[openib-general] Failed multicast join with new multicast module

[Hal Rosenstock]

Hi Sean,

I just (finally) updated to include the new multicast module (I'm at the
latest svn) and I see one anomaly.

When a multicast join fails (properly) for a group (e.g. status 0x0600),
it appears to be continually retried and it never gives up. The join
request is being denied for some IPv6 groups as they are not previously
created. In about 100 usecs after the ERR_REQ_INSUFFICIENT_COMPONENTS
status is returned, the module appears to rerequest and never give up.

I forget exactly what the strategy for this was before the multicast
module was introduced: whether it was exponential backoff up to some
limit, or whether it was linear up to some retry count.

Also, in looking at the new multicast code, I see the following:

static int retry_timer = 5000; /* 5 sec */
module_param(retry_timer, int, 0444);
MODULE_PARM_DESC(retry_timer, "Time in ms between retried requests.");

static int retries = 3;
module_param(retries, int, 0444);
MODULE_PARM_DESC(retries, "Number of times to retry a request.");

so it appears that the multicast module has it's own retry strategy. Is
that true ? If so, does this interact with IPoIB's for rerequesting or
has that changed ? 

Thanks.

-- Hal