Re: [PATCH] Rework of /proc/stat

[mingming cao <cmm@us.ibm.com>]

Alan Cox wrote:

>>I've made a new version of IO statistics in kstat that remove the
>>previous limitations of MAX_MAJOR. I've made tests on my machine only, so could someone test it, please?
>>Feedback welcome.
>>
> 
> Any reason for preferring this over the sard patches in -ac ?
> 

Basically, statistic data are moved from the global kstat structure to 
the request_queue structures, and it is allocated/freed when the request 
queue is initialized and freed.  This way it is

1)self-controlled;
2)avoid the lookup step before the accounting, so it should be faster;
3)statistics implementation is not affected by the major/minor numbers;
4)able to gathering statistics for all disks while keep the memory needs 
minimized.