mbox series

[for-next,0/2] IB/hfi1: Diagnostic improvements

Message ID 20180926175835.2451.14284.stgit@scvm10.sc.intel.com (mailing list archive)
Headers show
Series IB/hfi1: Diagnostic improvements | expand

Message

Dennis Dalessandro Sept. 26, 2018, 5:59 p.m. UTC
Hi Doug and Jason,

We have a minor debugging aid from Kaike to dump out the PIO info for kernel
send contexts as well as a sizeable patch from Mitko that changes the way user
tools can access the HFI's BAR. We used to do this in a private interface in
the driver but to get out of staging we ripped that out in favor of using the
resource0 file. However, that is no longer going to work due to distro issues.
Details are in the commit message, but the gist of it is he has created a new
interface in debugfs.

---

Kaike Wan (1):
      IB/hfi1: Dump pio info for non-user send contexts

Mitko Haralanov (1):
      IB/hfi1: Add diagnostic debugfs interface


 drivers/infiniband/hw/hfi1/chip_registers.h |    4 
 drivers/infiniband/hw/hfi1/debugfs.c        |  502 +++++++++++++++++++++------
 drivers/infiniband/hw/hfi1/pio.c            |   25 +
 drivers/infiniband/hw/hfi1/pio.h            |    3 
 include/uapi/rdma/rdma_user_ioctl.h         |    4 
 5 files changed, 428 insertions(+), 110 deletions(-)

--
-Denny

Comments

Jason Gunthorpe Sept. 26, 2018, 7:50 p.m. UTC | #1
On Wed, Sep 26, 2018 at 10:59:39AM -0700, Dennis Dalessandro wrote:
> From: Kaike Wan <kaike.wan@intel.com>
> 
> This patch dumps the pio info for non-user send contexts to assist
> debugging in the field.
> 
> Reviewed-by: Mike Marciniczyn <mike.marciniszyn@intel.com>
> Reviewed-by: Mike Ruhl <michael.j.ruhl@intel.com>
> Signed-off-by: Kaike Wan <kaike.wan@intel.com>
> Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
> ---
>  drivers/infiniband/hw/hfi1/chip_registers.h |    4 ++
>  drivers/infiniband/hw/hfi1/debugfs.c        |   49 +++++++++++++++++++++++++++
>  drivers/infiniband/hw/hfi1/pio.c            |   25 ++++++++++++++
>  drivers/infiniband/hw/hfi1/pio.h            |    3 ++
>  4 files changed, 81 insertions(+), 0 deletions(-)

For some reason this patch did not go into patchworks, any idea why?

Jason
Dennis Dalessandro Sept. 26, 2018, 8:57 p.m. UTC | #2
On 9/26/2018 3:50 PM, Jason Gunthorpe wrote:
> On Wed, Sep 26, 2018 at 10:59:39AM -0700, Dennis Dalessandro wrote:
>> From: Kaike Wan <kaike.wan@intel.com>
>>
>> This patch dumps the pio info for non-user send contexts to assist
>> debugging in the field.
>>
>> Reviewed-by: Mike Marciniczyn <mike.marciniszyn@intel.com>
>> Reviewed-by: Mike Ruhl <michael.j.ruhl@intel.com>
>> Signed-off-by: Kaike Wan <kaike.wan@intel.com>
>> Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
>> ---
>>   drivers/infiniband/hw/hfi1/chip_registers.h |    4 ++
>>   drivers/infiniband/hw/hfi1/debugfs.c        |   49 +++++++++++++++++++++++++++
>>   drivers/infiniband/hw/hfi1/pio.c            |   25 ++++++++++++++
>>   drivers/infiniband/hw/hfi1/pio.h            |    3 ++
>>   4 files changed, 81 insertions(+), 0 deletions(-)
> 
> For some reason this patch did not go into patchworks, any idea why?

Looks like it never made it to the mailing list. I don't see it in the 
archives I usually use. I've had patches take an extra long time to make 
it to the list before. If it doesn't show up soon-ish I can re-send it. 
No clue what would be wrong though.

-Denny
Dennis Dalessandro Sept. 27, 2018, 11:30 a.m. UTC | #3
On 9/26/2018 4:57 PM, Dennis Dalessandro wrote:
> On 9/26/2018 3:50 PM, Jason Gunthorpe wrote:
>> On Wed, Sep 26, 2018 at 10:59:39AM -0700, Dennis Dalessandro wrote:
>>> From: Kaike Wan <kaike.wan@intel.com>
>>>
>>> This patch dumps the pio info for non-user send contexts to assist
>>> debugging in the field.
>>>
>>> Reviewed-by: Mike Marciniczyn <mike.marciniszyn@intel.com>
>>> Reviewed-by: Mike Ruhl <michael.j.ruhl@intel.com>
>>> Signed-off-by: Kaike Wan <kaike.wan@intel.com>
>>> Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
>>> ---
>>>   drivers/infiniband/hw/hfi1/chip_registers.h |    4 ++
>>>   drivers/infiniband/hw/hfi1/debugfs.c        |   49 
>>> +++++++++++++++++++++++++++
>>>   drivers/infiniband/hw/hfi1/pio.c            |   25 ++++++++++++++
>>>   drivers/infiniband/hw/hfi1/pio.h            |    3 ++
>>>   4 files changed, 81 insertions(+), 0 deletions(-)
>>
>> For some reason this patch did not go into patchworks, any idea why?
> 
> Looks like it never made it to the mailing list. I don't see it in the 
> archives I usually use. I've had patches take an extra long time to make 
> it to the list before. If it doesn't show up soon-ish I can re-send it. 
> No clue what would be wrong though.

Re-sent and looks like patchworks picked it up:
https://patchwork.kernel.org/patch/10617795/

-Denny