[VC++ 14.0 preview] New failures in develop regression tests
At my request, Tom Kent has kindly added VC++ 14.0 (toolset msvc-14.0) to his develop branch tests. See http://www.boost.org/development/tests/develop/developer/summary.html rightmost column. This compiler is available as part of the Visual Studio 2015 Preview. See http://www.visualstudio.com/en-us/downloads/visual-studio-2015-downloads-vs.... Several libraries are showing new test failures. Please check out the test results to see if your library has started to fail. If it is your problem, fix it or at least create a ticket for yourself:-) If it is a Microsoft problem, consider submitting a bug report and posting a notice on the Boost list so that other developers are aware of the problem. Also post a link to the bug report on https://connect.microsoft.com Knowing the Microsoft Connect URL for the bug report lets other boosters up vote the importance of the issue. This is known to have at least some influence on the priority given the bug report, and that means faster fixes for the bugs boosters care about. Thanks, --Beman
Beman Dawes <bdawes <at> acm.org> writes:
At my request, Tom Kent has kindly added VC++ 14.0 (toolset msvc-14.0) to his develop branch tests. [...] Several libraries are showing new test failures.
BOOST_MSVC_ENABLE_2014_JUN_CTP seems to be missing, visible in the Boost.Lockfree test failures (the default constructor of tagged_ptr must be noexcept). Many failures are because MSVC14 has problems with Boost.Assign's templated conversion operator, this is already fixed: https://connect.microsoft.com/VisualStudio/feedback/details/1007225/vc14- ctp4-boost-assign-broken
On Mon, Nov 17, 2014 at 11:25 AM, Marcel Raad <raad@teamviewer.com> wrote:
Beman Dawes <bdawes <at> acm.org> writes:
At my request, Tom Kent has kindly added VC++ 14.0 (toolset msvc-14.0) to his develop branch tests. [...] Several libraries are showing new test failures.
BOOST_MSVC_ENABLE_2014_JUN_CTP seems to be missing, visible in the Boost.Lockfree test failures (the default constructor of tagged_ptr must be noexcept).
I just did another run with BOOST_MSVC_ENABLE_2014_JUN_CTP defined. Does it look like it worked? I'm assuming this will be enabled by default for VC14 soon? Tom
On Tue, Nov 18, 2014 at 8:11 AM, Tom Kent <lists@teeks99.com> wrote:
On Mon, Nov 17, 2014 at 11:25 AM, Marcel Raad <raad@teamviewer.com> wrote:
Beman Dawes <bdawes <at> acm.org> writes:
At my request, Tom Kent has kindly added VC++ 14.0 (toolset msvc-14.0)
to
his develop branch tests. [...] Several libraries are showing new test failures.
BOOST_MSVC_ENABLE_2014_JUN_CTP seems to be missing, visible in the Boost.Lockfree test failures (the default constructor of tagged_ptr must be noexcept).
I just did another run with BOOST_MSVC_ENABLE_2014_JUN_CTP defined. Does it look like it worked?
No, still wrong. The easy way to tell is to look at Boost.config's config_info test. It should show only two remaining C++11 macros: BOOST_NO_CXX11_ADDRESSOF [no value] BOOST_NO_CXX11_CONSTEXPR [no value]
I'm assuming this will be enabled by default for VC14 soon?
Yes. The only thing BOOST_MSVC_ENABLE_2014_JUN_CTP is doing now seems to be causing problems. I'll post a separate message asking if anyone objects to removing the BOOST_MSVC_ENABLE_2014_JUN_CTP requirement. Thanks, --Beman
On Mon, Nov 17, 2014 at 11:15 AM, Beman Dawes <bdawes@acm.org> wrote:
At my request, Tom Kent has kindly added VC++ 14.0 (toolset msvc-14.0) to his develop branch tests.
See http://www.boost.org/development/tests/develop/developer/summary.html rightmost column.
This compiler is available as part of the Visual Studio 2015 Preview.
See
http://www.visualstudio.com/en-us/downloads/visual-studio-2015-downloads-vs....
Several libraries are showing new test failures. Please check out the test results to see if your library has started to fail. If it is your problem, fix it or at least create a ticket for yourself:-)
If it is a Microsoft problem, consider submitting a bug report and posting a notice on the Boost list so that other developers are aware of the problem. Also post a link to the bug report on https://connect.microsoft.com
Not sure if it's a Microsoft problem or not.. But the compiler version reported is 13.0.22310. Is that correct for what they are calling v14? < http://www.boost.org/development/tests/develop/developer/output/teeks99-08f-...
-- -- Rene Rivera -- Grafik - Don't Assume Anything -- Robot Dreams - http://robot-dreams.net -- rrivera/acm.org (msn) - grafikrobot/aim,yahoo,skype,efnet,gmail
[Beman Dawes]
If it is a Microsoft problem, consider submitting a bug report and posting a notice on the Boost list so that other developers are aware of the problem. Also post a link to the bug report on https://connect.microsoft.com Knowing the Microsoft Connect URL for the bug report lets other boosters up vote the importance of the issue. This is known to have at least some influence on the priority given the bug report, and that means faster fixes for the bugs boosters care about.
It would also be a good idea to mention Boost in the title. I can bring especially important bugs to the compiler team's attention. [Rene Rivera]
Not sure if it's a Microsoft problem or not.. But the compiler version reported is 13.0.22310. Is that correct for what they are calling v14?
2015's product version is 14 (skipping 13) and its compiler version is 19 (incrementing normally). Can you reproduce this manually? I can report this if you can tell me which command prompt (e.g. x64 native) plus a self-contained test case (e.g. inspecting _MSC_FULL_VER) mentions "13". STL
On Mon, Nov 17, 2014 at 12:24 PM, Stephan T. Lavavej < stl@exchange.microsoft.com> wrote:
[Beman Dawes]
If it is a Microsoft problem, consider submitting a bug report and posting a notice on the Boost list so that other developers are aware of the problem. Also post a link to the bug report on https://connect.microsoft.com Knowing the Microsoft Connect URL for the bug report lets other boosters up vote the importance of the issue. This is known to have at least some influence on the priority given the bug report, and that means faster fixes for the bugs boosters care about.
It would also be a good idea to mention Boost in the title. I can bring especially important bugs to the compiler team's attention.
[Rene Rivera]
Not sure if it's a Microsoft problem or not.. But the compiler version reported is 13.0.22310. Is that correct for what they are calling v14?
2015's product version is 14 (skipping 13) and its compiler version is 19 (incrementing normally).
OK.. Can you reproduce this manually? I can't, since I don't personally have VS (or a Windows machine that is recent enough to install it).
I can report this if you can tell me which command prompt (e.g. x64 native) plus a self-contained test case (e.g. inspecting _MSC_FULL_VER) mentions "13".
It uses _MSC_VER for the major and minor numbers < https://github.com/boostorg/predef/blob/develop/include/boost/predef/compiler/visualc.h#L51>. Did "_MSC_VER" stop being incremented? -- -- Rene Rivera -- Grafik - Don't Assume Anything -- Robot Dreams - http://robot-dreams.net -- rrivera/acm.org (msn) - grafikrobot/aim,yahoo,skype,efnet,gmail
On Mon, Nov 17, 2014 at 12:40 PM, Rene Rivera <grafikrobot@gmail.com> wrote:
On Mon, Nov 17, 2014 at 12:24 PM, Stephan T. Lavavej < stl@exchange.microsoft.com> wrote:
[Beman Dawes]
If it is a Microsoft problem, consider submitting a bug report and posting a notice on the Boost list so that other developers are aware of the problem. Also post a link to the bug report on https://connect.microsoft.com Knowing the Microsoft Connect URL for the bug report lets other boosters up vote the importance of the issue. This is known to have at least some influence on the priority given the bug report, and that means faster fixes for the bugs boosters care about.
It would also be a good idea to mention Boost in the title. I can bring especially important bugs to the compiler team's attention.
[Rene Rivera]
Not sure if it's a Microsoft problem or not.. But the compiler version reported is 13.0.22310. Is that correct for what they are calling v14?
2015's product version is 14 (skipping 13) and its compiler version is 19 (incrementing normally).
OK..
Can you reproduce this manually?
I can't, since I don't personally have VS (or a Windows machine that is recent enough to install it).
I can report this if you can tell me which command prompt (e.g. x64 native) plus a self-contained test case (e.g. inspecting _MSC_FULL_VER) mentions "13".
It uses _MSC_VER for the major and minor numbers < https://github.com/boostorg/predef/blob/develop/include/boost/predef/compiler/visualc.h#L51>. Did "_MSC_VER" stop being incremented?
Hm.. actually NM. I see what's going on. Because of the skipped product version, without skipping the compiler version, the formula no longer works. I guess I need to write a special case for >= 19. (PITA) -- -- Rene Rivera -- Grafik - Don't Assume Anything -- Robot Dreams - http://robot-dreams.net -- rrivera/acm.org (msn) - grafikrobot/aim,yahoo,skype,efnet,gmail
2014-11-17 21:15 GMT+04:00 Beman Dawes <bdawes@acm.org>: <...>
Several libraries are showing new test failures. Please check out the test results to see if your library has started to fail. If it is your problem, fix it or at least create a ticket for yourself:-)
If it is a Microsoft problem, consider submitting a bug report and posting a notice on the Boost list so that other developers are aware of the problem. Also post a link to the bug report on https://connect.microsoft.com
Looks like Variadic templates are still broken in VC++. Following code works on Сlang and GCC: template <typename... Ts, int... Is> std::string operator()(const std::tuple<Ts...>& tup, indices<Is...>) { std::ostringstream ost; ost << "( "; (void) (int []){0, (ost << printer()( std::get<Is>(tup) ), 0)... }; ost << ") "; return ost.str(); } But VC++ 14.0 gives a compilation error: ..\libs\variant\test\recursive_variant_test.cpp(88): error C2059: syntax error: '{' See https://github.com/boostorg/variant/blob/develop/test/recursive_variant_test... for a full test case. Another failure is in https://github.com/boostorg/variant/blob/develop/test/variant_reference_test... ..\libs\variant\test\variant_reference_test.cpp(88): error C2784: 'boost::add_reference<T>::type wknd_get(boost::variant<T,> &,int)' Unfortunately could not report those issues to https://connect.microsoft.com, web site does not respond.
At my request, Tom Kent has kindly added VC++ 14.0 (toolset msvc-14.0) to his develop branch tests.
Great thanks to both of you! -- Best regards, Antony Polukhin
[Antony Polukhin]
Looks like Variadic templates are still broken in VC++. (void) (int []){0, (ost << printer()( std::get<Is>(tup) ), 0)... };
I don't think it's variadic templates specifically, but attempting to brace-init the elements of an array.
Another failure is in https://github.com/boostorg/variant/blob/develop/test/variant_reference_test... ..\libs\variant\test\variant_reference_test.cpp(88): error C2784: 'boost::add_reference<T>::type wknd_get(boost::variant<T,> &,int)'
Is this a template argument deduction error? There should be more to this message.
Unfortunately could not report those issues to https://connect.microsoft.com, web site does not respond.
As a favor to Boost, I am willing to file compiler bugs for these issues, but I (and the compiler team) would like to have minimal test cases, i.e. small source files with no Boost includes. The compiler team can deal with preprocessed files but digging through tons of source takes longer. STL
participants (6)
-
Antony Polukhin -
Beman Dawes -
Marcel Raad -
Rene Rivera -
Stephan T. Lavavej -
Tom Kent