Mailing-List: contact dev-help@apr.apache.org; run by ezmlm
Precedence: bulk
Received-SPF: pass (athena.apache.org: domain of trawick@gmail.com designates
 74.125.83.50 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type:content-transfer-encoding;
        b=Dz6cvK9r+MBJOvFfYNybJ111ruoxPI6Pt6acuEQlDwfs4QdUjlUrDmGHu/MkmZHaU8
         PcgLTMxVr3LJT++hEADX/SyNOMb/cD7oUvy7G+J+0YxaDV0mK9l6SOB+xJ/zB/U/NWA7
         tG+26htKTJcNBL1KJTvbBHjBXZ1B9V38bSa9A=
MIME-Version: 1.0
In-Reply-To: <b4e5ce321003291453lac32287gf570cdd486f9c78b@mail.gmail.com>
References: <b4e5ce321001061747v14dcd735ja917999ea317cc11@mail.gmail.com>
	 <b4e5ce321001062106g6e0d2da4g72fcd8a09aeec399@mail.gmail.com>
	 <b4e5ce321001161730g242e0076u6485e5ccb841ffdc@mail.gmail.com>
	 <b4e5ce321002042014u43d926f4m29b75a4d81c8ad59@mail.gmail.com>
	 <b4e5ce321003291019l1371bfd1o1f4c6696fa96b8e5@mail.gmail.com>
	 <38BDBF8B-717D-4FD4-814F-EB141452AD03@apache.org>
	 <b4e5ce321003291453lac32287gf570cdd486f9c78b@mail.gmail.com>
Date: Fri, 9 Apr 2010 17:01:29 -0400
Message-ID: <g2xcc67648e1004091401tcc9f6a2en8ccdd21b4f81b655@mail.gmail.com>
Subject: Re: [PATCH] bug in pollset_wakeup() + nocopy
From: Jeff Trawick <trawick@gmail.com>
To: APR Developer List <dev@apr.apache.org>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

On Mon, Mar 29, 2010 at 5:53 PM, Neil Conway <nrc@cs.berkeley.edu> wrote:
> On Mon, Mar 29, 2010 at 2:14 PM, Nick Kew <niq@apache.org> wrote:
>> I don't see the patch in this post. =A0Is it small/simple/clear enough t=
o
>> review in a brief-ish session?
>
> Yep, should be very straightforward. Attached are two versions of the
> patch (one for the 1.4.x branch, one for trunk). The reasoning for the
> fix is simple (APR_POLLSET_NOCOPY =3D=3D don't pass a stack-allocated
> pollfd_t to pollset_add()), and discussed in earlier emails:
>
> http://markmail.org/message/izj3zpc65sckzuao

finally committed to 1.4.x-trunk

Neil, you mentioned earlier in the thread:

>Note that if you want to backport this bug fix to the 1.4 branch, the
>previous version of the patch should be used. But perhaps the easiest
>route is to first backport the pollcb_wakeup() change, and then apply
>this version of the patch.

As I understand it, the pollcb changes you refer to are for a new API,
which can't be added to 1.4.x at this point; is that  "new API"
understanding correct?  Anyway, there's no reason it can't go to
1.5.x; care to post a patch to get 1.5.x caught up?