Intel NetEffect NE020 E10G81GP use

Hi all, I have tested two cards Intel NetEffect NE020 E10G81GP with RDMA over Ethernet; results are OK in latency and bandwidth. Are there other protocols supported by cards,that offer the same performances? (TCP Offload, STORAGE, MPI). Socket Direct Protocol (SDO) is not good for those

[PATCH for-2.6.38 1/2] IB/ipoib: remove LRO support

towards moving from LRO to GRO, revert commit af40da894e9 IPoIB: add LRO support. Also elimiate the ethtool set_flags callback which isn't needed anymore, and have the declaration of restart_syscall to be imported through direct inclusion of linux/sched.h CC: Ben Hutchings

[PATCH for-2.6.38 2/2] IB/ipoib: add GRO support

add GRO support Signed-off-by: Or Gerlitz ogerl...@voltaire.com --- tested with 2.6.37-rc5 drivers/infiniband/ulp/ipoib/ipoib_cm.c |1 + drivers/infiniband/ulp/ipoib/ipoib_ib.c |2 +- drivers/infiniband/ulp/ipoib/ipoib_main.c |2 ++ 3 files changed, 4 insertions(+), 1

Re: ibnetdiscover issue

Tom Ammon wrote: Eli, Is there a quick workaround we could put in place? I want to map out our fabric, and I especially need the spine GUIDs on the GD4200 because I'm going to be doing up/down routing and want to specify the root GUIDs. I can also submit a support case to Voltaire, if you

Re: opensm configuration question

Ira, Thanks! I misunderstood up/down, thanks for clearing that up. That helps a lot. I set the console to local, and then tried telnetting to localhost port 1 but got a connection refused. Should I set it to socket instead? Tom On 12/14/2010 05:12 PM, Ira Weiny wrote: On Tue, 14 Dec

rdma provider module references

Hey Roland, I notice that if I have a user rdma application running that has an rdma connection using iw_cxgb3, then the iw_cxgb3 module reference count is bumped and thus it cannot be unloaded. However when I have an NFSRDMA connection that utilizes iw_cxgb3, the module reference count is

Re: opensm configuration question

On Wed, 15 Dec 2010 08:12:46 -0800 Tom Ammon tom.am...@utah.edu wrote: Ira, Thanks! I misunderstood up/down, thanks for clearing that up. That helps a lot. I set the console to local, and then tried telnetting to localhost port 1 but got a connection refused. Should I set it to

Re: rdma provider module references

I notice that if I have a user rdma application running that has an rdma connection using iw_cxgb3, then the iw_cxgb3 module reference count is bumped and thus it cannot be unloaded. However when I have an NFSRDMA connection that utilizes iw_cxgb3, the module reference count is not

Re: [PATCH for-2.6.38 1/2] IB/ipoib: remove LRO support

Thanks for doing this Or, I've been meaning to switch from LRO to GRO myself. One question: have the declaration of restart_syscall to be imported through direct inclusion of linux/sched.h Is this because linux/inet_lro.h indirectly included linux/sched.h, and we're losing that include? -

Re: [PATCH for-2.6.38 1/2] IB/ipoib: remove LRO support

Roland Dreier rdre...@cisco.com wrote: Is this because linux/inet_lro.h indirectly included linux/sched.h, and we're losing that include? Yes, see http://www.spinics.net/lists/netdev/msg149809.html -- To unsubscribe from this list: send the line unsubscribe linux-rdma in the body of a message

Re: rdma provider module references

On 12/15/2010 11:09 AM, Roland Dreier wrote: I notice that if I have a user rdma application running that has an rdma connection using iw_cxgb3, then the iw_cxgb3 module reference count is bumped and thus it cannot be unloaded. However when I have an NFSRDMA connection that

Add check for libpthread during configure

Add a check for libpthread during librdmacm configure. This will add libpthread to the list of libraries that librdmacm is linked to. Currently librdmacm gets libpthread implicitly through libibverbs, but this breaks when using a linker that does not implicitly link with such dependencies; eg the

[PATCH] infiniband-diags: Verify timeout value specified to diagnostics

Verify timeout value specified to diagnostics Signed-off-by: Ira Weiny wei...@hype137.llnl.gov --- infiniband-diags/src/ibdiag_common.c | 10 +++--- 1 files changed, 7 insertions(+), 3 deletions(-) diff --git a/infiniband-diags/src/ibdiag_common.c b/infiniband-diags/src/ibdiag_common.c

RE: Add check for libpthread during configure

thanks - applied -- 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

RE: Intel NetEffect NE020 E10G81GP use

Hi Andrea, I think you meant RDMA over TCP/IP (iWarp) and not RDMA over Converged Ethernet (RoCE) as Intel NetEffect cards do not support the latter. You may safely try to run any MPI benchmarks using MVAPICH2, Open MPI or Intel MPI. There is more information on iWarp, Intel NetEffect cards

Re: Possible bug on SRP device delete

On Mon, 2010-12-13 at 17:32 +0100, torn5 wrote: I have been suggested to the RDMA mailing list as it might be an ib_srp bug: 5) If I echo one line of ibsrpdm -c to /sys/class/infiniband_srp/srp-mthca0-1/add_target so to add a srp disk to my system, what