Re: [Fwd: Re: [openib-general] Solaris IPoIB MTU with OpenSM]

2005-03-15 Thread Hal Rosenstock
Hi Nitin,

On Tue, 2005-03-15 at 16:15, Nitin Hande wrote:
> This is cool, I have got Solaris IPoIB happily working with the
> OpenSM now. It plumbs, pings and snoops on 0x pkey. 

Great. That's good news. I'll work on a real fix for this now.

> On other hand, on my linux node, if I try to use 8001 partition and
> configure IB interface with IP addr (same time while ib0 is using 0x
> pkey), I get the following error, you may want to investigate that
> 
> [EMAIL PROTECTED] ~]# echo 0x8001 > /sys/class/net/ib0/create_child
> [EMAIL PROTECTED] ~]# ifconfig ib0.8001 10.10.1.1
> [EMAIL PROTECTED]: multicast join failed for
> ff12:401b:8001:0:0:0::, status -22
>  ~]# ib0.8001: multicast join failed for ff12:401b:8001:0:0:0::,
> status -22

I will look into this but I suspect this is caused by the response to
some request in the join "flow" to be more than 1 RMPP packet. Remember
that OpenSM is currently hamstrung in this manner until there is
sufficient RMPP for SA GetTableResps.

Thanks.

-- Hal

___
openib-general mailing list
openib-general@openib.org
http://openib.org/mailman/listinfo/openib-general

To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general


Re: [Fwd: Re: [openib-general] Solaris IPoIB MTU with OpenSM]

2005-03-15 Thread Roland Dreier
Nitin> On other hand, on my linux node, if I try to use 8001
Nitin> partition and configure IB interface with IP addr (same
Nitin> time while ib0 is using 0x pkey), I get the following
Nitin> error, you may want to investigate that

I think this is probably an OpenSM issue (does OpenSM support multiple
partitions?).  On my fabric, running Topspin's embedded SM on a
switch, I can do:

# modprobe ib_ipoib
# echo 0x8001 > /sys/class/net/ib0/create_child
# ifconfig ib0.8001 up

on both systems.  On system #1 I have:

# ifconfig ib0.8001
ib0.8001  Link encap:UNSPEC  HWaddr 
00-13-04-06-FE-80-00-00-00-00-00-00-00-00-00-00
  inet6 addr: fe80::202:c901:7fc:c711/64 Scope:Link
  UP BROADCAST RUNNING MULTICAST  MTU:2044  Metric:1
  RX packets:0 errors:0 dropped:0 overruns:0 frame:0
  TX packets:4 errors:0 dropped:0 overruns:0 carrier:0
  collisions:0 txqueuelen:128
  RX bytes:0 (0.0 b)  TX bytes:300 (300.0 b)

and on system #2 I'm able to do:

# ping6 -I ib0.8001 fe80::202:c901:7fc:c711
PING fe80::202:c901:7fc:c711(fe80::202:c901:7fc:c711) from 
fe80::202:c901:78c:e461 ib0.8001: 56 data bytes
64 bytes from fe80::202:c901:7fc:c711: icmp_seq=1 ttl=64 time=4.56 ms
64 bytes from fe80::202:c901:7fc:c711: icmp_seq=2 ttl=64 time=0.077 ms
64 bytes from fe80::202:c901:7fc:c711: icmp_seq=3 ttl=64 time=0.065 ms

 - R.
___
openib-general mailing list
openib-general@openib.org
http://openib.org/mailman/listinfo/openib-general

To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general


Re: [Fwd: Re: [openib-general] Solaris IPoIB MTU with OpenSM]

2005-03-15 Thread Nitin Hande
Hal,

On Fri, 2005-03-04 at 12:53, Hal Rosenstock wrote:
> Hi again Nitin,
> 
> Finally got a chance to work on this. I have a workaround for you for
> now. Real patch later... Let me know if this does the trick for you. It
> did for me.
> 
> -- Hal
> 
> Index: osm_sa_mcmember_record.c
> ===
> --- osm_sa_mcmember_record.c  (revision 1953)
> +++ osm_sa_mcmember_record.c  (working copy)
> @@ -1522,9 +1522,11 @@
>if ((IB_MCR_COMPMASK_PROXY & comp_mask) &&
>(p_rcvd_rec->proxy_join != p_mgrp->mcmember_rec.proxy_join)) goto Exit;
>  
> +#if 0
>/* if defined MUST match exactly !*/
>if ((IB_MCR_COMPMASK_MTU_SEL & comp_mask) &&
>((p_rcvd_rec->mtu >> 6) != (p_mgrp->mcmember_rec.mtu >> 6))) goto Exit;
> +#endif
>  
>if ((IB_MCR_COMPMASK_MTU & comp_mask) &&
>((p_rcvd_rec->mtu & 0x3F) != (p_mgrp->mcmember_rec.mtu & 0x3F))) goto 
> Exit;
This is cool, I have got Solaris IPoIB happily working with the OpenSM
now. It plumbs, pings and snoops on 0x pkey. Here is some output:

[EMAIL PROTECTED] ~]# cat /etc/path_to_inst | grep ibd
"/[EMAIL PROTECTED],60/[EMAIL PROTECTED]/pci15b3,[EMAIL PROTECTED]/[EMAIL 
PROTECTED],,ipib" 0 "ibd"
"/[EMAIL PROTECTED],60/[EMAIL PROTECTED]/pci15b3,[EMAIL PROTECTED]/[EMAIL 
PROTECTED],,ipib" 1 "ibd"
[EMAIL PROTECTED] ~]# ifconfig ibd0
ibd0: flags=1000843 mtu 2044 index
3
inet 192.168.100.111 netmask ff00 broadcast 192.168.100.255
ipib 0:0:0:16:fe:80:0:0:0:0:0:0:0:2:c9:1:9:76:51:d1 
[EMAIL PROTECTED] ~]# ping 192.168.100.112
192.168.100.112 is alive
[EMAIL PROTECTED] ~]# snoop -d ibd1
192.168.100.112 -> *ARP C Who is 192.168.100.111,
192.168.100.111 ?
192.168.100.111 -> 192.168.100.112 ARP R 192.168.100.111,
192.168.100.111 is 0:0:0:16:fe:80:0:0:0:0:0:0:0:2:c9:1:9:76:51:d1
192.168.100.111 -> 192.168.100.112 ICMP Echo request (ID: 641 Sequence
number: 0)
192.168.100.112 -> 192.168.100.111 ICMP Echo reply (ID: 641 Sequence
number: 0)

This is fantastic. Thanks Hal !..

BTW, I have not tested it with multiple GetTable reponse - RMPP packet.
 
On other hand, on my linux node, if I try to use 8001 partition and
configure IB interface with IP addr (same time while ib0 is using 0x
pkey), I get the following error, you may want to investigate that

[EMAIL PROTECTED] ~]# echo 0x8001 > /sys/class/net/ib0/create_child
[EMAIL PROTECTED] ~]# ifconfig ib0.8001 10.10.1.1
[EMAIL PROTECTED]: multicast join failed for
ff12:401b:8001:0:0:0::, status -22
 ~]# ib0.8001: multicast join failed for ff12:401b:8001:0:0:0::,
status -22
[EMAIL PROTECTED] ~]# ib0.8001: multicast join failed for
ff12:401b:8001:0:0:0::, status -22
0.8001: multicast join failed for ff12:401b:8001:0:0:0::, status
-22
0.8001: multicast join failed for ff12:401b:8001:0:0:0::, status
-22
b0.8001: multicast join failed for ff12:401b:8001:0:0:0::,
status -22
b0.8001: multicast join failed for ff12:401b:8001:0:0:0::,
status -22
b0.8001: multicast join failed for ff12:401b:8001:0:0:0::,
status -22
b0.8001: multicast join failed for ff12:401b:8001:0:0:0::,
status -22
0.8001: multicast join failed for ff12:401b:8001:0:0:0::, status
-22
b0.8001: multicast join failed for ff12:401b:8001:0:0:0::,
status -22
b0.8001: multicast join failed for ff12:401b:8001:0:0:0::,
status -22
b0.8001: multicast join failed for ff12:401b:8001:0:0:0::,
status -22
b0.8001: multicast join failed for ff12:401b:8001:0:0:0::,
status -22
0.8001: multicast join failed for ff12:401b:8001:0:0:0::, status
-22
b0.8001: multicast join failed for ff12:401b:8001:0:0:0::,
status -22

Thanks
Nitin



> 
> 
> -Forwarded Message-
> 
> From: Hal Rosenstock <[EMAIL PROTECTED]>
> To: Nitin Hande <[EMAIL PROTECTED]>
> Cc: openib , Tom Duffy <[EMAIL PROTECTED]>
> Subject: Re: [openib-general] Solaris IPoIB MTU with OpenSM
> Date: 24 Feb 2005 08:42:23 -0500
> 
> Hi Nitin,
> 
> On Wed, 2005-02-23 at 17:19, Nitin Hande wrote:
> > Hal, 
> > 
> > [comments below]
> > On Wed, 2005-02-23 at 02:19, Hal Rosenstock wrote:
> > > On Tue, 2005-02-22 at 22:56, Nitin Hande wrote:
> > > > So I tried the latest patches and preliminarily things seem to be
> > > > working fine. 
> > > 
> > > Yipee.
> > [snip..]
> > > 
> > > > 
> > > > So after this test above, I try to run snoop on the solaris interface
> > > > and get the following error message from the layer below IPoIB:
> > > > 
> > > > Feb 22 19:50:25 dongon.SFBay.Sun.COM ibd: [ID 517869 kern.info] NOTICE:
> > > > ibd0: HCA GUID 0002c901097651d0 port 1 PKEY  Could not get list of
> > > > IBA multicast groups
> > > > 
> > > > My preliminary assumption is that OpenSm is not returning the list of
> > > > multicast groups that the ibd interface has joined. I will look at the
> > > > MAD's tomorrow and try to ascertain that.
> > > 
> > > How does S10 request this ? Re