Messages in this thread | | | Date | Wed, 11 Feb 2004 10:12:32 -0500 (EST) | From | Alan Stern <> | Subject | Re: [PATCH] Slight optimisation of the uhci-hcd init code |
| |
On Wed, 11 Feb 2004 eric.piel@tremplin-utc.net wrote:
> > I'm not sure exactly what the ffffffff value means -- maximum utilization? > Yes, more exactly it means that the last 64 times the idle task was called the > bus master was detected as being used.
By "detected as being used", do you mean that the device was actually carrying out bus transactions or only that it was enabled as a bus master? And if it's the first, do you mean that the bus activity was occurring just as the idle task happened to check or that some bus activity had occurred since the last time the idle task checked?
USB controllers are more or less permanently enabled, but they only create bus activity when they actually have some work to do. If the only USB device you have connected is a mouse then there should be a small amount of bus activity occuring once every millisecond, with a little more than that every 32 milliseconds (or whatever polling period your mouse uses -- look in /proc/bus/usb/devices), and even more when you actually move the mouse or press a button.
> Well, my laptop seems fine. When I unplug the mouse the bus master activity goes > to 0000000. I've very little idea about how the bus master has to be used and > even less about the uhci host controller. However, at a first glance I thought > it was weird to see full bus master activity when all that is connected is a > mouse not being used. Do you think it's normal?
Since I don't really understand the value being reported, I can't say. You can judge for yourself from the explanation above.
Alan Stern
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |