MIME-Version: 1.0
Date:   Thu, 18 Nov 2021 10:34:44 +0000
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
From:   yajun.deng@linux.dev
Message-ID: <aecc489431c38f170097ae1eb37a7d22@linux.dev>
Subject: Re: [PATCH] refcount: introduce refcount_is_one() helper function
To:     "Peter Zijlstra" <peterz@infradead.org>
Cc:     will@kernel.org, boqun.feng@gmail.com, linux-kernel@vger.kernel.org
In-Reply-To: <YZYSdJh7xpkHJjoa@hirez.programming.kicks-ass.net>
References: <YZYSdJh7xpkHJjoa@hirez.programming.kicks-ass.net>
 <20211118074200.GF174703@worktop.programming.kicks-ass.net>
 <20211118035328.5730-1-yajun.deng@linux.dev>
 <a4b7401c06d3baa5d3d0a43d7aa6c62c@linux.dev>
Precedence: bulk

November 18, 2021 4:44 PM, "Peter Zijlstra" <peterz@infradead.org> wrote:=
=0A=0A> On Thu, Nov 18, 2021 at 08:12:56AM +0000, yajun.deng@linux.dev wr=
ote:=0A> =0A>> November 18, 2021 3:42 PM, "Peter Zijlstra" <peterz@infrad=
ead.org> wrote:=0A>> =0A>> On Thu, Nov 18, 2021 at 11:53:28AM +0800, Yaju=
n Deng wrote:=0A>> =0A>> There are many cases where it is necessary to de=
termine if refcount is one,=0A>> introduce refcount_is_one() helper funct=
ion for these cases.=0A>> =0A>> Give me one that is not racy?=0A>> =0A>> =
The following determine refcount is one, 35 count.=0A> =0A> Very good, no=
w get me one that isn't broken :-)=0A=0ASorry, I didn't understand what i=
s the 'isn't broken'=E3=80=82