<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=us-ascii">
<META content="MSHTML 6.00.2900.2963" name=GENERATOR></HEAD>
<BODY>
<DIV dir=ltr align=left><FONT face=Arial size=2>
<DIV><FONT face=Arial><FONT size=2><SPAN class=562051118-31102006>
<DIV><SPAN class=562051118-31102006></SPAN><FONT face=Arial><FONT size=2><FONT
color=#0000ff><FONT face=Tahoma><STRONG>From:</STRONG> Chris
Youb<BR><B>Sent:</B> Tuesday, October 31, 2006 12:13 PM<BR><B>To:</B>
openib-general@openib.org<BR><B>Subject:</B> [openib-general] OFED SRP initiator
always sends CM REJ in response to CM
REP</FONT><BR><BR></FONT></FONT></FONT><FONT size=2><FONT
color=#0000ff>Abstract: <BR> We are developing SRP target code and testing
it with the OFED 1.1 SRP initiator. The OFED SRP initiator sends us a CM
REQ (IB 12.6.5) and we respond with CM REP (12.6.8). However, instead of
the expected CM RTU (12.6.9) we ALWAYS receive a CM REJ (12.6.7) with status
0x1C == Reason 28 (12.6.7.2). <BR><BR>Software Setup: <BR>- SUSE 10.0 <BR>- OFED
1.1 <BR>- Mellanox card with 3.5.00 firmware <BR><BR>Details:
<BR> Initially we suspected our response values in the CM REP packet.
There is nothing obvious to us, and anything we weren't sure about we tried a
number of combinations. This applies to the SRP private data as well.
<BR> We also took a look at .<I class=moz-txt-slash><SPAN
class=moz-txt-tag>/</SPAN>openib-1.1/drivers/infiniband/core<SPAN
class=moz-txt-tag>/</SPAN></I> cm.c and cma.c. The function
cma.c:cma_rep_recv looked like a possibility but there's little debug
output. Aside from putting in printk's and recompiling and installing is
there an easier way to debug? <BR><BR> </FONT></DIV></FONT>
<DIV><FONT color=#0000ff size=2>Chris,</FONT></DIV>
<DIV><FONT color=#0000ff size=2></FONT> </DIV>
<DIV><FONT size=+0><FONT size=+0><FONT size=2><FONT color=#0000ff>I<SPAN
class=562051118-31102006> reviewed the packets and found at least 1
problem.</SPAN></FONT></FONT></FONT></FONT></DIV>
<DIV><FONT size=+0><FONT size=+0><FONT color=#0000ff size=2><SPAN
class=562051118-31102006></SPAN></FONT></FONT></FONT> </DIV>
<DIV><SPAN class=562051118-31102006><FONT color=#0000ff>The REQ has responder
resources=0x4, yet the REP has initiator depth=7. The REP must provide an
initiator depth <= the responder resources in the REQ<SPAN
class=390521218-31102006>.</SPAN></FONT></SPAN></DIV>
<DIV><SPAN class=562051118-31102006><FONT color=#0000ff><SPAN
class=390521218-31102006></SPAN></FONT></SPAN> </DIV>
<DIV><SPAN class=562051118-31102006><FONT color=#0000ff><SPAN
class=390521218-31102006>Some other non-fatal issues:</SPAN></FONT></SPAN></DIV>
<DIV><SPAN class=562051118-31102006><FONT color=#0000ff><SPAN
class=390521218-31102006>Target Ack Delay is a bit high (0x1f -> 2.4
hours). This will basically cause initiator to compute QP Ack timeouts of
2.4 hours. This value should represent the time internal to the CA from
receipt of a message to sending of the ACK. See IBTA 12.7.33 for more
info.</SPAN></FONT></SPAN></DIV>
<DIV><FONT size=+0><FONT size=+0><FONT color=#0000ff size=2><SPAN
class=562051118-31102006></SPAN></FONT></FONT></FONT> </DIV>
<DIV><FONT size=+0><FONT size=+0><FONT color=#0000ff size=2><SPAN
class=562051118-31102006>Todd
Rimmer</SPAN></FONT></FONT></FONT></DIV></SPAN></FONT></FONT></DIV></FONT></DIV>
<DIV> </DIV>
<DIV class=Section1><FONT face=Arial color=#0000ff
size=2></FONT> </DIV></BODY></HTML>