dmitri@dmitri-laptop:~$ gcc -msse2 -mpreferred-stack-boundary=2 -fPIC  
-O2 -shared himd.c -o libhimd.so
dmitri@dmitri-laptop:~$ objdump -d libhimd.so

000003f0 <himd_sqrt>:
  3f0:	55                   	push   %ebp
  3f1:	89 e5                	mov    %esp,%ebp
  3f3:	57                   	push   %edi
  3f4:	56                   	push   %esi
  3f5:	83 ec 10             	sub    $0x10,%esp
  3f8:	8b 45 0c             	mov    0xc(%ebp),%eax
  3fb:	8b 4d 08             	mov    0x8(%ebp),%ecx
  3fe:	89 c7                	mov    %eax,%edi
  400:	83 e7 01             	and    $0x1,%edi
  403:	8d 50 ff             	lea    -0x1(%eax),%edx
  406:	0f 44 d0             	cmove  %eax,%edx
  409:	85 d2                	test   %edx,%edx
  40b:	7e 34                	jle    441 <himd_sqrt+0x51>
  40d:	8d 75 e8             	lea    -0x18(%ebp),%esi
  410:	31 c0                	xor    %eax,%eax
  412:	8d b6 00 00 00 00    	lea    0x0(%esi),%esi
  418:	dd 04 c1             	fldl   (%ecx,%eax,8)
  41b:	dd 5d e8             	fstpl  -0x18(%ebp)
  41e:	dd 44 c1 08          	fldl   0x8(%ecx,%eax,8)
  422:	dd 5d f0             	fstpl  -0x10(%ebp)
  425:	66 0f 51 06          	sqrtpd (%esi),%xmm0
  429:	66 0f 29 06          	movapd %xmm0,(%esi)
  42d:	dd 45 e8             	fldl   -0x18(%ebp)
  430:	dd 1c c1             	fstpl  (%ecx,%eax,8)
  433:	dd 45 f0             	fldl   -0x10(%ebp)
  436:	dd 5c c1 08          	fstpl  0x8(%ecx,%eax,8)
  43a:	83 c0 02             	add    $0x2,%eax
  43d:	39 c2                	cmp    %eax,%edx
  43f:	7f d7                	jg     418 <himd_sqrt+0x28>
  441:	85 ff                	test   %edi,%edi
  443:	74 17                	je     45c <himd_sqrt+0x6c>
  445:	8d 04 d1             	lea    (%ecx,%edx,8),%eax
  448:	dd 00                	fldl   (%eax)
  44a:	dd 5d e8             	fstpl  -0x18(%ebp)
  44d:	66 0f 51 45 e8       	sqrtpd -0x18(%ebp),%xmm0
  452:	66 0f 29 45 e8       	movapd %xmm0,-0x18(%ebp)
  457:	dd 45 e8             	fldl   -0x18(%ebp)
  45a:	dd 18                	fstpl  (%eax)
  45c:	83 c4 10             	add    $0x10,%esp
  45f:	5e                   	pop    %esi
  460:	5f                   	pop    %edi
  461:	5d                   	pop    %ebp
  462:	c3                   	ret

dmitri@dmitri-laptop:~$ sudo cp libhimd.so /usr/lib
dmitri@dmitri-laptop:~$ ldd a.out
	linux-gate.so.1 =>  (0xb7f77000)
	libhimd.so => /usr/lib/libhimd.so (0xb7f41000)
	libc.so.6 => /lib/tls/i686/cmov/libc.so.6 (0xb7de3000)
	/lib/ld-linux.so.2 (0xb7f5d000)
dmitri@dmitri-laptop:~$ ./a.out
1.000000 1.414214 1.732051
dmitri@dmitri-laptop:~$ gcc -v
Используются внутренние спецификации.
Целевая архитектура: i486-linux-gnu
Параметры конфигурации: ../src/configure -v --with-pkgversion='Ubuntu  
4.3.2-1ubuntu12'  
--with-bugurl=file:///usr/share/doc/gcc-4.3/README.Bugs  
--enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr  
--enable-shared --with-system-zlib --libexecdir=/usr/lib  
--without-included-gettext --enable-threads=posix --enable-nls  
--with-gxx-include-dir=/usr/include/c++/4.3 --program-suffix=-4.3  
--enable-clocale=gnu --enable-libstdcxx-debug --enable-objc-gc  
--enable-mpfr --enable-targets=all --enable-checking=release  
--build=i486-linux-gnu --host=i486-linux-gnu --target=i486-linux-gnu
Модель многопотоковости: posix
gcc версия 4.3.2 (Ubuntu 4.3.2-1ubuntu12)