kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Alex Williamson <alex.williamson@redhat.com>
To: Jike Song <jike.song@intel.com>
Cc: pbonzini@redhat.com, guangrong.xiao@linux.intel.com,
	kwankhede@nvidia.com, cjia@nvidia.com, kevin.tian@intel.com,
	kvm@vger.kernel.org
Subject: Re: [v4 1/3] vfio: add vfio_group_notify support
Date: Tue, 15 Nov 2016 16:11:20 -0700	[thread overview]
Message-ID: <20161115161120.43d1aef8@t450s.home> (raw)
In-Reply-To: <1479209747-5564-2-git-send-email-jike.song@intel.com>

On Tue, 15 Nov 2016 19:35:45 +0800
Jike Song <jike.song@intel.com> wrote:

> A vfio_group may be or may not be attached to a KVM instance,
> if it is, the user of vfio_group might also want to know which
> KVM instance it is attached to, and when it will detach. In VFIO
> there are already external APIs for KVM to get/put vfio_group,
> by providing a similar vfio_group_notify, KVM can notify the
> vfio_group about attaching events.
> 
> 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  | 30 ++++++++++++++++++++++++++++++
>  include/linux/vfio.h |  4 ++++
>  2 files changed, 34 insertions(+)
> 
> diff --git a/drivers/vfio/vfio.c b/drivers/vfio/vfio.c
> index 0afb58e..b149ced 100644
> --- a/drivers/vfio/vfio.c
> +++ b/drivers/vfio/vfio.c
> @@ -86,6 +86,8 @@ struct vfio_group {
>  	struct mutex			unbound_lock;
>  	atomic_t			opened;
>  	bool				noiommu;
> +	struct kvm			*kvm;
> +	struct blocking_notifier_head	notifier;
>  };
>  
>  struct vfio_device {
> @@ -1015,6 +1017,34 @@ static long vfio_ioctl_check_extension(struct vfio_container *container,
>  	return ret;
>  }
>  
> +void vfio_group_notify(struct vfio_group *group, struct kvm *kvm)
> +{
> +	group->kvm = kvm;
> +	blocking_notifier_call_chain(&group->notifier,
> +				VFIO_GROUP_NOTIFY_SET_KVM, kvm);
> +}
> +EXPORT_SYMBOL_GPL(vfio_group_notify);

This shouldn't be called vfio_group_notify() if it's specific to kvm.
vfio_group_set_kvm() perhaps.

> +
> +static void vfio_group_register_notifier(struct vfio_group *group,
> +					 struct notifier_block *nb)
> +{
> +	blocking_notifier_chain_register(&group->notifier, nb);
> +
> +	/*
> +	 * The attaching of kvm and vfio_group might already happen, so
> +	 * here we replay once on registration.
> +	 */
> +	if (group->kvm)
> +		blocking_notifier_call_chain(&group->notifier,
> +					VFIO_GROUP_NOTIFY_SET_KVM, group->kvm);
> +}
> +
> +static void vfio_group_unregister_notifier(struct vfio_group *group,
> +					 struct notifier_block *nb)
> +{
> +	blocking_notifier_chain_unregister(&group->notifier, nb);
> +}
> +
>  /* hold write lock on container->group_lock */
>  static int __vfio_container_attach_groups(struct vfio_container *container,
>  					  struct vfio_iommu_driver *driver,
> diff --git a/include/linux/vfio.h b/include/linux/vfio.h
> index d4ce14d..ec9f74f 100644
> --- a/include/linux/vfio.h
> +++ b/include/linux/vfio.h
> @@ -100,6 +100,9 @@ extern void vfio_unregister_iommu_driver(
>  extern long vfio_external_check_extension(struct vfio_group *group,
>  					  unsigned long arg);
>  
> +struct kvm;
> +extern void vfio_group_notify(struct vfio_group *group, struct kvm *kvm);
> +
>  /*
>   * Sub-module helpers
>   */
> @@ -149,6 +152,7 @@ extern int vfio_unpin_pages(struct device *dev, unsigned long *user_pfn,
>  			    int npage);
>  
>  #define VFIO_IOMMU_NOTIFY_DMA_UNMAP	1
> +#define VFIO_GROUP_NOTIFY_SET_KVM	2
>  
>  extern int vfio_register_notifier(struct device *dev,
>  				  struct notifier_block *nb);


  reply	other threads:[~2016-11-15 23:11 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-11-15 11:35 [v4 0/3] plumb kvm/vfio to notify kvm:group attaching/detaching Jike Song
2016-11-15 11:35 ` [v4 1/3] vfio: add vfio_group_notify support Jike Song
2016-11-15 23:11   ` Alex Williamson [this message]
2016-11-16  3:02     ` Jike Song
2016-11-15 11:35 ` [v4 2/3] vfio_register_notifier: also register on the group notifier Jike Song
2016-11-15 23:11   ` Alex Williamson
2016-11-16  3:01     ` Jike Song
2016-11-16  3:43       ` Alex Williamson
2016-11-16  9:14         ` Kirti Wankhede
2016-11-16  9:37           ` Jike Song
2016-11-16 10:44             ` Kirti Wankhede
2016-11-16 17:48               ` Alex Williamson
2016-11-16 19:12                 ` Kirti Wankhede
2016-11-16 19:45                   ` Alex Williamson
2016-11-17  5:24                     ` Jike Song
2016-11-17  6:03                       ` Alex Williamson
2016-11-17  6:27                         ` Jike Song
2016-11-17 12:31                         ` Jike Song
2016-11-17 16:22                           ` Alex Williamson
2016-11-18 10:39                             ` Jike Song
2016-11-15 11:35 ` [v4 3/3] kvm: notify vfio on attaching and detaching Jike Song

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=20161115161120.43d1aef8@t450s.home \
    --to=alex.williamson@redhat.com \
    --cc=cjia@nvidia.com \
    --cc=guangrong.xiao@linux.intel.com \
    --cc=jike.song@intel.com \
    --cc=kevin.tian@intel.com \
    --cc=kvm@vger.kernel.org \
    --cc=kwankhede@nvidia.com \
    --cc=pbonzini@redhat.com \
    /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).