Problems compiling fflas_fpack-2.2.2 in sage 7.4

[Florian Hanisch]

Hi,

I am currently trying to compile sage 7.4 on opensuse 13.2 from the source. Unfortunately, the process stops quite early, as far as I see while compiling fflas_ffpack-2.2.2. The relevant part of the corresponding logfile seems to be:

../../../fflas-ffpack/fflas/fflas_simd/simd256.inl: In static member function 'static Simd256i_base::vect_t Simd256i_base::sll128(Simd256i_base::vect_t)':
../../../fflas-ffpack/fflas/fflas_simd/simd256.inl:94:85: error: there are no arguments to '_mm256_bslli_epi128' that depend on a template parameter, so a declaration of '_mm256_bslli_epi128' must be available [-fpermissive]
  static INLINE CONST vect_t sll128(const vect_t a) { return _mm256_bslli_epi128(a, s); }
                                                                                     ^
../../../fflas-ffpack/fflas/fflas_simd/simd256.inl:94:85: note: (if you use '-fpermissive', G++ will accept your code, but allowing the use of an undeclared name is deprecated)
../../../fflas-ffpack/fflas/fflas_simd/simd256.inl: In static member function 'static Simd256i_base::vect_t Simd256i_base::srl128(Simd256i_base::vect_t)':
../../../fflas-ffpack/fflas/fflas_simd/simd256.inl:102:85: error: there are no arguments to '_mm256_bsrli_epi128' that depend on a template parameter, so a declaration of '_mm256_bsrli_epi128' must be available [-fpermissive]
  static INLINE CONST vect_t srl128(const vect_t a) { return _mm256_bsrli_epi128(a, s); }
                                                                                     ^
Makefile:585: recipe for target 'fflas_L1_inst.lo' failed


As far as I can see, this (or a closely related) issue was already observed some month ago: https://trac.sagemath.org/ticket/17635  (e.g. comment: 178).

Unfortunately, I can not really get from that discussion what I have to do to resolve the issue. I somehow got the impression that I may have to update my gcc version etc (which is currently  4.8.3) but I am not sure. Can anybody confirm that this is indeed the problem or does anybody know, what I have to do to fix this issue?

Thanks a lot !

[Dima Pasechnik]

You might try forcing buildung the gcc supplied by Sage.Set SAGE_INSTALL_GCC=yes
and rerun make.

[Florian Hanisch]

Thanks a lot. That seems to work ! Sage 7.4 was compiled without problems (at least as far as I can see now) during the weekend.

Taking a look at fflas_ffpack-2.2.2.log, it seems that error posted in my initial question still occurs but no longer interrupts the installation process. Anyway, the log-file confirms at the end that the package was successfully installed.   

[Dima Pasechnik]

On Monday, November 7, 2016 at 10:36:47 AM UTC, Florian Hanisch wrote:

Thanks a lot. That seems to work ! Sage 7.4 was compiled without problems (at least as far as I can see now) during the weekend.

Taking a look at fflas_ffpack-2.2.2.log, it seems that error posted in my initial question still occurs but no longer interrupts the installation process. Anyway, the log-file confirms at the end that the package was successfully installed.   

it could be that the log was not cleaned, and so you see the previous attempts there, as well, with the later attempts appended at the end.