[openib-general] RMPP Message Format Errors

Eitan Zahavi eitan at mellanox.co.il
Fri Sep 16 12:24:53 PDT 2005


Hal Rosenstock wrote:
> Hi Eitan,
> 
> On Fri, 2005-09-16 at 06:02, Hal Rosenstock wrote:
> 
>>On Thu, 2005-09-15 at 04:01, Eitan Zahavi wrote:
>>
>>>Now lets go back to the test:
>>>
>>>I use a machine connected through a single switch (IS3) to itself.
>>>
>>>I use osmtest -f c to get Nodes,Ports and PathRecords from the SM.
> 
> 
>>>Now I go to the SA client section:
>>>
>>> From osmtest log I see:
>>>
>>>NodeInfoRec:
>>>Aug 21 14:46:56 [4017F6C0] -> __osmv_send_sa_req: Waiting for async
> 
> event.
> 
>>>Aug 21 14:46:56 [40D87BB0] -> osm_mad_pool_get: [
>>>Aug 21 14:46:56 [40D87BB0] -> osm_vendor_get: [
>>>Aug 21 14:46:56 [40D87BB0] -> osm_vendor_get: Acquiring UMAD for
> 
> p_madw = 0x807b8a4, size = 256.
> 
>>>Aug 21 14:46:56 [40D87BB0] -> osm_vendor_get: Acquired UMAD
> 
> 0x807c198, size = 256.
> 
>>>Aug 21 14:46:56 [40D87BB0] -> osm_vendor_get: ]
>>>Aug 21 14:46:56 [40D87BB0] -> osm_mad_pool_get: Acquired p_madw =
> 
> 0x807b898, p_mad = 0x807c1d0, size = 256.
> 
>>>Aug 21 14:46:56 [40D87BB0] -> osm_mad_pool_get: ]
>>>Aug 21 14:46:56 [40D87BB0] -> __osmv_sa_mad_rcv_cb: [
>>>Aug 21 14:46:56 [40D87BB0] -> __osmv_sa_mad_rcv_cb: Count = 1 = 200
> 
> / 112 (88)
> 
>>>Aug 21 14:46:56 [40D87BB0] -> osmtest_query_res_cb: [
>>>Aug 21 14:46:56 [40D87BB0] -> osmtest_query_res_cb: ]
>>>Aug 21 14:46:56 [40D87BB0] -> __osmv_sa_mad_rcv_cb: ]
>>>I wonder how come the received MAD is only of 256 bytes. I expected
> 
> it to be of headers + data = 56 + 336 = 392byte.
> 
> I see the problem: The 200 byte response is a short RMPP packet and the
> length is not being properly handled in the OpenIB OpenSM vendor layer.
> Patch for this shortly.
But I see the error also on multi segment RMPP transaction...
The __osmv_sa_mad_rcv_cb reports it got 256 byte mad.
> 
> BTW, are you running with OpenSM from osm-1.8.0-merge or the trunk for
> this ?
No I'm using the trunk! I was happy to delete my copy of the merge branch.
> 
> -- Hal
> 




More information about the general mailing list