Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Problems compiling with Intel C++ (OneAPI) #3390

Closed
gkthiruvathukal opened this issue Apr 17, 2023 · 12 comments
Closed

Problems compiling with Intel C++ (OneAPI) #3390

gkthiruvathukal opened this issue Apr 17, 2023 · 12 comments
Labels
question

Comments

@gkthiruvathukal
Copy link

I'm working with Intel's OneAPI compiler and running into a handful of errors when building fmt. The language level supported is C++ 17. This is the 2023.1 release of the Intel tools.

$ cmake -S . -B build
-- The CXX compiler identification is IntelLLVM 2023.1.0
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: /opt/intel/oneapi/compiler/latest/linux/bin/icpx - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Module support is disabled.
-- Version: 9.1.0
-- Build type:
-- CXX_STANDARD: 17
-- Performing Test has_std_17_flag
-- Performing Test has_std_17_flag - Success
-- Performing Test has_std_1z_flag
-- Performing Test has_std_1z_flag - Success
-- Required features: cxx_variadic_templates
-- Build spdlog: 1.9.2
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD
-- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Success
-- Found Threads: TRUE
-- Build type: Release
-- scn version: 1.1.2
-- SCN_PEDANTIC: OFF
-- SCN_WERROR: OFF
-- The C compiler identification is IntelLLVM 2023.1.0
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working C compiler: /opt/intel/oneapi/compiler/latest/linux/bin/icx - skipped
-- Detecting C compile features
-- Detecting C compile features - done
-- Found Python: /home/linuxbrew/.linuxbrew/bin/python3.11 (found version "3.11.3") found components: Interpreter
-- Configuring done (20.9s)
-- Generating done (0.1s)
-- Build files have been written to: /home/gkt/Work/unoapi-dpcpp-examples/build

# gkt @ linode5 in ~/Work/unoapi-dpcpp-examples on git:main x [15:47:21]
$ cmake --build build
[  2%] Building CXX object _deps/fmt-build/CMakeFiles/fmt.dir/src/format.cc.o
In file included from /home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/src/format.cc:8:
In file included from /home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format-inl.h:29:
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:3122:38: error: implicit instantiation of undefined template 'fmt::detail::dragonbox::float_info<long double>'
    const auto f = basic_fp<typename info::carrier_uint>(converted_value);
                                     ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:3230:13: note: in instantiation of function template specialization 'fmt::detail::format_float<long double>' requested here
  int exp = format_float(convert_float(value), precision, fspecs, buffer);
            ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:3394:20: note: in instantiation of function template specialization 'fmt::detail::write<char, fmt::appender, long double, 0>' requested here
    return detail::write(out, value, specs, locale);
                   ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:4082:10: note: in instantiation of member function 'fmt::detail::vformat_to(buffer<char> &, basic_string_view<char>, basic_format_args<basic_format_context<detail::buffer_appender<type_identity_t<char>>, type_identity_t<char>>>, locale_ref)::format_handler::on_format_specs' requested here
  struct format_handler : error_handler {
         ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/src/format.cc:34:23: note: in instantiation of function template specialization 'fmt::detail::vformat_to<char>' requested here
template FMT_API void vformat_to(buffer<char>&, string_view,
                      ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1289:54: note: template is declared here
template <typename T, typename Enable = void> struct float_info;
                                                     ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1408:46: error: implicit instantiation of undefined template 'fmt::detail::dragonbox::float_info<long double>'
    using carrier_uint = typename dragonbox::float_info<Float>::carrier_uint;
                                             ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1401:63: note: in instantiation of function template specialization 'fmt::detail::basic_fp<unsigned long long>::assign<long double, 0>' requested here
  template <typename Float> FMT_CONSTEXPR basic_fp(Float n) { assign(n); }
                                                              ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:3145:31: note: in instantiation of function template specialization 'fmt::detail::basic_fp<unsigned long long>::basic_fp<long double>' requested here
    fp normalized = normalize(fp(converted_value));
                              ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:3230:13: note: in instantiation of function template specialization 'fmt::detail::format_float<long double>' requested here
  int exp = format_float(convert_float(value), precision, fspecs, buffer);
            ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:3394:20: note: in instantiation of function template specialization 'fmt::detail::write<char, fmt::appender, long double, 0>' requested here
    return detail::write(out, value, specs, locale);
                   ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:4082:10: note: in instantiation of member function 'fmt::detail::vformat_to(buffer<char> &, basic_string_view<char>, basic_format_args<basic_format_context<detail::buffer_appender<type_identity_t<char>>, type_identity_t<char>>>, locale_ref)::format_handler::on_format_specs' requested here
  struct format_handler : error_handler {
         ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/src/format.cc:34:23: note: in instantiation of function template specialization 'fmt::detail::vformat_to<char>' requested here
template FMT_API void vformat_to(buffer<char>&, string_view,
                      ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1289:54: note: template is declared here
template <typename T, typename Enable = void> struct float_info;
                                                     ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1412:48: warning: overflow in expression; result is 2147483647 with type 'int' [-Winteger-overflow]
    const auto significand_mask = implicit_bit - 1;
                                               ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1413:14: error: no matching function for call to 'bit_cast'
    auto u = bit_cast<carrier_uint>(n);
             ^~~~~~~~~~~~~~~~~~~~~~
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:297:22: note: candidate template ignored: requirement 'sizeof(int) == sizeof(long double)' was not satisfied [with To = carrier_uint, From = long double]
FMT_CONSTEXPR20 auto bit_cast(const From& from) -> To {
                     ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:450:13: note: candidate template ignored: requirement 'sizeof(int) > sizeof(long double)' was not satisfied [with To = carrier_uint, From = long double]
inline auto bit_cast(const From& from) -> To {
            ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1415:43: error: no matching function for call to 'exponent_mask'
    auto biased_e = static_cast<int>((u & exponent_mask<Float>()) >>
                                          ^~~~~~~~~~~~~~~~~~~~
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1355:16: note: candidate template ignored: substitution failure [with Float = long double]: implicit instantiation of undefined template 'fmt::detail::dragonbox::float_info<long double>'
constexpr auto exponent_mask() ->
          ~~~~ ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1411:47: warning: shift count >= width of type [-Wshift-count-overflow]
    const auto implicit_bit = carrier_uint(1) << num_float_significand_bits;
                                              ^  ~~~~~~~~~~~~~~~~~~~~~~~~~~
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1408:46: error: implicit instantiation of undefined template 'fmt::detail::dragonbox::float_info<long double>'
    using carrier_uint = typename dragonbox::float_info<Float>::carrier_uint;
                                             ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:3164:42: note: in instantiation of function template specialization 'fmt::detail::basic_fp<unsigned __int128>::assign<long double, 0>' requested here
                                     : f.assign(converted_value);
                                         ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:3230:13: note: in instantiation of function template specialization 'fmt::detail::format_float<long double>' requested here
  int exp = format_float(convert_float(value), precision, fspecs, buffer);
            ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:3394:20: note: in instantiation of function template specialization 'fmt::detail::write<char, fmt::appender, long double, 0>' requested here
    return detail::write(out, value, specs, locale);
                   ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:4082:10: note: in instantiation of member function 'fmt::detail::vformat_to(buffer<char> &, basic_string_view<char>, basic_format_args<basic_format_context<detail::buffer_appender<type_identity_t<char>>, type_identity_t<char>>>, locale_ref)::format_handler::on_format_specs' requested here
  struct format_handler : error_handler {
         ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/src/format.cc:34:23: note: in instantiation of function template specialization 'fmt::detail::vformat_to<char>' requested here
template FMT_API void vformat_to(buffer<char>&, string_view,
                      ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1289:54: note: template is declared here
template <typename T, typename Enable = void> struct float_info;
                                                     ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1412:48: warning: overflow in expression; result is 2147483647 with type 'int' [-Winteger-overflow]
    const auto significand_mask = implicit_bit - 1;
                                               ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1413:14: error: no matching function for call to 'bit_cast'
    auto u = bit_cast<carrier_uint>(n);
             ^~~~~~~~~~~~~~~~~~~~~~
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:297:22: note: candidate template ignored: requirement 'sizeof(int) == sizeof(long double)' was not satisfied [with To = carrier_uint, From = long double]
FMT_CONSTEXPR20 auto bit_cast(const From& from) -> To {
                     ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:450:13: note: candidate template ignored: requirement 'sizeof(int) > sizeof(long double)' was not satisfied [with To = carrier_uint, From = long double]
inline auto bit_cast(const From& from) -> To {
            ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1415:43: error: no matching function for call to 'exponent_mask'
    auto biased_e = static_cast<int>((u & exponent_mask<Float>()) >>
                                          ^~~~~~~~~~~~~~~~~~~~
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1355:16: note: candidate template ignored: substitution failure [with Float = long double]: implicit instantiation of undefined template 'fmt::detail::dragonbox::float_info<long double>'
constexpr auto exponent_mask() ->
          ~~~~ ^
/home/gkt/Work/unoapi-dpcpp-examples/build/_deps/fmt-src/include/fmt/format.h:1411:47: warning: shift count >= width of type [-Wshift-count-overflow]
    const auto implicit_bit = carrier_uint(1) << num_float_significand_bits;
                                              ^  ~~~~~~~~~~~~~~~~~~~~~~~~~~
4 warnings and 7 errors generated.
gmake[2]: *** [_deps/fmt-build/CMakeFiles/fmt.dir/build.make:76: _deps/fmt-build/CMakeFiles/fmt.dir/src/format.cc.o] Error 1
gmake[1]: *** [CMakeFiles/Makefile2:372: _deps/fmt-build/CMakeFiles/fmt.dir/all] Error 2
gmake: *** [Makefile:156: all] Error 2
@gkthiruvathukal
Copy link
Author

I neglected to mention that version 8.1.0 worked for me just fine. I was trying to upgrade to 9.1.0 and am seeing the above errors.

@mfuntowicz
Copy link

@gkthiruvathukal thanks for opening up this issue, I'm running into the exact same on my side 😁.

@gkthiruvathukal
Copy link
Author

@mfuntowicz My pleasure! Thank you for following up. I need this to work for an online resource I am creating for OneAPI.

@vitaut
Copy link
Contributor

vitaut commented Apr 19, 2023 •
edited
Loading

What is the long double format on your platform?

@gkthiruvathukal
Copy link
Author

@vitaut Are you looking for sizeof? Or are you asking something architectural in nature?

@gkthiruvathukal
Copy link
Author

@vitaut Happy to run a minimum viable example to get whatever info you need.

@vitaut
Copy link
Contributor

vitaut commented Apr 20, 2023

Either the documentation of the format (is it IEEE754 type and which one) or the output of the program from #3365 (comment) which prints various numeric "limits".

@gkthiruvathukal
Copy link
Author 

$ dpcpp -o nlimits nlimits.cpp
icpx: warning: use of 'dpcpp' is deprecated and will be removed in a future release. Use 'icpx -fsycl' [-Wdeprecated]

# gkt @ linode5 in ~/Work/unoapi-dpcpp-examples on git:main x [22:43:05]
$ ./nlimits
e
1
size:16

1
1
0
0
1
1
1
1
0
1
1
1
0
64
18
21
2
-16381
-4931
16384
4932
0
0

@vitaut
Copy link
Contributor

vitaut commented Apr 22, 2023

This looks like normal IEEE754 binary128 which is supported. Have you tried building on the latest commit (c039389)?

@mfuntowicz
Copy link

Thanks @vitaut I confirm it works on my side with latest commit (c039389)

@vitaut vitaut closed this as completed Apr 23, 2023
@gkthiruvathukal
Copy link
Author

@vitaut Will there be be an updated release?

@vitaut
Copy link
Contributor

vitaut commented Apr 24, 2023

Yes, we are working on a new release.

@vitaut vitaut added the question label May 8, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
question
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@vitaut @gkthiruvathukal @mfuntowicz