Date: Wed, 11 Apr 2012 12:02:05 -0400 From: John Baldwin <jhb@freebsd.org> To: Andrey Zonov <andrey@zonov.org> Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r233925 - in head: sys/kern sys/sys sys/vm usr.bin/kdump usr.bin/ktrace Message-ID: <201204111202.05409.jhb@freebsd.org> In-Reply-To: <4F848E4B.8020302@zonov.org> References: <201204051713.q35HDE3d067735@svn.freebsd.org> <4F848E4B.8020302@zonov.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tuesday, April 10, 2012 3:47:23 pm Andrey Zonov wrote: > On 05.04.2012 21:13, John Baldwin wrote: > > Author: jhb > > Date: Thu Apr 5 17:13:14 2012 > > New Revision: 233925 > > URL: http://svn.freebsd.org/changeset/base/233925 > > > > Log: > > Add new ktrace records for the start and end of VM faults. This gives > > a pair of records similar to syscall entry and return that a user can > > use to determine how long page faults take. The new ktrace records are > > enabled via the 'p' trace type, and are enabled in the default set of > > trace points. > > > > Reviewed by: kib > > MFC after: 2 weeks > > > > Hi John, > > Thanks a lot, this change is very useful! > > And while you are here I would like to show you a patch which adds a > "wmesg" to the context switch tracing. It is not finished, it's just a > concept. Please review it and if you are interesting in that I'll > finish it and will test more widely. Ah, this looks fairly neat. However, I think you will need a few changes: 1) You will want to make kdump handle either the old or new size of struct ktr_csw. (The size of the record is in the header, so as long as you have a 'struct ktr_csw_old' you can handle this fairly easily.) 2) cs->wmesg is never NULL. Instead, it should probably just always print it out. 3) condvar's have a valid wmesg. -- John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201204111202.05409.jhb>