Re: [RFC] [Patch 4/4] lock contention tracking slimmed down

[Martin Peschke <mp3@de.ibm.com>]

Peter Zijlstra wrote:
> On Wed, 2007-06-06 at 23:34 +0200, Martin Peschke wrote:
> 
>> +struct statistic_info lock_stat_info[_LOCK_STAT_NUMBER] = {
>> +	[LOCK_STAT_CONT] = {
>> +		.name	  = "contentions",
>> +		.x_unit	  = "instruction_pointer",
>> +		.y_unit	  = "occurrence",
>> +		.defaults = "type=sparse entries=4",
>> +		.flags	  = STATISTIC_FLAGS_LABEL,
>> +	},
>> +	[LOCK_STAT_WAIT_READ] = {
>> +		.name	  = "wait_read",
>> +		.x_unit	  = "nanoseconds",
>> +		.y_unit	  = "occurrence",
>> +		.defaults = "type=utilisation",
>> +	},
>> +	[LOCK_STAT_WAIT_WRITE] = {
>> +		.name	  = "wait_write",
>> +		.x_unit	  = "nanoseconds",
>> +		.y_unit	  = "occurrence",
>> +		.defaults = "type=utilisation",
>> +	},
>> +	[LOCK_STAT_HOLD_READ] = {
>> +		.name	  = "hold_read",
>> +		.x_unit	  = "nanoseconds",
>> +		.y_unit	  = "occurrence",
>> +		.defaults = "type=utilisation",
>> +	},
>> +	[LOCK_STAT_HOLD_WRITE] = {
>> +		.name	  = "hold_write",
>> +		.x_unit	  = "nanoseconds",
>> +		.y_unit	  = "occurrence",
>> +		.defaults = "type=utilisation",
>>  	}
>>  };
> 
> You're parsing strings in-kernel to setup data structures?

Might be surprising (or revolting?)... but, yes, I do.

.defaults reflects the developers opinion on the way of data
accumululation that might make most sense in most cases.
The current implementation of lib/statistic.c allows users to change
setting that we see in .defaults. If one needs to track more than
4 contention points, for example, then they can change this limit by
passing "entries=4" to the user interface. So there is some parsing
code anyway. Passing default setting in a similar fashion has just been
convenient for the developer.