From mboxrd@z Thu Jan 1 00:00:00 1970 Received: by 2002:a5d:4308:0:0:0:0:0 with SMTP id h8-v6csp1123537wrq; Fri, 13 Jul 2018 14:16:50 -0700 (PDT) X-Google-Smtp-Source: AAOMgpcs7iNBhJDPvMHoeWq1s4zczaesK96RnG1jj2+V9P9L9Pu0yjwgB/csLxrMtqtWsK+Oa3U1 X-Received: by 2002:aed:2686:: with SMTP id q6-v6mr7399215qtd.199.1531516610531; Fri, 13 Jul 2018 14:16:50 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1531516610; cv=none; d=google.com; s=arc-20160816; b=LTtC7gH8tNOEMzgbodbP7bIvG18XzCSXRS97O4gblNIBt+H3vCHaV1cfnGbd2SkRRR nzVGTo6Z43ezdeM8MZxsa29GeZDY+rtF96xSKm3tQttW1J9sQiMwZCcVVXw9eiFZC8EN 94sXTUeFd5DRRk3Ew86pwxaHoctCnnzI4iXNcekq1yL92+WgOaB94RG69fi56zvG+BNs SXpxEXURTACGf1TjLcHwiTAqTAVGItBL9A9xRoulC0L/tff6Lsk/pU/SM2iKXOnNBQwj NIjJe601ocuQgqn+Y9rAU+HAYQ0si8Iy9BSCLgH994Eb2+kA3Qs9jDtrpGQOGg1DNKOG XhGA== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:cc:list-subscribe:list-help:list-post:list-archive :list-unsubscribe:list-id:precedence:subject:user-agent:in-reply-to :content-disposition:mime-version:references:message-id:to:from:date :arc-authentication-results; bh=jm1xvYuS04qE07FS9NfyTZn1/7Sj1/c0Vuh4rCe+Iuc=; b=n+fu2cWCGpchE9at0ADowfuTQFbk6BeV8TquFOJAHutsEMn77oZCqPyez1G3X47IKd 2PkKxhh+dGW1JVlJD/LceXZS08hZ2J5ELdVUCU2MHUHsmClMpCCbVAqXIsh5DfXnr49n MjxCTdCpCms8mURIaa8TqBx141LL7y8EhZz9meFmsscNU/idQX3OB1zLqEWYrjvzw+dj jM4R8mEVLL4yuNSmQ7SWWsiUA5WlbVAVGxtPYVeHPKZKVhqEi2YA7OeIFzTfTDoCgcxr nwYKzfqOQ043xTANsMN+qTgUsfMCr6OI9Js/vNmXB777KwjYRE5z0RYdGYjbDIHM02Ll 2gyQ== ARC-Authentication-Results: i=1; mx.google.com; spf=pass (google.com: domain of qemu-arm-bounces+alex.bennee=linaro.org@nongnu.org designates 2001:4830:134:3::11 as permitted sender) smtp.mailfrom="qemu-arm-bounces+alex.bennee=linaro.org@nongnu.org"; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=redhat.com Return-Path: Received: from lists.gnu.org (lists.gnu.org. [2001:4830:134:3::11]) by mx.google.com with ESMTPS id d195-v6si12636882qke.83.2018.07.13.14.16.50 for (version=TLS1 cipher=AES128-SHA bits=128/128); Fri, 13 Jul 2018 14:16:50 -0700 (PDT) Received-SPF: pass (google.com: domain of qemu-arm-bounces+alex.bennee=linaro.org@nongnu.org designates 2001:4830:134:3::11 as permitted sender) client-ip=2001:4830:134:3::11; Authentication-Results: mx.google.com; spf=pass (google.com: domain of qemu-arm-bounces+alex.bennee=linaro.org@nongnu.org designates 2001:4830:134:3::11 as permitted sender) smtp.mailfrom="qemu-arm-bounces+alex.bennee=linaro.org@nongnu.org"; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=redhat.com Received: from localhost ([::1]:39199 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fe5Qo-00053A-0a for alex.bennee@linaro.org; Fri, 13 Jul 2018 17:16:50 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:47304) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fe5Qh-000533-UB for qemu-arm@nongnu.org; Fri, 13 Jul 2018 17:16:44 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1fe5Qe-0003mT-PT for qemu-arm@nongnu.org; Fri, 13 Jul 2018 17:16:43 -0400 Received: from mx1.redhat.com ([209.132.183.28]:42870) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1fe5Qe-0003m0-G9; Fri, 13 Jul 2018 17:16:40 -0400 Received: from smtp.corp.redhat.com (int-mx08.intmail.prod.int.phx2.redhat.com [10.5.11.23]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id F30D432B674; Fri, 13 Jul 2018 21:16:38 +0000 (UTC) Received: from localhost (ovpn-116-12.gru2.redhat.com [10.97.116.12]) by smtp.corp.redhat.com (Postfix) with ESMTP id 37EE119489; Fri, 13 Jul 2018 21:16:38 +0000 (UTC) Date: Fri, 13 Jul 2018 18:16:36 -0300 From: Eduardo Habkost To: Thomas Huth Message-ID: <20180713211636.GS31657@localhost.localdomain> References: <1531470464-21522-1-git-send-email-thuth@redhat.com> <1531470464-21522-2-git-send-email-thuth@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1531470464-21522-2-git-send-email-thuth@redhat.com> X-Fnord: you can see the fnord User-Agent: Mutt/1.9.2 (2017-12-15) X-Scanned-By: MIMEDefang 2.84 on 10.5.11.23 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.38]); Fri, 13 Jul 2018 21:16:39 +0000 (UTC) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] [fuzzy] X-Received-From: 209.132.183.28 Subject: Re: [Qemu-arm] [PATCH v2 01/16] qom/object: Add a new function object_initialize_child() X-BeenThere: qemu-arm@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Peter Maydell , Alistair Francis , qemu-devel@nongnu.org, Markus Armbruster , Beniamino Galvani , qemu-arm@nongnu.org, Paolo Bonzini , Subbaraya Sundeep , Andreas =?iso-8859-1?Q?F=E4rber?= Errors-To: qemu-arm-bounces+alex.bennee=linaro.org@nongnu.org Sender: "Qemu-arm" X-TUID: +W0optLgEXT8 On Fri, Jul 13, 2018 at 10:27:29AM +0200, Thomas Huth wrote: > A lot of code is using the object_initialize() function followed by a call > to object_property_add_child() to add the newly initialized object as a child > of the current object. Both functions increase the reference counter of the > new object, but many spots that call these two functions then forget to drop > one of the superfluous references. So the newly created object is often not > cleaned up correctly when the parent is destroyed. In the worst case, this > can cause crashes, e.g. because device objects are not correctly removed from > their parent_bus. > > Since this is a common pattern between many code spots, let's introdcue a > new function that takes care of calling all three required initialization > functions, first object_initialize(), then object_property_add_child() and > finally object_unref(). > > And while we're at object.h, also fix some copy-n-paste errors in the > comments there ("to store the area" --> "to store the error"). > > Signed-off-by: Thomas Huth > --- > include/qom/object.h | 23 +++++++++++++++++++++-- > qom/object.c | 15 +++++++++++++++ > 2 files changed, 36 insertions(+), 2 deletions(-) > > diff --git a/include/qom/object.h b/include/qom/object.h > index f3d2308..3362db0 100644 > --- a/include/qom/object.h > +++ b/include/qom/object.h > @@ -749,6 +749,25 @@ int object_set_propv(Object *obj, > void object_initialize(void *obj, size_t size, const char *typename); > > /** > + * object_initialize_child: > + * @parentobj: The parent object to add a property to > + * @propname: The name of the property > + * @childobj: A pointer to the memory to be used for the object. > + * @size: The maximum size available at @obj for the object. > + * @type: The name of the type of the object to instantiate. > + * @errp: If an error occurs, a pointer to an area to store the error > + * > + * This function will initialize an object. The memory for the object should > + * have already been allocated. The object will then be added as child property > + * to a parent with object_property_add_child() function. The returned object > + * has a reference count of 1 (for the "child<...>" property from the parent), > + * so the object will get finalized automatically when the parent gets removed. > + */ > +void object_initialize_child(Object *parentobj, const char *propname, > + void *childobj, size_t size, const char *type, > + Error **errp); I wonder if we should deprecate object_initialize() and support only object_initialize_child() later. Initializing an object contained inside another one without making it a child of the parent object is a recipe for trouble. Reviewed-by: Eduardo Habkost -- Eduardo From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:47315) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1fe5Qk-00053N-0W for qemu-devel@nongnu.org; Fri, 13 Jul 2018 17:16:46 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1fe5Qj-0003oW-2M for qemu-devel@nongnu.org; Fri, 13 Jul 2018 17:16:46 -0400 Date: Fri, 13 Jul 2018 18:16:36 -0300 From: Eduardo Habkost Message-ID: <20180713211636.GS31657@localhost.localdomain> References: <1531470464-21522-1-git-send-email-thuth@redhat.com> <1531470464-21522-2-git-send-email-thuth@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1531470464-21522-2-git-send-email-thuth@redhat.com> Subject: Re: [Qemu-devel] [PATCH v2 01/16] qom/object: Add a new function object_initialize_child() List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Thomas Huth Cc: qemu-devel@nongnu.org, Peter Maydell , Paolo Bonzini , qemu-arm@nongnu.org, Markus Armbruster , Beniamino Galvani , Subbaraya Sundeep , Alistair Francis , "Edgar E. Iglesias" , Andreas =?iso-8859-1?Q?F=E4rber?= On Fri, Jul 13, 2018 at 10:27:29AM +0200, Thomas Huth wrote: > A lot of code is using the object_initialize() function followed by a call > to object_property_add_child() to add the newly initialized object as a child > of the current object. Both functions increase the reference counter of the > new object, but many spots that call these two functions then forget to drop > one of the superfluous references. So the newly created object is often not > cleaned up correctly when the parent is destroyed. In the worst case, this > can cause crashes, e.g. because device objects are not correctly removed from > their parent_bus. > > Since this is a common pattern between many code spots, let's introdcue a > new function that takes care of calling all three required initialization > functions, first object_initialize(), then object_property_add_child() and > finally object_unref(). > > And while we're at object.h, also fix some copy-n-paste errors in the > comments there ("to store the area" --> "to store the error"). > > Signed-off-by: Thomas Huth > --- > include/qom/object.h | 23 +++++++++++++++++++++-- > qom/object.c | 15 +++++++++++++++ > 2 files changed, 36 insertions(+), 2 deletions(-) > > diff --git a/include/qom/object.h b/include/qom/object.h > index f3d2308..3362db0 100644 > --- a/include/qom/object.h > +++ b/include/qom/object.h > @@ -749,6 +749,25 @@ int object_set_propv(Object *obj, > void object_initialize(void *obj, size_t size, const char *typename); > > /** > + * object_initialize_child: > + * @parentobj: The parent object to add a property to > + * @propname: The name of the property > + * @childobj: A pointer to the memory to be used for the object. > + * @size: The maximum size available at @obj for the object. > + * @type: The name of the type of the object to instantiate. > + * @errp: If an error occurs, a pointer to an area to store the error > + * > + * This function will initialize an object. The memory for the object should > + * have already been allocated. The object will then be added as child property > + * to a parent with object_property_add_child() function. The returned object > + * has a reference count of 1 (for the "child<...>" property from the parent), > + * so the object will get finalized automatically when the parent gets removed. > + */ > +void object_initialize_child(Object *parentobj, const char *propname, > + void *childobj, size_t size, const char *type, > + Error **errp); I wonder if we should deprecate object_initialize() and support only object_initialize_child() later. Initializing an object contained inside another one without making it a child of the parent object is a recipe for trouble. Reviewed-by: Eduardo Habkost -- Eduardo