Date: Sat, 21 Aug 2010 00:33:43 -0700 (PDT) From: Doug Barton <dougb@FreeBSD.org> To: Andriy Gapon <avg@icyb.net.ua> Cc: freebsd-current@FreeBSD.org Subject: Re: Latest intr problems Message-ID: <alpine.BSF.2.00.1008210028530.1718@qbhto.arg> In-Reply-To: <4C6F7BD1.4030009@icyb.net.ua> References: <alpine.BSF.2.00.1008201701001.19740@qbhto.arg> <4C6F772A.5020703@icyb.net.ua> <4C6F7BD1.4030009@icyb.net.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, 21 Aug 2010, Andriy Gapon wrote: >> I think that for such amount of data it is better to use links (perhaps a >> service like pastebin) rather than inlining it. No problem: http://people.freebsd.org/~dougb/intr-out.txt >> BTW, it seems that there are no followups/comments on results of the old dtrace >> script, so I am not sure if there is any point in continuing to post it. >> It is useless personally for me. Ok. >> Back to the data. >> Could you please report results of >> procstat -k 10 >> procstat -k 11 > > Could you please remind what the "old dtrace script" is? :-) http://people.freebsd.org/~dougb/dtrace-script.txt > Output of sysctl dev.cpu (normal and when the problem hits) > > Another dtrace script: > profile:::profile-1001 > { > @stacks[curthread->td_oncpu, pid, tid, stack()] = count(); > } > END > { > trunc(@stacks, 20); > printa(@stacks); > } All of the above is in that intr-out.txt file. The last 2 (sysctl and the 2nd dtrace script) are actually while the intr cpu% was going down (i.e., the system was recovering) but I'll run it again next time if you like. Thanks! Doug -- Improve the effectiveness of your Internet presence with a domain name makeover! http://SupersetSolutions.com/ Computers are useless. They can only give you answers. -- Pablo Picasso
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.BSF.2.00.1008210028530.1718>