Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934363AbWKUHjx (ORCPT ); Tue, 21 Nov 2006 02:39:53 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S934364AbWKUHjx (ORCPT ); Tue, 21 Nov 2006 02:39:53 -0500 Received: from brick.kernel.dk ([62.242.22.158]:36384 "EHLO kernel.dk") by vger.kernel.org with ESMTP id S934363AbWKUHjw (ORCPT ); Tue, 21 Nov 2006 02:39:52 -0500 Date: Tue, 21 Nov 2006 08:39:39 +0100 From: Jens Axboe To: Alan Stern Cc: "Paul E. McKenney" , Oleg Nesterov , Kernel development list Subject: Re: [patch] cpufreq: mark cpufreq_tsc() as core_initcall_sync Message-ID: <20061121073938.GI8055@kernel.dk> References: <20061120201334.GE8055@kernel.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1672 Lines: 41 On Mon, Nov 20 2006, Alan Stern wrote: > On Mon, 20 Nov 2006, Jens Axboe wrote: > > > > > Must we introduce memory allocations in srcu_read_lock()? It makes it > > > > much harder and nastier for me to use. I'd much prefer a failing > > > > init_srcu(), seems like a much better API. > > > > > > Paul agrees with you that allocation failures in init_srcu() should be > > > passed back to the caller, and I certainly don't mind doing so. > > > > > > However we can't remove the memory allocation in srcu_read_lock(). That > > > was the point which started this whole thread: the per-cpu allocation > > > cannot be done statically, and some users of a static SRCU structure can't > > > easily call init_srcu() early enough. > > > > > > Once the allocation succeeds, the overhead in srcu_read_lock() is minimal. > > > > It's not about the overhead, it's about a potentially problematic > > allocation. > > I'm not sure what you mean by "problematic allocation". If you > successfully call init_srcu_struct then the allocation will be taken care > of. Later calls to srcu_read_lock won't experience any slowdowns or > problems. That requires init_srcu_struct() to return the error. If it does that, I'm fine with it. > Does this answer your objection? If not, can you explain in more detail > what other features you would like? It does, if the allocation failure in init_srcu_struct() is signalled. -- Jens Axboe - 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/