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

Disable sse41, sse42, avx, avx2 on 32-bit #397

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Disable sse41, sse42, avx, avx2 on 32-bit #397

wants to merge 1 commit into from

Conversation

vbraun
Copy link

@vbraun vbraun commented Jun 2, 2024 •
edited
Loading

On a AMD Ryzen 5 3600 with Debian 12 32-bit, linbox fails to build with

make[1]: Leaving directory '/var/lib/buildbot/worker/sage_git/build/local/var/tmp/sage/build/linbox-1.7.0+sage-2024-05-18/src/linbox/algorithms/matrix-blas3'
Making all in polynomial-matrix
make[1]: Entering directory '/var/lib/buildbot/worker/sage_git/build/local/var/tmp/sage/build/linbox-1.7.0+sage-2024-05-18/src/linbox/algorithms/polynomial-matrix'
make[1]: Nothing to be done for 'all'.
make[1]: Leaving directory '/var/lib/buildbot/worker/sage_git/build/local/var/tmp/sage/build/linbox-1.7.0+sage-2024-05-18/src/linbox/algorithms/polynomial-matrix'
make[1]: Entering directory '/var/lib/buildbot/worker/sage_git/build/local/var/tmp/sage/build/linbox-1.7.0+sage-2024-05-18/src/linbox/algorithms'
/usr/bin/bash ../../libtool  --tag=CXX   --mode=compile g++ -std=gnu++11 -DHAVE_CONFIG_H -I. -I../..  -I../.. -DDISABLE_COMMENTATOR -O2 -march=native -mfpmath=sse -Wall -DNDEBUG -UDEBUG -DFFLAS_COMPILED -DFFPACK_COMPILED  -I/var/lib/buildbot/worker/sage_git/build/local/include -pthread -I/var/lib/buildbot/worker/sage_git/build/local/include  -I/var/lib/buildbot/worker/sage_git/build/local/include -I/var/lib/buildbot/worker/sage_git/build/local/include  -g -O2 -c -o diophantine-solver.lo diophantine-solver.C
libtool: compile:  g++ -std=gnu++11 -DHAVE_CONFIG_H -I. -I../.. -I../.. -DDISABLE_COMMENTATOR -O2 -march=native -mfpmath=sse -Wall -DNDEBUG -UDEBUG -DFFLAS_COMPILED -DFFPACK_COMPILED -I/var/lib/buildbot/worker/sage_git/build/local/include -pthread -I/var/lib/buildbot/worker/sage_git/build/local/include -I/var/lib/buildbot/worker/sage_git/build/local/include -I/var/lib/buildbot/worker/sage_git/build/local/include -g -O2 -c diophantine-solver.C  -fPIC -DPIC -o .libs/diophantine-solver.o
In file included from ../../linbox/matrix/dense-matrix.h:37,
                 from ../../linbox/solutions/methods.h:26,
                 from ../../linbox/algorithms/rational-solver.h:41,
                 from ../../linbox/algorithms/diophantine-solver.h:29,
                 from diophantine-solver.C:26:
../../linbox/vector/vector-traits.h:153:29: warning: 'template<class _Arg1, class _Arg2, class _Result> struct std::binary_function' is deprecated [-Wdeprecated-declarations]
  153 |                 public std::binary_function<const std::pair<size_t, Element>&, const std::pair<size_t, Element>&, bool > {
      |                             ^~~~~~~~~~~~~~~
In file included from /usr/include/c++/12/string:48,
                 from /usr/include/c++/12/bits/locale_classes.h:40,
                 from /usr/include/c++/12/bits/ios_base.h:41,
                 from /usr/include/c++/12/ios:42,
                 from /usr/include/c++/12/ostream:38,
                 from /usr/include/c++/12/iostream:39,
                 from ../../linbox/linbox-config.h:45,
                 from diophantine-solver.C:25:
/usr/include/c++/12/bits/stl_function.h:131:12: note: declared here
  131 |     struct binary_function
      |            ^~~~~~~~~~~~~~~
In file included from ../../linbox/algorithms/polynomial-matrix/fft.h:43,
                 from ../../linbox/algorithms/polynomial-matrix/matpoly-mult-fft-wordsize-fast.inl:33,
                 from ../../linbox/algorithms/polynomial-matrix/matpoly-mult-fft.h:369,
                 from ../../linbox/algorithms/polynomial-matrix/polynomial-matrix-domain.h:34,
                 from ../../linbox/algorithms/polynomial-matrix/order-basis.h:41,
                 from ../../linbox/algorithms/block-massey-domain.h:42,
                 from ../../linbox/algorithms/lifting-container.h:45,
                 from ../../linbox/algorithms/./dixon-solver/./dixon-solver-dense.inl:26,
                 from ../../linbox/algorithms/./dixon-solver/dixon-solver-dense.h:363,
                 from ../../linbox/algorithms/rational-solver.h:583:
../../linbox/algorithms/polynomial-matrix/fft-integral.inl: In member function 'void LinBox::FFT_base<Field, Simd, typename std::enable_if<(Field::is_elt_integral_v && typename Simd::is_same_element<Field>::value)>::type>::DIF_core_laststeps(Element*, size_t, size_t, const Element*, const Element*, const simd_vect_t&, const simd_vect_t&, LinBox::FFTSimdHelper<4>) const':
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:703:49: error: incomplete type 'Simd128<long long unsigned int>' {aka 'Simd128_impl<true, true, false, 8>'} used in nested name specifier
  703 |                         V7 = Simd128<uint64_t>::mulx (V4, Wp);
      |                                                 ^~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl: In member function 'void LinBox::FFT_base<Field, Simd, typename std::enable_if<(Field::is_elt_integral_v && typename Simd::is_same_element<Field>::value)>::type>::DIF_core_laststeps(Element*, size_t, size_t, const Element*, const Element*, const simd_vect_t&, const simd_vect_t&, LinBox::FFTSimdHelper<8>) const':
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:821:49: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
  821 |                         V3 = Simd256<uint64_t>::permute128<0x20> (V1, V2);
      |                                                 ^~~~~~~~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:822:49: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
  822 |                         V4 = Simd256<uint64_t>::permute128<0x31> (V1, V2);
      |                                                 ^~~~~~~~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:842:48: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
  842 |                         Q = Simd256<uint64_t>::mulx (V4, W2p);
      |                                                ^~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl: In member function 'void LinBox::FFT_base<Field, Simd, typename std::enable_if<(Field::is_elt_integral_v && typename Simd::is_same_element<Field>::value)>::type>::DIT_core_firststeps(Element*, size_t&, size_t&, const Element*&, const Element*&, const simd_vect_t&, const simd_vect_t&, LinBox::FFTSimdHelper<4>) const':
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:1213:49: error: incomplete type 'Simd128<long long unsigned int>' {aka 'Simd128_impl<true, true, false, 8>'} used in nested name specifier
 1213 |                         T2 = Simd128<uint64_t>::mulx (T1, Wp);
      |                                                 ^~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl: In member function 'void LinBox::FFT_base<Field, Simd, typename std::enable_if<(Field::is_elt_integral_v && typename Simd::is_same_element<Field>::value)>::type>::DIT_core_firststeps(Element*, size_t&, size_t&, const Element*&, const Element*&, const simd_vect_t&, const simd_vect_t&, LinBox::FFTSimdHelper<8>) const':
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:1356:49: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
 1356 |                         V3 = Simd256<uint64_t>::unpacklo_intrinsic(V6,V7);
      |                                                 ^~~~~~~~~~~~~~~~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:1357:49: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
 1357 |                         V4 = Simd256<uint64_t>::unpackhi_intrinsic(V6,V7);
      |                                                 ^~~~~~~~~~~~~~~~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:1373:48: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
 1373 |                         Q = Simd256<uint64_t>::mulx (V5, Wp);
      |                                                ^~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:1382:49: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
 1382 |                         V6 = Simd256<uint64_t>::unpacklo_intrinsic (V2, V7);
      |                                                 ^~~~~~~~~~~~~~~~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:1392:49: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
 1392 |                         V6 = Simd256<uint64_t>::unpacklo_intrinsic (V1, V2);
      |                                                 ^~~~~~~~~~~~~~~~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:1393:49: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
 1393 |                         V7 = Simd256<uint64_t>::unpackhi_intrinsic (V1, V2);
      |                                                 ^~~~~~~~~~~~~~~~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:1394:49: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
 1394 |                         V3 = Simd256<uint64_t>::permute128<0x20> (V6, V7);
      |                                                 ^~~~~~~~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:1395:49: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
 1395 |                         V4 = Simd256<uint64_t>::permute128<0x31> (V6, V7);
      |                                                 ^~~~~~~~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:1403:49: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
 1403 |                         V1 = Simd256<uint64_t>::permute128<0x20> (V3, V4);
      |                                                 ^~~~~~~~~~
../../linbox/algorithms/polynomial-matrix/fft-integral.inl:1404:49: error: incomplete type 'Simd256<long long unsigned int>' {aka 'Simd256_impl<true, true, false, 8>'} used in nested name specifier
 1404 |                         V2 = Simd256<uint64_t>::permute128<0x31> (V3, V4);
      |                                                 ^~~~~~~~~~
make[1]: *** [Makefile:680: diophantine-solver.lo] Error 1
make[1]: Leaving directory '/var/lib/buildbot/worker/sage_git/build/local/var/tmp/sage/build/linbox-1.7.0+sage-2024-05-18/src/linbox/algorithms'
make: *** [Makefile:716: all-recursive] Error 1

This is because the CPU supports sse41, 42, avx, and avx2 but simd128_int64.inl is only included on 64-bit. Most 32-bit machines nowadays are probably modern CPUs for CI and not >10 year old hardware.

Probably also makes not much sense to try to hardware-accelerate 64-bit on 32-bit OS. I'm proposing to just disable sse41+ on 32-bit.

See also:

@vbraun
Disable sse41, sse42, avx, avx2 on 32-bit
621a3b0 
This can potentially be provided by the CPU, but will not compile
since simd128_int64.inl is 64-bit only. Probably also makes not much
sense to try to hardware-accelerate 64-bit on 32-bit OS.
This was referenced Jun 2, 2024
Merged
Closed
vbraun pushed a commit to vbraun/sage that referenced this pull request Jun 3, 2024
sagemathgh-38136: Fix building fflas, linbox on 32-bit by disabling S…
552af2c 
…SE, AVX

    
Fixes sagemath#38000

Upstream PR at linbox-team/fflas-ffpack#397
    
URL: sagemath#38136
Reported by: Volker Braun
Reviewer(s):
vbraun pushed a commit to vbraun/sage that referenced this pull request Jun 4, 2024
sagemathgh-38136: Fix building fflas, linbox on 32-bit by disabling S…
436efdd 
…SE, AVX

    
Fixes sagemath#38000

Upstream PR at linbox-team/fflas-ffpack#397
    
URL: sagemath#38136
Reported by: Volker Braun
Reviewer(s):
vbraun pushed a commit to vbraun/sage that referenced this pull request Jun 5, 2024
sagemathgh-38136: Fix building fflas, linbox on 32-bit by disabling S…
223b854 
…SE, AVX

    
Fixes sagemath#38000

Upstream PR at linbox-team/fflas-ffpack#397
    
URL: sagemath#38136
Reported by: Volker Braun
Reviewer(s):
vbraun pushed a commit to vbraun/sage that referenced this pull request Jun 8, 2024
sagemathgh-38136: Fix building fflas, linbox on 32-bit by disabling S…
0947a25 
…SE, AVX

    
Fixes sagemath#38000

Upstream PR at linbox-team/fflas-ffpack#397
    
URL: sagemath#38136
Reported by: Volker Braun
Reviewer(s):
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
1 participant
@vbraun