Date: Fri, 26 Mar 2004 14:04:57 -0500 From: John Baldwin <jhb@FreeBSD.org> To: "M. Warner Losh" <imp@bsdimp.com> Cc: tanimura@tanimura.dyndns.org Subject: Re: Is MTX_CONTESTED evil? Message-ID: <200403261404.57741.jhb@FreeBSD.org> In-Reply-To: <20040326.113751.14204546.imp@bsdimp.com> References: <200403260230.i2Q2UTNZ097932@green.homeunix.org> <xzpekrgf103.fsf@dwp.des.no> <20040326.113751.14204546.imp@bsdimp.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Friday 26 March 2004 01:37 pm, M. Warner Losh wrote: > In message: <xzpekrgf103.fsf@dwp.des.no> > > des@des.no (Dag-Erling Sm=F8rgrav) writes: > : "Brian F. Feldman" <green@FreeBSD.org> writes: > : > Well, that certainly explains the blitz of crashes I had to fix > : > recently! Since INVARIANTS and WITNESS are on by default, it would > : > make sense to make ADAPTIVE_MUTEX default to catch more bugs. > : > : Making ADAPTIVE_MUTEXES default will not catch more bugs, it will just > : piss off users. > > I think that if we don't fix the class of bugs that AM points out, > we'll ruin our reputation as a quality system in the long run. Maybe > they are so pervasive that we don't want to turn it on by default just > yet, but there will come a time we turn it on, just like we turned on > WITNESS and INVARIANTS to get better coverage on our testing. Right, let's not just gratuitously break everyone right yet. It's actually= an=20 option largely b/c I want us to be able to benchmark it once our locking is= =20 farther along to see if it helps more than it hurts. =2D-=20 John Baldwin <jhb@FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve" =3D http://www.FreeBSD.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200403261404.57741.jhb>