Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261751AbVDZT3L (ORCPT ); Tue, 26 Apr 2005 15:29:11 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261752AbVDZT3L (ORCPT ); Tue, 26 Apr 2005 15:29:11 -0400 Received: from relay.2ka.mipt.ru ([194.85.82.65]:49833 "EHLO 2ka.mipt.ru") by vger.kernel.org with ESMTP id S261751AbVDZT3F (ORCPT ); Tue, 26 Apr 2005 15:29:05 -0400 Date: Tue, 26 Apr 2005 23:28:12 +0400 From: Evgeniy Polyakov To: dtor_core@ameritech.net Cc: dmitry.torokhov@gmail.com, netdev@oss.sgi.com, Greg KH , Jamal Hadi Salim , Kay Sievers , Herbert Xu , James Morris , Guillaume Thouvenin , linux-kernel@vger.kernel.org, Andrew Morton , Thomas Graf , Jay Lan Subject: Re: [1/1] connector/CBUS: new messaging subsystem. Revision number next. Message-ID: <20050426232812.0c7bb3a4@zanzibar.2ka.mipt.ru> In-Reply-To: References: <20050411125932.GA19538@uganda.factory.vocord.ru> <20050426202437.234e7d45@zanzibar.2ka.mipt.ru> <20050426203023.378e4831@zanzibar.2ka.mipt.ru> <20050426220713.7915e036@zanzibar.2ka.mipt.ru> <20050426223126.37b7aea1@zanzibar.2ka.mipt.ru> <20050426224833.3b6a0792@zanzibar.2ka.mipt.ru> Reply-To: johnpol@2ka.mipt.ru Organization: MIPT X-Mailer: Sylpheed-Claws 0.9.12b (GTK+ 1.2.10; i386-pc-linux-gnu) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-1.7.5 (2ka.mipt.ru [194.85.82.65]); Tue, 26 Apr 2005 23:28:26 +0400 (MSD) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2121 Lines: 51 On Tue, 26 Apr 2005 14:06:36 -0500 Dmitry Torokhov wrote: > On 4/26/05, Evgeniy Polyakov wrote: > > On Tue, 26 Apr 2005 13:42:10 -0500 > > Dmitry Torokhov wrote: > > > Yes, that woudl work, although I would urge you to implement a message > > > queue for callbacks (probably limit it to 1000 messages or so) to > > > allow bursting. > > > > It already exist, btw, but not exactly in that way - > > we have skb queue, which can not be filled from userspace > > if pressure is so strong so work queue can not be scheduled. > > It is of course different and is influenced by other things > > but it handles bursts quite well - it was tested on both > > SMP and UP machines with continuous flows of forks with > > shape addon of new running tasks [both fith fork bomb and not], > > so I think it can be called real bursty test. > > > > Ok, hear me out and tell me where I am wrong: > > By default a socket can receive at least 128 skbs with 258-byte > payload, correct? That means that user of cn_netlink_send, if started > "fresh", 128 average - size connector messages. If sender does not > want to wait for anything (unlike your fork tests that do schedule) > that means that 127 of those 128 messages will be dropped, although > netlink would deliver them in time just fine. > > What am I missing? Concider netlink_broadcast - it delivers skb to the kernel listener directly to the input callback - no queueing actually, the same is for netlink_unicast - so in-kernel users are always called synchronously. send() [sys_send()] system call(which btw as syscall does schedule too) ends up in netlink_sendmsg and thus either netlink_broadcast and netlink_unicast, which is called directly as we have seen. > -- > Dmitry Evgeniy Polyakov Only failure makes us experts. -- Theo de Raadt - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/