On 10/1/2015 12:51 AM, ira.weiny@xxxxxxxxx wrote: > From: Ira Weiny <ira.weiny@xxxxxxxxx> > > This has mild testing and still has some FIXMEs etc. I'm submitting to help > decide if we wish to remove the MAD snooping infrastructure. I think this has a ways to go before it has equivalent functionality to madeye. I'm also not sure that this is a replacement for MAD snooping. There's some overlap but they are somewhat complementary right now. IMO, at a minimum, the snooping interface should not be removed until this has equivalent functionality. > I believe this to be a more efficient debugging mechanism as it is most always > available (if debugfs is available) Snooping could be exposed to user space via user_mad and madeye could be ported to user space. > and should have limited performance impact when off. This is true for snooping too. -- Hal > > Ira Weiny (5): > IB/MAD: add send path trace points > IB/MAD: add recv path trace point > IB/MAD Add agent trace points > IB/mad: Add umad trace points > IB/mad: Add SMP tracing > > drivers/infiniband/core/mad.c | 72 +++++++ > drivers/infiniband/core/user_mad.c | 5 + > include/trace/events/ib_mad.h | 385 +++++++++++++++++++++++++++++++++++++ > include/trace/events/ib_umad.h | 312 ++++++++++++++++++++++++++++++ > 4 files changed, 774 insertions(+) > create mode 100644 include/trace/events/ib_mad.h > create mode 100644 include/trace/events/ib_umad.h > -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html