[RFC] Thread CPU load estimation for ULE, not sampled by hardclock()

2012-03-09 Thread Alexander Motin
Hi. At this time ULE scheduler uses hardclock() ticks via sched_tick() to estimate threads CPU load. That strictly limits it's best possible precision and that is one more reason to call hardclock() on every HZ tick, even when there is no callouts scheduled. I've made a patch to get CPU load

Re: mfid, raid monitoring daemon

2012-03-09 Thread Damien Fleuriot
On 3/8/12 7:08 PM, Sean Bruno wrote: > I'm trying to decide if I should cram "mfid" for mfi(4) controllers into > the src tree or if we should package it up into a ports package. I > suspect that either one is acceptible, but it seems to make more sense > to put it into the src tree since mfiutil

Strong host model in IPv6?

2012-03-09 Thread Alex Yong
Hi all, I've been playing around with IPv6 networking on FreeBSD release 8.2 and found that there seems to be no strong incoming host model as specified in RFC 1122. I've spotted that in IPv4 there is the sysctl "net.inet.ip.check_interface" which defaults to set, but I've been unable to find any

Re: DTrace built-in variables

2012-03-09 Thread Dr. Baud
    I'll give it a try, thanks. On Thu, Mar 8, 2012 at 8:27 AM, Dr. Baud wrote: > > >     Are the build-in variables cpu and curcpu supported in a 8.1-RELEASE > DTrace? > >     Dr. Unfortunately not.  However, you can approximate them with (IIRC) curthread->td_oncpu. _

Re: mfid, raid monitoring daemon

2012-03-09 Thread Vincent Hoffman
On 09/03/2012 11:52, Damien Fleuriot wrote: > On 3/8/12 7:08 PM, Sean Bruno wrote: >> I'm trying to decide if I should cram "mfid" for mfi(4) controllers into >> the src tree or if we should package it up into a ports package. I >> suspect that either one is acceptible, but it seems to make more s

Re: mfid, raid monitoring daemon

2012-03-09 Thread Damien Fleuriot
On 3/9/12 5:53 PM, Vincent Hoffman wrote: > On 09/03/2012 11:52, Damien Fleuriot wrote: >> On 3/8/12 7:08 PM, Sean Bruno wrote: >>> I'm trying to decide if I should cram "mfid" for mfi(4) controllers into >>> the src tree or if we should package it up into a ports package. I >>> suspect that eit

Re: Strong host model in IPv6?

2012-03-09 Thread Doug Barton
On 3/9/2012 7:02 AM, Alex Yong wrote: > I've been playing around with IPv6 networking on FreeBSD release 8.2 and > found that there seems to be no strong incoming host model as specified in > RFC 1122. First, you're infinitely more likely to get a useful response if you send your message to freebs