SDP kernel modules do use rdma CMA. You can get a little info from the kernel module by setting debug: echo 1 > /sys/modules/ib_sdp/debug_level Use dmesg to read the output. This might give you enough info to get to the root of the problem.
Before you start adding printf()s, you can rebuild with CONFIG_INFINIBAND_SDP_DEBUG_DATA defined. I just stick a #define at the top of sdp.h. Then: echo 1 > /sys/modules/ib_sdp/data_debug_level JIm -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Steve Wise Sent: Tuesday, January 08, 2008 12:39 PM To: Craig Prescott Cc: [email protected] Subject: Re: [ofa-general] SDP and iWARP Craig Prescott wrote: > Steve Wise wrote: >> >> >> Craig Prescott wrote: >>> >>> Hi; >>> >>> Before the holidays, I tried a netperf SDP_STREAM >>> test between a pair of Chelsio S310-SR iWARP cards >>> using OFED 1.2, which crashed the netperf client >>> host. Using OFED-1.3-20080107-0942 on both hosts >>> the crash no longer happens, which is great, but it >>> isn't working quite yet: >> >> Hey Craig, >> >> SDP currently isn't suppored over the Chelsio RNICs. No testing has >> been done. >> > > Hi Steve; > > I understand - just following up on the short thread from > before the holidays where I started looking at this. > That was sooo last year. :) > I'm interested to try to make it SDP work anyway on these RNICs > for a project we are working on. I am willing to put in some > time to work on this and (if successful) do some testing and > report back (if anyone is interested), and would be grateful > for any advice or help from this list. > Ok. First make sure the sdp kernel module uses the rdma cma. Then I'd add printk hooks in cma.c, addr.c, and iwcm.c to see what's going on and where things are failing. Also a wire trace is good if we're getting that far (like at least doing arp resolution). I haven't looked at the sdp code at all, so you'll have to do the leg work. Thanks! Steve. > Cheers, > Craig _______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general _______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
