<br><font size=2 face="sans-serif">Michael,</font>
<br>
<br><font size=2><tt>> > > Which FW revision do you have?<br>
> > ><br>
> > The "ibstat" command shows:<br>
> > <br>
> > CA type: MT25204<br>
> > Number of ports: 1<br>
> > Firmware version: 1.0.800<br>
> > Hardware version: a0<br>
> > Node GUID: 0x0002c90200216dc4<br>
> > System image GUID: 0x0002c90200216dc7<br>
> > <br>
> > -Don Albert-<br>
> > <br>
> <br>
> Yes, that's the latest revision. Hmm.<br>
> <br>
</tt></font>
<br><font size=2><tt>What about the other thing I mentioned in my first
message: the "lspci" command complains about the board
slot that the HCA is plugged into:</tt></font>
<br>
<br><font size=2 face="Courier New"> pcilib: Resource 2 in
/sys/bus/pci/devices/0000:03:00.0/resource has a 64-bit address, ignoring</font>
<br><font size=2 face="Courier New"> ....</font>
<br><font size=2 face="Courier New"> 03:00.0 InfiniBand: Mellanox
Technologies MT25204 [InfiniHost III Lx HCA] (rev 20)</font>
<br><font size=2><tt><br>
I also found out that on this machine the HCA is plugged into a 16X PCI-e
slot, which is different than the other machine which is working, where
the slot is 8X.</tt></font>
<br>
<br><font size=2><tt>Bear in mind, however, that both machines were previously
working with the 2.6.9-34 kernel with the backport patches and the OpenIB
svn 6500 code. Did something happen in 2.6.16, or am I missing a
patch?</tt></font>
<br>
<br><font size=2><tt> -Don Albert-</tt></font>
<br>