create just a simple C++ file with a call to sqrt()
> or pow() and have a CMakeLists.txt with the following lines,
> 
> include (CheckFunctionExists)
> check_function_exists(sqrt HAVE_SQRT)
> 
> cmake fails to find sqrt (or pow, log, exp, etc). In the sqrt case,
> CMakeError.log shows
> 
> /usr/share/cmake/Modules/CheckFunctionExists.c:3: warning: conflicting
> types for built-in function 'sqrt'
> Linking C executable cmTryCompileExec
> /usr/bin/cmake -E cmake_link_script
> CMakeFiles/cmTryCompileExec.dir/link.txt --verbose=1
> /usr/bin/gcc -DCHECK_FUNCTION_EXISTS=sqrt -fPIC
> CMakeFiles/cmTryCompileExec.dir/CheckFunctionExists.c.o -o cmTryCompileExec
> -rdynamic
> CMakeFiles/cmTryCompileExec.dir/CheckFunctionExists.c.o: In function
> 'main':
> CheckFunctionExists.c:(.text+0x15): undefined reference to 'sqrt'
> 
> What is the workaround for this, in the situation where the function is
> built in, but I want to test for it on other platforms?

set(CMAKE_REQUIRED_INCLUDES math.h)
set(CMAKE_REQUIRED_LIBRARIES m)

Eike
-- 


signature.asc
Description: This is a digitally signed message part.
--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake

Re: [CMake] How to check the operating system is MacOSX?

2012-09-19 Thread Doug

Perhaps use check_c_source_compiles?

## Check for specific apple platforms
macro(check_apple_platform)
  if(APPLE)
include(CheckCSourceCompiles)
check_c_source_compiles("
  #include \"TargetConditionals.h\"
  int main(int argc, char *argv[]) {
  #if !TARGET_OS_MAC
-1
  #endif
  }
" OS_OSX)
check_c_source_compiles("
  #include \"TargetConditionals.h\"
  int main(int argc, char *argv[]) {
  #if !TARGET_OS_IPHONE
-1
  #endif
  }
" OS_IPHONE)
  endif()
endmacro()

check_apple_platform()
if(OS_OSX)
  message("Found OSX")
else()
  message("Found no OSX")
endif()
if(OS_IPHONE)
  message("Found IPHONE")
else()
  message("Found no IPHONE")
endif()

~
Doug.

On Wed, Sep 19, 2012 at 4:52 AM, Eric Wing  wrote:
> On 9/18/12, Stefan Reuschl  wrote:
>> How do you crosscompile for iOS?
>> If using a CMake toolchain you could simply set a variable in that file, I
>> which you could check later...
>
> Right now I'm not brave enough to try using CMake for iOS and Android.
> I maintain separate projects. But someday I would like to do away with
> that.
>
> -Eric
> --
> Beginning iPhone Games Development
> http://playcontrol.net/iphonegamebook/
> --
>
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at 
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the CMake FAQ at: 
> http://www.cmake.org/Wiki/CMake_FAQ
>
> Follow this link to subscribe/unsubscribe:
> http://www.cmake.org/mailman/listinfo/cmake
--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake

[CMake] check_function_exists failing for gcc built-in functions?

2012-09-19 Thread Jack Stalnaker

Hi,

I'm having trouble with check_function_exists() when it comes to built-in
functions in gcc. If I create just a simple C++ file with a call to sqrt()
or pow() and have a CMakeLists.txt with the following lines,

include (CheckFunctionExists)
check_function_exists(sqrt HAVE_SQRT)

cmake fails to find sqrt (or pow, log, exp, etc). In the sqrt case,
CMakeError.log shows

/usr/share/cmake/Modules/CheckFunctionExists.c:3: warning: conflicting
types for built-in function 'sqrt'
Linking C executable cmTryCompileExec
/usr/bin/cmake -E cmake_link_script
CMakeFiles/cmTryCompileExec.dir/link.txt --verbose=1
/usr/bin/gcc -DCHECK_FUNCTION_EXISTS=sqrt -fPIC
CMakeFiles/cmTryCompileExec.dir/CheckFunctionExists.c.o -o cmTryCompileExec
-rdynamic
CMakeFiles/cmTryCompileExec.dir/CheckFunctionExists.c.o: In function
'main':
CheckFunctionExists.c:(.text+0x15): undefined reference to 'sqrt'

What is the workaround for this, in the situation where the function is
built in, but I want to test for it on other platforms?

Thanks
--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake

Re: [CMake] CMake and GDB

2012-09-19 Thread GOO Creations

Stupid, stupid me. Yes it works, I just ran GDB with "start" instead of 
"run".

Thanks for your help Nils.

Chris


On 2012/09/19 04:56 PM, Nils Gladitz wrote:
"cmake -DCMAKE_BUILD_TYPE=Debug" should be enough to get gdb 
debuggable binaries.
Is the process which you intend to debug running when you ask gdb for 
a backtrace?
You will only be able to get a backtrace if the process has been 
started and has not yet exited.


e.g.:
gdb ./mybin
break main
run
backtrace

Nils

On 09/19/2012 04:43 PM, GOO Creations wrote:

Hi,

I'm trying to debug my program using GDB. I've done the following:

-DCMAKE_BUILD_TYPE=Debug
SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fPIC -g")
SET(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -g")

But whenever I run GDB and do a backtrace, I get a message: "No Stack".
I've done a Google search and apparently the -g flag is never passed 
to g++.
Is -g automatically added when my build type is "Debug"? Does anyone 
know why GDB gives me this error?


Thanks
Chris
--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html


Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ


Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake





--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake

Re: [CMake] CMake and GDB

2012-09-19 Thread Nils Gladitz

"cmake -DCMAKE_BUILD_TYPE=Debug" should be enough to get gdb debuggable 
binaries.
Is the process which you intend to debug running when you ask gdb for a 
backtrace?
You will only be able to get a backtrace if the process has been started 
and has not yet exited.


e.g.:
gdb ./mybin
break main
run
backtrace

Nils

On 09/19/2012 04:43 PM, GOO Creations wrote:

Hi,

I'm trying to debug my program using GDB. I've done the following:

-DCMAKE_BUILD_TYPE=Debug
SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fPIC -g")
SET(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -g")

But whenever I run GDB and do a backtrace, I get a message: "No Stack".
I've done a Google search and apparently the -g flag is never passed 
to g++.
Is -g automatically added when my build type is "Debug"? Does anyone 
know why GDB gives me this error?


Thanks
Chris
--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html


Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ


Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake



--
Nils Gladitz, B.Sc.
DICOM, Konnektivität und Entwicklung

Scivis wissenschaftliche Bildverarbeitung GmbH
Bertha-von-Suttner-Str. 5
D-37085 Göttingen
GERMANY
Handelsregister Nr. / Trade Register No. B3100 Göttingen
Geschäftsführer / Managing Directors Dr. Gernot Ebel, Dr. Uwe Engeland

Tel: 0049 (0)551 634181-28
E-Mail: glad...@scivis.de
Web: www.scivis.de

--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake

[CMake] CMake and GDB

2012-09-19 Thread GOO Creations


Hi,

I'm trying to debug my program using GDB. I've done the following:

-DCMAKE_BUILD_TYPE=Debug
SET(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fPIC -g")
SET(CMAKE_CXX_FLAGS_DEBUG "${CMAKE_CXX_FLAGS_DEBUG} -g")

But whenever I run GDB and do a backtrace, I get a message: "No Stack".
I've done a Google search and apparently the -g flag is never passed to g++.
Is -g automatically added when my build type is "Debug"? Does anyone 
know why GDB gives me this error?


Thanks
Chris
--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake

[CMake] Different library versions for sub projects?

2012-09-19 Thread norulez

Hello,

is it possible to use different libraries for sub projects?
For example if a sub project needs dynamic linked boost 1.48 library and a 
other sub project needs static linked boost 1.51 library.

The same could be for Qt projects too.

Thanks in advance
Best Regards
--

Powered by www.kitware.com

Visit other Kitware open-source projects at 
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at: 
http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake

Re: [CMake] CMake/Ninja support in Qt Creator

2012-09-19 Thread Peter Kümmel

On 19.09.2012 10:19, Wouter van Kleunen wrote:

Peter,

yes I figured it would be something like that, compilations per seconds.
But did the ninja developers remove this option ?

No, I've added it ;), but the official ninja binary is not a 1.0 version,
it's two months old.

I tried the ninja version you uploaded to github and this also works fine.

Wouter

On 19.09.2012 09:48, Wouter van Kleunen wrote:

Peter,

I guess removing just the build directory also seems to work.

I agree with your objections of the location the user file, the source
directory should stay clean. An inattentive moment and you have the file
in your version control.

Anyway, it seems to build now using an older version of ninja. However
when I use the 1.0.0 version of ninja, I get the following error:

ninja: FATAL: unknown placeholder '%o' in $NINJA_STATUS

This is needed for printing the compile rate: have a look at
"Compile Output", there you can see the compiled files per second rate.

With the ninja downloaded from here:
https://github.com/martine/ninja/downloads

This is not a 1.0.0 build, it needs -V for printing the version, which is
pre 1.0.
And even 1.0.0 has an ugly bug. Please select the ninja in the bin dir.
I've also uploaded a ninja binary on github.

Peter

Wouter

Yes, starting from scratch helps in most cases. But with cmake
we are used to simply clean the build dir, which doesn't help here.

It is a problem that creator's .user file is stored beside
the CMakeLists.txt: you can't check it in because of absolute paths,
a clean of the repository removes it, a buggy setup
is remember by creator, and it violates the principle
"don't touch the the source". I told them, but they don't care.

Peter

On 19.09.2012 08:08, Wouter van Kleunen wrote:

Peter,

i had 2.8.9-pre3 before, i upgraded to 2.8.9. But only after deleting
the CMakeLists user file from qtcreator, deleting the build directory
and restarting qtcreator I was able to get the option to build with
Ninja.

Anyway, it seems to work fine now, and the building is much faster.

Wouter

Op 18-9-2012 22:50, Peter Kümmel schreef:

Hi Wouter,

you need the latest cmake version, 2.8.9.

Under Tools/Options/Build&Run/CMake
the ninja binary could be set, or have it PATH.

Peter

On 18.09.2012 22:29, Wouter van Kleunen wrote:

So how do you make it to use ninja ? Should I enable ninja somewhere
?
Because when I run cmake from the gui it now still generates
nmakefiles.

Wouter

Op 17-9-2012 18:08, Peter Kümmel schreef:

Qt Creator 2.6 is beta now.

On github I've uploaded a patched version with Ninja support for
CMake
based projects:

https://github.com/syntheticpp/qt-creator/downloads

When CDB is installed

http://doc-snapshot.qt-project.org/qtcreator-2.5/creator-debugger-engines.html#supported-cdb-versions

debugging works even with the 32bit/64bit compilers of Visual
Studio
Express 2012.

Currently I'm trying to bring the patches upstream, but for 2.6 it
seems too late.

Peter
--

Visit other Kitware open-source projects at
http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the CMake FAQ at:
http://www.cmake.org/Wiki/CMake_FAQ

Follow this link to subscribe/unsubscribe:
http://www.cmake.org/mailman/listinfo/cmake