From mboxrd@z Thu Jan 1 00:00:00 1970 From: Kent Overstreet Subject: Re: [PATCH 22/25] Generic dynamic per cpu refcounting Date: Thu, 29 Nov 2012 11:29:25 -0800 Message-ID: <20121129192925.GH15094@google.com> References: <1354121029-1376-1-git-send-email-koverstreet@google.com> <1354121029-1376-23-git-send-email-koverstreet@google.com> <20121129185720.GE15094@google.com> <20121129185953.GW16230@one.firstfloor.org> <20121129191214.GG15094@google.com> <20121129192003.GX16230@one.firstfloor.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-kernel@vger.kernel.org, linux-aio@kvack.org, linux-fsdevel@vger.kernel.org, zab@redhat.com, bcrl@kvack.org, jmoyer@redhat.com, axboe@kernel.dk, viro@zeniv.linux.org.uk To: Andi Kleen Return-path: Content-Disposition: inline In-Reply-To: <20121129192003.GX16230@one.firstfloor.org> Sender: owner-linux-aio@kvack.org List-Id: linux-fsdevel.vger.kernel.org On Thu, Nov 29, 2012 at 08:20:03PM +0100, Andi Kleen wrote: > > The trick is that we don't watch for the refcount hitting 0 until we're > > shutting down - so this only works if you keep track of your initial > > refcount. As long as we're not shutting down, we know the refcount can't > > hit 0 because we haven't released the initial refcount. > > This seems dangerous to me: assume you have one CPU which always > does get and another does put. So there may be 2^32 such operations > without a kill and you wrap for real in a way that does not get > corrected. I don't know how to write a proof that it works (and I should... I haven't done any real math in ages, argh) but try working out some examples to see what happens: cpu 0 does 2^32 gets, cpu 1 does 2^32 - 1 puts, actual ref should be 1: cpu 0 ref: 0 cpu 1 ref: 1 (it started at 0, and subtracted 1 2^32 - 1 times) cpu 0 does 2^32 + 1 gets, cpu 1 does 2^32 puts, again ref should be 1: cpu 0 ref: 1 cpu 1 ref: 0 There's some kind of symmetry going on here, and if I'd been awake more in college I could probably say exactly why it works, but it does. -- To unsubscribe, send a message with 'unsubscribe linux-aio' in the body to majordomo@kvack.org. For more info on Linux AIO, see: http://www.kvack.org/aio/ Don't email: aart@kvack.org