Hi,
On Dec 16, 2012, at 9:32 PM, Hal Rosenstock wrote:
> Hi,
>
> On 12/16/2012 7:03 AM, Jens Domke wrote:
>> Hello Hal,
>>
>> On Dec 15, 2012, at 5:44 AM, Hal Rosenstock wrote:
>>
>>> Hi,
>>>
>>> On 12/14/2012 3:32 PM, Jens Domke wrote:
>>>> Hello Hal,
>>>>
>>>> On Dec 15, 2012, at 3:58 AM, Hal Rosenstock wrote:
>>>>
>>>>> Hi,
>>>>>
>>>>> On 12/14/2012 1:24 PM, Jens Domke wrote:
>>>>>> Hello Hal,
>>>>>>
>>>>>> On Dec 15, 2012, at 1:42 AM, Hal Rosenstock wrote:
>>>>>>
>>>>>>> Hi again,
>>>>>>>
>>>>>>> On 12/14/2012 10:17 AM, Jens Domke wrote:
>>>>>>>> Hello Hal,
>>>>>>>>
>>>>>>>> thank you for the fast response. I will try to clarify some points.
>>>>>>>>
>>>>>>>>>> d) OpenMPI runs are executed with "--mca
>>>>>>>>>> btl_openib_ib_path_record_service_level 1"
>>>>>>>>>
>>>>>>>>> I'm not familiar with what DFSSSP does to figure out SLs exactly but
>>>>>>>>> there should be no need to set this. The proper SL for querying the SA
>>>>>>>>> for PathRecords, etc. is always in PortInfo.SMSL. In the case of
>>>>>>>>> DFSSSP
>>>>>>>>> (and other QoS based routing algorithms), it calculates that and the
>>>>>>>>> SM
>>>>>>>>> pushes this into each port. That should be used. It's possible that
>>>>>>>>> SL1
>>>>>>>>> is not a valid SL for port <-> SA querying using DFSSSP.
>>>>>>>> The OpenMPI parameter btl_openib_ib_path_record_service_level does not
>>>>>>>> specify the SL for querying the PathRecords.
>>>>>>>> It just enables the functionality. And the ompi processes use the
>>>>>>>> PortInfo.SMSL to send the request.
>>>>>>>> For the request "port -> SA" every 0<=SL<=7 was used in the test, and
>>>>>>>> the SA received the requests.
>>>>>>>>>
>>>>>>>>>> e) kernel 2.6.32-220.13.1.el6.x86_64
>>>>>>>>>>
>>>>>>>>>> As far as I understand the whole system:
>>>>>>>>>> 1. the OMPI processes are sending MAD requests
>>>>>>>>>> (SubnAdmGet:PathRecord) to the OpenSM
>>>>>>>>>> 2. the SA receives the request on QP1
>>>>>>>>>
>>>>>>>>> There is the SL in the query itself. This should be the SMSL that the
>>>>>>>>> SM
>>>>>>>>> set for that port.
>>>>>>>> Hmm, there you might have a point. I think I saw that the query itself
>>>>>>>> had SL=0 specified.
>>>>>>>> In fact OpenMPI sets everthing to 0 except for slid and dlid.
>>>>>>>>>
>>>>>>>>>> 3. SA asks the routing algorithm (like LASH, DFSSSP or Torus_2QoS)
>>>>>>>>>> about a special service level for the slid/dlid path
>>>>>>>>>
>>>>>>>>> This is a (potentially) different SL (for MPI<->MPI port
>>>>>>>>> communication)
>>>>>>>>> than the one the query used and is the one returned inside the
>>>>>>>>> PathRecord attribute/data.
>>>>>>>> Yes, it can be different, but DFSSSP sets the same SL, because the SM
>>>>>>>> is running on a port which is also used for MPI comm.
>>>>>>>
>>>>>>> With DFSSSP are all SLs same from source port to get to any destination
>>>>>>> ?
>>>>>> No, not necessarily. In general DFSSSP does not enforce SL(LID1->LID2)
>>>>>> == SL(LID2->LID1) or SL(LID1->LID2) == SL(LID1->LID3).
>>>>>
>>>>> If SL(LID1->LID2) != SL(LID2->LID1), that's not a reversible path.
>>>> True. But i don't think that the SA asks the DFSSSP routing about the SL
>>>> for the reversible path.
>>>> So, the SA could use any SL which is a valid SL, even if the DFSSSP would
>>>> recommend another SL.
>>>>
>>>> I just read the IB Specs and it says, that "SL specified in the received
>>>> packet is used as the SL in the response packet" for MAD packets.
>>>> So, its most likely, that there is a mismatch in the way how OMPI does the
>>>> setup of the PathRequest and the way how the SA does build the respond
>>>> packet.
>>>> OMPI always specifies SL=0 (lets say SL_a) inside of the PathRequest
>>>> packet,
>>>
>>> So CompMask in the query has the SL bit on and SL is set to 0 inside the
>>> SubAdmGet of PatchRecord ?
>>
>> No, the CompMask didn't had the SL bit and the SL was set to 0.
>
> That means the SL in the request is wildcarded so the SA/SM fills in a
> valid one in the response.
Ok.
>
>> I tried to follow the path of the SL bit (IB_PR_COMPMASK_SL) and the only
>> reference I found was in osm_sa_path_record.c
>> The SA just treats the SL in the PathRequest as a "I would like to use this
>> SL" in case the SL bit is set.
>> But the routing engine can overwrite the requested SL before the reply is
>> send.
>>
>> Nevertheless, I have changed the code of OMPI so that it sets the SL bit in
>> the CompMask and sets the SL to SMSL for the PathRequest, so that SL_a ==
>> SL_b.
>> Sadly, the reply send by the SA does not leave the node (for SL_b>0). Only
>> if I change the SL to 0 in the MAD right before umad_send is called by the
>> SA, the paket is able to leave the node and reaches the OMPI process.
>
> Are you sure the response doesn't leave the SA node or it's not received
> at the requester (OMPI node) ?
No, I'm not sure. Is there any possibility to check that? As far as I know,
ibdump does not show MAD pakets which leave a port, it only shows the pakets
when they are received on the other end.
>
>>
>>>
>>>> and sends the packet on SL_b (PortInfo.SMSL).
>>>
>>> Good.
>>>
>>>> The SA uses p_mad_addr->addr_type.gsi.service_level, which is SL_b, for
>>>> the response.
>>>> If SL_b is not 0, then the packet can't reach the OMPI process. Right?
>>>
>>> Depends. It may be that both SLs work but maybe not.
>>>
>>>> If I analyse this correctly, then there are two bugs. One is in OMPI, that
>>>> it does not specify the SL within the PathRequest in a appropriate way
>>>> (which would be a SL suggested by DFSSSP for the reversible path). And the
>>>> second bug is that the SA uses the SL, on which the PathRequest packet was
>>>> send, and not the SL specified within the packet.
>>>> What do you think?
>>>
>>> Yes, it might be better to wildcard the SL in the query. The only
>>> scenario that would fail with the query you are making if there's no SL
>>> 0 path between the src/dest LIDs or GIDs in the OMPI PathRecord query.
>>> If that's the case, SA should return MAD status 0xc (status code 3 -
>>> ERR_NO_RECORDS). But the response doesn't make it back to the requester
>>> OMPI node so it's not even getting that far.
>>
>> Yes, exactly. So, do you have an idea why the response hands in the SA node?
>> I have no inside of the underlying layer (kernel driver and fireware). Maybe
>> there are some implementations, which prevent the SA from sending MADs back
>> on SL>0?
>
> If you're sure this response doesn't get out of the SA node, please
> contact Mellanox support with the details.
Ok, I can do this, if it turns out to be true.
>
>>>
>>>> I can try to change the PathRequest of OMPI tomorrow, so that it matches
>>>> addr_type.gsi.service_level.
>>>> Maybe, with this change the packets of the SA will reach the OMPI process
>>>> on a SL>0.
>>>>>
>>>>>>>
>>>>>>>>>
>>>>>>>>>> 4. SA sends the PathRecord back to the OMPI process via umad_send in
>>>>>>>>>> libvendor/osm_vendor_ibumad.c
>>>>>>>>>
>>>>>>>>> By the response reversibility rule, I think this is returned on the SL
>>>>>>>>> of the original query but haven't verified this in the code base yet.
>>>>>>>> Ok, I was not aware of that rule. But if this is true, then the SA
>>>>>>>> should also be able to send via SL>0.
>>>>>>>
>>>>>>> I doubled checked and indeed the SA response does use the SL that the
>>>>>>> incoming request was received on.
>>>>>>>
>>>>>>>>>
>>>>>>>>>> The osm_vendor_send() function builds the MAD packet with the
>>>>>>>>>> following attributes:
>>>>>>>>>> /* GS classes */
>>>>>>>>>> umad_set_addr_net(p_vw->umad, p_mad_addr->dest_lid,
>>>>>>>>>> p_mad_addr->addr_type.gsi.remote_qp,
>>>>>>>>>> p_mad_addr->addr_type.gsi.service_level,
>>>>>>>>>> IB_QP1_WELL_KNOWN_Q_KEY);
>>>>>>>>>> So, the SL is the same like the one which was used by the OMPI
>>>>>>>>>> process. The Q_Key matches the Q_key on the OMPI process, and
>>>>>>>>>> remote_qp and dest_lid is correct, too.
>>>>>>>>>> Afterwards umad_send(…) is used to send the reply with the
>>>>>>>>>> PathRecord, and this send does not work (except for SL=0).
>>>>>>>>>
>>>>>>>>> By not working, what do you mean ? Do you mean it's not received at
>>>>>>>>> the
>>>>>>>>> requester with no message in the OpenSM log or not received at the
>>>>>>>>> OpenSM or something else ? It could be due to the wrong SL being used
>>>>>>>>> in
>>>>>>>>> the original request (forcing it to SL 1). That could cause it not to
>>>>>>>>> be
>>>>>>>>> received at the SM or the response not to make it back to the
>>>>>>>>> requester
>>>>>>>>> from the SA if the SL used is not "reversible".
>>>>>>>> By "not working" I mean, that the MPI process does not receive any
>>>>>>>> response from the SA.
>>>>>>>> I get messages from the MPI process like the following:
>>>>>>>> [rc011][[14851,1],1][connect/btl_openib_connect_sl.c:301:get_pathrecord_info]
>>>>>>>> No response from SA after 20 retries
>>>>>>>> The log of OpenSM shows that the SA received the PathRequest query,
>>>>>>>> dumps the query into the log, and sends the reply back.
>>>>>>>> And I think I was some messages in the log about "…1 outstanding MAD…".
>>>>>>>>>
>>>>>>>>>> If I look into the MAD before it is send, then it looks like this:
>>>>>>>>>> Breakpoint 2, umad_send (fd=9, agentid=2, umad=0x7fffe8012530,
>>>>>>>>>> length=120, timeout_ms=0, retries=3)
>>>>>>>>>> at src/umad.c:791
>>>>>>>>>> 791 if (umaddebug > 1)
>>>>>>>>>> (gdb) p *mad
>>>>>>>>>> $1 = {agent_id = 2, status = 0, timeout_ms = 0, retries = 3, length
>>>>>>>>>> = 0, addr = {qpn = 1325427712, qkey = 384,
>>>>>>>>>> lid = 4096, sl = 6 '\006', path_bits = 0 '\000', grh_present = 0
>>>>>>>>>> '\000', gid_index = 0 '\000',
>>>>>>>>>> hop_limit = 0 '\000', traffic_class = 0 '\000', gid = '\000'
>>>>>>>>>> <repeats 15 times>, flow_label = 0,
>>>>>>>>>> pkey_index = 0, reserved = "\000\000\000\000\000"}, data =
>>>>>>>>>> 0x7fffe8012530 "\002"}
>>>>>>>>>
>>>>>>>>> Is this the PathRecord query on the OpenMPI side or the response on
>>>>>>>>> the
>>>>>>>>> OpenSM side ? SL is 6 rather than 1 here.
>>>>>>>> This is the response on the OpenSM side (inside the umad_send
>>>>>>>> function, right before it is written to the device with write(fd, …).
>>>>>>>> SL=6 indicates, that the MPI process was sending the request on SL 6.
>>>>>>>
>>>>>>> What is SMSL for the requester ? Was it SL 6 ?
>>>>>> Yes, it was SL 6.
>>>>>> Here is a content of a similar packet which was received by the SA. I
>>>>>> have used ibdump on the port where the OpenSM was running:
>>>>>> ======================================================================================
>>>>>> No. Time Source Destination Protocol
>>>>>> Length Info
>>>>>> 785 14.352168 LID: 384 LID: 4140 InfiniBand
>>>>>> 290 UD Send Only SubnAdmGet(PathRecord)
>>>>>>
>>>>>> Frame 785: 290 bytes on wire (2320 bits), 290 bytes captured (2320 bits)
>>>>>> Arrival Time: Dec 13, 2012 18:09:44.437633332 JST
>>>>>> Epoch Time: 1355389784.437633332 seconds
>>>>>> [Time delta from previous captured frame: 4.332020528 seconds]
>>>>>> [Time delta from previous displayed frame: 4.332020528 seconds]
>>>>>> [Time since reference or first frame: 14.352168681 seconds]
>>>>>> Frame Number: 785
>>>>>> Frame Length: 290 bytes (2320 bits)
>>>>>> Capture Length: 290 bytes (2320 bits)
>>>>>> [Frame is marked: False]
>>>>>> [Frame is ignored: False]
>>>>>> [Protocols in frame: erf:infiniband]
>>>>>> Extensible Record Format
>>>>>> [ERF Header]
>>>>>> Timestamp: 0x50c99b587008bcf2
>>>>>> [Header type]
>>>>>> .001 0101 = type: INFINIBAND (21)
>>>>>> 0... .... = Extension header present: 0
>>>>>> 0000 0100 = flags: 4
>>>>>> .... ..00 = capture interface: 0
>>>>>> .... .1.. = varying record length: 1
>>>>>> .... 0... = truncated: 0
>>>>>> ...0 .... = rx error: 0
>>>>>> ..0. .... = ds error: 0
>>>>>> 00.. .... = reserved: 0
>>>>>> record length: 306
>>>>>> loss counter: 0
>>>>>> wire length: 290
>>>>>> InfiniBand
>>>>>> Local Route Header
>>>>>> 0110 .... = Virtual Lane: 0x06
>>>>>> .... 0000 = Link Version: 0
>>>>>> 0110 .... = Service Level: 6
>>>>>> .... 00.. = Reserved (2 bits): 0
>>>>>> .... ..10 = Link Next Header: 0x02
>>>>>> Destination Local ID: 19
>>>>>> 0000 0... .... .... = Reserved (5 bits): 0
>>>>>> .... .000 0100 1000 = Packet Length: 72
>>>>>> Source Local ID: 16
>>>>>> Base Transport Header
>>>>>> Opcode: 100
>>>>>> 1... .... = Solicited Event: True
>>>>>> .1.. .... = MigReq: True
>>>>>> ..00 .... = Pad Count: 0
>>>>>> .... 0000 = Header Version: 0
>>>>>> Partition Key: 65535
>>>>>> Reserved (8 bits): 0
>>>>>> Destination Queue Pair: 0x000001
>>>>>> 0... .... = Acknowledge Request: False
>>>>>> .000 0000 = Reserved (7 bits): 0
>>>>>> Packet Sequence Number: 0
>>>>>> DETH - Datagram Extended Transport Header
>>>>>> Queue Key: 2147549184
>>>>>> Reserved (8 bits): 0
>>>>>> Source Queue Pair: 0x00380050
>>>>>> MAD Header - Common Management Datagram
>>>>>> Base Version: 0x01
>>>>>> Management Class: 0x03
>>>>>> Class Version: 0x02
>>>>>> Method: Get() (0x01)
>>>>>> Status: 0x0000
>>>>>> Class Specific: 0x0000
>>>>>> Transaction ID: 0x0010000f38005000
>>>>>> Attribute ID: 0x0035
>>>>>> Reserved: 0x0000
>>>>>> Attribute Modifier: 0x00000000
>>>>>> MAD Data Payload:
>>>>>> 000000000000000000000000000000000000000000000000...
>>>>>> Illegal RMPP Type (0)!
>>>>>> RMPP Type: 0x00
>>>>>> RMPP Type: 0x00
>>>>>> 0000 .... = R Resp Time: 0x00
>>>>>> .... 0000 = RMPP Flags: Unknown (0x00)
>>>>>> RMPP Status: (Normal) (0x00)
>>>>>> RMPP Data 1: 0x00000000
>>>>>> RMPP Data 2: 0x00000000
>>>>>> SMASubnAdmGet(PathRecord)
>>>>>> SM_Key (Verification Key): 0x0000000000000000
>>>>>> Attribute Offset: 0x0000
>>>>>> Reserved: 0x0000
>>>>>> Component Mask: 0x0000003000000000
>>>>>> Attribute (PathRecord)
>>>>>> PathRecord
>>>>>> DGID: :: (::)
>>>>>> SGID: ::0.15.0.16 (::0.15.0.16)
>>>>>> DLID: 0x0000
>>>>>> SLID: 0x0000
>>>>>> 0... .... = RawTraffic: 0x00
>>>>>> .... 0000 0000 0000 0000 0000 = FlowLabel: 0x000000
>>>>>> HopLimit: 0x00
>>>>>> TClass: 0x00
>>>>>> 0... .... = Reversible: 0x00
>>>>>> .000 0000 = NumbPath: 0x00
>>>>>> P_Key: 0x0000
>>>>>> .... .... .... 0000 = SL: 0x0000
>>>>>> 00.. .... = MTUSelector: 0x00
>>>>>> ..00 0000 = MTU: 0x00
>>>>>> 00.. .... = RateSelector: 0x00
>>>>>> ..00 0000 = Rate: 0x00
>>>>>> 00.. .... = PacketLifeTimeSelector: 0x00
>>>>>> ..00 0000 = PacketLifeTime: 0x00
>>>>>> Preference: 0x00
>>>>>> Variant CRC: 0xad4e
>>>>>> ======================================================================================
>>>>>
>>>>> And the SubnAdmGetResp(PathRecord) is not seen ? If not, it doesn't get
>>>>> out that machine and the issue is internal to that machine. It could be
>>>>> because of the underlying issue which hangs OpenSM when some IB program
>>>>> tried to unregister from the MAD layer but there were outstanding work
>>>>> completions. That's based on your original email earlier this AM.
>>>> No, the SubnAdmGetResp does not show up, if I use ibdump on the OMPI side
>>>> and the SA uses a SL>0.
>>>
>>> Can ibdump be used to capture output on the SM port ?
>>
>> Yes, that works quite well, despite the warning in the ibdump manual.
>> But I have started ibdump before opensm, maybe that makes a difference, not
>> sure.
>>
>> Regards,
>> Jens
>>
>> PS: I have seen a small bug. Not sure if its a bug in wireshark or ibdump,
>> but the response received by the OMPI node isn't shown correctly. The
>> PathRecord contains an offset which is either missing in the dump or is not
>> treated correctly be wireshark. But it causes wireshark to show the
>> PathRecord data with wrong values.
>> Maybe you could redirect this to the developer of ibdump, so that he can
>> check/fix it.
>
> Are you referring to the fields after the SA AttributeOffset or
> something else ?
Yes, after the SMASubnAdmGet Attribute Offset. Here an example:
I get on the OMPI side:
SMASubnAdmGetResp(PathRecord)
SM_Key (Verification Key): 0x0000000000000000
Attribute Offset: 0x0008
Reserved: 0x0000
Component Mask: 0x0000803000000000
Attribute (PathRecord)
PathRecord
DGID: ::8:f104:399:ebb5:fe80:0 (::8:f104:399:ebb5:fe80:0)
SGID: ::8:f104:399:ecd5:4:8 (::8:f104:399:ecd5:4:8)
DLID: 0x0000
SLID: 0x0000
0... .... = RawTraffic: 0x00
.... 0000 1000 0000 1111 1111 = FlowLabel: 0x0080ff
HopLimit: 0xff
TClass: 0x00
0... .... = Reversible: 0x00
.000 0011 = NumbPath: 0x03
P_Key: 0x8486
.... .... .... 0000 = SL: 0x0000
00.. .... = MTUSelector: 0x00
..00 0000 = MTU: 0x00
00.. .... = RateSelector: 0x00
..00 0000 = Rate: 0x00
00.. .... = PacketLifeTimeSelector: 0x00
..00 0000 = PacketLifeTime: 0x00
Preference: 0x00
But it should show (see the difference in SLID, DLID, SL which are now correct):
SMASubnAdmGetResp(PathRecord)
SM_Key (Verification Key): 0x0000000000000000
Attribute Offset: 0x0008
Reserved: 0x0000
Component Mask: 0x0000803000000000
Attribute (PathRecord)
PathRecord
DGID: ::8:f104:399:ebb5 (::8:f104:399:ebb5)
SGID: fe80::8:f104:399:ecd5 (fe80::8:f104:399:ecd5)
DLID: 0x0004
SLID: 0x0008
0... .... = RawTraffic: 0x00
.... 0000 0000 0000 0000 0000 = FlowLabel: 0x000000
HopLimit: 0x00
TClass: 0x00
1... .... = Reversible: 0x01
.000 0000 = NumbPath: 0x00
P_Key: 0xffff
.... .... .... 0011 = SL: 0x0003
10.. .... = MTUSelector: 0x02
..00 0100 = MTU: 0x04
10.. .... = RateSelector: 0x02
..00 0110 = Rate: 0x06
10.. .... = PacketLifeTimeSelector: 0x02
..01 0010 = PacketLifeTime: 0x12
Preference: 0x00
Regards,
Jens
>
> -- Hal
>
>>>
>>> -- Hal
>>>
>>>>>
>>>>>>>
>>>>>>> One would need to walk the SLToVLMappingTables from requester (OMPI
>>>>>>> port) to SA and back to see whether SL6 would even have a chance of
>>>>>>> working (not dropping) aside from whether it's really the correct SL to
>>>>>>> use.
>>>>>> All SL2VL tables look the same. I checked the output of OpenSM.
>>>>>> SL: | 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
>>>>>> 13 | 14 | 15 |
>>>>>> VL: | 0x0 |0x1 |0x2 |0x3 |0x4 |0x5 |0x6 |0x7 |0x0 |0x1 |0x2 |0x3 |0x4
>>>>>> |0x5 |0x6 |0x7 |
>>>>>> But this is also as expected, because I have set the QoS in the opensm
>>>>>> config as follows:
>>>>>> qos_sl2vl 0,1,2,3,4,5,6,7,0,1,2,3,4,5,6,7
>>>>>> This was set for "default", "CA" and "Switch external ports". I have not
>>>>>> touched the config for "Switch Port 0" and "Router ports", they
>>>>>> remained: qos_[sw0 | rtr]_sl2vl (null)
>>>>>
>>>>> That works as long as all links have (at least) 8 data VLs (VLCap 4).
>>>> Yes, all VL_CAP show 4 in the OpenSM log file.
>>>>
>>>> Regards
>>>> Jens
>>>>
>>>>
>>>>
>>>>>
>>>>> -- Hal
>>>>>
>>>>>> Regards
>>>>>> Jens
>>>>>>
>>>>>>>
>>>>>>> -- Hal
>>>>>>>
>>>>>>>>>
>>>>>>>>>> The output of OpenMPI or OpenSM's log file don't show any useful
>>>>>>>>>> information for this problem, even with higher debug levels.
>>>>>>>>>
>>>>>>>>> So nothing interesting logged relative to the PathRecord queries ?
>>>>>>>> In the OpenSM log, only that it was received, how the request looks
>>>>>>>> like, and that it was send back.
>>>>>>>> And a few "outstanding MADs" a few lines later in the log.
>>>>>>>>>
>>>>>>>>>> So, right now I'm stuck, and have no idea if there is an error in
>>>>>>>>>> the kernel driver, the HCA firmware or something completely
>>>>>>>>>> different. Or if umad_send basically does not support SL>0.
>>>>>>>>>> A workaround for the moment is to set the SL in the
>>>>>>>>>> umad_set_addr_net(...) call to 0.
>>>>>>>>>
>>>>>>>>> So SL 0 works between all nodes and SA for querying/responses. Wonder
>>>>>>>>> if
>>>>>>>>> that's how SMSL is set by DFSSSP.
>>>>>>>> No, the SMSL set by DFSSSP is different from 0, I have checked this.
>>>>>>>> In our case (OpenSM running on a compute node), it sets the same SL,
>>>>>>>> which is used
>>>>>>> for MPI<->MPI traffic, to ensure deadlock freedom.
>>>>>>>>
>>>>>>>> Regards
>>>>>>>> Jens
>>>>>>>>
>>>>>>>> --------------------------------
>>>>>>>> Dipl.-Math. Jens Domke
>>>>>>>> Researcher - Tokyo Institute of Technology
>>>>>>>> Satoshi MATSUOKA Laboratory
>>>>>>>> Global Scientific Information and Computing Center
>>>>>>>> 2-12-1-E2-7 Ookayama, Meguro-ku,
>>>>>>>> Tokyo, 152-8550, JAPAN
>>>>>>>> Tel/Fax: +81-3-5734-3876
>>>>>>>> E-Mail: domke.j...@m.titech.ac.jp
>>>>>>>> --------------------------------
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>> --
>>>>>>> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
>>>>>>> the body of a message to majord...@vger.kernel.org
>>>>>>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>>>>>>
>>>>>> --------------------------------
>>>>>> Dipl.-Math. Jens Domke
>>>>>> Researcher - Tokyo Institute of Technology
>>>>>> Satoshi MATSUOKA Laboratory
>>>>>> Global Scientific Information and Computing Center
>>>>>> 2-12-1-E2-7 Ookayama, Meguro-ku,
>>>>>> Tokyo, 152-8550, JAPAN
>>>>>> Tel/Fax: +81-3-5734-3876
>>>>>> E-Mail: domke.j...@m.titech.ac.jp
>>>>>> --------------------------------
>>>>>>
>>>>>>
>>>>>
>>>>> --
>>>>> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
>>>>> the body of a message to majord...@vger.kernel.org
>>>>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>>>>
>>>> --------------------------------
>>>> Dipl.-Math. Jens Domke
>>>> Researcher - Tokyo Institute of Technology
>>>> Satoshi MATSUOKA Laboratory
>>>> Global Scientific Information and Computing Center
>>>> 2-12-1-E2-7 Ookayama, Meguro-ku,
>>>> Tokyo, 152-8550, JAPAN
>>>> Tel/Fax: +81-3-5734-3876
>>>> E-Mail: domke.j...@m.titech.ac.jp
>>>> --------------------------------
>>>>
>>>>
>>>
>>> --
>>> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
>>> the body of a message to majord...@vger.kernel.org
>>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>>
>> --------------------------------
>> Dipl.-Math. Jens Domke
>> Researcher - Tokyo Institute of Technology
>> Satoshi MATSUOKA Laboratory
>> Global Scientific Information and Computing Center
>> 2-12-1-E2-7 Ookayama, Meguro-ku,
>> Tokyo, 152-8550, JAPAN
>> Tel/Fax: +81-3-5734-3876
>> E-Mail: domke.j...@m.titech.ac.jp
>> --------------------------------
>>
>>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majord...@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
