From mboxrd@z Thu Jan 1 00:00:00 1970 From: Or Gerlitz Subject: Re: [PATCH 19/22] IB/ipoib: make sure we reap all our ah on shutdown Date: Thu, 12 Feb 2015 20:35:09 +0200 Message-ID: <54DCF25D.3050307@mellanox.com> References: <14ee2e7737123c7d37465ac52be5c026240c9985.1423703861.git.dledford@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <14ee2e7737123c7d37465ac52be5c026240c9985.1423703861.git.dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Doug Ledford , linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, roland-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org Cc: Or Gerlitz , Erez Shitrit List-Id: linux-rdma@vger.kernel.org On 02/12/2015 03:43 AM, Doug Ledford wrote: > The introduction of garbage collection for neighbors in the ipoib stack > caused a leak of resources. In particular, an ah can have a refcount > from a mcast struct, from a neigh struct, and from a path struct. When > we put an ah ref, the ah gets moved to a dead list to be reaped later if > the refcount goes to 0. see my reply on this patch in it's singleton version which you sent couple of weeks ago -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html