1 |
Hi Bryan - |
2 |
|
3 |
Thanks! I inserted the rules, and restarted udev. The permissions were |
4 |
wrong on the components, but I manually changed them and it works! |
5 |
|
6 |
Thanks again for your help, |
7 |
Brian |
8 |
|
9 |
On Jan 2, 2008 4:41 PM, Bryan Green <bryan.d.green@××××.gov> wrote: |
10 |
|
11 |
> "Brian Budge" writes: |
12 |
> > Hi Bryan - |
13 |
> > |
14 |
> > I don't seem to have a 40-ib.rules in any of my /etc/udev/rules.d on any |
15 |
> > node. |
16 |
> |
17 |
> Aha. That file is part of sys-cluster/openib-drivers, which you don't |
18 |
> have |
19 |
> installed. You can use the infiniband support that is part of the kernel, |
20 |
> but the driver versions won't match your openib userspace software |
21 |
> versions |
22 |
> (not necessarily a problem), and you'll be missing the startup scripts. |
23 |
> |
24 |
> Older versions of the files are installed by the openib-files ebuild. |
25 |
> That |
26 |
> ebuild is currently incompatible with openib-userspace, though it perhaps |
27 |
> shouldn't be. I guess I could fix that so you could install openib-files. |
28 |
> But if possible, I'd recommend turning off the kernel builtin drivers, and |
29 |
> emerge openib-drivers. |
30 |
> |
31 |
> In the mean time, just installing '40-ib.rules' might help, but I'm not |
32 |
> sure. |
33 |
> |
34 |
> #### /etc/udev/rules.d/40-ib.rules #### |
35 |
> KERNEL=="umad*", NAME="infiniband/%k" |
36 |
> KERNEL=="issm*", NAME="infiniband/%k" |
37 |
> KERNEL=="ucm*", NAME="infiniband/%k", MODE="0666" |
38 |
> KERNEL=="uverbs*", NAME="infiniband/%k", MODE="0666" |
39 |
> KERNEL=="uat", NAME="infiniband/%k", MODE="0666" |
40 |
> KERNEL=="ucma", NAME="infiniband/%k", MODE="0666" |
41 |
> KERNEL=="rdma_cm", NAME="infiniband/%k", MODE="0666" |
42 |
> ######## |
43 |
> |
44 |
> > My /sys/class/infiniband directory contains mthca0, which contains: |
45 |
> > > ls -la /sys/class/infiniband/mthca0/ |
46 |
> > total 0 |
47 |
> > drwxr-xr-x 3 root root 0 Jan 2 20:54 . |
48 |
> > drwxr-xr-x 3 root root 0 Jan 2 20:54 .. |
49 |
> > -r--r--r-- 1 root root 4096 Jan 2 21:07 board_id |
50 |
> > lrwxrwxrwx 1 root root 0 Jan 3 00:01 device -> |
51 |
> > ../../../devices/pci0000:20/0000:20:0a.0/0000:21:00.0 |
52 |
> > -r--r--r-- 1 root root 4096 Jan 2 21:07 fw_ver |
53 |
> > -r--r--r-- 1 root root 4096 Jan 2 21:07 hca_type |
54 |
> > -r--r--r-- 1 root root 4096 Jan 2 21:07 hw_rev |
55 |
> > -rw-r--r-- 1 root root 4096 Jan 2 21:07 node_desc |
56 |
> > -r--r--r-- 1 root root 4096 Jan 2 21:07 node_guid |
57 |
> > -r--r--r-- 1 root root 4096 Jan 2 21:06 node_type |
58 |
> > drwxr-xr-x 3 root root 0 Jan 2 21:07 ports |
59 |
> > lrwxrwxrwx 1 root root 0 Jan 3 00:01 subsystem -> |
60 |
> > ../../../class/infiniband |
61 |
> > -r--r--r-- 1 root root 4096 Jan 2 21:07 sys_image_guid |
62 |
> > --w------- 1 root root 4096 Jan 2 20:54 uevent |
63 |
> |
64 |
> Is this what is shown on the node that does not have '/dev/infiniband'? |
65 |
> |
66 |
> What about '/sys/class/infiniband_verbs/'? |
67 |
> |
68 |
> -- |
69 |
> gentoo-cluster@g.o mailing list |
70 |
> |
71 |
> |