netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* Re: [PATCH 5/6 v2] IB: IP address based RDMA connection manager
       [not found] <ORSMSX401FRaqbC8wSA0000000e@orsmsx401.amr.corp.intel.com>
@ 2006-03-13 15:43 ` Roland Dreier
  2006-03-13 17:11   ` Sean Hefty
  0 siblings, 1 reply; 3+ messages in thread
From: Roland Dreier @ 2006-03-13 15:43 UTC (permalink / raw)
  To: Sean Hefty; +Cc: netdev, linux-kernel, openib-general

It seems that cma_detach_from_dev():

 > +static void cma_detach_from_dev(struct rdma_id_private *id_priv)
 > +{
 > +	list_del(&id_priv->list);
 > +	if (atomic_dec_and_test(&id_priv->cma_dev->refcount))
 > +		wake_up(&id_priv->cma_dev->wait);
 > +	id_priv->cma_dev = NULL;
 > +}

doesn't need to do atomic_dec_and_test(), because it is never dropping
the last reference to id_priv (and in fact if it was, the last line
would be a use-after-free bug).

Does it make sense to replace it with:

	static void cma_detach_from_dev(struct rdma_id_private *id_priv)
	{
		list_del(&id_priv->list);
		/*
		 * cma_detach_from_dev() will never be dropping the last
		 * reference to id_priv, so no need to test here.
		 */
		atomic_dec(&id_priv->cma_dev->refcount);
		id_priv->cma_dev = NULL;
	}

on my x86_64 build that's worth

	add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-40 (-40)
	function                                     old     new   delta
	cma_detach_from_dev                          106      66     -40

 - R.

^ permalink raw reply	[flat|nested] 3+ messages in thread

* RE: [PATCH 5/6 v2] IB: IP address based RDMA connection manager
  2006-03-13 15:43 ` [PATCH 5/6 v2] IB: IP address based RDMA connection manager Roland Dreier
@ 2006-03-13 17:11   ` Sean Hefty
  0 siblings, 0 replies; 3+ messages in thread
From: Sean Hefty @ 2006-03-13 17:11 UTC (permalink / raw)
  To: 'Roland Dreier'; +Cc: netdev, linux-kernel, openib-general

> > +static void cma_detach_from_dev(struct rdma_id_private *id_priv)
> > +{
> > +	list_del(&id_priv->list);
> > +	if (atomic_dec_and_test(&id_priv->cma_dev->refcount))
> > +		wake_up(&id_priv->cma_dev->wait);
> > +	id_priv->cma_dev = NULL;
> > +}
>
>doesn't need to do atomic_dec_and_test(), because it is never dropping
>the last reference to id_priv (and in fact if it was, the last line
>would be a use-after-free bug).

It's dropping the reference on cma_dev, as opposed to id_priv.

- Sean

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH 5/6 v2] IB: IP address based RDMA connection manager
       [not found] <ORSMSX401FRaqbC8wSA0000001e@orsmsx401.amr.corp.intel.com>
@ 2006-03-13 17:26 ` Roland Dreier
  0 siblings, 0 replies; 3+ messages in thread
From: Roland Dreier @ 2006-03-13 17:26 UTC (permalink / raw)
  To: Sean Hefty; +Cc: netdev, linux-kernel, openib-general

    Sean> It's dropping the reference on cma_dev, as opposed to
    Sean> id_priv.

Duh, sorry.

 - R.

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2006-03-13 17:26 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
     [not found] <ORSMSX401FRaqbC8wSA0000000e@orsmsx401.amr.corp.intel.com>
2006-03-13 15:43 ` [PATCH 5/6 v2] IB: IP address based RDMA connection manager Roland Dreier
2006-03-13 17:11   ` Sean Hefty
     [not found] <ORSMSX401FRaqbC8wSA0000001e@orsmsx401.amr.corp.intel.com>
2006-03-13 17:26 ` Roland Dreier

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).