[openib-general] performance counters in /sys

[Ronald G. Minnich]

On Fri, 20 May 2005, Grant Grundler wrote:

> Not entirely. One could fill unimplemented values with -1 or 0.

or use s-expressions a la supermon. That's worked the best for us in 
widely varying environments. These fixed-format tables of the type 
that /proc delivers are painful. 

> For monitoring performance, a single /sys file that lists all devices
> is the lightest weight and is acceptable according to
> Documentation/filesystems/sysfs.txt:
> 
> | Attributes should be ASCII text files, preferably with only one value
> | per file. It is noted that it may not be efficient to contain only
> | value per file, so it is socially acceptable to express an array of
> | values of the same type.

yes, that is the accepted thing nowadays. And, boy, is it SLOW. SLOW, 
SLOW, SLOW. 

Open file, read ONE value, close file. Takes a long time. So we've seen 
that for a number of values you really can't sample at interesting rates. 

Of course, you can speed it up a bit -- keep your files open, right?  
open once, read, fseek 0, read, ...

Except then, if you need to unload the module, well, you can't: you have
the /sys files open and the module can't really be unloaded.

Of course, if you only need values very 10 seconds or so, /sys is ok. But 
if you're look at 100 or 1khz, monitoring, which we do sometimes here, 
/sys is entirely inadequate. /sys is wonderful, and all, but it's not for 
high speed monitoring.

If I understood what to look for from the driver better (I lack knowledge 
in this IB area :-) I could write a supermon kernel module for it. Again, 
this only matters if you need performance, I'm not sure what people need 
here.

ron