From mboxrd@z Thu Jan 1 00:00:00 1970 From: Yevgeny Kliteynik Subject: Re: [PATCH] opensm/osm_mcast_mgr.c: fix osm_mcast_mgr_compute_max_hops for managed switch Date: Wed, 23 Dec 2009 16:29:45 +0200 Message-ID: <4B322959.10500@dev.mellanox.co.il> References: <4B2F4405.9000300@dev.mellanox.co.il> <20091222212348.GF26940@me> Reply-To: kliteyn-LDSdmyG8hGV8YrgS2mwiifqBs+8SCbDb@public.gmane.org Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20091222212348.GF26940@me> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Sasha Khapyorsky Cc: Linux RDMA List-Id: linux-rdma@vger.kernel.org Hi Sasha, On 22/Dec/09 23:23, Sasha Khapyorsky wrote: > Hi Yevgeny, > > On 11:46 Mon 21 Dec , Yevgeny Kliteynik wrote: >> If a certain mcast group has only one member, and this >> member is a switch, OSM fails to select this switch >> when trying to create spanning tree for this mcast group: >> >> ERR 0A08: Unable to locate a suitable switch for group 0xC000 >> ERR 0A17: Unable to create spanning tree (IB_ERROR) for mlid 0xc000 >> >> Signed-off-by: Yevgeny Kliteynik >> --- >> opensm/opensm/osm_mcast_mgr.c | 16 ++++++++++------ >> 1 files changed, 10 insertions(+), 6 deletions(-) >> >> diff --git a/opensm/opensm/osm_mcast_mgr.c b/opensm/opensm/osm_mcast_mgr.c >> index 697fb58..16f578b 100644 >> --- a/opensm/opensm/osm_mcast_mgr.c >> +++ b/opensm/opensm/osm_mcast_mgr.c >> @@ -204,6 +204,12 @@ static float osm_mcast_mgr_compute_max_hops(osm_sm_t * sm, cl_qlist_t * l, >> >> OSM_LOG_ENTER(sm->p_log); >> >> + if (cl_qlist_count(l) == 0) { >> + /* We should be here if there aren't any ports in the group */ >> + max_hops = 10001; /* see later - we use it to realize no hops */ >> + goto Exit; >> + } >> + > > Could we reach this condition? > >> From the caller code: > > num_ports = cl_qlist_count(&port_list); > if (num_ports == 0) { > OSM_LOG(sm->p_log, OSM_LOG_VERBOSE, > "MLID 0x%X has no members - nothing to do\n", > mbox->mlid); > goto Exit; > } You're fight, we can't reach this condition anyway. So the fix should be just removing the special treatment of the problematic case. Patch shortly. -- Yevgeny > Sasha > >> /* >> For each member of the multicast group, compute the >> number of hops to its base LID. >> @@ -215,12 +221,10 @@ static float osm_mcast_mgr_compute_max_hops(osm_sm_t * sm, cl_qlist_t * l, >> max_hops = hops; >> } >> >> - if (max_hops == 0) >> - /* >> - We should be here if there aren't any ports in the group. >> - */ >> - max_hops = 10001; /* see later - we use it to realize no hops */ >> - >> + /* Note that at this point we might get (max_hops == 0), >> + which means that there's only one member in the mcast >> + group, and it's the current switch */ >> +Exit: >> OSM_LOG_EXIT(sm->p_log); >> return (float)max_hops; >> } >> -- >> 1.5.1.4 >> >> -- >> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in >> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org >> More majordomo info at http://vger.kernel.org/majordomo-info.html >> > -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html