Date: Tue, 9 Mar 2010 23:55:35 +0100 From: Fabien Thomas <fabien.thomas@netasq.com> To: David Christensen <davidch@broadcom.com> Cc: pyunyh@gmail.com, Ian FREISLICH <ianf@clue.co.za>, Ryan Stone <rysto32@gmail.com>, current@freebsd.org Subject: Re: dev.bce.X.com_no_buffers increasing and packet loss Message-ID: <EDAEC324-1741-4503-B94A-FE7551442E3A@netasq.com> In-Reply-To: <bc2d971003091430g236806edy4cf3bb873665fc5@mail.gmail.com> References: <20100305210435.GF14818@michelle.cdnetworks.com> <20100305175639.GB14818@michelle.cdnetworks.com> <E1NnVaT-0003Ft-3p@clue.co.za> <E1Nnc4d-0003mB-6e@clue.co.za> <E1Nne0Q-0003uZ-OR@clue.co.za> <E1Noulp-0007Rc-Ro@clue.co.za> <20100309212139.GO1311@michelle.cdnetworks.com> <5D267A3F22FD854F8F48B3D2B52381933AF90EED69@IRVEXCHCCR01.corp.ad.broadcom.com> <20100309214012.GQ1311@michelle.cdnetworks.com> <5D267A3F22FD854F8F48B3D2B52381933AF90EEDA7@IRVEXCHCCR01.corp.ad.broadcom.com> <bc2d971003091430g236806edy4cf3bb873665fc5@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
If you are on head/stable_7/stable_8 you can also do quick test with top = mode pmcstat -S unhalted-cycles -T = (http://wiki.freebsd.org/PmcTools/PmcTop). For more in depth post processing with source code (c+asm) you can = output to Kcachegrind (http://wiki.freebsd.org/PmcTools/PmcKcachegrind). Fabien >> What's the traffic look like? Jumbo, standard, short frames? Any >> good ideas on profiling the code? I haven't figured out how to use >> the CPU TSC but there is a free running timer on the device that >> might be usable to calculate where the driver's time is spent. >>=20 >> Dave >=20 > In my experience hwpmc is the best and easiest way to profile anything > on FreeBSD. Here's something I sent to a different thread a couple of > months ago explaining how to use it: >=20 > 1) If device hwpmc is not compiled into your kernel, kldload hwpmc(you > will need the HWPMC_HOOKS option in either case) > 2) Run pmcstat to begin taking samples(make sure that whatever you are > profiling is busy doing work first!): >=20 > pmcstat -S unhalted-cycles -O /tmp/samples.out >=20 > The -S option specifies what event you want to use to trigger > sampling. The unhalted-cycles is the best event to use if your > hardware supports it; pmc will take a sample every 64K non-idle CPU > cycles, which is basically equivalent to sampling based on time. If > the unhalted-cycles event is not supported by your hardware then the > instructions event will probably be the next best choice(although it's > nowhere near as good, as it will not be able to tell you, for example, > if a particular function is very expensive because it takes a lot of > cache misses compared to the rest of your program). One caveat with > the unhalted-cycles event is that time spent spinning on a spinlock or > adaptively spinning on a MTX_DEF mutex will not be counted by this > event, because most of the spinning time is spent executing an hlt > instruction that idles the CPU for a short period of time. >=20 > Modern Intel and AMD CPUs offer a dizzying array of events. They're > mostly only useful if you suspect that a particular kind of event is > hurting your performance and you would like to know what is causing > those events. For example, if you suspect that data cache misses are > causing you problems you can take samples on cache misses. > Unfortunately on some of the newer CPUs(namely the Core2 family, > because that's what I'm doing most of my profiling on nowadays) I find > it difficult to figure out just what event to use to profile based on > cache misses. man pmc will give you an overview of pmc, and there are > manpages for every CPU family supported(eg man pmc.core2) >=20 > 3) After you've run pmcstat for "long enough"(a proper definition of > long enough requires a statistician, which I most certainly am not, > but I find that for a busy system 10 seconds is enough), Control-C it > to stop it*. You can use pmcstat to post-process the samples into > human-readable text: >=20 > pmcstat -R /tmp/samples.out -G /tmp/graph.txt >=20 > The graph.txt file will show leaf functions on the left and their > callers beneath them, indented to reflect the callchain. It's not too > easy to describe and I don't have sample output available right now. >=20 >=20 > Another interesting tool for post-processing the samples is > pmcannotate. I've never actually used the tool before but it will > annotate the program's source to show which lines are the most > expensive. This of course needs unstripped modules to work. I think > that it will also work if the GNU "debug link" is in the stripped > module pointing to the location of the file with symbols. >=20 >=20 > * Here's a tip I picked up from Joseph Koshy's blog: to collect > samples for a fixed period of time(say 1 minute), have pmcstat run the > sleep command: >=20 > pmcstat -S unhalted-cycles -O /tmp/samples.out sleep 60 > _______________________________________________ > freebsd-current@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-current > To unsubscribe, send any mail to = "freebsd-current-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?EDAEC324-1741-4503-B94A-FE7551442E3A>