Re: [PATCH] Export URB statistics for powertop.

On Wed, Oct 03, 2007 at 03:37:26PM -0700, David Brownell wrote:
> > powertop currently tracks interrupts generated by uhci and ehci,
> 
> OHCI too I'd hope...
> 
> 
> > but it has no way of telling which USB device to blame interrupts on.
> > This patch exports the number of URBs that complete for a given device.
> 
> Note that there's not a one-to-one correspondence between URBs
> and interrupts...
> 
> Plus, both OHCI and EHCI drivers include logic to reduce the number
> of interrupts issued per URB.  So for example usb-storage will probably
> get an IRQ every two or three URBs with OHCI, and once per scatterlist
> request (I've seen several dozen URBs per scatterlist) with EHCI.
> 
> So, powertop shouldn't assume one URB per IRQ ...

Sure.  Arjan really only wanted a measure of the activity of a
particular USB device.  An URB count isn't the best way to measure this
(since big and small requests will be counted equally).  I tried to
argue that he really wanted a better statistic, but he said (to quote
the IRC conversation I had with him this morning):

"It's not about interrupts; it's about finding activity levels.  No
measure will be perfect and I don't care.  All I want is a pretty good
measure and urb count already is that and it's simple."

Sarah

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel