From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from moutng.kundenserver.de ([212.227.126.183]:59671 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752580AbXJVTLg convert rfc822-to-8bit (ORCPT ); Mon, 22 Oct 2007 15:11:36 -0400 From: Arnd Bergmann Subject: Re: [PATCH 1/2] irq_flags_t: intro and core annotations Date: Mon, 22 Oct 2007 21:10:34 +0200 References: <20071020235546.GB1825@martell.zuzino.mipt.ru> <20071022152912.GA841@linux-mips.org> <20071022112110.105b8e11.akpm@linux-foundation.org> In-Reply-To: <20071022112110.105b8e11.akpm@linux-foundation.org> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8BIT Content-Disposition: inline Message-Id: <200710222110.34768.arnd@arndb.de> Sender: linux-arch-owner@vger.kernel.org To: Andrew Morton Cc: Ralf Baechle , Alexey Dobriyan , Al Viro , torvalds@linux-foundation.org, viro@zeniv.linux.org.uk, linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org List-ID: On Monday 22 October 2007, Andrew Morton wrote: > Yes, it's always been ugly that we use unsigned long for this rather than > abstracting it properly. > > However I'd prefer that we have some really good reason for introducing > irq_flags_t now.  Simply so that I don't needlessly spend the next two > years wrestling with literally thousands of convert-to-irq_flags_t patches > and having to type "please use irq_flags_t here" in hundreds of patch > reviews. (snivel, wimper) On a related note, should we encourage the use of spin_lock() and spin_lock_irq() instead of spin_lock_irqsave() where possible? On some architectures, accessing the interrupt flag is a heavyweight operation, especially when running under a hypervisor, so a number of drivers could benefit from being converted to not save the flags at all instead of just changing the type of the flags variable. Arnd <><