Re: [Qemu-devel] [PATCH 02/20] nbd: correctly propagate errors

[Paolo Bonzini <pbonzini@redhat.com>]

Il 10/02/2014 08:38, Fam Zheng ha scritto:
>> >      if (s->client.is_unix) {
>> > -        sock = unix_socket_outgoing(qemu_opt_get(s->socket_opts, "path"));
>> > +        sock = unix_connect(qemu_opt_get(s->socket_opts, "path"), errp);
> Why not use unix_connect_opts like below?

Right!

>> >      } else {
>> > -        sock = tcp_socket_outgoing_opts(s->socket_opts);
>> > +        sock = inet_connect_opts(s->socket_opts, errp, NULL, NULL);
>> >          if (sock >= 0) {
>> >              socket_set_nodelay(sock);
>> >          }
>> > @@ -255,17 +251,19 @@ static int nbd_open(BlockDriverState *bs, QDict *options, int flags,
>> >      BDRVNBDState *s = bs->opaque;
>> >      char *export = NULL;
>> >      int result, sock;
>> > +    Error *local_err = NULL;
>> >
>> >      /* Pop the config into our state object. Exit if invalid. */
>> > -    result = nbd_config(s, options, &export);
>> > -    if (result != 0) {
>> > -        return result;
>> > +    nbd_config(s, options, &export, &local_err);
>> > +    if (local_err) {
> Isn't error_is_set() better here?

No, error_is_set(&foo) is the same as foo != NULL.

So we use error_is_set only with Error**, which really should never 
happen because you miss errors if the errp is NULL. :)

Paolo

>> > +        error_propagate(errp, local_err);
>> > +        return -EINVAL;
>> >      }