From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Jiafei.Pan@freescale.com" Subject: RE: [PATCH] net: use hardware buffer pool to allocate skb Date: Thu, 16 Oct 2014 02:30:05 +0000 Message-ID: <43e8b86d47784ff78df70fc1b6e4b965@BLUPR03MB517.namprd03.prod.outlook.com> References: <1413343571-33231-1-git-send-email-Jiafei.Pan@freescale.com> <20141015113323.5321b2f7@uryu.home.lan> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 8BIT Cc: "davem@davemloft.net" , "jkosina@suse.cz" , "netdev@vger.kernel.org" , "LeoLi@freescale.com" , "linux-doc@vger.kernel.org" , "Jiafei.Pan@freescale.com" To: Stephen Hemminger Return-path: In-Reply-To: <20141015113323.5321b2f7@uryu.home.lan> Content-Language: en-US Sender: linux-doc-owner@vger.kernel.org List-Id: netdev.vger.kernel.org > -----Original Message----- > From: Stephen Hemminger [mailto:stephen@networkplumber.org] > Sent: Wednesday, October 15, 2014 5:33 PM > To: Pan Jiafei-B37022 > Cc: davem@davemloft.net; jkosina@suse.cz; netdev@vger.kernel.org; Li Yang-Leo- > R58472; linux-doc@vger.kernel.org > Subject: Re: [PATCH] net: use hardware buffer pool to allocate skb > > Since an skb can sit forever in an application queue, you have created > an easy way to livelock the system when enough skb's are waiting to be > read. I think there is no possible to livelock the system, because in my patch The function __netdev_alloc_skb will try to allocate hardware block buffer Firstly if dev->alloc_hw_skb is set, but it will continue allocate normal skb buffer if the hardware block buffer allocation fails.