Date: Thu, 25 Mar 2004 21:30:29 -0500 From: "Brian F. Feldman" <green@FreeBSD.org> To: John Baldwin <jhb@FreeBSD.org> Cc: arch@FreeBSD.org Subject: Re: Is MTX_CONTESTED evil? Message-ID: <200403260230.i2Q2UTNZ097932@green.homeunix.org> In-Reply-To: Message from John Baldwin <jhb@FreeBSD.org> of "Tue, 23 Mar 2004 17:24:45 EST." <200403231724.45923.jhb@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
John Baldwin <jhb@FreeBSD.org> wrote: > On Tuesday 23 March 2004 03:06 pm, Dag-Erling Sm=C3=B8rgrav wrote: > > des@des.no (Dag-Erling Sm=C3=B8rgrav) writes: > > > John Baldwin <john@baldwin.cx> writes: > > > > Adaptive mutexes work just fine, but they aren't on by default. > > > > > > No, they don't "work just fine", unless of course they are *suppose= d* > > > to cause frequent panics. > > > > s/panic/freeze/ > = > They worked just fine on sparc64, alpha, and i386 when they were develo= ped and = > nothing has changed since then. However, since they increase the chanc= es of = > "near concurrency" on multiple CPUs (i.e. one CPU grabbing a lock right= after = > another released it) they expose races and thus bugs in code that uses = > mutexes improperly. The fault is not in adaptive mutexes, but in the o= ther = > broken code, just as compile failures aren't the result of the tinderbo= x = > itself being broken. :-) 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 ma= ke = ADAPTIVE_MUTEX default to catch more bugs. -- = Brian Fundakowski Feldman \'[ FreeBSD ]''''''''= ''\ <> green@FreeBSD.org \ The Power to Serv= e! \ Opinions expressed are my own. \,,,,,,,,,,,,,,,,,,= ,,,,\
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200403260230.i2Q2UTNZ097932>