Update FP(Y)LLL

[malb]

Update FP(y)LLL to newest upstream releases

• ​https://github.com/fplll/fplll/releases/download/5.3.2/fplll-5.3.2.tar.gz
• ​https://github.com/fplll/fpylll/releases/download/0.5.1dev/fpylll-0.5.1dev.tar.gz

[malb]

New commits:

​0593646

update FPLLL to 5.3.1 and FPYLLL to 0.5.1dev

[tscrim]

LGTM.

(Do you want this to be in 9.1 instead of 9.0? There might be some time to still sneak it in.)

[malb]

I'd love to see this in 9.0 but I figured that's already on beta9.

[tscrim]

Then we'll set it to 9.0 and see if it gets in. Worst case is it doesn't, but since there no point in intentionally delaying it, we might as well set 9.0 as the target.

[malb]

Thanks, make sense.

[vbraun]

On a Debian 9 x86_64 buildbot this causes

[...] 
sage: A = random_matrix(ZZ, 4, 4) ## line 4357 ##
sage: B = random_matrix(ZZ, 2, 3) ## line 4358 ##
sage: B._solve_iml(A) ## line 4359 ##
sage: A._solve_iml(B, right=False) ## line 4363 ##
sage: A._solve_iml(B, right=True) ## line 4367 ##
sage: A = random_matrix(ZZ, 2000, 2000) ## line 4374 ##
sage: B = random_matrix(ZZ, 2000, 2000) ## line 4375 ##
sage: t0 = walltime() ## line 4376 ##
sage: alarm(2); A._solve_iml(B)  # long time ## line 4377 ##
**********************************************************************
Traceback (most recent call last):
  File "/var/lib/buildbot/slave/sage2_git/build/local/lib/python2.7/site-packages/sage/doctest/forker.py", line 2504, in __call__
    doctests, extras = self._run(runner, options, results)
  File "/var/lib/buildbot/slave/sage2_git/build/local/lib/python2.7/site-packages/sage/doctest/forker.py", line 2553, in _run
    result = runner.run(test)
  File "/var/lib/buildbot/slave/sage2_git/build/local/lib/python2.7/site-packages/sage/doctest/forker.py", line 897, in run
    return self._run(test, compileflags, out)
  File "/var/lib/buildbot/slave/sage2_git/build/local/lib/python2.7/site-packages/sage/doctest/forker.py", line 710, in _run
    got = got.decode('utf-8')
  File "/var/lib/buildbot/slave/sage2_git/build/local/lib/python2.7/encodings/utf_8.py", line 16, in decode
    return codecs.utf_8_decode(input, errors, True)
MemoryError
**********************************************************************
File "src/sage/modules/free_module_integer.py", line 702, in sage.modules.free_module_integer.FreeModule_submodule_with_basis_integer.voronoi_cell
Failed example:
    V.volume()           # long time
Exception raised:
    Traceback (most recent call last):
      File "/var/lib/buildbot/slave/sage2_git/build/local/lib/python2.7/site-packages/sage/doctest/forker.py", line 681, in _run
        self.compile_and_execute(example, compiler, test.globs)
      File "/var/lib/buildbot/slave/sage2_git/build/local/lib/python2.7/site-packages/sage/doctest/forker.py", line 1123, in compile_and_execute
        exec(compiled, globs)
      File "<doctest sage.modules.free_module_integer.FreeModule_submodule_with_basis_integer.voronoi_cell[6]>", line 1, in <module>
        V.volume()           # long time
      File "sage/misc/cachefunc.pyx", line 1948, in sage.misc.cachefunc.CachedMethodCaller.__call__ (build/cythonized/sage/misc/cachefunc.c:10273)
        w = self._instance_call(*args, **kwds)
      File "sage/misc/cachefunc.pyx", line 1824, in sage.misc.cachefunc.CachedMethodCaller._instance_call (build/cythonized/sage/misc/cachefunc.c:9758)
        return self.f(self._instance, *args, **kwds)
      File "/var/lib/buildbot/slave/sage2_git/build/local/lib/python2.7/site-packages/sage/geometry/polyhedron/base.py", line 6498, in volume
        triangulation = self.triangulate(engine=engine, **kwds)
      File "/var/lib/buildbot/slave/sage2_git/build/local/lib/python2.7/site-packages/sage/geometry/polyhedron/base.py", line 3630, in triangulate
        return pc.triangulate()
      File "/var/lib/buildbot/slave/sage2_git/build/local/lib/python2.7/site-packages/sage/geometry/triangulation/point_configuration.py", line 1088, in triangulate
        return self.placing_triangulation()
      File "/var/lib/buildbot/slave/sage2_git/build/local/lib/python2.7/site-packages/sage/geometry/triangulation/point_configuration.py", line 2021, in placing_triangulation
        simplices = [frozenset(self.contained_simplex(large=True))]
      File "/var/lib/buildbot/slave/sage2_git/build/local/lib/python2.7/site-packages/sage/geometry/triangulation/point_configuration.py", line 1922, in contained_simplex
        ker = matrix(edges).right_kernel().matrix()
      File "sage/matrix/matrix2.pyx", line 4216, in sage.matrix.matrix2.Matrix.right_kernel (build/cythonized/sage/matrix/matrix2.c:29513)
        M = self.right_kernel_matrix(*args, **kwds)
      File "sage/matrix/matrix2.pyx", line 3820, in sage.matrix.matrix2.Matrix.right_kernel_matrix (build/cythonized/sage/matrix/matrix2.c:28153)
        format, M = self._right_kernel_matrix(algorithm=algorithm, proof=proof)
      File "sage/matrix/matrix_rational_dense.pyx", line 1408, in sage.matrix.matrix_rational_dense.Matrix_rational_dense._right_kernel_matrix (build/cythonized/sage/matrix/matrix_rational_dense.c:13855)
        K = A._rational_kernel_iml().transpose().change_ring(QQ)
      File "sage/matrix/matrix_integer_dense.pyx", line 3892, in sage.matrix.matrix_integer_dense.Matrix_integer_dense._rational_kernel_iml (build/cythonized/sage/matrix/matrix_integer_dense.c:31814)
        sig_on()
    SignalError: Segmentation fault
**********************************************************************
1 item had failures:
   1 of  12 in sage.modules.free_module_integer.FreeModule_submodule_with_basis_integer.voronoi_cell
    [97 tests, 1 failure, 27.71 s]
----------------------------------------------------------------------
sage -t --long src/sage/matrix/matrix_integer_dense.pyx  # MemoryError in doctesting framework
sage -t --long src/sage/modules/free_module_integer.py  # 1 doctest failed
----------------------------------------------------------------------

---

New commits:

​0593646

update FPLLL to 5.3.1 and FPYLLL to 0.5.1dev

[malb]

I can reproduce it locally now.

[malb]

The failures are in _solve_imp and _solve_flint when interrupted by using alarm(2), so perhaps FPLLL screws with the signal handler, somehow? Not clear what in FPLLL would, though. From the REPL it works:

sage:             sage: A = random_matrix(ZZ, 2000, 2000)
....:             sage: B = random_matrix(ZZ, 2000, 2000)
....:             sage: t0 = walltime()
....:             sage: alarm(2); A._solve_iml(B)  # long time
....:             
---------------------------------------------------------------------------
AlarmInterrupt                            Traceback (most recent call last)
<ipython-input-2-92190be6680a> in <module>()
      2 B = random_matrix(ZZ, Integer(2000), Integer(2000))
      3 t0 = walltime()
----> 4 alarm(Integer(2)); A._solve_iml(B)  # long time

/home/malb/software/sage-fplll/local/lib/python3.7/site-packages/sage/matrix/matrix_integer_dense.pyx in sage.matrix.matrix_integer_dense.Matrix_integer_dense._solve_iml (build/cythonized/sage/matrix/matrix_integer_dense.c:35877)()
   4447         try:
   4448             verbose('Calling solver n = %s, m = %s'%(n,m))
-> 4449             sig_on()
   4450             nonsingSolvLlhsMM(solu_pos, n, m, mself, mB, mp_N, mp_D)
   4451             sig_off()

AlarmInterrupt: 

Any ideas?

[embray]

Ticket retargeted after milestone closed

[git]

Branch pushed to git repo; I updated commit sha1. New commits:

​6c0e664	Merge branch 'master' into t/28886/update_fp_y_lll
​8dfc58c	update to fplll 5.3.2

[malb]

Updated FPLLL to (seemingly) fix the issue identified.

[tscrim]

Then positive review.