Erik Huelsmann wrote:
>> Why does the ipv6_supported variable exist at all in the below code?
>
> Optimization, I believe. (Does that need optimization?!)
Look carefully. It's not optimizing anything!
It would be if ipv6_supported was declared static, but it is not!
Max.
> ==============================================================================
>> {
>> apr_sockaddr_t *sa;
>> apr_status_t status;
>> int family = APR_INET;
>> int ipv6_supported = APR_HAVE_IPV6;
>>
>> /* Make sure we have IPV6 support first before giving
>> apr_sockaddr_info_get
>> APR_UNSPEC, because it may give us back an IPV6 address even if we
>> can't
>> create IPV6 sockets. */
>>
>> #if APR_HAVE_IPV6
>> if (ipv6_supported)
>> {
>> #ifdef MAX_SECS_TO_LINGER
>> status = apr_socket_create(sock, APR_INET6, SOCK_STREAM, pool);
>> #else
>> status = apr_socket_create(sock, APR_INET6, SOCK_STREAM,
>> APR_PROTO_TCP, pool);
>> #endif
>> if (status != 0)
>> ipv6_supported = 0;
>> else
>> {
>> apr_socket_close(*sock);
>> family = APR_UNSPEC;
>> }
>> }
>> #endif
>>
> ==============================================================================
>>
>> Couldn't it be removed entirely with *NO* change to the functioning of
>> the
>> code at all, like so:
>>
>>
> ==============================================================================
>> {
>> apr_sockaddr_t *sa;
>> apr_status_t status;
>> int family = APR_INET;
>>
>> /* Make sure we have IPV6 support first before giving
>> apr_sockaddr_info_get
>> APR_UNSPEC, because it may give us back an IPV6 address even if we
>> can't
>> create IPV6 sockets. */
>>
>> #if APR_HAVE_IPV6
>> #ifdef MAX_SECS_TO_LINGER
>> status = apr_socket_create(sock, APR_INET6, SOCK_STREAM, pool);
>> #else
>> status = apr_socket_create(sock, APR_INET6, SOCK_STREAM,
>> APR_PROTO_TCP, pool);
>> #endif
>> if (status == 0)
>> {
>> apr_socket_close(*sock);
>> family = APR_UNSPEC;
>> }
>> #endif
>>
> ==============================================================================
>>
>>
>> Thoughts?
>>
>>
>> Max.
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
>> For additional commands, e-mail: dev-help@subversion.tigris.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Mar 22 12:03:45 2005