search for: reached_ready

...s(+), 4 deletions(-) diff --git a/generator/states.c b/generator/states.c index 202c305..a4eecbd 100644 --- a/generator/states.c +++ b/generator/states.c @@ -110,6 +110,12 @@ send_from_wbuf (struct nbd_connection *conn) /*----- End of prologue. -----*/ /* STATE MACHINE */ { + READY: + conn->reached_ready = true; + if (conn->cmds_to_issue) + SET_NEXT_STATE (%ISSUE_COMMAND.START); + return 0; + DEAD: if (conn->sock) { conn->sock->ops->close (conn->sock); diff --git a/lib/internal.h b/lib/internal.h index 67bd52a..2d6ad9d 100644 --- a/lib/internal.h +++ b/lib/internal....

...Returns true if the connection is connecting or handshaking int nbd_[unlocked_]aio_is_processing (conn) Returns true if the connection is issuing or replying to commands (but not in the READY state) I will post these later this morning. I believe they will let you get rid of the reached_ready flag, as well as making this patch more correct because you probably want to avoid issuing commands on a connection which is closed or dead. You'd use a test like: if (nbd_unlocked_aio_is_ready (conn)) { // call nbd_internal_run } else if (nbd_unlocked_aio_is_processing (conn)) {...

On v1, we discussed whether cmds_to_issue needed to be a list, since it never had more than one element. I played with the idea of making it a list, and allowing the client to queue up new commands regardless of whether the state machine is currently in READY. I also polished up the tmp demo into a bit more full-fledged example file, worth including since it also let me discover a hard-to-hit race