diff options
author | Moni Shoua <monis@Voltaire.COM> | 2009-01-28 15:15:56 -0800 |
---|---|---|
committer | Roland Dreier <rolandd@cisco.com> | 2009-01-28 15:15:56 -0800 |
commit | 270b8b85134c299799dddec624ceeb5671330131 (patch) | |
tree | a558f73a67af4cf9bb4f4e27cf7cd0eaa27ee2b4 /drivers/infiniband/hw | |
parent | f3b8436ad9a8ad36b3c9fa1fe030c7f38e5d3d0b (diff) | |
download | op-kernel-dev-270b8b85134c299799dddec624ceeb5671330131.zip op-kernel-dev-270b8b85134c299799dddec624ceeb5671330131.tar.gz |
IB/mthca: Fix dispatch of IB_EVENT_LID_CHANGE event
When snooping a PortInfo MAD, its client_reregister bit is checked.
If the bit is ON then a CLIENT_REREGISTER event is dispatched,
otherwise a LID_CHANGE event is dispatched. This way of decision
ignores the cases where the MAD changes the LID along with an
instruction to reregister (so a necessary LID_CHANGE event won't be
dispatched) or the MAD is neither of these (and an unnecessary
LID_CHANGE event will be dispatched).
This causes problems at least with IPoIB, which will do a "light"
flush on reregister, rather than the "heavy" flush required due to a
LID change.
Fix this by dispatching a CLIENT_REREGISTER event if the
client_reregister bit is set, but also compare the LID in the MAD to
the current LID. If and only if they are not identical then a
LID_CHANGE event is dispatched.
Signed-off-by: Moni Shoua <monis@voltaire.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Yossi Etigin <yosefe@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Diffstat (limited to 'drivers/infiniband/hw')
-rw-r--r-- | drivers/infiniband/hw/mthca/mthca_mad.c | 25 |
1 files changed, 19 insertions, 6 deletions
diff --git a/drivers/infiniband/hw/mthca/mthca_mad.c b/drivers/infiniband/hw/mthca/mthca_mad.c index 6404495..5648659 100644 --- a/drivers/infiniband/hw/mthca/mthca_mad.c +++ b/drivers/infiniband/hw/mthca/mthca_mad.c @@ -104,7 +104,8 @@ static void update_sm_ah(struct mthca_dev *dev, */ static void smp_snoop(struct ib_device *ibdev, u8 port_num, - struct ib_mad *mad) + struct ib_mad *mad, + u16 prev_lid) { struct ib_event event; @@ -114,6 +115,7 @@ static void smp_snoop(struct ib_device *ibdev, if (mad->mad_hdr.attr_id == IB_SMP_ATTR_PORT_INFO) { struct ib_port_info *pinfo = (struct ib_port_info *) ((struct ib_smp *) mad)->data; + u16 lid = be16_to_cpu(pinfo->lid); mthca_update_rate(to_mdev(ibdev), port_num); update_sm_ah(to_mdev(ibdev), port_num, @@ -123,12 +125,15 @@ static void smp_snoop(struct ib_device *ibdev, event.device = ibdev; event.element.port_num = port_num; - if (pinfo->clientrereg_resv_subnetto & 0x80) + if (pinfo->clientrereg_resv_subnetto & 0x80) { event.event = IB_EVENT_CLIENT_REREGISTER; - else - event.event = IB_EVENT_LID_CHANGE; + ib_dispatch_event(&event); + } - ib_dispatch_event(&event); + if (prev_lid != lid) { + event.event = IB_EVENT_LID_CHANGE; + ib_dispatch_event(&event); + } } if (mad->mad_hdr.attr_id == IB_SMP_ATTR_PKEY_TABLE) { @@ -196,6 +201,8 @@ int mthca_process_mad(struct ib_device *ibdev, int err; u8 status; u16 slid = in_wc ? in_wc->slid : be16_to_cpu(IB_LID_PERMISSIVE); + u16 prev_lid = 0; + struct ib_port_attr pattr; /* Forward locally generated traps to the SM */ if (in_mad->mad_hdr.method == IB_MGMT_METHOD_TRAP && @@ -233,6 +240,12 @@ int mthca_process_mad(struct ib_device *ibdev, return IB_MAD_RESULT_SUCCESS; } else return IB_MAD_RESULT_SUCCESS; + if ((in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_SUBN_LID_ROUTED || + in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE) && + in_mad->mad_hdr.method == IB_MGMT_METHOD_SET && + in_mad->mad_hdr.attr_id == IB_SMP_ATTR_PORT_INFO && + !ib_query_port(ibdev, port_num, &pattr)) + prev_lid = pattr.lid; err = mthca_MAD_IFC(to_mdev(ibdev), mad_flags & IB_MAD_IGNORE_MKEY, @@ -252,7 +265,7 @@ int mthca_process_mad(struct ib_device *ibdev, } if (!out_mad->mad_hdr.status) { - smp_snoop(ibdev, port_num, in_mad); + smp_snoop(ibdev, port_num, in_mad, prev_lid); node_desc_override(ibdev, out_mad); } |