From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [PATCH 01/12] netvm: Prevent a stream-specific deadlock Date: Fri, 11 May 2012 01:10:34 -0400 (EDT) Message-ID: <20120511.011034.557833140906762226.davem@davemloft.net> References: <1336658065-24851-1-git-send-email-mgorman@suse.de> <1336658065-24851-2-git-send-email-mgorman@suse.de> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: akpm@linux-foundation.org, linux-mm@kvack.org, netdev@vger.kernel.org, linux-nfs@vger.kernel.org, linux-kernel@vger.kernel.org, Trond.Myklebust@netapp.com, neilb@suse.de, hch@infradead.org, a.p.zijlstra@chello.nl, michaelc@cs.wisc.edu, emunson@mgebm.net To: mgorman@suse.de Return-path: In-Reply-To: <1336658065-24851-2-git-send-email-mgorman@suse.de> Sender: owner-linux-mm@kvack.org List-Id: netdev.vger.kernel.org From: Mel Gorman Date: Thu, 10 May 2012 14:54:14 +0100 > It could happen that all !SOCK_MEMALLOC sockets have buffered so > much data that we're over the global rmem limit. This will prevent > SOCK_MEMALLOC buffers from receiving data, which will prevent userspace > from running, which is needed to reduce the buffered data. > > Fix this by exempting the SOCK_MEMALLOC sockets from the rmem limit. > > Signed-off-by: Peter Zijlstra > Signed-off-by: Mel Gorman This introduces an invariant which I am not so sure is enforced. With this change it is absolutely required that once a socket becomes SOCK_MEMALLOC it must never _ever_ lose that attribute. Otherwise we can end up liberating global rmem tokens which we never actually took. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: email@kvack.org