I have ran the tests on Windows - Intel Fortran and gfortran compiler. The results differ from the ones reported here. If you are interested I can prepare a report.
(One curious thing: Intel Fortran reported that it provides no support for subnormal numbers ...)
yes, thank you, please send the test outputs, and I'll add them.
For Intel and GCC, the test outputs should contain the compiler
version and the flags, so I think I need nothing else.
Although, ... I haven't used MS systems for years.
Are there other layers underneath the compiler, like mingw?
Maybe you want to provide the details of those too.
Thanks again
I have run the program on:
Cygwin using gfortran 9.3.0
Plain Windows using Intel Fortran 18.0
The attached tarballs contain the output from the three versions per platform - 32, 64 and 128 bits reals. Note that I did not attempt to build and run it using the 32-bits versions of the compilers, I only used the 64-bits version.
I could add a third platform - MinGW-w64/MSYS2, to use its full name.
Details of Cygwin:
$ uname -a
CYGWIN_NT-10.0 L02223 3.1.4(0.340/5/3) 2020-02-19 08:49 x86_64 Cygwin
$ gfortran -v
Using built-in specs.
COLLECT_GCC=gfortran
COLLECT_LTO_WRAPPER=/usr/lib/gcc/x86_64-pc-cygwin/9.3.0/lto-wrapper.exe
Target: x86_64-pc-cygwin
Configured with: /cygdrive/i/szsz/tmpp/gcc/gcc-9.3.0-2.x86_64/src/gcc-9.3.0/configure --srcdir=/cygdrive/i/szsz/tmpp/gcc/gcc-9.3.0-2.x86_64/src/gcc-9.3.0 --prefix=/usr --exec-prefix=/usr --localstatedir=/var --sysconfdir=/etc --docdir=/usr/share/doc/gcc --htmldir=/usr/share/doc/gcc/html -C --build=x86_64-pc-cygwin --host=x86_64-pc-cygwin --target=x86_64-pc-cygwin --without-libiconv-prefix --without-libintl-prefix --libexecdir=/usr/lib --enable-shared --enable-shared-libgcc --enable-static --enable-version-specific-runtime-libs --enable-bootstrap --enable-__cxa_atexit --with-dwarf2 --with-tune=generic --enable-languages=c,c++,fortran,lto,objc,obj-c++ --enable-graphite --enable-threads=posix --enable-libatomic --enable-libgomp --enable-libquadmath --enable-libquadmath-support --disable-libssp --enable-libada --disable-symvers --with-gnu-ld --with-gnu-as --with-cloog-include=/usr/include/cloog-isl --without-libiconv-prefix --without-libintl-prefix --with-system-zlib --enable-linker-build-id --with-default-libstdcxx-abi=gcc4-compatible --enable-libstdcxx-filesystem-ts
Thread model: posix
gcc version 9.3.0 (GCC)
Details of Windows:
ver
Microsoft Windows [Version 10.0.17134.1425]
Intel(R) MPI Library 2018 Update 4 for Windows* Target Build Environment for Intel(R) 64 applications
Copyright 2007-2018 Intel Corporation.
Copyright (C) 1985-2018 Intel Corporation. All rights reserved.
Intel(R) Compiler 18.0 Update 5 (package 274)
As I was wokring on this anyway, I also ran the program on MinGW-ww64/MSYS2. The results are attached. Note that my installation contains an older version of gfortran(7.3.0) than my Cygwin installation (9.3.0).
Details of MinGW-w64/MSYS2:
$ uname -a
MINGW64_NT-10.0 L02223 2.10.0(0.325/5/3) 2018-02-09 15:25 x86_64 Msys
I added these results under 13-MAY-2020, please have a look. The only bit of data I'd like to add for those entries is the CPU arch, if you know, e.g. dmesg, etc.
Thank you
forgot to say - I attributed those results to you. Let me know if you'd rather your name was not on the page.
I could not find the dmesg command in Cygwin (it has not been installed ;)), but I did find the details of the CPU:
Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, Rev. 40457
Should this be of interest, the BIOS version is:
HPQOEM - 0 P70 Ver. 01.23 HP - 10017
(This is the information reported by the Total Commander utility; using my name in the report is fine)
It's Kaby Lake: https://ark.intel.com/content/www/us/en/ark/products/97185/intel-core-i7-7700hq-processor-6m-cache-up-to-3-80-ghz.html
I updated the page.
Let me know if I can now close this ticket.
Thanks again for your tests and data
You're welcome - as far as I am concerned, you can close this ticket.