Re: [openib-general] smpdump and current MAD layer

2004-12-02 Thread Sean Hefty
Hal Rosenstock wrote: So solicited MAD responses cannot currently be snooped nor can unsolicited ones for which an agent is registered (Since SMA and PMA are currently firmware based, the latter is not an issue for the current implementation). I've gotten a start on adding in the snooping support.

[openib-general] mthca Page Allocation Failures

2004-12-02 Thread Hal Rosenstock
Hi, I'm seeing this a lot more lately. Not sure when this started. It happens on first starting the driver after boot. It seems to occur when there is more going on at system startup (like fixing the disk after a previous freeze). -- Hal Dec 2 14:36:17 localhost kernel: ib_mthca: Mellanox

Re: [openib-general] smpdump and current MAD layer

2004-12-02 Thread Sean Hefty
Hal Rosenstock wrote: I'd like to place the snooping code in as few places as possible, but still be able to snoop locally processed MADs. Ideally a MAD should be snooped exactly once, which requires some extra care when handling QP errors. Snooping in the completion handling allows the MAD

[openib-general] [PATCH] Add vendor OUI support to MAD layer

2004-12-02 Thread Hal Rosenstock
mad: Add vendor OUI support to MAD layer Also, in ib_register_mad_agent, if registration request supplied, make sure that class supplied in request is consistent with the QP type. On sending, it is the responsibility of the client to put the OUI in the MAD when sending a vendor MAD with OUI