Hi Michael,<br><br>Thanks for taking a look,<br><br>Would be grateful if you could advise as I'm not a coder and have spotted the following difference in the Free Queue Pair Number procedure between arbel.c and hermon.c<br>
<br>arbel_bit_mask_free ( arbel->qp_inuse, qpn_offset );<br><br>hermon_bit_bit_mask_free ( hermon->qp_inuse, qpn_offset, 1 );<br><br>Is it worth me modify the arbel.c and trying again?<br><br>Regards<br><br>Lee<br>
<br><br><br><div class="gmail_quote">On 19 May 2011 09:34, Michael Brown <span dir="ltr"><<a href="mailto:mbrown@fensystems.co.uk" target="_blank">mbrown@fensystems.co.uk</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"><div><div></div><div>On Wednesday 18 May 2011 20:28:43 Lee Staples wrote:<br>
> Have just reinstalled SCST and built latest ipxe commit<br>
> (c49659c4f26e23f3fc234c2068786872554daa69) with the arbel queue pair<br>
> patch from last year.<br>
><br>
> Would be grateful if you could take a look at the error I'm getting and<br>
> advise as it appears to login successfully to the server<br>
><br>
> Could not open SAN device: Input/output error (<a href="http://ipxe.org/1d714039" target="_blank">http://ipxe.org/1d714039</a>)<br>
> srp boot failedCMRC 0x23354 shutting down<br>
> Arbel 0x215b4 issuing command 0021<br>
> CMRC 0x23354 send error: Operation canceled (<a href="http://ipxe.org/0b1360a0" target="_blank">http://ipxe.org/0b1360a0</a>)<br>
<br>
</div></div>Login is handled via a pair of management datagrams. The RC queue pair is not<br>
used until the first packet after login completes. Your client.log shows that<br>
the RC queue pair is transitioning into an error state as soon as the first<br>
send WQE is posted:<br>
<br>
QPN 0xd75405 context before doorbell:<br>
Arbel 0x215b4 issuing command 0022<br>
Arbel 0x215b4 QPN 0xd75405 context:<br>
00000000 : 30 00 19 00 ff 3e 3f 16-9f 0a 13 00 00 00 00 01<br>
...<br>
QPN 0xd75405 context after doorbell:<br>
Arbel 0x215b4 issuing command 0022<br>
Arbel 0x215b4 QPN 0xd75405 context:<br>
00000000 : 60 00 19 00 f8 3c 28 14-9f 0a 13 00 00 00 00 01<br>
...<br>
<br>
The first nibble of this hex dump is the queue state - 3="ready to send",<br>
6="error".<br>
<br>
Something must be wrong with either the queue pair context or the send WQE<br>
created by iPXE for Arbel. I have examined both in excruciating detail the<br>
last time I worked on this, and I couldn't find any problem.<br>
<br>
<br>
Itay: is there a debug version of the firmware available that would provide<br>
some indication of why the QP is transitioning to ERR?<br>
<font color="#888888"><br>
Michael<br>
</font></blockquote></div><br>