Date: Sat, 24 Aug 2013 19:25:16 +0300 From: Alexander Motin <mav@FreeBSD.org> To: Dmitry Morozovsky <marck@rinet.ru> Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r254787 - head/sys/kern Message-ID: <5218DE6C.4060700@FreeBSD.org> In-Reply-To: <alpine.BSF.2.00.1308242009140.21486@woozle.rinet.ru> References: <201308241441.r7OEfnFx060671@svn.freebsd.org> <alpine.BSF.2.00.1308241926410.21486@woozle.rinet.ru> <5218D22F.8000009@FreeBSD.org> <alpine.BSF.2.00.1308242009140.21486@woozle.rinet.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
On 24.08.2013 19:11, Dmitry Morozovsky wrote: > On Sat, 24 Aug 2013, Alexander Motin wrote: > >>>> MFprojects/camlock r254460: >>>> Remove locking from taskqueue_member(). The list of threads is static >>>> during the taskqueue life cycle, so there is no need to protect it, >>>> taking quite congested lock several more times for each ZFS I/O. >>> >>> Great, thanks! >>> >>> Any chances to MFC this to 9? >> >> I don't see any problem to do it after some time. > > Ah, so IIUC, it does not depend on other changes and could be tested on stock > stable/9 right now? If so, I'd merge this to our tinderbox/poudriere builder > and test. Yes, it is completely independent. Please, welcome. I am just not sure whether you notice much difference unless you have really a lot of IOPS. On my tests doing 90K IOPS over bunch of SSDs I've measured difference of several percents. -- Alexander Motin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5218DE6C.4060700>