<br><br><b><i>Dotan Barak <dotanb@mellanox.co.il></i></b> wrote:<blockquote class="replbq" style="border-left: 2px solid rgb(16, 16, 255); margin-left: 5px; padding-left: 5px;"> <meta http-equiv="Content-Type" content="text/html; charset=us-ascii"><title>Message</title> <meta content="MSHTML 6.00.3790.2491" name="GENERATOR"> <div> </div> <blockquote style="border-left: 2px solid rgb(0, 0, 255); padding-left: 5px; margin-left: 5px; margin-right: 0px;"> <div></div> <div class="OutlookMessageHeader" dir="ltr" align="left" lang="en-us"><font face="Tahoma" size="2">-----Original Message-----<br><b>From:</b> keshetti mahesh [mailto:k_mahesh85@yahoo.co.in] <br><b>Sent:</b> Monday, July 24, 2006 3:21 PM<br><b>To:</b> Dotan Barak<br><b>Subject:</b> RE: [openib-general] connection loss handling in mthca<br><br></font></div><br><br><b><i>Dotan Barak <dotanb@mellanox.co.il></i></b> wrote: <blockquote class="replbq" style="border-left: 2px solid
rgb(16, 16, 255); padding-left: 5px; margin-left: 5px;"> <div><font color="#0000ff" face="Arial" size="2"></font> </div> <blockquote style="border-left: 2px solid rgb(0, 0, 255); padding-left: 5px; margin-left: 5px; margin-right: 0px;"> <div></div> <div class="OutlookMessageHeader" dir="ltr" align="left" lang="en-us"><font face="Tahoma" size="2">-----Original Message-----<br><b>From:</b> keshetti mahesh [mailto:k_mahesh85@yahoo.co.in] <br><b>Sent:</b> Monday, July 24, 2006 2:40 PM<br><b>To:</b> Dotan Barak<br><b>Subject:</b> Re: [openib-general] connection loss handling in mthca<br><br></font></div><br><br><b><i>Dotan Barak <dotanb@mellanox.co.il></i></b> wrote: <blockquote class="replbq" style="border-left: 2px solid rgb(16, 16, 255); padding-left: 5px; margin-left: 5px;"> <div>Hi.<br><br>On Monday 24 July 2006 13:50, keshetti mahesh wrote:<br>> i have a query regarding the
handling of asynchronous events in mthca driver<br>> consider the situation, receiver has posted some 10 descriptors. and 5 out of them are completd successfully, after that connection is lost( in NIC level) due to some reason<br>> <br>> now,<br>> 1. how do the QP know about this(there is no IB specific event)<br>If the QP was the responder of an RDMA operation which failed, there should be an async event on the QP.<br>> 2. What about the remaining descriptors in the receiver side<br>> are completions will be generated for them <br>In case of an error, the QP state will be changed to error and all the WR (in SQ and RQ) will be flushed (with error)<br><br>where does it happen? in the interrupt handler or ??<br>i have gone through the mthca code <br> 1. there is no IQE or event corresponding to the connection
lose<br> 2. in the interrupt handlers only the event handler corresponding to that QP is called (no QP state change)<br><span class="779425211-24072006"><font color="#0000ff" face="Arial" size="2">[Dotan Barak] </font></span></div> <div><span class="779425211-24072006"><font color="#0000ff" face="Arial" size="2">When there is an error with the QP, the QP state is being changed by the HCA (Automatically).<br><br><br></font></span></div> <div><span class="779425211-24072006"><font color="#0000ff" face="Arial" size="2">The async event event occur only if the operation is an RDMA operation and the QP is the responder,</font></span></div> <div><span class="779425211-24072006"><font color="#0000ff" face="Arial" size="2">there should be completion with error after the QP had the problem (is there are WR in the
QP).</font></span></div> <div><span class="779425211-24072006"><font color="#0000ff" face="Arial" size="2"></font></span> </div> <div><span class="779425211-24072006"><font color="#0000ff" face="Arial" size="2">the event is an affiliated event (only for this QP), so only the event handler of this QP should get this event.</font></span></div> <div><span class="779425211-24072006"><font color="#0000ff" face="Arial" size="2"></font></span> </div> <div><span class="779425211-24072006"><font color="#0000ff" face="Arial" size="2">Dotan</font></span></div> <div><span class="779425211-24072006"><font color="#0000ff" face="Arial" size="2"></font></span> </div><span class="779425211-24072006"></span> <div><br><br>Dotan<br><br><br>ok, now what i can understand is <br> if we consider this case (i.e. connection lose) the HCA will automatically change the state of QP to error.<br>No
async event or error will be generated (this is not RDMA operation) . and <br> a completion with error code (which error code????) will be generated for the completion which is in process and all other outstanding WRs will be flushed.<br>is this OK???<br><br>??with which error status the WR in progress will be completed.<br><br>-Mahesh<br><br><br><span class="131413212-24072006"><font color="#0000ff" face="Arial" size="2">[Dotan Barak] what you understood is correct.</font></span></div> <div><span class="131413212-24072006"><font color="#0000ff" face="Arial" size="2"></font></span> </div> <div><span class="131413212-24072006"><font color="#0000ff" face="Arial" size="2">I cannot tell you the expected status of the completion if i don't know what you are doing </font></span></div> <div><span
class="131413212-24072006"><font color="#0000ff" face="Arial" size="2">(which opcodes do you use, if the QP which go to error is responder or requestor ...).</font></span></div> <div><span class="131413212-24072006"><font color="#0000ff" face="Arial" size="2"></font></span> </div> <div><span class="131413212-24072006"><font color="#0000ff" face="Arial" size="2">the first WR which fails will have a "meaningful" status and the rest of the completions status will be "flushed with error".</font></span></div> <div><span class="131413212-24072006"><font color="#0000ff" face="Arial" size="2"></font></span> </div> <div><span class="131413212-24072006"> <font color="#0000ff" face="Arial" size="2">Dotan</font></span></div></blockquote></blockquote></blockquote></blockquote></blockquote><br>let me put the whole thing again<br>A is the sender(who has posted some 5 descriptors) and B is the receiver(who has
posted the same 5 receive descriptors)<br><br>now the sender(A) HCA has detected the connection lose due to " TPT error for data buffer" on the receiver (B) side then <br>-the receiver(B) will be notified about this through an interrupt(affiliated asynchronous error) ??<br>-upon receiving the interrupt the receiver(B) HCA will transit the state of QP to error<br>-what happens to the WR s in progress at the both ends ..with which code the completion will be generated???<br><br>-Mahesh<br><br><br><p>
<hr size=1></hr>
Find out what India is talking about on <a href="http://us.rd.yahoo.com/mail/in/mailanswersshare/*http://in.answers.yahoo.com/">Yahoo! Answers India.</a>