Namik and Karli,
As a follow-up to my earlier email, which is likely still sitting in the
queue do to its large size [sorry ;-) ], I wanted to ask a question
regarding the iterative solvers.
Will the GUI support benchmarking the iterative solvers? I see they are
not part of an official benchmark. I would be happy to work on setting
that up and enhancing a version of the iterative.cpp tutorial to support a
benchmark reporting either elapsed time or GFLOPS, with the former being
trivial. We could easily normalize or convert the time to a score if you
want to go that route such that the > the number the better. The majority
of my ViennaCL interest is in solving Newton-Raphson problems where the
conjugate gradient solver is a great help to me. For selfish reasons,
bringing that to higher prominence through the public benchmarks would be
nice.
-Matt
On Fri, Aug 1, 2014 at 7:17 PM, Matthew Musto <matthew.mu...@gmail.com>
wrote:
> Karli,
>
> Before I get into the latest CMake drama. I wanted to ask a question
> about:
>
>
> >> 1.) Boost: Namik, can you please remove all use of Boost (sources plus
> >> build system files)? This is just required at very few locations, mostly
> >> sparse stuff, so it shouldn't be a big deal. I'll commit some matrix
> >> generation code soon, which will allow us to get rid of all the IO and
> >> the testdata files, which is the primary use of Boost.
>
> Does that mean that boost's compressed_matrix will not be supported by the
> gui benchmark algorithm? Should I be using ell_matrix instead for my
> sparse matrices? I ask, because the focus of all of my research and
> development is on sparse matrices and I have relied heavily upon
> compressed_matrix up till now. Which is why I was so distraught to find
> out at ublas will be broken in boost 1.56. But that is a conversation for
> another day.
>
> Now onto CMake builds. Historically, I was apparently incorrectly
> downloading the zip file and making my own folder structure. Cloning makes
> things much much easier and allowed me to download libarchive
> automatically. I realize zlib is not working yet, but I thought I would
> send you my CMake output with a clean build directory for your reference in
> case you see other issues.
>
> Thanks,
> -Matt
>
> The C compiler identification is MSVC 18.0.30501.0
>
> The CXX compiler identification is MSVC 18.0.30501.0
>
> Check for working C compiler using: Visual Studio 12 2013 Win64
>
> Check for working C compiler using: Visual Studio 12 2013 Win64 -- works
>
> Detecting C compiler ABI info
>
> Detecting C compiler ABI info - done
>
> Check for working CXX compiler using: Visual Studio 12 2013 Win64
>
> Check for working CXX compiler using: Visual Studio 12 2013 Win64 -- works
>
> Detecting CXX compiler ABI info
>
> Detecting CXX compiler ABI info - done
>
> CMake Error at CMakeLists.txt:56 (add_subdirectory):
> The source directory
>
> C:/viennacl-benchmark-gui-master/viennacl-benchmark-gui/external/zlib
>
>
> does not contain a CMakeLists.txt file.
>
> Could NOT find ZLIB (missing: ZLIB_LIBRARY ZLIB_INCLUDE_DIR)
>
> Could NOT find BZip2 (missing: BZIP2_LIBRARIES BZIP2_INCLUDE_DIR)
>
...
removed a bunch of noise...
...
> Checking support for ARCHIVE_CRYPTO_SHA384_OPENSSL
>
> Checking support for ARCHIVE_CRYPTO_SHA384_OPENSSL -- not found
>
> Checking support for ARCHIVE_CRYPTO_SHA512_OPENSSL
>
> Checking support for ARCHIVE_CRYPTO_SHA512_OPENSSL -- not found
>
> Checking support for ARCHIVE_CRYPTO_MD5_LIBMD
>
> Checking support for ARCHIVE_CRYPTO_MD5_LIBMD -- not found
>
> Checking support for ARCHIVE_CRYPTO_RMD160_LIBMD
>
> Checking support for ARCHIVE_CRYPTO_RMD160_LIBMD -- not found
>
> Checking support for ARCHIVE_CRYPTO_SHA1_LIBMD
>
> Checking support for ARCHIVE_CRYPTO_SHA1_LIBMD -- not found
>
> Checking support for ARCHIVE_CRYPTO_SHA256_LIBMD
>
> Checking support for ARCHIVE_CRYPTO_SHA256_LIBMD -- not found
>
> Checking support for ARCHIVE_CRYPTO_SHA512_LIBMD
>
> Checking support for ARCHIVE_CRYPTO_SHA512_LIBMD -- not found
>
> Checking support for ARCHIVE_CRYPTO_MD5_WIN
>
> Checking support for ARCHIVE_CRYPTO_MD5_WIN -- found
>
> Checking support for ARCHIVE_CRYPTO_SHA1_WIN
>
> Checking support for ARCHIVE_CRYPTO_SHA1_WIN -- found
>
> Checking support for ARCHIVE_CRYPTO_SHA256_WIN
>
> Checking support for ARCHIVE_CRYPTO_SHA256_WIN -- not found
>
> Checking support for ARCHIVE_CRYPTO_SHA384_WIN
>
> Checking support for ARCHIVE_CRYPTO_SHA384_WIN -- not found
>
> Checking support for ARCHIVE_CRYPTO_SHA512_WIN
>
> Checking support for ARCHIVE_CRYPTO_SHA512_WIN -- not found
>
> Generating
> C:/viennacl-benchmark-gui-master/build/external/libarchive/libarchive/test/list.h
>
> CMake Warning (dev) at external/libarchive/cat/CMakeLists.txt:29
> (GET_TARGET_PROPERTY):
> Policy CMP0026 is not set: Disallow use of the LOCATION target property.
> Run "cmake --help-policy CMP0026" for policy details. Use the cmake_policy
> command to set the policy and suppress this warning.
>
> The LOCATION property should not be read from target "bsdcat". Use the
> target name directly with add_custom_command, or use the generator
> expression $<TARGET_FILE>, as appropriate.
>
> This warning is for project developers. Use -Wno-dev to suppress it.
>
> Generating
> C:/viennacl-benchmark-gui-master/build/external/libarchive/cat/test/list.h
>
> CMake Warning (dev) at external/libarchive/tar/CMakeLists.txt:42
> (GET_TARGET_PROPERTY):
> Policy CMP0026 is not set: Disallow use of the LOCATION target property.
> Run "cmake --help-policy CMP0026" for policy details. Use the cmake_policy
> command to set the policy and suppress this warning.
>
> The LOCATION property should not be read from target "bsdtar". Use the
> target name directly with add_custom_command, or use the generator
> expression $<TARGET_FILE>, as appropriate.
>
> This warning is for project developers. Use -Wno-dev to suppress it.
>
> Generating
> C:/viennacl-benchmark-gui-master/build/external/libarchive/tar/test/list.h
>
> CMake Warning (dev) at external/libarchive/cpio/CMakeLists.txt:39
> (GET_TARGET_PROPERTY):
> Policy CMP0026 is not set: Disallow use of the LOCATION target property.
> Run "cmake --help-policy CMP0026" for policy details. Use the cmake_policy
> command to set the policy and suppress this warning.
>
> The LOCATION property should not be read from target "bsdcpio". Use the
> target name directly with add_custom_command, or use the generator
> expression $<TARGET_FILE>, as appropriate.
>
> This warning is for project developers. Use -Wno-dev to suppress it.
>
> Generating
> C:/viennacl-benchmark-gui-master/build/external/libarchive/cpio/test/list.h
>
> Boost version: 1.54.0
>
> Found OPENCL: C:/Program Files (x86)/AMD APP SDK/2.9/lib/x86_64/OpenCL.lib
>
> Configuring incomplete, errors occurred!
>
> See also
> "C:/viennacl-benchmark-gui-master/build/CMakeFiles/CMakeOutput.log".
>
> See also
> "C:/viennacl-benchmark-gui-master/build/CMakeFiles/CMakeError.log".
>
>
>
>
> On Fri, Aug 1, 2014 at 11:12 AM, Karl Rupp <r...@iue.tuwien.ac.at> wrote:
>
>> Hi guys,
>>
>> okay, so the libarchive and zlib issues should be (mostly) fixed. Both
>> can be linked statically. Namik just pushed a version which allows to
>> also build it dynamically. Will this work sufficiently well? We will
>> have to make sure Windows finds these dlls, which is cumbersome... Any
>> ideas how to ensure this? Copying the .dll files to the .exe?
>>
>> There are three things left to make everything really smoothly:
>>
>> 1.) Boost: Namik, can you please remove all use of Boost (sources plus
>> build system files)? This is just required at very few locations, mostly
>> sparse stuff, so it shouldn't be a big deal. I'll commit some matrix
>> generation code soon, which will allow us to get rid of all the IO and
>> the testdata files, which is the primary use of Boost.
>>
>> 2.) viennacl/: I'll add viennacl-dev as a submodule in external/, so
>> that a user does not have to specify the path manually with CMake.
>>
>> 3.) libarchive+zlib: It seems like in the current build setup the zlib
>> is not enabled in libarchive, as it tries to run 'gzip' as a fallback.
>> Namik, can you please have a look at that?
>>
>> I'll be online in our IRC channel during the next hours, so let's get
>> this fixed quickly and enjoy a nice build system then :-)
>>
>> Best regards,
>> Karli
>>
>>
>>
>>
>>
>> (message truncated)
>>
>
------------------------------------------------------------------------------
Want fast and easy access to all the code in your enterprise? Index and
search up to 200,000 lines of code with a free copy of Black Duck
Code Sight - the same software that powers the world's largest code
search on Ohloh, the Black Duck Open Hub! Try it now.
http://p.sf.net/sfu/bds
_______________________________________________
ViennaCL-devel mailing list
ViennaCL-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/viennacl-devel
