Re: [PATCH v5] IB/core: Trace points for diagnosing completion queue issues

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 




> On Nov 8, 2019, at 9:21 AM, Jason Gunthorpe <jgg@xxxxxxxx> wrote:
> 
> On Thu, Nov 07, 2019 at 03:54:34PM -0500, Chuck Lever wrote:
>> Signed-off-by: Chuck Lever <chuck.lever@xxxxxxxxxx>
>> Reviewed-by: Parav Pandit <parav@xxxxxxxxxxxx>
>> drivers/infiniband/core/Makefile |    2 
>> drivers/infiniband/core/cq.c     |   36 +++++
>> drivers/infiniband/core/trace.c  |   14 ++
>> include/rdma/ib_verbs.h          |   11 +-
>> include/trace/events/rdma_core.h |  251 ++++++++++++++++++++++++++++++++++++++
>> 5 files changed, 307 insertions(+), 7 deletions(-)
>> create mode 100644 drivers/infiniband/core/trace.c
>> create mode 100644 include/trace/events/rdma_core.h
>> 
>> Changes since v4:
>> - Removed __ib_poll_cq, uninlined ib_poll_cq
> 
> Oh I don't think uninlining is such a good idea, isn't this rather
> performance sensitive?

So would you rather go with v4?

Not clear to me that the compiler won't do the right thing,
at least for __ib_process_cq.


--
Chuck Lever







[Index of Archives]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Photo]     [Yosemite News]     [Yosemite Photos]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux