For example this post: http://lists.freebsd.org/pipermail/freebsd-current/2004-May/026871.html
VH On Mon, 30 May 2005, Christopher Faylor wrote: > On Mon, May 30, 2005 at 08:00:52PM +0200, Vaclav Haisman wrote: > >I am not sure, that is why I wrote "probably". But from what I see there is > >already exclusive access guaranteed by the mx.lock() call, unless of course I > >am completely misunderstanding something. > > I don't think you are. I missed this too when I saw your patch and had > the same reaction that Corinna did. > > I think this patch should be ok to apply. > > >I can tell you that "lock; cmpxchg" pair of instruction is really not as > >cheap > >as it looks. Especially on SMP systems. It takes above 100 of cycles on > >contemporary CPUs. > > Do you have a reference which states this? 100s of cycles sounds like an > incredible amount of overhead. > > cgf >