Date: Mon, 11 Nov 2002 00:42:30 -0500 (EST) From: Jeff Roberson <jroberson@chesapeake.net> To: Matthew Jacob <mjacob@feral.com> Cc: Jake Burkholder <jake@locore.ca>, "Alan L. Cox" <alc@imimic.com>, <jeff@FreeBSD.org>, <cvs-committers@FreeBSD.org>, <cvs-all@FreeBSD.org> Subject: Re: cvs commit: src/sys/vm uma_dbg.c Message-ID: <20021111003812.C97372-100000@mail.chesapeake.net> In-Reply-To: <Pine.BSF.4.21.0211102058500.64258-100000@beppo>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 10 Nov 2002, Matthew Jacob wrote: > > > > I guess getting rid of the ZONE_LOCK in uma_dbg_getslab is also needed > :-) > > I think this is the right thing. The reason that I wasn't holding the lock before is because the code in uma_dbg_ doesn't actually move things on and off of the free list. It just marks them as in a bucket or not, so that frees to seperate per cpu caches can be caught. I was assuming that the spin lock would be enough to keep the caches coherent, but I forgot that we don't do atomic writes to 8bit ints on many platforms. Feel free to add the appropriate locks to uma_dbg. If you want to punt on it I can address this when I get back in town. Thanks, Jeff To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20021111003812.C97372-100000>