GMP 6.0.0a on AMD A8-6600K

[tg at gmplib.org]

Marc Glisse <marc.glisse at inria.fr> writes:

  This seems to fix it. I thought that removing the line would work as
  well, but it doesn't. Torbjörn, you know best how that whole machinery
  is supposed to work, any opinion?
  
That fix makes sense.  I am not sure what "removing the line" means.

  diff --git a/configure.ac b/configure.ac
  --- a/configure.ac
  +++ b/configure.ac
  @@ -1831,17 +1831,17 @@ case $host in
         *noavx)
          gcc_cflags_noavx="-mno-avx";;
       esac
  
       case $host in
         X86_64_PATTERN)
          cclist_64="gcc cc"
          gcc_64_cflags="$gcc_cflags -m64"
  -       gcc_64_cflags_optlist="cpu arch"
  +       gcc_64_cflags_optlist="cpu arch noavx"
          CALLING_CONVENTIONS_OBJS_64='amd64call.lo amd64check$U.lo'
          SPEED_CYCLECOUNTER_OBJ_64=x86_64.lo
          cyclecounter_size_64=2
  
          cclist_x32="gcc cc"
          gcc_x32_cflags="$gcc_cflags -mx32"
          gcc_x32_cflags_optlist="$gcc_64_cflags_optlist"
          CALLING_CONVENTIONS_OBJS_x32="$CALLING_CONVENTIONS_OBJS_64"
  _______________________________________________
  gmp-bugs mailing list
  gmp-bugs at gmplib.org
  https://gmplib.org/mailman/listinfo/gmp-bugs
  

-- 
Torbjörn
Please encrypt, key id 0xC8601622