[ofa-general] Problem while running ib tests
Karun Sharma
karun.sharma at qlogic.com
Tue Sep 22 22:52:42 PDT 2009
I am referring to ib_send_bw, ib_read_bw, ib_write_bw etc.
Usage:
ib_read_bw start a server and wait for connection
ib_read_bw <host> connect to server at <host>
Options:
-p, --port=<port> listen on/connect to port <port> (default 18515)
-d, --ib-dev=<dev> use IB device <dev> (default first device found)
-i, --ib-port=<port> use port <port> of IB device (default 1)
-m, --mtu=<mtu> mtu size (256 - 4096. default for hermon is 2048)
-o, --outs=<num> num of outstanding read/atom(default 4)
-s, --size=<size> size of message to exchange (default 65536)
-a, --all Run sizes from 2 till 2^23
-t, --tx-depth=<dep> size of tx queue (default 100)
-n, --iters=<iters> number of exchanges (at least 2, default 1000)
-u, --qp-timeout=<timeout> QP timeout, timeout value is 4 usec * 2 ^(timeout), default 14
-S, --sl=<sl> SL (default 0)
-x, --gid-index=<index> test uses GID with GID index taken from command line (for RDMAoE index should be 0)
-b, --bidirectional measure bidirectional bandwidth (default unidirectional)
-V, --version display version number
-e, --events sleep on CQ events (default poll)
-F, --CPU-freq do not fail even if cpufreq_ondemand module is loaded
Regards
Karun
-----Original Message-----
From: Ira Weiny [mailto:weiny2 at llnl.gov]
Sent: Tuesday, September 22, 2009 9:26 PM
To: Karun Sharma
Cc: Sneha Mistry; general at lists.openfabrics.org
Subject: Re: [ofa-general] Problem while running ib tests
On Tue, 22 Sep 2009 04:54:22 -0500
Karun Sharma <karun.sharma at qlogic.com> wrote:
> Please use "-F" option while running the tests. It will ignore the "Conflicting CPU frequency" errors. You will still see these messages on your screen, but with "-F", you will also see the results.
Which commands are you referring to with the "-F" option?
I just did a pull from git://git.openfabrics.org/~mst/perftest.git and I don't see a -F for the commands there. Have these tools moved? Or are you speaking of other tools?
I also run into this problem and I have resorted to turning off cpuspeed.
Thanks,
Ira
>
> Regards
> Karun
>
> -----Original Message-----
> From: general-bounces at lists.openfabrics.org [mailto:general-bounces at lists.openfabrics.org] On Behalf Of Sneha Mistry
> Sent: Tuesday, September 22, 2009 2:35 PM
> To: general at lists.openfabrics.org
> Subject: [ofa-general] Problem while running ib tests
>
> Hi,
>
> I have two Dual port HCAa card.Which I have installed in
> same PC.
>
> I am using OpenSuse 10.3 and installed OFED 1.4.
>
> If I try to run any IB bandwidth test or latency test it end us with warning
> "Conflicting CPU frequency values detected: 2394.000000 != 1596.000000".
>
> Is it an OFED related problem or Linux kernel problem.
>
> HCA are connected back to back and output of ibstat is as given below.
> ---------------------------------------------------------------------------------------------------------
> CA 'mthca0'
> CA type: MT25208 (MT23108 compat mode)
> Number of ports: 2
> Firmware version: 4.8.200
> Hardware version: 20
> Node GUID: 0x0002c90200283734
> System image GUID: 0x0002c90200283737
> Port 1:
> State: Active
> Physical state: LinkUp
> Rate: 20
> Base lid: 1
> LMC: 0
> SM lid: 1
> Capability mask: 0x02510a6a
> Port GUID: 0x0002c90200283735
> Port 2:
> State: Initializing
> Physical state: LinkUp
> Rate: 20
> Base lid: 0
> LMC: 0
> SM lid: 0
> Capability mask: 0x02510a68
> Port GUID: 0x0002c90200283736
> CA 'mthca1'
> CA type: MT25208 (MT23108 compat mode)
> Number of ports: 2
> Firmware version: 4.8.200
> Hardware version: 20
> Node GUID: 0x0002c90200283730
> System image GUID: 0x0002c90200283733
> Port 1:
> State: Initializing
> Physical state: LinkUp
> Rate: 20
> Base lid: 0
> LMC: 0
> SM lid: 0
> Capability mask: 0x02510a68
> Port GUID: 0x0002c90200283731
> Port 2:
> State: Active
> Physical state: LinkUp
> Rate: 20
> Base lid: 3
> LMC: 0
> SM lid: 1
> Capability mask: 0x02510a68
> Port GUID: 0x0002c90200283732
> --------------------------------------------------------------------------------------------------------------------------------------
>
> Even ibibnetdiscover is detecting one hope.
> So there is nothing wrong if I try to test 2 HCA connected back to back.
>
> Please tell me what can be done for running ib tests.
>
> Thanks,
> SGM
> _______________________________________________
> general mailing list
> general at lists.openfabrics.org
> http://*lists.openfabrics.org/cgi-bin/mailman/listinfo/general
>
> To unsubscribe, please visit http://*openib.org/mailman/listinfo/openib-general
> _______________________________________________
> general mailing list
> general at lists.openfabrics.org
> http://*lists.openfabrics.org/cgi-bin/mailman/listinfo/general
>
> To unsubscribe, please visit http://*openib.org/mailman/listinfo/openib-general
>
--
Ira Weiny
Math Programmer/Computer Scientist
Lawrence Livermore National Lab
925-423-8008
weiny2 at llnl.gov
More information about the general
mailing list