On Tuesday, October 30, 2007 5:00 AM, Eric Saxe wrote:

> Li, Aubrey wrote:
>> Intrstat just reports device interrupt activity, not enough, I think.
>> We need to figure out which process is using this device,right?
>> 
> From the screen shots at linuxpowertop.org, it looks like it just has
> <interrupt> as the process name, and then the device name. 
Yes, it does.

> Does it currently display I/O requester information as well?
> 
No, no I/O requester information in the current powertop.
But I thought it'd better to figure out which process is using the
device, because there are still some interrupts not belonging to any
process. What's your thoughts?

> If need be, we can get at who's requesting the I/O through the DTrace
> I/O provider though. 
> 
We can do that if it's necessary.

>>>    - Event information (timer / callout activity)
>>>       - Can use existing callout related DTrace probes to report on
>>> who's waking up and why (kernel and user activity).
>>> 
>> 
>> I'm looking at this now. Is it already done?
>> 
>> 
> 
> In a way, yes. The DTrace sdt provider provides callout-start and
> callout-end probes. Check out chapter 22 of the DTrace answerbook:
> http://docs.sun.com/app/docs/doc/817-6223
> 

Thanks for the link. Actually I'm checking this guide these days, ;-)

>>> I've added the necessary DTrace probes...
>>> 
>> You are so quick, after 3 days of SUN Tech days China, the main
>> function of powertop are almostly ready, ;-).
>> Did you commit into PowerTop repo?
>> 
> No. I'll commit the DTrace probes (hopefully today). I have a
> "scaffolding" in powertop.c for plugging in the Solaris specific code.
> I'll send a follow up message with that attached, and you can see if
> what you have is further along. :)
> 
Thanks,
-Aubrey

Reply via email to