Date: Mon, 12 Jul 2010 15:49:39 +0300 From: Alexander Motin <mav@FreeBSD.org> To: John Baldwin <jhb@freebsd.org> Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r209900 - head/sys/kern Message-ID: <4C3B0F63.6000905@FreeBSD.org> In-Reply-To: <201007120813.19223.jhb@freebsd.org> References: <201007111647.o6BGlk0O033551@svn.freebsd.org> <201007120813.19223.jhb@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
John Baldwin wrote: > On Sunday, July 11, 2010 12:47:46 pm Alexander Motin wrote: >> Author: mav >> Date: Sun Jul 11 16:47:45 2010 >> New Revision: 209900 >> URL: http://svn.freebsd.org/changeset/base/209900 >> >> Log: >> Remove interval validation from cpu_tick_calibrate(). As I found, check >> was needed at preliminary version of the patch, where number of CPU ticks >> was divided strictly on 16 seconds. Final code instead uses real interval >> duration, so precise interval should not be important. Same time aliasing >> issues around second boundary causes false positives, periodically logging >> useless "t_delta ... too long/short" messages when HZ set below 256. > > Hmm, did you ask phk@ about this? Yes. He agreed that code should be reconsidered. > I notice that the printfs only trigger if > you have bootverbose enabled, so they were not affecting normal users as most > people do not run their systems with bootverbose enabled. Yes, but they produce major part of kernel messages on my own systems. I've got tired seeing them for years. It would be fine if they were usable, but, as I have told here and on current@, with preset implementation they were absolutely meaningless. -- Alexander Motin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4C3B0F63.6000905>