Messages in this thread | | | From | Jean Delvare <> | Subject | Re: [PATCH] IPMI: Add parameter to limit CPU usage in kipmid | Date | Thu, 14 Jan 2010 15:02:18 +0100 |
| |
Le samedi 19 décembre 2009 14:56, Jean Delvare a écrit : > Le jeudi 17 décembre 2009 23:08, Corey Minyard a écrit : > > I cleaned up the patch some. I added some state results to the list of > > things to be busy waited for (should improve performance a bit) and > > changed the schedule_timeout_interruptible back to 1, since that's > > what it's supposed to be. And I added some comments. > > Thanks for doing this! > > > I did some testing on my system here. On my system, kipmid uses almost > > no CPU normally. If I set the kipmid_max_busy_us value to 500, the > > interface was more than 5 times slower. I had to set the value up to > > 35000 for it to go back to the normal performance, and it was pretty > > linear between the two values. So this is definiately not for all > > systems. > > > > Can you try this out to make sure its ok? > > I don't have any hardware where I can test this myself. But hopefully > Martin does?
I've had the latest version of this patch tested by one of my colleagues who has access to the relevant hardware, results are OK. So I think we're ready to go, can we finally have this patch pushed upstream?
Thanks, -- Jean Delvare Suse L3 -- 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/
| |