[ofa-general] IB performance stats (revisited)

[Mark Seger]

>>>There's now an experimental performance manager which has been discussed
>>>on the list. The performance data collected can be accessed.
>>>  
>>>      
>>>
>>alas, since I use this tool on commercial systems, I can't run it 
>>against experimental code.  perhaps when the experimental becomes real I 
>>can.
>>    
>>
>
>It should be in the OFED 1.3 timeframe. Also, there are vendor
>Performance Managers too.
>  
>
that would be good.  if I can detect the 1.3 stack I can change my 
monitoring accordingly

>>  I'll try to find the notes in the archives.
>>    
>>
>
>I can send you this if you can't find it
>  
>
that would be great if you can easily lay your hands on it.

>>Good point, but I guess I'm between a rock and a hard place.  imho: as 
>>long as the counters don't wrap this problem will never be solved.
>>    
>>
>
>It's the IBTA standard (rather than IETF style counters). I don't think
>it's going to change.
>  
>
yeah, but it also makes one wonder why non-wrapping counters were chosen 
when the IETF proved years ago that one needs wrapping counter to allow 
concurrent access to them.  sigh...
-mark