VirtualBox and FMA CPU extension #151

mathiasbrito · 2023-07-28T11:36:06Z

mathiasbrito
Jul 28, 2023

Hi everyone, is it possible to compile srsRAN in VirtualBox? I'm trying to and getting an error regarding FMA CPU extension not being present. Trying to compile through a related error. After some research, seems that VirtualBox does not support it. Is it possible to compile it in VirtualBox? Before I move on to another virtualized environment, have you tried other solutions like VMWare, KVM, etc?

Below is the error:

[ 15%] Building CXX object lib/srsvec/CMakeFiles/srsvec.dir/prod.cpp.o
[ 15%] Building CXX object lib/fapi_adaptor/mac/messages/CMakeFiles/srsran_mac_fapi_adaptors.dir/ssb.cpp.o
In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:111,
                 from /vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/simd.h:41,
                 from /vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/prod.cpp:26:
/usr/lib/gcc/x86_64-linux-gnu/10/include/fmaintrin.h: In function 'srsran::simd_cf_t srsran::srsran_simd_cf_prod(srsran::simd_cf_t, srsran::simd_cf_t)':
/usr/lib/gcc/x86_64-linux-gnu/10/include/fmaintrin.h:63:1: error: inlining failed in call to 'always_inline' '__m256 _mm256_fmadd_ps(__m256, __m256, __m256)': target specific option mismatch
   63 | _mm256_fmadd_ps (__m256 __A, __m256 __B, __m256 __C)
      | ^~~~~~~~~~~~~~~
In file included from /vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/prod.cpp:26:
/vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/simd.h:881:38: note: called from here
  881 |   ret.im            = _mm256_fmadd_ps(a.re, b.im, _mm256_mul_ps(a.im, b.re));
      |                       ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:111,
                 from /vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/simd.h:41,
                 from /vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/prod.cpp:26:
/usr/lib/gcc/x86_64-linux-gnu/10/include/fmaintrin.h:111:1: error: inlining failed in call to 'always_inline' '__m256 _mm256_fmsub_ps(__m256, __m256, __m256)': target specific option mismatch
  111 | _mm256_fmsub_ps (__m256 __A, __m256 __B, __m256 __C)
      | ^~~~~~~~~~~~~~~
In file included from /vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/prod.cpp:26:
/vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/simd.h:880:38: note: called from here
  880 |   ret.re            = _mm256_fmsub_ps(a.re, b.re, _mm256_mul_ps(a.im, b.im));
      |                       ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:111,
                 from /vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/simd.h:41,
                 from /vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/prod.cpp:26:
/usr/lib/gcc/x86_64-linux-gnu/10/include/fmaintrin.h:111:1: error: inlining failed in call to 'always_inline' '__m256 _mm256_fmsub_ps(__m256, __m256, __m256)': target specific option mismatch
  111 | _mm256_fmsub_ps (__m256 __A, __m256 __B, __m256 __C)
      | ^~~~~~~~~~~~~~~
In file included from /vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/prod.cpp:26:
/vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/simd.h:880:38: note: called from here
  880 |   ret.re            = _mm256_fmsub_ps(a.re, b.re, _mm256_mul_ps(a.im, b.im));
      |                       ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /usr/lib/gcc/x86_64-linux-gnu/10/include/immintrin.h:111,
                 from /vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/simd.h:41,
                 from /vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/prod.cpp:26:
/usr/lib/gcc/x86_64-linux-gnu/10/include/fmaintrin.h:63:1: error: inlining failed in call to 'always_inline' '__m256 _mm256_fmadd_ps(__m256, __m256, __m256)': target specific option mismatch
   63 | _mm256_fmadd_ps (__m256 __A, __m256 __B, __m256 __C)
      | ^~~~~~~~~~~~~~~
In file included from /vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/prod.cpp:26:
/vagrant/Docker-srsRAN-Compiled/srsRAN_Project/lib/srsvec/simd.h:881:38: note: called from here
  881 |   ret.im            = _mm256_fmadd_ps(a.re, b.im, _mm256_mul_ps(a.im, b.re));
      |                       ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
make[2]: *** [lib/srsvec/CMakeFiles/srsvec.dir/build.make:212: lib/srsvec/CMakeFiles/srsvec.dir/prod.cpp.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:9763: lib/srsvec/CMakeFiles/srsvec.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs....
[ 16%] Linking CXX static library libsrsran_mac_fapi_adaptors.a
[ 16%] Built target srsran_mac_fapi_adaptors
make: *** [Makefile:160: all] Error 2

Answered by xavierarteaga

Jul 28, 2023

Hi!
First of all, many thanks for your interest in srsRAN.

CMake should detect whether it supports FMA. Could you share the CMake console output from a fresh build directory?

If you don't need to worry about running it in real-time, re-run CMake with AUTO_DETECT_ISA=False.

For cleaning the CMake cache and re-running CMake:

$ rm CMakeCache.txt
$ cmake .. -DAUTO_DETECT_ISA=False

Let us know how it goes and close the issue if it compiles successfully.

View full answer

xavierarteaga · 2023-07-28T14:19:35Z

xavierarteaga
Jul 28, 2023
Collaborator

Hi!
First of all, many thanks for your interest in srsRAN.

CMake should detect whether it supports FMA. Could you share the CMake console output from a fresh build directory?

If you don't need to worry about running it in real-time, re-run CMake with AUTO_DETECT_ISA=False.

For cleaning the CMake cache and re-running CMake:

$ rm CMakeCache.txt
$ cmake .. -DAUTO_DETECT_ISA=False

Let us know how it goes and close the issue if it compiles successfully.

1 reply

mathiasbrito Jul 28, 2023
Author

I'll try your approach. Right now, I do not need Real-Time, but soon I'll. Below more details about the issue, FMA is detected, but seems that this is a VirtualBox issue, It would be interesting to know if anyone is using VirtualBox... F16C, FMA and BMI extensions are not supported.

To make it easy to reproduce, below is a Vagrantfile I used to start the VM:

Vagrant.configure("2") do |config|
    config.vm.box = "ubuntu/jammy64"
    config.vm.define "control-plane-1" do |node|
        node.vm.provider "virtualbox" do |v|
            v.memory = 4096
            v.cpus = 2
        end
    end
end

Thi is the output from CMAKE

-- The C compiler identification is GNU 11.3.0
-- The CXX compiler identification is GNU 11.3.0
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working C compiler: /usr/bin/cc - skipped
-- Detecting C compile features
-- Detecting C compile features - done
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: /usr/bin/c++ - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- Build type not specified: defaulting to Release.
-- Performing Test HAS_MAYBE_UNINITIALIZED
-- Performing Test HAS_MAYBE_UNINITIALIZED - Success
-- Performing Test HAVE_NON_VIRTUAL_DTOR
-- Performing Test HAVE_NON_VIRTUAL_DTOR - Success
-- Performing Test HAVE_SUGGEST_OVERRIDE
-- Performing Test HAVE_SUGGEST_OVERRIDE - Success
-- Performing Test HAVE_SHADOW
-- Performing Test HAVE_SHADOW - Success
-- Assertion level set to NORMAL
-- Could NOT find libdw (missing: LIBDW_LIBRARY LIBDW_INCLUDE_DIR)
-- Could NOT find libbfd (missing: LIBBFD_LIBRARY LIBBFD_INCLUDE_DIR)
-- Could NOT find libdwarf (missing: LIBDWARF_LIBRARY LIBDWARF_INCLUDE_DIR LIBELF_LIBRARY LIBELF_INCLUDE_DIR)
-- Found Backward: /home/vagrant/srsRAN_Project/cmake/modules
-- Backward-cpp found, but external libraries are missing.
-- Found PkgConfig: /usr/bin/pkg-config (found version "0.29.2")
-- Checking for module 'mbedtls'
--   No package 'mbedtls' found
-- MBEDTLS LIBRARIES: /usr/lib/x86_64-linux-gnu/libmbedcrypto.so
-- MBEDTLS STATIC LIBRARIES: /usr/lib/x86_64-linux-gnu/libmbedcrypto.a
-- MBEDTLS INCLUDE DIRS: /usr/include
-- Found MbedTLS: /usr/lib/x86_64-linux-gnu/libmbedcrypto.so
-- Checking for module 'fftw3f >= 3.0'
--   Found fftw3f , version 3.3.8
-- FFTW3F LIBRARIES: /usr/lib/x86_64-linux-gnu/libfftw3f.so
-- FFTW3F STATIC LIBRARIES: /usr/lib/x86_64-linux-gnu/libfftw3f.a
-- FFTW3F INCLUDE DIRS: /usr/include
-- Found FFTW3F: /usr/lib/x86_64-linux-gnu/libfftw3f.so
-- Found GTest: /usr/lib/x86_64-linux-gnu/cmake/GTest/GTestConfig.cmake (found version "1.11.0")
-- Found Threads: TRUE
-- UHD LIBRARIES UHD_LIBRARIES-NOTFOUND
-- UHD INCLUDE DIRS UHD_INCLUDE_DIRS-NOTFOUND
-- Could NOT find UHD (missing: UHD_LIBRARIES UHD_INCLUDE_DIRS)
-- Checking for module 'yaml-cpp'
--   Found yaml-cpp, version 0.7.0
-- YAMLCPP LIBRARIES: /usr/lib/x86_64-linux-gnu/libyaml-cpp.so
-- YAMLCPP INCLUDE DIRS: /usr/include
-- Found YAMLCPP: /usr/lib/x86_64-linux-gnu/libyaml-cpp.so
-- Performing Test HAVE_SSE
-- Performing Test HAVE_SSE - Success
-- SSE4.1 is enabled - target CPU must support it
-- Performing Test HAVE_AVX
-- Performing Test HAVE_AVX - Success
-- AVX is enabled - target CPU must support it
-- Performing Test HAVE_AVX2
-- Performing Test HAVE_AVX2 - Success
-- AVX2 is enabled - target CPU must support it
-- Performing Test HAVE_FMA
-- Performing Test HAVE_FMA - Success
-- FMA is enabled - target CPU must support it
-- Performing Test HAVE_AVX512
-- Performing Test HAVE_AVX512 - Failed
-- Performing Test HAVE_MARCH
-- Performing Test HAVE_MARCH - Success
-- Could NOT find Doxygen (missing: DOXYGEN_EXECUTABLE)
-- Checking for module 'sctp'
--   No package 'sctp' found
-- SCTP LIBRARIES: /usr/lib/x86_64-linux-gnu/libsctp.so
-- SCTP INCLUDE DIRS: /usr/include
-- Found SCTP: /usr/lib/x86_64-linux-gnu/libsctp.so
-- FFT_LIBRARIES: /usr/lib/x86_64-linux-gnu/libfftw3f.so
-- Building srsRAN version 23.5.0
-- Configuring done
-- Generating done
-- Build files have been written to: /home/vagrant/srsRAN_Project/build

And this is the error... after make

[ 17%] Building CXX object lib/ofh/ethernet/CMakeFiles/srsran_ofh_ethernet.dir/ethernet_factories.cpp.o
[ 17%] Building CXX object lib/srsvec/CMakeFiles/srsvec.dir/dot_prod.cpp.o
[ 17%] Building CXX object lib/ofh/ethernet/CMakeFiles/srsran_ofh_ethernet.dir/ethernet_transmitter_impl.cpp.o
[ 17%] Building CXX object lib/ofh/ethernet/CMakeFiles/srsran_ofh_ethernet.dir/ethernet_receiver_impl.cpp.o
[ 17%] Building CXX object lib/srsvec/CMakeFiles/srsvec.dir/modulus_square.cpp.o
[ 18%] Building CXX object lib/ofh/ethernet/CMakeFiles/srsran_ofh_ethernet.dir/vlan_ethernet_frame_builder_impl.cpp.o
[ 18%] Building CXX object lib/srsvec/CMakeFiles/srsvec.dir/prod.cpp.o
[ 18%] Building CXX object lib/ofh/ethernet/CMakeFiles/srsran_ofh_ethernet.dir/vlan_ethernet_frame_decoder_impl.cpp.o
[ 18%] Linking CXX static library libsrsran_ofh_ethernet.a
[ 18%] Built target srsran_ofh_ethernet
[ 18%] Building CXX object lib/ofh/serdes/CMakeFiles/srsran_ofh_message_serdes.dir/ofh_cplane_message_builder_impl.cpp.o
In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:99,
                 from /home/vagrant/srsRAN_Project/lib/srsvec/simd.h:41,
                 from /home/vagrant/srsRAN_Project/lib/srsvec/prod.cpp:26:
/usr/lib/gcc/x86_64-linux-gnu/11/include/fmaintrin.h: In function ‘srsran::simd_cf_t srsran::srsran_simd_cf_prod(srsran::simd_cf_t, srsran::simd_cf_t)’:
/usr/lib/gcc/x86_64-linux-gnu/11/include/fmaintrin.h:63:1: error: inlining failed in call to ‘always_inline’ ‘__m256 _mm256_fmadd_ps(__m256, __m256, __m256)’: target specific option mismatch
   63 | _mm256_fmadd_ps (__m256 __A, __m256 __B, __m256 __C)
      | ^~~~~~~~~~~~~~~
In file included from /home/vagrant/srsRAN_Project/lib/srsvec/prod.cpp:26:
/home/vagrant/srsRAN_Project/lib/srsvec/simd.h:881:38: note: called from here
  881 |   ret.im            = _mm256_fmadd_ps(a.re, b.im, _mm256_mul_ps(a.im, b.re));
      |                       ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:99,
                 from /home/vagrant/srsRAN_Project/lib/srsvec/simd.h:41,
                 from /home/vagrant/srsRAN_Project/lib/srsvec/prod.cpp:26:
/usr/lib/gcc/x86_64-linux-gnu/11/include/fmaintrin.h:111:1: error: inlining failed in call to ‘always_inline’ ‘__m256 _mm256_fmsub_ps(__m256, __m256, __m256)’: target specific option mismatch
  111 | _mm256_fmsub_ps (__m256 __A, __m256 __B, __m256 __C)
      | ^~~~~~~~~~~~~~~
In file included from /home/vagrant/srsRAN_Project/lib/srsvec/prod.cpp:26:
/home/vagrant/srsRAN_Project/lib/srsvec/simd.h:880:38: note: called from here
  880 |   ret.re            = _mm256_fmsub_ps(a.re, b.re, _mm256_mul_ps(a.im, b.im));
      |                       ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:99,
                 from /home/vagrant/srsRAN_Project/lib/srsvec/simd.h:41,
                 from /home/vagrant/srsRAN_Project/lib/srsvec/prod.cpp:26:
/usr/lib/gcc/x86_64-linux-gnu/11/include/fmaintrin.h:111:1: error: inlining failed in call to ‘always_inline’ ‘__m256 _mm256_fmsub_ps(__m256, __m256, __m256)’: target specific option mismatch
  111 | _mm256_fmsub_ps (__m256 __A, __m256 __B, __m256 __C)
      | ^~~~~~~~~~~~~~~
In file included from /home/vagrant/srsRAN_Project/lib/srsvec/prod.cpp:26:
/home/vagrant/srsRAN_Project/lib/srsvec/simd.h:880:38: note: called from here
  880 |   ret.re            = _mm256_fmsub_ps(a.re, b.re, _mm256_mul_ps(a.im, b.im));
      |                       ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from /usr/lib/gcc/x86_64-linux-gnu/11/include/immintrin.h:99,
                 from /home/vagrant/srsRAN_Project/lib/srsvec/simd.h:41,
                 from /home/vagrant/srsRAN_Project/lib/srsvec/prod.cpp:26:
/usr/lib/gcc/x86_64-linux-gnu/11/include/fmaintrin.h:63:1: error: inlining failed in call to ‘always_inline’ ‘__m256 _mm256_fmadd_ps(__m256, __m256, __m256)’: target specific option mismatch
   63 | _mm256_fmadd_ps (__m256 __A, __m256 __B, __m256 __C)
      | ^~~~~~~~~~~~~~~
In file included from /home/vagrant/srsRAN_Project/lib/srsvec/prod.cpp:26:
/home/vagrant/srsRAN_Project/lib/srsvec/simd.h:881:38: note: called from here
  881 |   ret.im            = _mm256_fmadd_ps(a.re, b.im, _mm256_mul_ps(a.im, b.re));
      |                       ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
make[2]: *** [lib/srsvec/CMakeFiles/srsvec.dir/build.make:216: lib/srsvec/CMakeFiles/srsvec.dir/prod.cpp.o] Error 1
make[1]: *** [CMakeFiles/Makefile2:8975: lib/srsvec/CMakeFiles/srsvec.dir/all] Error 2
make[1]: *** Waiting for unfinished jobs....
[ 18%] Building CXX object lib/ofh/serdes/CMakeFiles/srsran_ofh_message_serdes.dir/ofh_uplane_message_builder_impl.cpp.o
[ 18%] Building CXX object lib/ofh/serdes/CMakeFiles/srsran_ofh_message_serdes.dir/ofh_uplane_message_builder_dynamic_compression_impl.cpp.o
[ 18%] Building CXX object lib/ofh/serdes/CMakeFiles/srsran_ofh_message_serdes.dir/ofh_uplane_message_builder_static_compression_impl.cpp.o
[ 18%] Building CXX object lib/ofh/serdes/CMakeFiles/srsran_ofh_message_serdes.dir/ofh_uplane_message_decoder_impl.cpp.o
[ 18%] Building CXX object lib/ofh/serdes/CMakeFiles/srsran_ofh_message_serdes.dir/ofh_uplane_message_decoder_dynamic_compression_impl.cpp.o
[ 18%] Building CXX object lib/ofh/serdes/CMakeFiles/srsran_ofh_message_serdes.dir/ofh_uplane_message_decoder_static_compression_impl.cpp.o
[ 18%] Linking CXX static library libsrsran_ofh_message_serdes.a
[ 18%] Built target srsran_ofh_message_serdes
make: *** [Makefile:146: all] Error 2

Hope this help...

xavierarteaga · 2023-07-28T15:18:25Z

xavierarteaga
Jul 28, 2023
Collaborator

Correct, it detects FMA:

-- Performing Test HAVE_FMA
-- Performing Test HAVE_FMA - Success

Have you tried -DAUTO_DETECT_ISA=False?

1 reply

mathiasbrito Jul 28, 2023
Author

Thanks @xavierarteaga, -DAUTO_DETECT_ISA=False solved the problem.

xavierarteaga · 2023-07-28T15:21:18Z

xavierarteaga
Jul 28, 2023
Collaborator

It could be that fma is not included in -march=native.

You can check it running the following command:

$ gcc -march=native -Q --help=target

In any case, the -DAUTO_DETECT_ISA=False should stop the problem.

1 reply

mathiasbrito Jul 28, 2023
Author

This is the output of running the command gcc -march=native -Q --help=target

vagrant@ubuntu-jammy:~$ gcc -march=native -Q --help=target
The following options are target specific:
  -m128bit-long-double        		[enabled]
  -m16                        		[disabled]
  -m32                        		[disabled]
  -m3dnow                     		[disabled]
  -m3dnowa                    		[disabled]
  -m64                        		[enabled]
  -m80387                     		[enabled]
  -m8bit-idiv                 		[disabled]
  -m96bit-long-double         		[disabled]
  -mabi=                      		sysv
  -mabm                       		[enabled]
  -maccumulate-outgoing-args  		[disabled]
  -maddress-mode=             		long
  -madx                       		[disabled]
  -maes                       		[enabled]
  -malign-data=               		compat
  -malign-double              		[disabled]
  -malign-functions=          		0
  -malign-jumps=              		0
  -malign-loops=              		0
  -malign-stringops           		[enabled]
  -mamx-bf16                  		[disabled]
  -mamx-int8                  		[disabled]
  -mamx-tile                  		[disabled]
  -mandroid                   		[disabled]
  -march=                     		haswell
  -masm=                      		att
  -mavx                       		[enabled]
  -mavx2                      		[enabled]
  -mavx256-split-unaligned-load 	[disabled]
  -mavx256-split-unaligned-store 	[disabled]
  -mavx5124fmaps              		[disabled]
  -mavx5124vnniw              		[disabled]
  -mavx512bf16                		[disabled]
  -mavx512bitalg              		[disabled]
  -mavx512bw                  		[disabled]
  -mavx512cd                  		[disabled]
  -mavx512dq                  		[disabled]
  -mavx512er                  		[disabled]
  -mavx512f                   		[disabled]
  -mavx512ifma                		[disabled]
  -mavx512pf                  		[disabled]
  -mavx512vbmi                		[disabled]
  -mavx512vbmi2               		[disabled]
  -mavx512vl                  		[disabled]
  -mavx512vnni                		[disabled]
  -mavx512vp2intersect        		[disabled]
  -mavx512vpopcntdq           		[disabled]
  -mavxvnni                   		[disabled]
  -mbionic                    		[disabled]
  -mbmi                       		[enabled]
  -mbmi2                      		[enabled]
  -mbranch-cost=<0,5>         		3
  -mcall-ms2sysv-xlogues      		[disabled]
  -mcet-switch                		[disabled]
  -mcld                       		[disabled]
  -mcldemote                  		[disabled]
  -mclflushopt                		[disabled]
  -mclwb                      		[disabled]
  -mclzero                    		[disabled]
  -mcmodel=                   		[default]
  -mcpu=
  -mcrc32                     		[enabled]
  -mcx16                      		[enabled]
  -mdispatch-scheduler        		[disabled]
  -mdump-tune-features        		[disabled]
  -menqcmd                    		[disabled]
  -mf16c                      		[disabled]
  -mfancy-math-387            		[enabled]
  -mfentry                    		[disabled]
  -mfentry-name=
  -mfentry-section=
  -mfma                       		[disabled]
  -mfma4                      		[disabled]
  -mforce-drap                		[disabled]
  -mforce-indirect-call       		[disabled]
  -mfp-ret-in-387             		[enabled]
  -mfpmath=                   		sse
  -mfsgsbase                  		[enabled]
  -mfunction-return=          		keep
  -mfused-madd                		-ffp-contract=fast
  -mfxsr                      		[enabled]
  -mgeneral-regs-only         		[disabled]
  -mgfni                      		[disabled]
  -mglibc                     		[enabled]
  -mhard-float                		[enabled]
  -mharden-sls=               		none
  -mhle                       		[enabled]
  -mhreset                    		[disabled]
  -miamcu                     		[disabled]
  -mieee-fp                   		[enabled]
  -mincoming-stack-boundary=  		0
  -mindirect-branch-cs-prefix 		[disabled]
  -mindirect-branch-register  		[disabled]
  -mindirect-branch=          		keep
  -minline-all-stringops      		[disabled]
  -minline-stringops-dynamically 	[disabled]
  -minstrument-return=        		none
  -mintel-syntax              		-masm=intel
  -mkl                        		[disabled]
  -mlarge-data-threshold=<number> 	65536
  -mlong-double-128           		[disabled]
  -mlong-double-64            		[disabled]
  -mlong-double-80            		[enabled]
  -mlwp                       		[disabled]
  -mlzcnt                     		[enabled]
  -mmanual-endbr              		[disabled]
  -mmemcpy-strategy=
  -mmemset-strategy=
  -mmitigate-rop              		[disabled]
  -mmmx                       		[enabled]
  -mmovbe                     		[enabled]
  -mmovdir64b                 		[disabled]
  -mmovdiri                   		[disabled]
  -mmpx                       		[disabled]
  -mms-bitfields              		[disabled]
  -mmusl                      		[disabled]
  -mmwait                     		[enabled]
  -mmwaitx                    		[disabled]
  -mneeded                    		[disabled]
  -mno-align-stringops        		[disabled]
  -mno-default                		[disabled]
  -mno-fancy-math-387         		[disabled]
  -mno-push-args              		[disabled]
  -mno-red-zone               		[disabled]
  -mno-sse4                   		[disabled]
  -mnop-mcount                		[disabled]
  -momit-leaf-frame-pointer   		[disabled]
  -mpc32                      		[disabled]
  -mpc64                      		[disabled]
  -mpc80                      		[disabled]
  -mpclmul                    		[enabled]
  -mpcommit                   		[disabled]
  -mpconfig                   		[disabled]
  -mpku                       		[disabled]
  -mpopcnt                    		[enabled]
  -mprefer-avx128             		-mprefer-vector-width=128
  -mprefer-vector-width=      		none
  -mpreferred-stack-boundary= 		0
  -mprefetchwt1               		[disabled]
  -mprfchw                    		[disabled]
  -mptwrite                   		[disabled]
  -mpush-args                 		[enabled]
  -mrdpid                     		[disabled]
  -mrdrnd                     		[enabled]
  -mrdseed                    		[disabled]
  -mrecip                     		[disabled]
  -mrecip=
  -mrecord-mcount             		[disabled]
  -mrecord-return             		[disabled]
  -mred-zone                  		[enabled]
  -mregparm=                  		6
  -mrtd                       		[disabled]
  -mrtm                       		[disabled]
  -msahf                      		[enabled]
  -mserialize                 		[disabled]
  -msgx                       		[disabled]
  -msha                       		[disabled]
  -mshstk                     		[disabled]
  -mskip-rax-setup            		[disabled]
  -msoft-float                		[disabled]
  -msse                       		[enabled]
  -msse2                      		[enabled]
  -msse2avx                   		[disabled]
  -msse3                      		[enabled]
  -msse4                      		[enabled]
  -msse4.1                    		[enabled]
  -msse4.2                    		[enabled]
  -msse4a                     		[disabled]
  -msse5                      		-mavx
  -msseregparm                		[disabled]
  -mssse3                     		[enabled]
  -mstack-arg-probe           		[disabled]
  -mstack-protector-guard-offset=
  -mstack-protector-guard-reg=
  -mstack-protector-guard-symbol=
  -mstack-protector-guard=    		tls
  -mstackrealign              		[disabled]
  -mstringop-strategy=        		[default]
  -mstv                       		[enabled]
  -mtbm                       		[disabled]
  -mtls-dialect=              		gnu
  -mtls-direct-seg-refs       		[enabled]
  -mtsxldtrk                  		[disabled]
  -mtune-ctrl=
  -mtune=                     		haswell
  -muclibc                    		[disabled]
  -muintr                     		[disabled]
  -mvaes                      		[disabled]
  -mveclibabi=                		[default]
  -mvect8-ret-in-mem          		[disabled]
  -mvpclmulqdq                		[disabled]
  -mvzeroupper                		[enabled]
  -mwaitpkg                   		[disabled]
  -mwbnoinvd                  		[disabled]
  -mwidekl                    		[disabled]
  -mx32                       		[disabled]
  -mxop                       		[disabled]
  -mxsave                     		[enabled]
  -mxsavec                    		[disabled]
  -mxsaveopt                  		[disabled]
  -mxsaves                    		[disabled]

  Known assembler dialects (for use with the -masm= option):
    att intel

  Known ABIs (for use with the -mabi= option):
    ms sysv

  Known code models (for use with the -mcmodel= option):
    32 kernel large medium small

  Valid arguments to -mfpmath=:
    387 387+sse 387,sse both sse sse+387 sse,387

  Known choices for mitigation against straight line speculation with -mharden-sls=:
    all indirect-jmp none return

  Known indirect branch choices (for use with the -mindirect-branch=/-mfunction-return= options):
    keep thunk thunk-extern thunk-inline

  Known choices for return instrumentation with -minstrument-return=:
    call none nop5

  Known data alignment choices (for use with the -malign-data= option):
    abi cacheline compat

  Known vectorization library ABIs (for use with the -mveclibabi= option):
    acml svml

  Known address mode (for use with the -maddress-mode= option):
    long short

  Known preferred register vector length (to use with the -mprefer-vector-width= option):
    128 256 512 none

  Known stack protector guard (for use with the -mstack-protector-guard= option):
    global tls

  Valid arguments to -mstringop-strategy=:
    byte_loop libcall loop rep_4byte rep_8byte rep_byte unrolled_loop vector_loop

  Known TLS dialects (for use with the -mtls-dialect= option):
    gnu gnu2

  Known valid arguments for -march= option:
    i386 i486 i586 pentium lakemont pentium-mmx winchip-c6 winchip2 c3 samuel-2 c3-2 nehemiah c7 esther i686 pentiumpro pentium2 pentium3 pentium3m pentium-m pentium4 pentium4m prescott nocona core2 nehalem corei7 westmere sandybridge corei7-avx ivybridge core-avx-i haswell core-avx2 broadwell skylake skylake-avx512 cannonlake icelake-client rocketlake icelake-server cascadelake tigerlake cooperlake sapphirerapids alderlake bonnell atom silvermont slm goldmont goldmont-plus tremont knl knm intel geode k6 k6-2 k6-3 athlon athlon-tbird athlon-4 athlon-xp athlon-mp x86-64 x86-64-v2 x86-64-v3 x86-64-v4 eden-x2 nano nano-1000 nano-2000 nano-3000 nano-x2 eden-x4 nano-x4 k8 k8-sse3 opteron opteron-sse3 athlon64 athlon64-sse3 athlon-fx amdfam10 barcelona bdver1 bdver2 bdver3 bdver4 znver1 znver2 znver3 btver1 btver2 generic native

  Known valid arguments for -mtune= option:
    generic i386 i486 pentium lakemont pentiumpro pentium4 nocona core2 nehalem sandybridge haswell bonnell silvermont goldmont goldmont-plus tremont knl knm skylake skylake-avx512 cannonlake icelake-client icelake-server cascadelake tigerlake cooperlake sapphirerapids alderlake rocketlake intel geode k6 athlon k8 amdfam10 bdver1 bdver2 bdver3 bdver4 btver1 btver2 znver1 znver2 znver3

xavierarteaga · 2023-08-07T09:25:59Z

xavierarteaga
Aug 7, 2023
Collaborator

In conclusion, -march=native does not include -mfma in your case. Therefore, we will add all the flags of the instruction families regardless of -march=native.

This patch will fix:

diff --git a/CMakeLists.txt b/CMakeLists.txt
index 941663596..06f695cba 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -269,33 +269,23 @@ if (AUTO_DETECT_ISA)
 
     if (HAVE_AVX2)
         add_definitions(-DHAVE_AVX2)
-        if (NOT HAVE_MARCH)
-            set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mavx2")
-        endif(NOT HAVE_MARCH)
+        set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mavx2")
     endif (HAVE_AVX2)
     if (HAVE_AVX)
         add_definitions(-DHAVE_AVX)
-        if (NOT HAVE_MARCH)
-            set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mavx")
-        endif(NOT HAVE_MARCH)
+        set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mavx")
     endif (HAVE_AVX)
     if (HAVE_SSE)
         add_definitions(-DHAVE_SSE)
-        if (NOT HAVE_MARCH)
-            set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -msse4.2")
-        endif(NOT HAVE_MARCH)
+        set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -msse4.1")
     endif (HAVE_SSE)
     if (HAVE_FMA)
         add_definitions(-DHAVE_FMA)
-        if (NOT HAVE_MARCH)
-            set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mfma")
-        endif(NOT HAVE_MARCH)
+        set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mfma")
     endif (HAVE_FMA)
     if (HAVE_AVX512)
         add_definitions(-DHAVE_AVX512)
-        if (NOT HAVE_MARCH)
-            set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mavx512f -mavx512cd -mavx512bw -mavx512dq")
-        endif(NOT HAVE_MARCH)
+        set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mavx512f -mavx512cd -mavx512bw -mavx512dq")
     endif (HAVE_AVX512)
 else (AUTO_DETECT_ISA)
     unset(HAVE_SSE CACHE)

0 replies

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

VirtualBox and FMA CPU extension #151

{{title}}

Replies: 4 comments 3 replies

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

{{title}}

Select a reply

VirtualBox and FMA CPU extension #151

mathiasbrito Jul 28, 2023

Replies: 4 comments · 3 replies

xavierarteaga Jul 28, 2023 Collaborator

mathiasbrito Jul 28, 2023 Author

xavierarteaga Jul 28, 2023 Collaborator

mathiasbrito Jul 28, 2023 Author

xavierarteaga Jul 28, 2023 Collaborator

mathiasbrito Jul 28, 2023 Author

xavierarteaga Aug 7, 2023 Collaborator

mathiasbrito
Jul 28, 2023

Replies: 4 comments 3 replies

xavierarteaga
Jul 28, 2023
Collaborator

mathiasbrito Jul 28, 2023
Author

xavierarteaga
Jul 28, 2023
Collaborator

mathiasbrito Jul 28, 2023
Author

xavierarteaga
Jul 28, 2023
Collaborator

mathiasbrito Jul 28, 2023
Author

xavierarteaga
Aug 7, 2023
Collaborator