> Wei Yongjun wrote: > >>> Wei Yongjun wrote: >>> >>> >>> Interesting approach and I applaud for creativity, but it doesn't work if there is >>> data loss. >>> >>> I see only one way out of this mess and that's to remove the SHUTDOWN_PENDING >>> state and make that a modifier/flag that can be set on other states <= ESTABLISHED. >>> >>> >> Do you mean we should support multi send() with EOF flag? >> >> send(msg, SCTP_EOF); >> send(msg, SCTP_EOF); >> ... >> > As long as they are for different associations. SCTP_EOF is a way to do > shutdown() for 1-to-many sockets. If the second send() is for the same > association, it should fail. > > I just realized that this might actually work.... Autoclose timer isn't started > until association is up, so that takes care of any INIT retransmissions. > And the autoclose will do the right thing by entering SHUTDOWN_PENDING and > retransmitting DATA as needed. > > The only thing still missing is something that prevents the above multiple send > scenario. > > That's why I though that removing the SHUTDOWN_PENDING state and changing it into > a bit/flag would work. This way, sends would return failure because even though > we may be in COOKIE_WAIT, if the SHUTDOWN_PENDING flag is set, further writes are > disallowed. > > With this code, multiple writes with EOF for the same association will work, > if the init process takes a long time (due to retransmissions or long rto). > I got it, thank you very much. Wei Yongjun > -vlad > > >> >>> In other words, if the user has queued up data to the association, even if it is >>> not established, we should exhaust all retransmits before failing. >>> >>> >> shutdown after all the outdata is acked? >> >> >>> -vlad >>> >>> >>> >>>> /* If we are already past ASSOCIATE, the lower >>>> * layers are responsible for association cleanup. >>>> */ >>>> >>>> >>> >>> >> > > -- To unsubscribe from this list: send the line "unsubscribe linux-sctp" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
