Add Rmpfr_fpif_import_mem() and Rmpfr_fpif_export_mem(), new in 4.3.0-dev.
Allow atonv() to receive octal arguments - via new function oct2bin.
Add mpfrtoa_subn. (Tests added to mpfrtoa.t.)
In Makefile.PL, exclude "-lquadmath" from the "LIBS" argument passed to WriteMakefile() if ($Config{osname} =~/linux/i && $Config{nvtype} ne "__float128" && !$config{gnulibc_version}). We do this in an attempt to allow Math::MPFR to build on linux systems where libc is musl, not glibc. For an example of the problem see: https://www.cpantesters.org/cpan/report/65ddb9fa-a453-11f0-9829-bde118873113
In Makefile.PL make allowance for some cygwin build options. See (eg): https://www.cpantesters.org/cpan/report/41ed0a6e-6f74-1014-a0b5-c5df354b4780