Hi Hal, On 28 Feb 2006 14:41:58 -0500, Hal Rosenstock <[EMAIL PROTECTED]> wrote: > Hi again Fab, > > On Tue, 2006-02-21 at 01:10, Fabian Tillier wrote: > > Specifically, the problem relates to handling a 1X node trying to join > > the broadcast group, and what the retry policy should be. If the > > group already exists at 4X, the join should fail if the SM follows the > > compliance statements in the IB spec. > > Correct; it should fail with ERR_REQ_INVALID > > > Because the code allowed for > > the broadcast group not pre-existing (that is, a join could fail > > because the group wasn't created), it was unclear whether a failure of > > the join indicated that there was a setting incompatibility (1X vs. > > 4X), or just whether the group needed to be created. > > A join to a nonexistent group should return > ERR_REQ_INSUFFICIENT_COMPONENTS. > > Is this sufficient or is more needed ?
Hmm, this should allow the loop to end - an ERR_REQ_INVALID response to the join would break out of the loop. ERR_REQ_INVALID to the create goes to join... I had missed this status code, thanks! I think it will work, and will let you know if it doesn't. Thanks, - Fab _______________________________________________ openib-general mailing list [email protected] http://openib.org/mailman/listinfo/openib-general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
