From mboxrd@z Thu Jan  1 00:00:00 1970
From: Alex Williamson <alex.williamson@redhat.com>
Subject: Re: [v4 2/3] vfio_register_notifier: also register on the group
 notifier
Date: Tue, 15 Nov 2016 16:11:25 -0700
Message-ID: <20161115161125.7d75fb49@t450s.home>
References: <1479209747-5564-1-git-send-email-jike.song@intel.com>
        <1479209747-5564-3-git-send-email-jike.song@intel.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Cc: pbonzini@redhat.com, guangrong.xiao@linux.intel.com,
        kwankhede@nvidia.com, cjia@nvidia.com, kevin.tian@intel.com,
        kvm@vger.kernel.org
To: Jike Song <jike.song@intel.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:38638 "EHLO mx1.redhat.com"
        rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
        id S932927AbcKOXL0 (ORCPT <rfc822;kvm@vger.kernel.org>);
        Tue, 15 Nov 2016 18:11:26 -0500
In-Reply-To: <1479209747-5564-3-git-send-email-jike.song@intel.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On Tue, 15 Nov 2016 19:35:46 +0800
Jike Song <jike.song@intel.com> wrote:

> The user of vfio_register_notifier might care about not only
> iommu events but also vfio_group events, so also register the
> notifier_block on vfio_group.
> 
> Cc: Xiao Guangrong <guangrong.xiao@linux.intel.com>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Alex Williamson <alex.williamson@redhat.com>
> Signed-off-by: Jike Song <jike.song@intel.com>
> ---
>  drivers/vfio/vfio.c | 4 ++++
>  1 file changed, 4 insertions(+)
> 
> diff --git a/drivers/vfio/vfio.c b/drivers/vfio/vfio.c
> index b149ced..2c0eedb 100644
> --- a/drivers/vfio/vfio.c
> +++ b/drivers/vfio/vfio.c
> @@ -2065,6 +2065,8 @@ int vfio_register_notifier(struct device *dev, struct notifier_block *nb)
>  	else
>  		ret = -ENOTTY;
>  
> +	vfio_group_register_notifier(group, nb);
> +
>  	up_read(&container->group_lock);
>  	vfio_group_try_dissolve_container(group);
>  
> @@ -2102,6 +2104,8 @@ int vfio_unregister_notifier(struct device *dev, struct notifier_block *nb)
>  	else
>  		ret = -ENOTTY;
>  
> +	vfio_group_unregister_notifier(group, nb);
> +
>  	up_read(&container->group_lock);
>  	vfio_group_try_dissolve_container(group);
>  

You haven't addressed the error paths, if the iommu driver returns
error and therefore the {un}register returns error, what is the caller
to expect about the group registration?