Date: Fri, 21 Sep 2007 15:22:15 -0400 From: John Baldwin <jhb@freebsd.org> To: Andrew Thompson <thompsa@freebsd.org> Cc: cvs-src@freebsd.org, Alfred Perlstein <alfred@freebsd.org>, src-committers@freebsd.org, cvs-all@freebsd.org Subject: Re: cvs commit: src/sys/net ieee8023ad_lacp.c ieee8023ad_lacp.h if_lagg.c if_lagg.h Message-ID: <200709211522.16194.jhb@freebsd.org> In-Reply-To: <20070917203208.GA9614@heff.fud.org.nz> References: <200709150111.l8F1BQii070212@repoman.freebsd.org> <20070917161848.GB79417@elvis.mu.org> <20070917203208.GA9614@heff.fud.org.nz>
next in thread | previous in thread | raw e-mail | index | archive | help
On Monday 17 September 2007 04:32:08 pm Andrew Thompson wrote: > On Mon, Sep 17, 2007 at 09:18:48AM -0700, Alfred Perlstein wrote: > > * Andrew Thompson <thompsa@FreeBSD.org> [070914 18:11] wrote: > > > thompsa 2007-09-15 01:11:26 UTC > > > > > > FreeBSD src repository > > > > > > Modified files: (Branch: RELENG_6) > > > sys/net ieee8023ad_lacp.c ieee8023ad_lacp.h > > > if_lagg.c if_lagg.h > > > Log: > > > MFC > > > Change from a mutex to a read/write lock. This allows the tx port to be > > > selected simultaneously by multiple senders and transmit/receive is not > > > serialised between aggregated interfaces. > > > > Rad! :) > > And big thanks to you for MFCing it. Have you benchmarked this? Because rwlocks don't adaptively spin when a reader holds the lock, there are some cases where a mutex actually performs better than a rwlock. -- John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200709211522.16194.jhb>