why It will not OK in future, Do you have some point about that?<br><br><div><span class="gmail_quote">On 03 Mar 2006 08:27:49 -0500, <b class="gmail_sendername">Hal Rosenstock</b> <<a href="mailto:halr@voltaire.com">halr@voltaire.com
</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">On Fri, 2006-03-03 at 07:55, Devesh Sharma wrote:<br>> Hi Hal,<br>> Thanks for replying.
<br>><br>> On 03 Mar 2006 06:29:50 -0500, Hal Rosenstock <<a href="mailto:halr@voltaire.com">halr@voltaire.com</a>><br>> wrote:<br>> Hi Devesh,<br>><br>> On Thu, 2006-03-02 at 23:31, Devesh Sharma wrote:
<br>> > On 02 Mar 2006 08:35:04 -0500, Hal Rosenstock<br>> <<a href="mailto:halr@voltaire.com">halr@voltaire.com</a>><br>> > wrote:<br>> > Hi Devesh,<br>> >
<br>>
> On Thu, 2006-03-02
at 08:03, Devesh Sharma wrote:<br>> > > Hi All,<br>>
> > I have
another query regarding Opensm,<br>>
> > What is MLID
assignment policy?<br>>
> > Is there any
porvision that MLID assigned by<br>> Opensm may<br>> > always remain<br>>
> > in 0xC001
0xC0FF range, In case by underliying<br>> hardware only<br>> > 255<br>>
> > seprate
multicast groups are supported at a time?<br>> ><br>>
> OpenSM does not
overlay multiple groups (MGIDs) with<br>> the same<br>>
> characteristics on
the same MLID. It uses unique<br>> MLIDs per<br>> > group.<br>> ><br>> > This is fine, each group will have unique MLID though they<br>> have same<br>
> > characteristics, I am trying to know is,<br>> > as in unicast LID assignment there is a file maintained by<br>> SM, so for<br>> > MLID is there any such mapping file
<br>><br>> No.<br>><br>> > or this is maintained in some internal structure,<br>><br>> The MLIDs in use are maintained in an internal structure.<br>><br>> > or every create request with new MGID will be given One
<br>> higher MLID<br>> > from MLID assigned to previous create request and as the<br>> > switchInfo::MulticastFDBCap is reached, create request will<br>> fail?<br>><br>
> Sort of. MLIDs are also returned when the group is reclaimed<br>> some time<br>> after it is deleted (last full member leaves). That leaves<br>> holes in the<br>> table so it is not a simple increment. Requests fail when a
<br>> new group<br>> creation would cause there to be MLIDs in use greater than the<br>> lowest<br>> SwitchInfo:MulticastFDBCap in the subnet.<br>><br>> I'm not sure what your concern is here.
<br>><br>> Here I am finding the possibility of whether 16 bit MLID can be mapped<br>> on a 8 bit number, which is my requirement.<br><br>That depends on the size of the MFTs in the switches (and may be OK<br>today but not some time in the future).
<br><br>-- Hal<br><br>> Well thanks again for replying.<br>><br>> Devesh<br>><br>> -- Hal<br>><br>>
> MLIDs start at
0xC000 and are constrained by the<br>> least capable<br>> > switch<br>>
> (in terms of
SwitchInfo::MulticastFDBCap).<br>> ><br>>
> Are you just
asking or are you having some issue in<br>> this area<br>> > ?<br>> ><br>> > I am just asking about this.<br>> ><br>> > -- Hal
<br>> ><br>> ><br>> ><br>><br>><br><br></blockquote></div><br>