Re: [Patch 1/4] Delay accounting: Initialization

[Shailabh Nagar <nagar@watson.ibm.com>]

Peter Chubb wrote:
>>>>>>"Andrew" == Andrew Morton <akpm@osdl.org> writes:
> 
> 
> Andrew> Shailabh Nagar <nagar@watson.ibm.com> wrote:
> 
>>> + *ts = sched_clock();
> 
> 
> Andrew> I'm not sure that it's kosher to use sched_clock() for
> Andrew> fine-grained timestamping like this.  Ingo had issues with it
> Andrew> last time this happened?
> 
> It wasn't Ingo, it was Andi Kleen...  for my Microstate Accounting
> patches, which do very similar things to Shailabh's patchsetm, but
> using /proc and a system call instead (following Solaris's lead)
> 

Were these the comments from Andi to which you refer:
	http://www.uwsg.indiana.edu/hypermail/linux/kernel/0503.1/1237.html

The objections to microstate overhead seemed to stem from the syscall
overhead, not use of sched_clock() per se.


Andi, Ingo,

Are there problems with using sched_clock()for timestamping if one is prepared
to live with them not necessarily being nanosecond accurate ? I'm trying to search
the archives etc. but if you can respond with any quick comments, that'd be very
helpful.


Thanks,
Shailabh