From: "Philippe Mathieu-Daudé" <philmd@linaro.org>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: qemu-devel@nongnu.org,
"Manos Pitsidianakis" <manos.pitsidianakis@linaro.org>,
"Junjie Mao" <junjie.mao@hotmail.com>,
"Daniel P. Berrangé" <berrange@redhat.com>,
"Richard Henderson" <richard.henderson@linaro.org>,
"Markus Armbruster" <armbru@redhat.com>,
"Eduardo Habkost" <eduardo@habkost.net>
Subject: Re: [PATCH v2 10/13] qdev: make properties array "const"
Date: Tue, 22 Oct 2024 18:43:44 -0300 [thread overview]
Message-ID: <ab4af84d-a518-406b-8dbb-513d41aa4d61@linaro.org> (raw)
In-Reply-To: <CABgObfawNPijVE7FqdGpnpAZ2JM9puY6=PSivFR4GgbHTKE2Fg@mail.gmail.com>
On 22/10/24 02:23, Paolo Bonzini wrote:
> On Tue, Oct 22, 2024 at 6:31 AM Philippe Mathieu-Daudé
> <philmd@linaro.org> wrote:
>>> -void qdev_property_add_static(DeviceState *dev, Property *prop)
>>> +void qdev_property_add_static(DeviceState *dev, const Property *prop)
>>> {
>>> Object *obj = OBJECT(dev);
>>> ObjectProperty *op;
>>> @@ -980,7 +980,7 @@ void qdev_property_add_static(DeviceState *dev, Property *prop)
>>> field_prop_getter(prop->info),
>>> field_prop_setter(prop->info),
>>> prop->info->release,
>>> - prop);
>>> + (Property *)prop);
>>
>> I like the overall patch idea, but I'm not keen on casting
>> const to non-const. Should we adapt the callee -- here
>> object_property_add() -- to also take a const argument?
>
> This argument goes into prop->opaque and is passed to all
> accessor/resolver/finalizers functions. So it would be a much larger
> change because it needs to change all those functions from "void
> *opaque" to "const void *opaque".
>
> It would also be an issue because some finalizers write to the opaque
> for good reason:
>
> static void object_finalize_child_property(
> Object *obj, const char *name, void *opaque)
> {
> Object *child = opaque;
>
> if (child->class->unparent) {
> (child->class->unparent)(child);
> }
> child->parent = NULL; // <--- here
> object_unref(child);
> }
>
> So, it's not great but it seems necessary. At least keeping the const
> within qdev properties makes things "safer" within that realm.
Since it is only within qdev-properties.c, it is indeed reasonable to
accept. Maybe make it explicit via a well-named macro to do the cast?
/* NON_CONST_PROP: Specific macro to this file because ... */
#define NON_CONST_PROP(prop) (Property *)(prop)
Regards,
Phil.
next prev parent reply other threads:[~2024-10-22 21:44 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-21 16:35 [PATCH v2 00/13] rust: miscellaneous cleanups + QOM integration tests Paolo Bonzini
2024-10-21 16:35 ` [PATCH v2 01/13] meson: import rust module into a global variable Paolo Bonzini
2024-10-23 10:29 ` Manos Pitsidianakis
2024-10-21 16:35 ` [PATCH v2 02/13] meson: remove repeated search for rust_root_crate.sh Paolo Bonzini
2024-10-21 16:35 ` [PATCH v2 03/13] meson: pass rustc_args when building all crates Paolo Bonzini
2024-10-22 2:35 ` Junjie Mao
2024-10-22 15:35 ` Zhao Liu
2024-10-21 16:35 ` [PATCH v2 04/13] rust: do not use --no-size_t-is-usize Paolo Bonzini
2024-10-22 2:38 ` Junjie Mao
2024-10-23 4:24 ` Zhao Liu
2024-10-21 16:35 ` [PATCH v2 05/13] rust: remove uses of #[no_mangle] Paolo Bonzini
2024-10-23 10:48 ` Paolo Bonzini
2024-10-23 14:06 ` Zhao Liu
2024-10-23 14:13 ` Zhao Liu
2024-10-21 16:35 ` [PATCH v2 06/13] rust: modernize link_section usage for ELF platforms Paolo Bonzini
2024-10-23 15:31 ` Zhao Liu
2024-10-24 6:04 ` Paolo Bonzini
2024-10-21 16:35 ` [PATCH v2 07/13] rust: build integration test for the qemu_api crate Paolo Bonzini
2024-10-22 1:52 ` Junjie Mao
2024-10-24 17:23 ` Zhao Liu
2024-10-21 16:35 ` [PATCH v2 08/13] rust: cleanup module_init!, use it from #[derive(Object)] Paolo Bonzini
2024-10-22 2:02 ` Junjie Mao
2024-10-22 5:16 ` Paolo Bonzini
2024-10-22 6:00 ` Junjie Mao
2024-10-22 7:20 ` Paolo Bonzini
2024-10-22 20:32 ` Kevin Wolf
2024-10-23 6:46 ` Junjie Mao
2024-10-25 8:59 ` Zhao Liu
2024-10-25 9:12 ` Paolo Bonzini
2024-10-21 16:35 ` [PATCH v2 09/13] rust: clean up define_property macro Paolo Bonzini
2024-10-22 19:46 ` Kevin Wolf
2024-10-23 7:10 ` Paolo Bonzini
2024-10-23 10:38 ` Manos Pitsidianakis
2024-10-23 11:23 ` Paolo Bonzini
2024-10-25 9:17 ` Zhao Liu
2024-10-21 16:35 ` [PATCH v2 10/13] qdev: make properties array "const" Paolo Bonzini
2024-10-22 4:31 ` Philippe Mathieu-Daudé
2024-10-22 5:23 ` Paolo Bonzini
2024-10-22 21:43 ` Philippe Mathieu-Daudé [this message]
2024-10-23 7:06 ` Paolo Bonzini
2024-10-21 16:35 ` [PATCH v2 11/13] rust: make properties array immutable Paolo Bonzini
2024-10-25 11:27 ` Zhao Liu
2024-10-21 16:35 ` [PATCH v2 12/13] rust: provide safe wrapper for MaybeUninit::zeroed() Paolo Bonzini
2024-10-25 10:10 ` Zhao Liu
2024-10-21 16:35 ` [PATCH v2 13/13] rust: do not use TYPE_CHARDEV unnecessarily Paolo Bonzini
2024-10-25 10:05 ` Zhao Liu
2024-10-22 20:46 ` [PATCH v2 00/13] rust: miscellaneous cleanups + QOM integration tests Kevin Wolf
2024-10-23 7:14 ` Paolo Bonzini
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=ab4af84d-a518-406b-8dbb-513d41aa4d61@linaro.org \
--to=philmd@linaro.org \
--cc=armbru@redhat.com \
--cc=berrange@redhat.com \
--cc=eduardo@habkost.net \
--cc=junjie.mao@hotmail.com \
--cc=manos.pitsidianakis@linaro.org \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=richard.henderson@linaro.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).