From: Eric Blake <eblake@redhat.com>
To: "Marc-André Lureau" <marcandre.lureau@gmail.com>,
"Paolo Bonzini" <pbonzini@redhat.com>
Cc: qemu-devel <qemu-devel@nongnu.org>,
"Armbruster, Markus" <armbru@redhat.com>
Subject: Re: [Qemu-devel] [PATCH 2/3] qobject: assume base of a qobject is at offset 0
Date: Wed, 21 Mar 2018 12:08:56 -0500 [thread overview]
Message-ID: <e9d6720a-19d9-b5d5-19dc-5c1f9a7a1665@redhat.com> (raw)
In-Reply-To: <CAJ+F1C+NCwxnGSSE=5JAzf-grC3L4ebtZuXCXfxm9vv84FSPGQ@mail.gmail.com>
On 03/21/2018 11:59 AM, Marc-André Lureau wrote:
>> Hmm, perhaps by making it "struct {} base"?
>>
>> Or even:
>>
>> struct QObjectCommon {
>> QType type;
>> size_t refcnt;
>> }
>>
>> struct QObject {
>> QObjectCommon base;
>> }
>>
>> struct QString {
>> QObjectCommon base;
>> ...
>> }
>
> I fail to see what that solves. You are moving the problem to
> QObjectCommon: we would have to replace all QObject * user to take
> QObjectCommon *, and then we wouldn't solve the problem for
> QObjectCommon.
Here, QObjectCommon is never used outside of qobject.h; all existing
code continues to uwe QObject *. But the conversion from QString* or
QObject * to QObjectCommon * is trivial (&obj->base), and the conversion
from QObjectCommon * to the public QObject * is trivial (container_of
style cast). So the QOBJECT() macro becomes comparable to a C++
reinterpret_cast<>() that converts any descendent of QObjectCommon into
QObject, even if the original type was not QObject; and we never offer a
public API to convert anything into or out of QObjectCommon (it only
exists as a dummy type to make our other casts easier).
--
Eric Blake, Principal Software Engineer
Red Hat, Inc. +1-919-301-3266
Virtualization: qemu.org | libvirt.org
next prev parent reply other threads:[~2018-03-21 17:09 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-21 13:40 [Qemu-devel] [PATCH 0/3] RFC: simplify qobject refcount Marc-André Lureau
2018-03-21 13:40 ` [Qemu-devel] [PATCH 1/3] compiler: extend QEMU_GENERIC to handle more arguments Marc-André Lureau
2018-03-21 13:40 ` [Qemu-devel] [PATCH 2/3] qobject: assume base of a qobject is at offset 0 Marc-André Lureau
2018-03-21 14:01 ` Paolo Bonzini
2018-03-21 14:08 ` Marc-André Lureau
2018-03-21 14:19 ` Eric Blake
2018-03-21 14:21 ` Marc-André Lureau
2018-03-21 14:49 ` Paolo Bonzini
2018-03-21 15:29 ` Eric Blake
2018-03-21 15:34 ` Paolo Bonzini
2018-03-21 15:55 ` Marc-André Lureau
2018-03-21 16:00 ` Paolo Bonzini
2018-03-21 16:11 ` Marc-André Lureau
2018-03-21 16:23 ` Paolo Bonzini
2018-03-21 16:59 ` Marc-André Lureau
2018-03-21 17:08 ` Eric Blake [this message]
2018-03-21 17:23 ` Marc-André Lureau
2018-03-21 17:38 ` Eric Blake
2018-03-21 17:43 ` Paolo Bonzini
2018-03-21 14:01 ` Eric Blake
2018-03-21 14:02 ` Paolo Bonzini
2018-03-21 13:40 ` [Qemu-devel] [PATCH 3/3] qobject: replace qobject_incref/QINCREF qobject_decref/QDECREF Marc-André Lureau
2018-03-21 14:28 ` Eric Blake
2018-03-21 14:51 ` Marc-André Lureau
2018-03-21 15:36 ` Eric Blake
2018-03-21 14:09 ` [Qemu-devel] [PATCH 0/3] RFC: simplify qobject refcount Eric Blake
2018-03-21 14:19 ` Marc-André Lureau
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=e9d6720a-19d9-b5d5-19dc-5c1f9a7a1665@redhat.com \
--to=eblake@redhat.com \
--cc=armbru@redhat.com \
--cc=marcandre.lureau@gmail.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).