fix fcntl usage when set socket nonblock

[MOON-CLJ]

cc @everpcpc

from fcntl man page

       EINTR  For F_SETLKW, the command was interrupted by a signal; see
       signal(7).  For F_GETLK and F_SETLK, the command was interrupted
       by a signal  before  the
       lock was checked or acquired.  Most likely when
       locking a remote file (e.g., locking over NFS), but
       can sometimes happen locally.

       EINVAL For F_DUPFD, arg is negative or is
       greater than the maximum allowable value.
       For F_SETSIG, arg is not an allowable signal
       number.

so i dont think cmd F_GETFL & F_SETFL will encounter EINTR or EAGAIN,
so just move the do while loop will do the same thing as before.
i see redis source
code，https://github.com/antirez/redis/blob/565e139a5631a777254e222d1c50ea6d696e1a8e/src/anet.c#L77

it's just like

    if (fcntl(fd, F_SETFL, flags) == -1) {
        anetSetError(err, "fcntl(F_SETFL,O_NONBLOCK): %s", strerror(errno));
        return ANET_ERR;
    }

so i think it's safe to do so。

this pr include some other minor fix.

• str_port's init args from "\0" to "", i think "" just means "\0"
  for c char array literal.

• max_timeout 6 to 3，i think 6 is too much, for example
  m_connectTimeout is 300ms in douban's mc config.

[mckelvin]

Any reason for this?

[MOON-CLJ]

@mckelvin no particular reason，just think 6 as ratio is large。

[MOON-CLJ]

i will not insist if you think it's not okay。

[mckelvin]

I'd prefer to keep it unchanged. If you think it's essential, I think you should also bump the version later to make users noticed.

[MOON-CLJ]

@mckelvin okay，i'll do it later if i think it's necessary after more observations.

[MOON-CLJ]

@mckelvin updated