From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:56400)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <armbru@redhat.com>) id 1aw819-0004Yo-Fu
	for qemu-devel@nongnu.org; Fri, 29 Apr 2016 08:59:41 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <armbru@redhat.com>) id 1aw80x-0000P7-Rp
	for qemu-devel@nongnu.org; Fri, 29 Apr 2016 08:59:30 -0400
Received: from mx1.redhat.com ([209.132.183.28]:38160)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <armbru@redhat.com>) id 1aw80x-0000NE-NS
	for qemu-devel@nongnu.org; Fri, 29 Apr 2016 08:59:23 -0400
From: Markus Armbruster <armbru@redhat.com>
References: <1461879932-9020-1-git-send-email-eblake@redhat.com>
	<1461879932-9020-3-git-send-email-eblake@redhat.com>
	<87eg9o7r95.fsf@dusky.pond.sub.org> <57234F4B.1010701@redhat.com>
	<572350F4.7090700@redhat.com>
Date: Fri, 29 Apr 2016 14:59:10 +0200
In-Reply-To: <572350F4.7090700@redhat.com> (Eric Blake's message of "Fri, 29
	Apr 2016 06:17:56 -0600")
Message-ID: <87vb30y3j5.fsf@dusky.pond.sub.org>
MIME-Version: 1.0
Content-Type: text/plain
Subject: Re: [Qemu-devel] [PATCH v16 02/24] qapi: Guarantee NULL obj on
 input visitor callback error
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel/>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Eric Blake <eblake@redhat.com>
Cc: qemu-devel@nongnu.org, Michael Roth <mdroth@linux.vnet.ibm.com>

Eric Blake <eblake@redhat.com> writes:

> On 04/29/2016 06:10 AM, Eric Blake wrote:
>> On 04/29/2016 02:28 AM, Markus Armbruster wrote:
>>> Eric Blake <eblake@redhat.com> writes:
>>>
>>>> Our existing input visitors were not very consistent on errors
>>>> in a function taking 'TYPE **obj' (that is, start_struct(),
>>>> start_alternate(), type_str(), and type_any(). next_list() is
>>>> similar, except that since commit 08f9541, it can't fail).
>>>
>>> Multiple sentences in a parenthesis, ugh :)
>>>
>>>     Our existing input visitors were not very consistent on errors in a
>>>     function taking 'TYPE **obj'.  These are start_struct(),
>>>     start_alternate(), type_str(), and type_any().  next_list() is
>>>     similar, but can't fail (see commit since 08f9541).
>>>
>>> Can touch up on commit.
>>>
>> 
>> Yes, sounds better.
>
> Spoke too soon: "see commit since 08f9541" is too wordy, should be
> either "see commit 08f9541" or "since commit 08f9541"

I'm terribly prone to leaving unwanted words behind when tinkering with
a sentence...  Thanks!

>>>> @@ -51,10 +57,16 @@ void visit_start_alternate(Visitor *v, const char *name,
>>>>                             GenericAlternate **obj, size_t size,
>>>>                             bool promote_int, Error **errp)
>>>>  {
>>>> +    Error *err = NULL;
>>>> +
>>>>      assert(obj && size >= sizeof(GenericAlternate));
>>>>      if (v->start_alternate) {
>>>> -        v->start_alternate(v, name, obj, size, promote_int, errp);
>>>> +        v->start_alternate(v, name, obj, size, promote_int, &err);
>>>>      }
>>>> +    if (v->type == VISITOR_INPUT) {
>>>> +        assert(!err != !*obj);
>>>
>>> Could assert(v->start_alternate && !err != !*obj), to preempt "what if
>>> !v->start_alternate" worries.  If you like that, I can do it on commit.
>> 
>> Can't hurt :)

Done.