<br><font size=2 face="sans-serif">Roland,</font>
<br>
<br><font size=2 face="sans-serif">Here is what I mean (partial output
of "ibv_devinfo -v"):</font>
<br>
<br><font size=2 face="sans-serif">port: 1<br>
state: PORT_ACTIVE
(4)<br>
max_mtu: 2048 (4)<br>
active_mtu: 2048 (4)<br>
sm_lid: 4<br>
port_lid: 128<br>
port_lmc: 0x02<br>
max_msg_sz: 0x0<br>
port_cap_flags: 0x00000000<br>
max_vl_num: 0<br>
bad_pkey_cntr: 0x0<br>
qkey_viol_cntr: 0x0<br>
sm_sl: 0<br>
pkey_tbl_len: 1<br>
gid_tbl_len: 1<br>
subnet_timeout: 8<br>
init_type_reply: 0<br>
active_width: 12X (8)<br>
active_speed: 2.5 Gbps (1)<br>
phys_state: invalid physical
state (0)<br>
GID[ 0]: fe80:0000:0000:0003:0002:5500:1002:1b3d<br>
<br>
port: 2<br>
state: PORT_ACTIVE
(4)<br>
max_mtu: 2048 (4)<br>
active_mtu: 2048 (4)<br>
sm_lid: 4<br>
port_lid: 136<br>
port_lmc: 0x02<br>
max_msg_sz: 0x0<br>
port_cap_flags: 0x00000000<br>
max_vl_num: 0<br>
bad_pkey_cntr: 0x0<br>
qkey_viol_cntr: 0x0<br>
sm_sl: 0<br>
pkey_tbl_len: 1<br>
gid_tbl_len: 1<br>
subnet_timeout: 8<br>
init_type_reply: 0<br>
active_width: 12X (8)<br>
active_speed: 2.5 Gbps (1)<br>
phys_state: invalid physical
state (0)<br>
GID[ 0]: fe80:0000:0000:0002:0002:5500:1002:1b7d</font>
<br>
<br><font size=2 face="sans-serif">The phys_state above does not correspond
to what is in </font>
<br><font size=2 face="sans-serif">/sys/class/infiniband/ehca0/ports/1
</font>
<br>
<br><font size=2 face="sans-serif">However, the "phys state"
that ibstatus displays is indeed correct and as expected (see below).</font>
<br>
<br>
<br><font size=2 face="sans-serif">ibstatus ehca0<br>
Infiniband device 'ehca0' port 1 status:<br>
default gid: fe80:0000:0000:0000:0002:5500:0000:933d<br>
base lid: 0xb<br>
sm lid: 0x2<br>
state: 4:
ACTIVE<br>
phys state: 0: <unknown><br>
rate: 30
Gb/sec (12X)<br>
<br>
Infiniband device 'ehca0' port 2 status:<br>
default gid: fe80:0000:0000:0000:0002:5500:0000:937d<br>
base lid: 0x7<br>
sm lid: 0x2<br>
state: 4:
ACTIVE<br>
phys state: 0: <unknown><br>
rate: 30
Gb/sec (12X)</font>
<br>
<br><font size=2 face="sans-serif">The fix in dev_info.c for this problem
is elementary (replace "invalid physical state" with "unknown").
Since the sources for this</font>
<br><font size=2 face="sans-serif">program occurred in the examples directory
I presumed that this was just a sample program written as an example. That</font>
<br><font size=2 face="sans-serif">was why I asked if it was "supported"
or not.</font>
<br>
<br><font size=2 face="sans-serif">Pradeep</font>
<br><font size=2 face="sans-serif">pradeep@us.ibm.com</font>
<br>
<br><tt><font size=2>Roland Dreier <rdreier@cisco.com> wrote on 10/05/2006
10:53:41 AM:<br>
<br>
> Pradeep> On eHCA we find that there are some discrepencies
(for<br>
> Pradeep> example the phys_state) between the outputs
of<br>
> Pradeep> "ibv_devinfo -v" and the corresponding
output of<br>
> Pradeep> ibstatus. Indeed ibstatus does display what
is in the<br>
> Pradeep> /sys/class/infiniband/* files, and as expected.<br>
> <br>
> Could you give some details on the differences? Is ibv_devinfo
or<br>
> ibstatus giving the correct output?<br>
> <br>
> Pradeep> I would like to understand the status of
ibv_devinfo. Is<br>
> Pradeep> this a supported program -given that the
sources for this<br>
> Pradeep> is in the example directory? Also,
I find the source<br>
> Pradeep> (devinfo.c) does appear in the OFED 1.0
tar ball.<br>
> <br>
> I'm not sure what "supported" would mean exactly, but certainly
I<br>
> would like ibv_devinfo to work as well as possible. Unfortunately
I<br>
> can't fix anything without a more detailed report...<br>
> <br>
> - R.<br>
</font></tt>