[openib-general] [PATCH] Better IPoIB multicast handling

[Tom Duffy]

On Fri, 2004-10-22 at 11:15 -0700, Roland Dreier wrote: 
> This patch improves how IPoIB handles multicasts.  It should fix the
> crash that Andras saw; unfortunately I don't think it will help with
> Tom's crash (although I don't understand that crash so it might fix
> it).  Unfortunately it still probably doesn't work with some SMs.
> 
> Also, with this patch, multicast seems to work (tested only with
> "ping -I ib0 224.0.0.1") although the way I handle multicast
> neighbours needs cleanup.
> 
> Any test feedback is appreciated...

Still crashing on my sparc64.  Looks like qp is NULL.

And yes, port 1 is going to ACTIVE.

Unable to handle kernel NULL pointer dereference
tsk->{mm,active_mm}->context = 0000000000000625
tsk->{mm,active_mm}->pgd = fffff8007ce84000
              \|/ ____ \|/
              "@'/ .. \`@"
              /_| \__/ |_\
                 \__U_/
ts_ib_mad(8771): Oops [#1]
TSTATE: 0000004480009605 TPC: 0000000002018244 TNPC: 0000000002018248 Y: 00000000    Not tainted
TPC: <ib_modify_qp+0x4/0x40 [ib_core]>
g0: 0000000000000000 g1: 0000000081234568 g2: 0000000000000003 g3: 0000000000000000
g4: fffff8007c07a100 g5: 0000000000000008 g6: fffff8006b6f8000 g7: 0000000000000024
o0: fffff8007d551640 o1: 0000000000000020 o2: 0000000000000001 o3: fffff8006b6fb35e
o4: 0000000000008001 o5: 00000000000000f0 sp: fffff8006b6fa9e1 ret_pc: fffff8007f8ad220
RPC: <0xfffff8007f8ad220>
l0: 0000000000000001 l1: fffff8006c389800 l2: 0000000002173800 l3: fffff8007fe99818
l4: 0000000000022b8c l5: 00000000006cfd90 l6: 000000000000595b l7: fffff8007fe1bb50
i0: 0000000000000000 i1: fffff8007d551640 i2: 0000000000000040 i3: fffff8006b6fb360
i4: 0000000000000040 i5: fffff8007f977ae0 i6: fffff8006b6faaa1 i7: 000000000218ce10
I7: <ipoib_mcast_attach+0x70/0x120 [ib_ipoib]>
Caller[000000000218ce10]: ipoib_mcast_attach+0x70/0x120 [ib_ipoib]
Caller[000000000218b008]: ipoib_mcast_join_finish+0x148/0x3a0 [ib_ipoib]
Caller[000000000218b7b4]: ipoib_mcast_join_complete+0x174/0x1e0 [ib_ipoib]
Caller[000000000217e5bc]: _tsIbMulticastJoinResponse+0xdc/0x2e0 [ib_sa_client]
Caller[0000000002176328]: ib_client_query_callback+0x68/0xa0 [ib_client_query]
Caller[0000000002176fa0]: ib_client_mad_handler+0x60/0x100 [ib_client_query]
Caller[000000000216f2b8]: ib_mad_invoke_filters+0x98/0x120 [ib_mad]
Caller[000000000216f744]: ib_mad_dispatch+0xe4/0x1e0 [ib_mad]
Caller[000000000216fcf0]: ib_mad_work_thread+0x70/0x540 [ib_mad]
Caller[00000000020088b0]: _tsKernelQueueThread+0xf0/0x160 [ib_services]
Caller[0000000002008624]: _tsKernelThreadStart+0x84/0xa0 [ib_services]
Caller[0000000000417430]: kernel_thread+0x30/0x60
Caller[00000000020086d4]: tsKernelThreadStart+0x94/0xe0 [ib_services]
Instruction DUMP: 01000000  01000000  9de3bf40 <c45e0000> 90100018  9410001a  92100019  c258a110  9fc04000
TSTATE: 00000000f0009601 TPC: 0000000000516398 TNPC: 000000000051639c Y: 00000000    Not tainted
TPC: <__bzero+0x14c/0x274>
g0: fffff8007fe5ba90 g1: fffff80000000000 g2: 00000000b6db6db7 g3: 0000000000000000
g4: fffff8007fe1b3e0 g5: 0000000000000080 g6: fffff8007fe58000 g7: 000000000051 000
o0: fffff8007d7f3100 o1: 0000000000000000 o2: 0000000000000000 o3: 0000000000000f00
o4: 0000000000000080 o5: 0000000000000000 sp: fffff8007fe5b141 ret_pc: 0000000000425fc4
RPC: <pci_alloc_consistent+0x64/0x1a0>
l0: fffff8007d7f2000 l1: 0000000000000000 l2: 0000000000000000 l3: fffff8007fe73000
l4: 0000000000002000 l5: fffff8007c886020 l6: fffff8006c389b90 l7: fffff8007fe37fd0
i0: 0000000000000000 i1: 0000000000000001 i2: fffff8007fe5bacc i3: 00000000c7dc0000
i4: 0000000000000000 i5: 0000000000000000 i6: fffff8007fe5b201 i7: 00000000020549f4
I7: <mthca_MAD_IFC+0x14/0xe0 [ib_mthca]>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.openfabrics.org/pipermail/general/attachments/20041022/8b6be606/attachment.sig>