Date: Sun, 10 Nov 2002 08:14:55 -0800 (PST) From: Matthew Jacob <mjacob@feral.com> To: Doug Rabson <dfr@nlsystems.com> Cc: "Alan L. Cox" <alc@imimic.com>, Jeff Roberson <jroberson@chesapeake.net>, Andrew Gallatin <gallatin@cs.duke.edu>, alpha@FreeBSD.ORG, alc@FreeBSD.ORG, John Baldwin <jhb@FreeBSD.ORG> Subject: Re: alpha: top of tree kernel blooie Message-ID: <Pine.BSF.4.21.0211100813580.59293-100000@beppo> In-Reply-To: <200211101110.34153.dfr@nlsystems.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> > 2. Don't we compile by default for the older Alphas that lack byte > > manipulation instructions? Thus, a byte store is implemented by a > > read-modify-write sequence of instructions. Thus, two simultaneous > > uma_dbg_alloc()s on adjacent locations in us_freelist could cause > > corruption. > > It was exactly this kind of breakage that prompted me to write the > atomic_* functions in the first place. Note that it is often possible > to get corruption even on a UP machine if an interrupt happens mid > sequence. I survived a -j4 buildworld on the 4-way SMP with atomic_set_8 in the uma_dbg_alloc/uma_dbg_free. I think we'll call this one 'good' To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0211100813580.59293-100000>